Xerox University Microfilms: 300 North Zeeb Road Ann Arbor, Michigan 48106

INFORMATION TO USERS
T h i s material was produced from a microfilm copy of the original document. While
t h e most advanced technological means to photograph and reproduce this document
h a v e been used, the quality is heavily dependent upon the quality of the original
s u b m itte d .
T h e following explanation of techniques is provided to help you understand

m a rk in g s or patterns which may appear on this reproduction.
1. The sign or "target" for pages apparently lacking from the document
photographed is "Missing Page(s)". If it was possible to obtain the missing
page(s) or section, they are spliced into the film along with adjacent pages.
This may have necessitated cutting thru an image and duplicating adjacent
pages to insure you complete continuity.
2. When an image on the film is obliterated with a large round black mark, it
is an indication that the photographer suspected that the copy may have
moved during exposure and thus cause a blurred image. You will find a
good image of the page in the adjacent frame.
3. When a map, drawing or chart, etc., was part of the material being
photographed the photographer followed a definite method in
"sectioning" the material. It is customary to begin photoing at the upper
left hand corner of a large sheet and to continue photoing from left to
right in equal sections with a small overlap. If necessary, sectioning is
continued again — beginning below the first row and continuing on until
complete.
4. The majority of users indicate that the textual content is of greatest value,
however, a somewhat higher quality reproduction could be made from
"photographs" if essential to the understanding of the dissertation. Silver
prints of "photographs" may be ordered at additional charge by writing
the Order Department, giving the catalog number, title, author and
specific pages you wish reproduced.
5. PLEASE NOTE: Some pages may have indistinct print. Filmed as

received.
Xerox University Microfilms

300 North Zeeb Road
Ann Arbor, Michigan 48106
LD3907 '13 11,0517
-
.07 Gobetz, Wallace. 1916-
1952 A quantification, standardization, and
.05 validation of the Bender-gestalt test on
an adult population in terras of its abili
ty to differentiate normal and psychoneu
rotic levels of adjustment.
îi,409p. d i a g r s t a b l e s .
Thesis (Ph.D.) - N.Y.U., Graduate
School. 1952.
Bibliography: p.335-340.
l.°Adjustment (Psychology) 2.Gestalt

psychology. 3.Dissertations, Academic -
N.Y.U. - 1952. ( I.Title:Bender-gesta'lt
test.
Xerox University Microfilms, Ann Arbor, Michigan 48106
TH IS DISSERTATION HAS BEEN M IC R O FILM ED E XA C TLY AS RECEIVED.

XiIBRART tf
ten IOKK UNIVERSITY
UHIVERSITT HE»HTF
A QUANTIFICATION, STANDARDIZATION, AND VALIDATION OF THE
BENDER-GESTALT TEST ON AN ADULT POPULATION IN TERMS OF ITS ABILITY TO
DIFFERENTIATE NORMAL AND PSYCHONEUROTIC LEVELS OF ADJUSTMENT
WALLACE GOBETZ
A dissertation in the department of Psychology submitted

in partial fulfillment of the requirements
for the degree of Doctor of Philosophy
at New York University
April, 1952
ACKNOWLEDGMENT
The writer wishes to express his gratitude to Professor
Thomas N. Jenkins, whose incisive suggestions and perspicacious
observations contributed innumerable insights to the present in
vestigation and whose kindly encouragement gave impetus to the ex
ploration of promising leads. The deepest appreciation is expressed
Dr. William D. Glenn, Jr., director of the New York University Test
ing and Advisement Center and personal friend and mentor of many years,
who very kindly consented to the special testing of the World War II
veterans and non-veteran clients. Heartfelt thanks are likewise prof
fered to Dr. Bernard Locke, chief clinical psychologist of the Mental
Hygiene Unit of the Brooklyn Hegional Office of the Veterans Adminis
tration, and to Dr. Richard H. Paynter, chief clinical psychologist of
the Mental Hygiene Unit of the New York Regional Office of the Veterans
Administration, whose generous cooperation made the testing of neurotic
out-patients possible. Finally, the writer wishes to acknowledge his
great debt to his wife, Ida, whose devotion, patience, and warm under
standing served as a never-ending source of inspiration and encourage
ment and to whom this study is affectionately dedicated.
/
TABLE OF CONTENTS
Chapter Page
I. HISTORICAL BACKGROUND.................................. 1
II. STATEMENT OF THE PROBLEM............................... 38
III. PROCEDURE........... 65
A. The Test Materials ...... 65
B. Administration of the Test....... 67
C . The Subj acts .......... 75
IV. THE SCORING SYSTEM.......................... 83
A. Introduction ..... 83
B. Scoring Manual......... 86
V. RESULTS............................................... 121
A. Global Signs on the InitialTest and the Retest....... 121
B. Individual Figure Signs on the Initial Test and

the Retest................ 148
C. The Recall Test...... 184
D. Scoring of the Test Records of the Matched Criterion

Groups with the Selected Signs................. 206
E. Scoring of the Test Records of the Unmatched Criterion

Groups.............. 226
F. Effect of Education, Age, and Intelligence upon Test

Scores................ 245
G. Scoring of the Initial Test Records of the Matched

Criterion Groups with Pascaland Suttell's Signs...... 247
H. Results of the Cross-Validation Study............ 252
VI. . DISCUSSION....... 286
VII. SUMMARY AND CONCLUSIONS................................. 326
BIBLIOGRAPHY................................................. 335
APPENDICES............................................... ..... 341
ii
LIST OF TABLES
Table Page
I. AGE COMPOSITION OF THE CRITERION GROUPS.................. 79
II. EDUCATIONAL STATUS OF THE CRITERION GROUPS............... 80
III. MARITAL STATUS OF THE CRITERION GROUPS................... 81
IV. MILITARY SERVICE AFFILIATION OF THE CRITERION GROUPS...... 81
V. CRITERIA FOR SCORING SHORT, LONG, FLAT, AND TALL FIGURES.... 120
VI. MEANS AND STANDARD DEVIATIONS OF THE LENGTHS OF EACH OF THE

FIGURE REPRODUCTIONS ON THE INITIAL TEST FOR THE CRITERION
GROUPS PLUS CRITICAL RATIOS OF THE DIFFERENCES BETWEEN THE
MEANS OF THE TOTAL CONTROL AND TOTAL NEUROTIC GROUPS....... 122
VII. MEANS AND STANDARD DEVIATIONS OF THE LENGTHS OF EACH OF THE

FIGURE REPRODUCTIONS ON THE RETEST FOR THE CRITERION GROUPS
PLUS CRITICAL RATIOS OF THE DIFFERENCES BETWEEN THE MEANS
OF THE TOTAL CONTROL AND TOTAL NEUROTIC GROUPS............ 123
VIII. COMPARISON OF THE ACTUAL LENGTHS OF THE STIMULUS FIGURES

WITH THE MEAN LENGTHS OF THE FIGURE REPRODUCTIONS ON THE
INITIAL TEST AND RETEST FOR THE TOTAL CONTROL AND TOTAL
NEUROTIC GROUPS................................... 125
IX. MEANS AND STANDARD DEVIATIONS OF THE HEIGHTS OF EACH OF

THE FIGURE REPRODUCTIONS ON THE INITIAL TEST FOR THE CRI
TERION GROUPS PLUS CRITICAL RATIOS OF THE DIFFERENCES BE
TWEEN THE MEANS OF THE TOTAL CONTROL AND TOTAL NEUROTIC
GROUPS........................................... 128
X. MEANS AND STANDARD DEVIATIONS OF THE HEIGHTS OF EACH OF

THE FIGURE REPRODUCTIONS ON THE RETEST FOR THE CRITERION
GROUPS PLUS CRITICAL RATIOS OF THE DIFFERENCES BETWEEN
THE MEANS OF THE TOTAL CONTROL AND TOTAL NEUROTIC GROUPS... 129
XI. COMPARISON OF THE ACTUAL HEIGHTS OF THE STIMULUS FIGURES

WITH THE MEAN HEIGHTS OF THE FIGURE REPRODUCTIONS ON THE
INITIAL TEST AND THE RETEST FOR THE TOTAL CONTROL AND
TOTAL NEUROTIC GROUPS........................... 131
XII. PERCENTAGE DISTRIBUTION OF SHORT FIGURES (REPRODUCTIONS AT

OR BELOW THE 10TH PERCENTILE OF THE TOTAL CONTROL.GROUP IN
LENGTH) FOR THE CRITERION GROUPS ON THE INITIAL TEST. 133
- iii -
Table Page
XIII. PERCENTAGE DISTRIBUTION OF SHORT FIGURES (REPRODUCTIONS AT

OR BELOW THE 10TH PERCENTILE OF THE TOTAL CONTROL GROUP IN
LENGTH) FOR THE CRITERION GROUPS ON THE RETEST............ 133
XIV. PERCENTAGE DISTRIBUTION OF SHORT FIGURES (REPRODUCTIONS AT

LENGTH) FOR THE CRITERION GROUPS ON THE INITIAL TEST...... 134
XV. PERCENTAGE DISTRIBUTIONS OF LONG FIGURES (REPRODUCTIONS AT

OR ABOVE THE 90TH PERCENTILE OF THE TOTAL CONTROL GROUP IN
LENGTH) FOR THE CRITERION GROUPS ON THE RETEST............ 134
XVI. PERCENTAGE DISTRIBUTION OF FLAT FIGURES (REPRODUCTIONS AT

HEIGHT) FOR THE CRITERION GROUPS ON THE INITIAL TEST...... 135
XVII. PERCENTAGE DISTRIBUTION OF FLAT FIGURES (REPRODUCTIONS AT

HEIGHT) FOR THE CRITERION GROUPS ON THE RETEST........... 135
XVIII. PERCENTAGE DISTRIBUTION OF TALL FIGURES (REPRODUCTIONS AT

HEIGHT) FOR THE CRITERION GROUPS ON THE INITIAL TEST...... 136
XVIX. PERCENTAGE DISTRIBUTION OF TALL FIGURES (REPRODUCTIONS AT

HEIGHT) FOR THE CRITERION GROUPS ON THE RETEST........... 136
XX. PERCENTAGE DISTRIBUTION OF THE NUMBER OF SEQUENTIAL CONNEC

TIONS FOR THE CRITERION GROUPS ON THE INITIAL TEST....... 138
XXI. PERCENTAGE DISTRIBUTION OF THE NUMBER OF SEQUENTIAL CONNEC

TIONS FOR THE CRITERION GROUPS ON THE RETEST............. 138
XXII. MEDIANS OF THE TIME REQUIRED TO REPRODUCE EACH FIGURE ON

THE INITIAL TEST FOR THE CRITERION GROUPS PLUS STANDARD
DEVIATIONS FOR THE TOTAL CONTROL AND TOTAL NEUROTIC GROUPS
AND CRITICAL RATIOS OF THE DIFFERENCES BETWEEN MEDIANS OF
THE TOTAL CONTROL AND TOTAL NEUROTIC GROUPS.............. 140
XXIII. MEDIANS OF THE TIME REQUIRED TO REPRODUCE EACH FIGURE ON

THE RETEST FOR THE CRITERION GROUPS PLUS STANDARD DEVIA
TIONS FOR THE TOTAL CONTROL AND TOTAL NEUROTIC GROUPS AND
CRITICAL RATIOS OF THE DIFFERENCES BETWEEN MEDIANS OF THE
TOTAL CONTROL AND TOTAL NEUROTIC GROUPS................. 141
- iv -
Table Page
XXIV. MEDIANS OF THE TOTAL TIME REQUIRED TO REPRODUCE ALL THE

FIGURES ON THE INITIAL TEST ALONE AND THE RETEST ALONE FOR
THE CRITERION GROUPS PLUS STANDARD DEVIATIONS FOR THE TOTAL
CONTROL AND TOTAL NEUROTIC GROUPS AND CRITICAL RATIOS OF
THE DIFFERENCES BETWEEN MEDIANS OF THE TOTAL CONTROL AND
TOTAL NEUROTIC GROUPS....................................... 143
XXV. PERCENTAGE DISTRIBUTION OF THE AMOUNT OF SPACE FILLED (COHESION)

FOR THE CRITERION GROUPS ON THE INITIALTEST............ 145
XXVI. PERCENTAGE DISTRIBUTION OF THE AMOUNT OF SPACE FILLED (COHESION)

FOR THE CRITERION GROUPS ON THE RETEST................. 145
XXVII. PERCENTAGE INCIDENCE OF THE GLOBAL SIGNS ON THE INITIAL TEST

FOR THE CRITERION GROUPS PLUS CRITICAL RATIOS OF THE DIFFER
ENCES BETWEEN THE TOTAL CONTROL AND TOTAL NEUROTIC GROUPS
WHICH ARE SIGNIFICANT AT THE 556 LEVEL ORBETTER............... 146
XXVIII. PERCENTAGE INCIDENCE OF THE GLOBAL SIGNS ON THE RETEST FOR THE
CRITERION GROUPS PLUS CRITICAL RATIOS OF THE DIFFERENCES BE
TWEEN THE TOTAL CONTROL AND TOTAL NEUROTIC GROUPS WHICH ARE SIG
NIFICANT AT THE 5% LEVEL OR BETTER......................... 147
XXIX. PERCENTAGE INCIDENCE OF THE INDIVIDUAL FIGURE SCORING SIGNS ON

THE INITIAL TEST FOR THE CRITERION GROUPS PLUS CRITICAL RATIOS
OF THE DIFFERENCES BETWEEN THE TOTAL CONTROL AND TOTAL NEUROT
IC GROUPS WHICH ARE SIGNIFICANT AT THE 5% LEVEL OR BETTER 150-8
XXIX-A. ANALYSIS OF THE SCORING CATEGORIES INTO INDIVIDUAL FIGURE

SCORING SIGNS OCCURRING 2056 OR MORE AND 556 OR LESS OF THE
CASES IN BOTH THE TOTAL CONTROL AND TOTAL NEUROTIC GROUPS ON
THE INITIAL TEST........................... 159-61
XXX. PERCENTAGE INCIDENCE OF THE INDIVIDUAL FIGURE SCORING SIGNS ON

THE RETEST FOR THE CRITERION GROUPS PLUS CRITICAL RATIOS OF THE
DIFFERENCES BETWEEN THE TOTAL CONTROL AND TOTAL NEUROTIC GROUPS
WHICH ARE SIGNIFICANT AT THE 556 LEVEL OR BETTER............. 162-70
XXX-A. ANALYSIS OF THE SCORING CATEGORIES INTO INDIVIDUAL FIGURE SCOR

ING SIGNS OCCURRING IN 2056 OR MORE AND 556 OR LESS OF THE CASES
IN BOTH THE TOTAL CONTROL AND TOTAL NEUROTIC GROUPS ON THE RE
TEST..................................................... 171-3
XXXI. BREAKDOWN OF THE TOTAL NUMBER OF SIGNS TABULATED FOR THE INI
TIAL TEST AND THE RETEST ON EACH OF THE BENDER-GESTALT FIGURES. 175
- v -
Table Page
XXXII. BREAKDOWN OF THE TOTAL NUMBER OF SIGNS SIGNIFICANT AT THE

% LEVEL ON THE INITIAL TEST AND THE RETEST FOR EACH OF THE
BENDER-GESTALT FIGURES................................. 175
XXZIII. PERCENTAGE INCIDENCE AND CRITICAL RATIOS OF THE DIFFER

ENCES BETWEEN PERCENTAGES FOR THE TOTAL CONTROL. AND TOTAL
NEUROTIC GROUPS OF TENTATIVELY SELECTED INDIVIDUAL FIGURE
SCORING SIGNS AND GLOBAL SIGNS WHICH MEET THE ITEM VALIDITY
CRITERIA OF (l) DISCRIMINATION AT THE % LEVEL ON EITHER
THE INITIAL TEST OR THE RETEST AND THE 10* LEVEL ON THE
OTHER OR (2) DISCRIMINATION AT THE 1% LEVEL ON EITHER THE
INITIAL TEST OR THE RETEST AND CONSISTENCY IN THE DIREC
TION OF THE DIFFERENCE ON THE OTHER..................... 177-8
XXXIV. AGE, GRADE, AND MARITAL STATUS DISTRIBUTIONS FOR THE

MATCHED CRITERION GROUPS (BELOW 70 AND NEUROTIC), INCLUD
ING MEANS AND STANDARD DEVIATIONS FOR AGE AND GRADE STATUS. 179
XXXV. PERCENTAGE INCIDENCE AND CRITICAL RATIOS OF THE DIFFERENCES

BETWEEN PERCENTAGES FOR THE MATCHED CRITERION GROUPS (BELOW
70 AND NEUROTIC) OF THE TENTATIVELY SELECTED INDIVIDUAL
FIGURE SCORING SIGNS AND GLOBAL SIGNS................... 182-3
XXXVI. FREQUENCY OF RECALL AND ORDER OF RECALL OF EACH OF THE TEST

FIGURES ON THEIMMEDIATE RECALL TESTFOR THE BELOW 70 GROUP 186
XXXVII. FREQUENCY OF RECALL AND ORDER OF RECALL OF EACH OF THE TEST

FIGURES ON THE IMMEDIATE RECALL TEST FOR THE 1 ABOVE 70
GROUP................................................ 186
XXXVIII. FREQUENCY OF RECALL AND ORDER OF RECALL OF EACH OF THE TEST

FIGURES ON THE IMMEDIATE RECALL TEST FOR THE 2 ABOVE 70
GROUP................................................ 187
XXXIX. FREQUENCY OF RECALL AND ORDER OF RECALL OF EACH OF THE TEST

FIGURES ON THE IMMEDIATE RECALL TEST FOR THE TOTAL CONTROL
GROUP................................................ 187
XL. FREQUENCY OF RECALL AND ORDER OF RECALL OF EACH OF THE TEST

FIGURES ON THE IMMEDIATE RECALL TEST FOR THE TOTAL NEUROTIC
GROUP................................................ 188
XLI. FREQUENCY OF RECALL AND ORDER OF RECALL, EXPRESSED AS PER

CENTAGES, OF EACH OF THE TEST FIGURES ON THE IMMEDIATE RE
CALL TEST FORTHE TOTAL CONTROL GROUP.................... 188
Table Page
XLII. FREQUENCY OF RECALL AND ORDER OF RECALL, EXPRESSED AS PER

CENTAGES, OF EACH OF THE TEST FIGURES ON THE IMMEDIATE RE
CALL TEST FOR THE TOTAL NEUROTIC GROUP..................... 189
XLIII. FREQUENCY OF TOTAL RECALL, EXPRESSED AS RANKS, OF EACH OF

THE TEST FIGURES ON THE IMMEDIATE RECALL TEST FOR THE BE
LOW 70, 1 ABOVE 70, 2 ABOVE 70, TOTAL CONTROL, AND TOTAL
NEUROTIC CRITERION GROUPS............................. ' 189
XLIV. INTERCORRELATIONS OF FREQUENCIES OF TOTAL RECALL (EXPRESSED

AS RANKS) OF THE TEST FIGURES ON THE IMMEDIATE RECALL TEST
FOR THE CRITERION GROUPS.............................. 189
XLV. FREQUENCY OF RECALL IN FIRST POSITION, EXPRESSED AS RANKS,

OF EACH OF THE TEST FIGURES ON THE IMMEDIATE RECALL TEST
FOR THE BELOW 70, 1 ABOVE 70, 2 ABOVE 70, TOTAL CONTROL,
AND TOTAL NEUROTIC CRITERION GROUPS.................... 192
XLVI. INTERCORRELATIONS OF FREQUENCIES (EXPRESSED AS RANKS) OF

THE TEST FIGURES IN FIRST POSITION ON THE IMMEDIATE RECALL
TEST FOR THE CRITERION GROUPS.......................... 192
XLVII. FREQUENCY OF TOTAL RECALL AND FREQUENCY OF RECALL IN FIRST

POSITION, EXPRESSED AS RANKS, OF EACH OF THE TEST FIGURES
ON THE IMMEDIATE RECALL TEST FOR THE BELOW 70, 1 ABOVE 70,
2 ABOVE 70, TOTAL CONTROL AND TOTAL NEUROTIC CRITERION
GROUPS............................................... 194
XLVIII. CORRELATIONS OF FREQUENCY OF TOTAL RECALL AND FREQUENCY OF

RECALL IN FIRST POSITION FOR THE BELOW 70, 1 ABOVE 70, 2
ABOVE 70, TOTAL CONTROL AND TOTAL NEUROTIC CRITERION GROUPS 194
XLIX. DISTRIBUTION OF THE TOTAL NUMBER OF FIGURES RECALLED, MEANS,

AND STANDARD DEVIATIONS ON THE IMMEDIATE RECALL TEST FOR
THE CRITERION GROUPS.................................. 196
L. FREQUENCY OF RECALL, EXPRESSED IN PERCENTAGES, OF EACH OF

THE TEST FIGURES ON THE IMMEDIATE RECALL TEST FOR THE BELOW
70, 1 ABOVE 70, 2 ABOVE 70, TOTAL CONTROL AND TOTAL NEUROT
IC CRITERION GROUPS PLUS CRITICAL RATIOS OF THE DIFFERENCES
BETWEEN THE TOTAL CONTROL AND TOTAL NEUROTIC GROUPS..... 198
LI. PERCENTAGE INCIDENCE OF DISTORTIONS OF EACH OF THE BENDER-

GESTALT FIGURES ON THE IMMEDIATE RECALL TEST FOR THE
MATCHED CRITERION GROUPS (BELOW 70 AND NEUROTIC)........ 200
- vii -
Table Page
LII. PERCENTAGE INCIDENCE OF CONFUSIONS FOR ALL COMBINATIONS OF

FIGURES APPEARING ON THE IMMEDIATE RECALL TEST SHOWN BY
SUBJECTS IN THE MATCHED CRITERION GROUPS (BELOW 70 AND
NEUROTICS)............................................... 203
LIII. PERCENTAGE INCIDENCE OF THE TENTATIVE SIGNS ON THE IMMEDI

ATE RECALL TEST FOR THE MATCHED CRITERION GROUPS (BELOW 70
AND NEUROTIC) PLUS CRITICAL RATIOS OF THE DIFFERENCES BE
TWEEN THE GROUPS WHICH ARE SIGNIFICANT AT THE 10£ LEVEL OR
BETTER.................................... 20
LIV. PERCENTAGE INCIDENCE OF THE TENTATIVE SIGNS ON THE IMMEDI

ATE RECALL TEST FOR THE BELOW 70 AND TOTAL NEUROTIC GROUPS
PLUS CRITICAL RATIOS OF THE DIFFERENCES BETWEEN THE GROUPS
WHICH ARE SIGNIFICANT AT THE 10* LEVEL OR BETTER........ 205
LV. PERCENTAGE INCIDENCE OF THE TENTATIVE SIGNS ON THE IMMEDI

ATE RECALL TEST FOR THE CRITERION GROUPS PLUS CRITICAL
RATIOS OF THE DIFFERENCES BETWEEN THE TOTAL CONTROL AND
TOTAL NEUROTIC GROUPS WHICH ARE SIGNIFICANT AT THE 5%
LEVEL OR BETTER ................................. 205
LVI. SCORING OF THE INITIAL TEST RECORD OF EACH CASE IN THE BE
LOW 70 MATCHED CRITERION GROUP ON THE BASIS OF THE SELECTED
SIGNS SHOWN TO HAVE CONSISTENT VALIDITY ON THE INITIAL TEST
AND THE RETEST........................................... 343-7
LVII. SCORING OF THE INITIAL TEST RECORD OF EACH CASE IN THE

NEUROTIC MATCHED CRITERION (SOUP ON THE BASIS OF THE
SELECTED SIGNS SHOWN TO HAVE CONSISTENT VALIDITY ON THE
INITIAL TEST AND THE RETEST.... ...................... 348-50
LVIII. SCORING OF THE RECALL TEST RECORD OF EACH CASE IN THE BE
LOW 70 MATCHED CRITERION (SOUP ON THE BASIS OF SELECTED
SIGNS SHOWN TO BE SIGNIFICANT AT THE 5)6 LEVEL.. 351-2
LIX. SCORING OF THE RECALL TEST RECORD OF EACH CASE IN THE

NEUROTIC MATCHED CRITERION (SOUP ON THE BASIS OF SELECTED
SIGNS SHOWN TO BE SIGNIFICANT AT THE 556 LEVEL..... 353-4
LX. SCORING OF THE RETEST RECORD OF EACH CASE IN THE BELOW 70

MATCHED CRITERION (SOUP ON THE BASIS OF THE SELECTED SIGNS
SHOWN TO HAVE CONSISTENT VALIDITY ON THE INITIAL TEST AND
THE RETEST........................................... 355-7
- viii -
Table Page
LXI. SCORING OF THE RETEST RECORD OF EACH CASE IN THE NEUROTIC

HATCHED CRITERION GROUP ON THE BASIS OF THE SELECTED SIGNS
SHOWN TO HAVE CONSISTENT VALIDITY ON THE INITIAL TEST AND
THE RETEST.............................................. 358-60
T.YTT. DISTRIBUTION OF SCORES, CUMULATIVE PERCENTAGES, MEANS, AND

STANDARD DEVIATIONS ON THE INITIAL TEST AND THE RETEST FOR
THE MATCHED CRITERION GROUPS (BELOW 70 AND NEUROTIC)......... 208
LXIII. INITIAL, RETEST, AND COMBINED SCORES OF EACH CASE IN THE

BELOW 70 MATCHED CRITERION GROUP.......................... 361-2
LXIV. INITIAL, RETEST, AND COMBINED SCORES OF EACH CASE IN THE

NEUROTIC MATCHED CRITERION GROUP........................ 363-4
LXV. DISTRIBUTION OF COMBINED INITIAL AND RETEST SCORES, CUMULA

TIVE PERCENTAGES, MEANS, AND STANDARD DEVIATIONSFOR THE
MATCHED CRITERION GROUPS (BELOW 70 AND NEUROTIC)............ 215
LXVI. DISTRIBUTION OF SCORES, CUMULATIVE PERCENTAGES, MEANS, AND

STANDARD DEVIATIONS ON THE RECALL TEST FOR THE HATCHED CRI
TERION GROUPS (BELOW 70 AND NEUROTIC)...................... 218
LXVII. INITIAL, RETEST, RECALL, AND TOTAL SCORES OF EACH CASE IN

THE BELOW 70 MATCHED CRITERION GROUP.................... ' 365-6
LXVIII. INITIAL, RETEST, RECALL, AND TOTAL SCORES OF EACH CASE IN

THE NEUROTIC MATCHED CRITERION GROUP.................... 367-8
LXIX. DISTRIBUTION OF TOTAL SCORES (INITIAL PLUS RETEST PLUS RE

CALL) , CUMULATIVE PERCENTAGES, MEANS, AND STANDARD DEVIA
TIONS FOR THE MATCHED CRITERION GROUPS (BELOW 70 AND
NEUROTIC)................................................ 221
LXX. DISTRIBUTION OF COMBINED INITIAL AND RECALL SCORES, CUMULA

TIVE PERCENTAGES, MEANS, AND STANDARD DEVIATIONS FOR THE
MATCHED CRITERION GROUPS (BELOW 70 AND NEUROTIC)............. 224
LXXI. DISTRIBUTION OF COMBINED RETEST AND RECALL SCORES, CUMULA

MATCHED CRITERION GROUPS (BELOW 70 AND NEUROTIC)............. 223
LXXII. DISTRIBUTION OF SCORES, CUMULATIVE PERCENTAGES, MEANS, AND

THE UNMATCHED CRITERION GROUPS (BELOW 70 AND NEUROTIC) 227
T.YTTTT. INITIAL, RETEST, AND COMBINED SCORES OF EACH CASE IN THE

BELOW 70 UNMATCHED CRITERION GROUP........................ 369-70
- ix -
Table Page
LXXIV. INITIAL, RETEST, AND COMBINED SCORES OF EACH CASE IN THE

NEUROTIC UNMATCHED CRITERION GROUP...................... 371
LXXV. DISTRIBUTION OF COMBINED INITIAL AND RETEST SCORES, CUMULA

UNMATCHED CRITERION GROUPS(BELOW 70 ANDNEUROTIC)........... 231
LXXVI. DISTRIBUTION OF SCORES, MEANS, AND STANDARD DEVIATIONS ON

THE INITIAL TEST AND THE RETEST FOR THE BELOW 70, 1 ABOVE
70, 2 ABOVE 70, TOTAL CONTROL, AND TOTAL NEUROTIC CRITERION
GROUPS................................................... 234
LXXVII. DISTRIBUTION OF COMBINED INITIAL AND RETEST SCORES, MEANS,

AND STANDARD DEVIATIONS FOR THE BELOW 70, 1 ABOVE 70, 2
ABOVE 70, TOTAL CONTROL, AND TOTALNEUROTICCRITERION GROUPS 235
LXXVIII. INITIAL, RETEST, AND COMBINED SCORES OF EACH CASE IN THE 1

ABOVE 70 CRITERION GROUP................................ 372-3
LXXIX. INITIAL, RETEST, AND COMBINED SCORES OF EACH CASE IN THE 2

ABOVE 70 CRITERION GROUP................................ 374-5
I,TXT. MEANS, DIFFERENCES BETWEEN MEANS, AND CRITICAL RATIOS OF THE

DIFFERENCES BETWEEN MEANS OF THE INITIAL TEST SCORES, RETEST
SCORES, AND COMBINED INITIAL AND RETEST SCORES FOR THE
MATCHED AND UNMATCHED CRITERION GROUPS AND FOR THE TOTAL CRI
TERION GROUPS............................................. 244
LXXXI. CORRELATIONS OF GRADE, AGE, AND INTELLIGENCE TEST SCORES

WITH SCORES ON THE INITIAL TEST AND THE RETEST FOR THE TOTAL
CONTROL AND TOTAL NEUROTIC CRITERION GROUPS................. 246
LXXXII. SCORING OF THE INITIAL TEST RECORD OF EACH CASE IN THE BELOW
70 MATCHED CRITERION GROUP, USING PASCAL AND SUTTELL'S SIGNS 376-87
LXXXIII. SCORING OF THE INITIAL TEST RECORD OF EACH CASE IN THE

NEUROTIC MATCHED CRITERION GROUP, USING PASCAL AND SUTTELL'S
SIGNS................................................... 388-99
LXXXIV. DISTRIBUTION OF THE HAW SCORES, MEANS, AND STANDARD DEVIA

TIONS ON THE INITIAL TEST RECORD FOR THE MATCHED CRITERION
GROUPS (BELOW 70 AND NEUROTIC) SCORED ACCORDING TO PASCAL
AND SUTTELL'S SYSTEM............................. 249
LXXXV. RAW SCORE MEANS, DIFFERENCES BETWEEN THE MEANS, AND CRITICAL
RATIOS OF THE DIFFERENCES BETWEEN MEANS ON THE INITIAL TEST
RECORD FOR THE MATCHED CRITERION GROUPS (BELOW 70 AND NEUROT
IC) SCORED ACCORDING TO PASCAL AND SUTTELL'S SYSTEM........... 250
- x -
Table Page
LXXXVI. EDUCATIONAL STATUS, AGE COMPOSITION, AND MARITAL STATUS OF

THE CROSS-VALIDATION GROUPS (TAC SATISFACTORY ADJUSTMENT
AND TAC UNSATISFACTORY ADJUSTMENT)...................... 257
LXXXVII. SCORING OF THE INITIAL TEST RECORD OF EACH CASE IN THE

TOTAL MALE TAC SATISFACTORY ADJUSTMENT GROUP ON THE BASIS
OF THE SELECTED SIGNS SHOWN TO HAVE CONSISTENT VALIDITY ON
THE INITIAL TEST AND THE RETEST............. 1 ........... 400-1
LXXXVIII. SCORING OF THE INITIAL TEST RECORD OF EACH CASE IN THE

TOTAL MALE TAC UNSATISFACTORY ADJUSTMENT GROUP ON THE BASIS
OF THE SELECTED SIGNS SHOWN TO HAVE CONSISTENT VALIDITY ON
THE INITIAL TEST AND THE RETEST......................... 402-3
LXXXIX. SCORING OF THE RETEST RECORD OF EACH CASE IN THE TOTAL MALE
TAC SATISFACTORY ADJUSTMENT GROUP ON THE BASIS OF THE
SELECTED SIGNS SHOWN TO HAVE CONSISTENTVALIDITY ON THE
INITIAL TEST AND THE RETEST............................ 404-5
XC. SCORING OF THE RETEST RECORD OF EACH CASE IN THE TOTAL MALE
TAC UNSATISFACTORY ADJUSTMENT GROUP CHI THE BASIS OF THE
SELECTED SIGNS SHOWN TO HAVE CONSISTENTVALIDITY ON THE
INITIAL TEST AND THE RETEST............................ 406-7
XCI. DISTRIBUTION OF SCORES, CUMULATIVE PERCENTAGES, MEANS, AND

THE MATCHED MALE CROSS-VALIDATION GROUPS (TAC SATISFACTORY
ADJUSTMENT AND TAC UNSATISFACTORY ADJUSTMENT)........... 258
XCII. INITIAL, RETEST, AND COMBINED INITIAL AND RETEST SCORES

FOR EACH CASE IN THE TOTAL MALE TAC SATISFACTORY ADJUST
MENT GROUP (MATCHED AND UNMATCHED) 408
XCIII. INITIAL, RETEST, AND COMBINED INITIAL AND RETEST SCORES

FOR EACH CASE IN THE TOTAL MALE TAC UNSATISFACTORY ADJUST
MENT GROUP (MATCHED AND UNMATCHED)..................... '409
XCIV. DISTRIBUTION OF SCORES, CUMULATIVE PERCENTAGES, MEANS, AND

STANDARD DEVIATIONS OF THE COMBINED INITIAL AND RETEST
SCORES FOR THE MATCHED MALE CROSS-VALIDATION GROUPS (TAC
SATISFACTORY ADJUSTMENT AND TAC UNSATISFACTORY ADJUSTMENT) 261
XCV. DISTRIBUTION OF SCORES, CUMULATIVE PERCENTAGES, MEANS, AND

STANDARD DEVIATIONS ON THE RECALL TEST FOR THE MATCHED
MALE CROSS-VALIDATION GROUPS (TAC SATISFACTORY ADJUSTMENT
AND TAC UNSATISFACTORY ADJUSTMENT)..................... 263
- ad -
Page
DISTRIBUTION OF SCORES, CUMULATIVE PERCENTAGES, MEANS, AND

THE TOTAL MALE CROSS-VALIDATION GROUPS (TAC SATISFACTORY

SCORES FOR THE TOTAL MALE CROSS-VALIDATION GROUPS (TAC

THE TOTAL FEMALE CROSS-VALIDATION GROUPS (TAC SATISFACTORY
DISTRIBUTION OF SCORES, CUMULATIVE PERCENTAGES, MEANS AND

SCORES FOR THE TOTAL FEMALE CROSS-VALIDATION GROUPS (TAC
SATISFACTORY AND TAC UNSATISFACTORY ADJUSTMENT)......... 271

THE TOTAL MALE AND FEMALE CROSS-VALIDATION GROUPS (TAC

SCORES FOR THE TOTAL MALE AND FEMALE CROSS-VALIDATION
GROUPS (TAC SATISFACTORY ADJUSTMENT AND TAC UNSATISFACTORY
ADJUSTMENT).......................................... 278

STANDARD DEVIATIONS ON THE RECALL TEST FOR THE TOTAL MALE
AND FEMALE CROSS-VALIDATION GROUPS (TAC SATISFACTORY ADJUST
MENT AND TAC UNSATISFACTORY ADJUSTMENT)................ 282
MEANS, DIFFERENCES BETWEEN MEANS, AND CRITICAL RATIOS OF

THE DIFFERENCES BETWEEN MEANS OF THE INITIAL TEST SCORES,
RETEST SCORES, AND COMBINED INITIAL AND RETEST SCORES FOR
THE CROSS-VALIDATION GROUPS........................... 283
PERCENTAGE INCIDENCE OF THE SIGNS USED IN THE FINAL SCOR

ING SYSTEM FOR THE TAC TOTAL MALE SATISFACTORY ADJUSTMENT
AND TAC TOTAL MALE UNSATISFACTORY ADJUSTMENT GROUPS ON THE
INITIAL TEST AND THE RETEST........................... 285
- xii -
CHAPTER I
HISTORICAL BACKGROUND
Although the literature on the Bender-Gestalt test, vhlch
is the subject of the present Investigation, is relatively limited and
unimpressive from the point of view of extensiveness, it is fundamental
to a proper appreciation of this test and the work that has been done
with it to understand fully the theoretical and clinical climate which
gave rise to it and which has served to gain for it increasing accep
tance and application. It is the purpose of this section to relate the
origin and development of the Bender-Gestalt to developments in the
field of projective personality testing in general and to relevant ap
proaches to the study of perception, expressive movement, and graphic
media in particular.
In recent years there has been an increasing emphasis on the
factor of motivation as an influence on the perceptual processes. A per
ception is not a literal copy of the stimulus, and individual differences
in perceiving cannot be explained solely in terms of the objective stimu
lus pattern, primary configural tendencies, or idiosyncrasies in sensory,
neural, or motor capacities. We frequently perceive what we expect to
perceive or what we want to perceive— a point exemplified in the experi
ments by Bruner and Goodman (31)» showing the tendency of poor children
to overestimate the size of coins to a significantly greater degree than
children of wealthy families, by Marks (65), demonstrating that the skin

- 2 -
color judgments by Negro college students of their classmates were marked
ly Influenced by their own skin color, and by Coffin (33), illustrating
the influence of prestige suggestion upon perceptual responses to the
Rorschach. Needs, wants, values, and tensions can be significant or
ganizing factors in perception, particularly when attention i3 focused
upon idiosyncratic variation rather than group generalization. As Gard
ner Murphy expresses it: (71) "...motives in the broad sense control per
ceptual responses exactly as they control motor responses" (p.348) and
"... whenever our needs differ we literally see differently. Much of
the process of individual perception depends on the force of past wants.
the person's need to disentangle and restructure in terms of the situa
tions with which he has had to cope." (p.339)
It follows that the more unstructured and ambiguous the stimu
lus situation, the greater the likelihood that any given subject will pro
ject his needs into the situation in an effort to give it meaning and
form and henee the greater the individual variation in perceptual re
sponse that may be anticipated. To quote Murphy (71): "The outer world
can never be so completely unstructured as to make perception depend sole
ly upon the perceiver, but it can never be so sharply and clearly organized
as to obliterate individual differences among perceivers." (p.353)• Struc
turation has become a key concept in the development of projective tech
niques on the assumption that the more amorphous the test situation, the
more opportunity the subject has to reveal his own individual habits, at
titudes, wishes, fears, and behavior tendencies. L. K. Frank's (41) char
acterization of projective personality tests makes this assumption quite
explicit:
- 3 -
A projective method for the study of personality involves

the presentation of a stimulus situation designed or chosen
because it will mean to the subject not what the experimen-
tor has arbitrarily decided it should mean (as in most psy
chological experiments using standardized stimuli in order
to be 'objective') but rather whatever it must mean to the
personality who gives it, or imposes upon it, his private,
idiosyncratic meaning and organization, (p. 403).
Hutt (55) classifies personality tests in terms of degree of
structuration as follows:
1. Structured. "Tests in which the test material consists

of conventional, culturally crystallized questions to
which the subject must respond in one of a very few,
fixed ways." Examples: Bernreuter, Minnesota Multi-
phasis Personality Inventory.
2. Partially structured, "...test stimulus remains rela

tively structured or conventionalized, but the subject
may respond in a relatively free or individual manner
to this stimulus. Test material will generally be rec
ognized by all subjects and interpreted, in so far as
superficial or primary perceptions are concerned, by
them in a fairly uniform way. However, their responses
will reflect certain aspects of their own personality
needs, since their responses are not restricted to any
designated, channelized alternatives. Since this type
of test invites, and in fact requires, the interaction
of the subject and the stimulus material, it presumably
is influenced by the subject's personality. On the other
hand the stimulus is still a conventional situation and
so may limit the response to those alternatives which
are possible within that cultural framework, at least
insofar as the subject still retains his contact with
and 'normal* perception of reality....All of these tests
may be termed projective personality tests because they
elicit responses dependent in part upon the projection
by the subject of his personal interpretation or inter
action into the stimulus." Examples: Thematic Apper
ception Test, Kent-Rosanoff Word Association, incom
plete sentences, Bender-Gestalt.
3* Unstructured. "Neither the stimulus nor the response

is conventionalized. There is no restriction of the
subject's use of the stimulus or his response to it.
They require an active, creative process by the sub
ject, since the stimulus is vague and diffuse and can
be responded to meaningfully only after the subject has
interpreted it in his own way and according to his own
- A -
needs. These are projective personality devices be

cause the response is directly attributable to the
interaction of the subject with the stimulus.n Ex
amples: Stem's Cloud Pictures, Rorschach.
It is not surprising that perceptual tests have proved them
selves to be a fruitful avenue of approach to the study of both the
normal and the abnormal personality. According to Murphy (71):
There is a primary reason why clinics have found

that they need perceptual tests. Serious sampling
errors are always encountered if personality is
judged from a small number of behavior instances;
by working at higher level we obtain at one stroke
so many expressions of personality dispositions and
of their interrelations that reliability is almost
guaranteed. Because outer expression is more vari
able than inner organization, the consistency of
personality is higher with reference to the way one
sees than with reference to the way one does. (p. 669)
The alteration of the perceptual processes is perhaps most
unequivocally observed in organic involvement and in the psychoses.
Harrower (45) showed 30 patients with cerebral lesions Rubin's vase
and modifications of it in which either the vase or the profile was
enhanced and found a marked perseveration of the set engendered by the
first perception and such rigidity of set that the less dominant figure
could not be seen at will despite its having been seen in drawings where
it was dominant. Schilder (S3) noted that it took from 200 ms. to 500
ms. for patients with acute confusions following head injuries to re
cognize objects recognized by normal subjects in 40 ms., and Piotrowski
(78, 79) has described peculiarities of response characteristic of or
ganic* on the Rorschach. With regard to the functional psychoses,
Schilder (04) and Angyal (9) argue that the perception of the body is
altered, Street (88) has found that both manic-depressives and schizo
phrenics perceive the partially drawn figures of his Gestalt Completion

- 5 -
Test less readily than normals, and D. E. Cameron (32) reports that
manic-depressives see fluctuations in an ambiguous figure with great
difficulty. Hunt and Cofer (54), after reviewing the literature, con
clude that perceptual processes show greater evidence of deficit in
functional and organic psychoses than do receptor thresholds and add:
"Furthermore, the deficit appears to reside in the central processes
initiated by receptor action, particularly in the response to patterns
and in the alteration of sets or attitudes." (p. 992-993)*
Although the interest of the clinical psychologist and the
student of personality in perception developed some time after the ex
perimental investigation of perception in the laboratory, the diagnos
tics and therapeutic possibilities of graphic-motor behavior as re
vealed in artistic productions were recognized by the clinician long
before such behavior was made the object of systematic experimental
investigation. Anastasi and Foley (3, 4, 5, 6) in a series of articles
survey the literature on artistic behavior in the abnormal— a litera
ture in which theory and speculation are almost as numerous as undis
puted fact. They distinguish between studies which (1) describe and
analyze spontaneous productions, which are "largely observational and
generally follow a case study procedure", and (2) those using experi
mental technique, in which the "drawing or other objects are produced
in response to specific directions by the experimenter and under con
ditions which are more or less controlled, especially with reference to
materials, time limits, presence of observers, and general surroundings".
The analysis of spontaneous productions is influenced principally by
the theoretical preconceptions of the investigator, while the experi

- 6 -
mental investigations are subject to such criticisms as (1) inadequate
control data, (2) variation in the nature and extent of control of ex
perimental conditions, (3) insufficient data an the background of the
subjects, (4) failure to report results fully or systematically, (5)
absence of quantitative data (frequently only the author*s conclusion
is given along with a few illustrative cases), and (6) interpretations
phrased in "Gestalt or psychoanalytic terns which are relatively vague,
ill-defined, complex, and mentalistic". (p.232).
Anastasi and Foley (8) endeavored to overcome the objections
to previous studies by conducting a carefully controlled experimental
investigation of the drawing behavior of adult psychotics in compari
son with that of a normal control group. 340 white, normal adults and
340 institutionalized patients, equated for age, sex, marital status,
educational background, vocational level, artistic training, geogra
phical distribution, and national background, were asked to make four
drawings: a free choice, a portrayal of danger, a man, and a copy of
a stylized floral design in six colors. The drawings were then clas
sified and scored on the basis of a 92-item check list of special pe
culiarities. While significant differences between normal and abnormal
drawings were found with respect to such characteristics as persevera
tion, scribbling or scrawling lines, omission of essential details,
decorative over-elaboration, disproportion or displacement of parts,
incoherence or chaotic organization, stereotypy of content or technique,
inclusion of writing, and gross distortions, the differences were not
as great as the extravagant statements of previous investigators might
have led one to believe, and in a high percentage of comparisons the
differences were neglible. Referring to the check list, the authors

- 7 -
comment: "None of the items on the check list occurs with a suffi
ciently high frequency in the abnormal group to be regarded as a reg
ular feature of abnormal drawings* Each of the special peculiarities
in the check list was observed in only a small minority of the abnor
mal sampling*... Nearly 60% of the frequences were under 10. The
larger frequencies, furthermore, were obtained in those characteristics
which were non-discriminating.”
Interest in graphic behavior has not, of course, been limited
to the study of art and art forms, since it has been recognized in the
past two decades that virtually all forms of graphic expression may have
implications for personality evaluation and diagnosis. Allport and Ver
non (2) in their study of expressive movement, which they define as
"these aspects of movement which are distinctive enough to differenti
ate one individual from another”, included such graphic indicators as
(1) drawing of circles the estimated size of 25-cent and 50-cent coins,
(2) drawing of rectangles the estimated size of dollar bills, (3) cir
cles drawn on paper, with right hand, (4) circles drawn on paper, with
4
left hand, (5) circles drawn with crayon, right and left hand, (6)
squares drawn on blackboard, right hand, (7) squares drawn in sandbox,
right and left feet, (8) parallel lines drawn on paper (pressure board,
right and left hand, (9) length of self-rating checks on rating sheet,
(10) copying prose paragraph, (11) writing sentences and signature on
pressure board, (12) writing "eee*s" on pressure board, (13) writing
sentence with crayon on large surface, (14) writing "one, two, three*•••
twenty" on blackboard, and (15) writing "eee*s" in sand with pointer
attached to feet. These two investigators were primarily interested in

- 8 -
the intra-individual consistency of expressive movement as reflected
in such objective measures as speed, area, proportion of unoccupied
space, fewness or crowdedness of lines, length, and point and grip
pressure and made no attempt to apply their findings directly to psy
chodiagnosis, Of interest to the clinical psychologist who uses gra
phic devices as projective techniques, however, is the following quo
tation, which offers some theoretical foundation for this practice:
Hotor acts are not so specific as to be meaningless,

and being organized they must reflect to a large de
gree the organization of the total brain field. There
are degrees of unity in movement, just as there are
degrees of unity in mental life and in personality.
It is surely not unreasonable to assume that insofar
as personality is organized, expressive movement is
haxmonious and self-consistent, and insofar as person
ality is unintegrated, expressive movement is self
contradictory. (p. 181-182).
In marked contrast to the approach of Allport and Vernon is
that of Werner Wolff (95, 96, 97* 98, 99, 100, 101) whose interest in
the objective measurement of expressive movements is secondary to the
interpretation of those movements, on the assumption that expressive
movements are correlated with personality structure and behavior. Al
though he studied many aspects of expressive behavior such as posture,
gesture, facial expression, tone of voice, and gait, it is significant
in connection with the present investigation that he regards graphic
movements as a projection of expressive movement on paper. Grapholo
gists, of course, have long maintained that handwriting is an expres
sion of personality, and in the past decade the growing belief of many
psychologists that "there may be something in it" has paralleled the
surging movement toward projective devices and away from traditional
personality inventories on the implicit assumption that a person reveals

- 9 -
himself in whatever he does. While the validity studies are confusing
and contradictory, certain graphologists who have been willing to spe
cify in detail how they arrived at their conclusions have achieved re
sults well beyond chance. Saudek (82), for example, has emphasized
that no one sign means anything by itself and has attempted to show
how as many as fifteen or twenty may be brought to bear on a single
trait. Considering 10 signs for the trait, honesty-deceitfulness,
and using a critical score of 4 or more unfavorable signs, he was
able to pick out a high percentage of dishonest employees in 18 busi
ness firms. Murphy (71) cites a case study (p. 695-700) in which the
Rorschach, a brush-painting technique, and graphological analysis ex
hibited a high degree of agreement with each other and with the clini
cal history, and Bell (16), after reviewing the literature, concludes:
No conclusive evidence for the validity of hand

writing as a measure of personality can be drawn
from the experiments to date. Many of them show
weaknesses in their design, yet many of the bet
ter experiments do not yield strong support for
the claims of graphology. On the other hand,
matching experiments seem to be building up evi
dence that handwriting is capable of giving infor
mation of value to the psychologist. If further
experimentation yields confirmation of handwriting*s
worth as a diagnostic medium, an especially useful
tool will be available for measurement, since one
of the chief advantages of handwriting is the ready
availability of specimens for analysis, (p. 316).
A significant step forward in the use of graphic expression
as a vehicle for the assessment of personality has been the use of Gooden-
ough*s Draw-a-Man Test as a projective device. Originally devised as
a non-verbal intelligence test for children, it has been widely used in
psychological clinics in the pastten years as a diagnostic supplement
to such tests as the Rorschach and Thematic Apperception Test. Brill (28)
- 10 -
Bender (21), and Des Lauriers and Halpern (36) report projec
tive applications of the test, but the most comprehensive treat
ment of figure drawing as a method of personality investiga
tion is provided by Karen Machover (63). She disagrees with
Anastasi's and Foley's (5) conclusion that differentiations
through drawings can be made only in the presence of extreme
mental disorders or with individuals who offer personalized
or bizarre productions and argues that figure drawing is an
instrument capable of yielding a subtle and comprehensive per
sonality analysis.
Machover proceeds on the hypothesis that "the fig
ure drawn is related to the individual who is drawing with the
same intimacy characterizing that individual's gait, his hand
writing, or any other of his expressive movements", (p. 5).
She emphasizes pattern interpretation rather than a check list
of signs and takes the point of view that "the differentia
tion of normal from abnormal by means of drawing analysis is a
less real problem than the effectiveness of the method in de
termining the personality and dynamics of the behavior of an
individual" (p. 24). She bases her interpretation upon "thou
sands of drawings" which she has seen in the past fifteen years
in her work at the psychiatric division of King's County Hospi
tal and adds that her "formulation owes much to established pro
jective methods of personality analysis and to psychoanalytical
theory". She speaks vaguely of success that is "much better than

- 11 -
chance" in matching her interpretations with case records, but nowhere
does she present the quantitative findings regarding validity and re
liability demanded by the critical student. It is true that she states
that "the brief survey of drawing analysis here presented is essentially
preliminary and tentative" (p. 103), but it is also true that she ex
presses more confidence than .caution in her actual interpretation of
ilJustrative cases. The elusive factors of intuition and clinical in
sight ostensibly contribute heavily to her success in interpretation,
raising some question as to the communicability of the method. Machover
argues that "graduate psychology students have been able to acquire a
grasp of basic principles involved in the method after some orientation
lectures" (p. 27) but she also adds that "As with all projective tech
niques which seek to grasp as a whole and yet analyze the complex
pattern of personality organization, the technical psychological knowl
edge and the clinical sagacity of the analyst are indispensable to
the translation of the language of the method into the traits and the
behavioral dynamics and trends of the individual studies" (p. 10A).*
* It has recently come to the writer's attention that Machover has de
veloped 88 signs for men and 98 signs for women which are said to
indicate "field orientation" in perception, indicating that her ex
pressed opposition to the "sign approach" has apparently been modi
fied in the interests of objectivity and communicability. These
signs are to appear in a forthcoming book by H. A. Witkin, S. Wapner,
M. Hertzman, K. Machover, H. Lewis and P. Bretnall, tentatively en
titled Perception and Personality, to be published by Harper Brothers,
• New York. In an unpublished doctoral dissertation (The S f a M H t y of
the Individual's Mode of Perception, and of Perception-Personal!tv
Relationships. April, 1951, New York University Library), Gerald
Bauman used 32 of the 88 signs for men and 40 of the 98 signs for
women and reports correlations of .94 and .95 between these"partial"
scores and Machover1s total Figure Drawing score.
- 12 -
Another Interesting development in the area of graphic
projective techniques is the "myokinetic psychodiagnosis" of Mira
(66, 67). Subjects are requested to draw horizontal lines from left
to right and right to left and vertical lines upwards and downwards
separately with the left and right hands and subsequently are asked
to draw zigzag lines, a chain, a "staircase", and a "top of a castle",
sometimes bimanually and sometimes unimanually. A number of quanti
tative measures (e.g., length of lines, variability as determined by
the arithmetic and algebraic sums of the differences between the
lengths of the lines and the lengths of the models) and qualitative
indicators (e.g., straightness, load, and regularity of lines, me
ticulousness, clumsiness) are evaluated for various clinical diagnos
tic groups (normals, epileptics, obsessed, agitated depression, re
tarded depression, suicidal, schizophrenic, elated, psychopathic
personality, and organic) and some significant differences are re
ported. However, although data are reported on a total of 144 cases,
the number of cases in the diagnostic groups is small, ranging from
5 in the organic group to 35 in the normal group, so that conclu
sions as to the clinical usefulness of the test must necessarily be
tentative.
In contrast to the investigations which have been reviewed up
to this point— investigations which have emphasized either perceptual
or motor response— we now come to a consideration of a group of studies
that employ tests which call for an integrated perceptual-motor per
formance and which are presumed to shed light on personality dynamics

- 13 -
and emotional adjustment. Laboratory studies of such tasks as mirror
drawing and the reproduction of designs are familiar to students of
the experimental psychology of perception and learning, but envisage-
ment of visual-motor performance as a diagnostic clinical device is a
comparatively new conception, nurtured in the soil of the findings re
garding motivational factors in perception, expressive movement, and
graphic productions. Bender (20) is generally given credit for taking
the experiments on the reproduction of visually perceived form whether
from copy or from memory, out of the laboratory and into the clinical
setting, and there is little question that she did more than any one
individual to see the diagnostic possibilities of the method and popu
larize its use, but it may be profitable to review some of the earlier,
though much less extensive, work along these lines in order to evaluate
her contribution in its proper perspective.
Sapas (80) exposed each of nine geometrical figures for 10
seconds and then required 60 normal adults and 126 abnormal adults to
draw them from memory. He reported characteristic deviations from the
models by the different types of cases, suggesting that such deviations
might have diagnostic value. He noted that the subject's approach was
a significant differentiating feature; melancholics, for example, showed
great inhibition in beginning to draw while manics were willing to draw
and talk, their approach not differing appreciably from normals.
Foster (40) exposed the card containing the two 1916 Stanford-
Binet designs (Drawing Designs from Memory X,3) for 15 seconds and then
asked 533 mental hospital patients to reproduce them from memory. The
most frequent changes were addition of irrelevant detail and alteration

- 14 -
in size. Syphilic psychotics, schizophrenics, and paranoid patients
elaborate the drawings or introduce fanciful lines having little or no
relation to the original, while arteriosclerotics, seniles, manic-
depressives, and mental defectives are apt to omit lines, producing
figures which are sketchy and scanty in detail. Arteriosclerotics
and seniles frequently fail to attempt the drawings— a finding that
agrees with that of Sapas in suggesting that the subject's approach
to the task may assqme as much diagnostic significance as the repro
ductions themselves. Foster makes quite a point of the size differ
ences, noting that size is exaggerated by syphilitic, schizophrenic,
and unclassified paranoid conditions and diminished in alcoholic
psychoses, manic-depressive psychosis, and mental deficiency. She
reports means but no standard deviations or tests of significance
for the size of one of the two figures for the several clinical
groups, but inspection of the differences suggests that they prob
ably are not significant. The height of the stimulus figure is
4.4 cm., and the height of the reproductions ranges from 3.8 cm.
for the mentally defective and "not insane" groups to 4.4 cm. for
the schizophrenics. Moreover, she does not give the frequency of the
reported deviations in form and embellishment for any of her groups
nor does she provide illustrations of these changes, so that it
would be difficult for the clinician to make any practical use of
her results.
Becker (15) asked his subjects (60 schizophrenics and 41
normals) to reproduce three figures (a church drawing and two simple
geometrical designs) from memory after a 10-second exposure for each

- 15 -
and reported as differentiating features the following characteristics
of schizophrenic performances (l) lack of integration or coordinated
plan, (2) a relatively high proportion of errors, and (3) idiosyncra
tic variations. It was his observation that the motor factor tends
to outweigh the visual in schizophrenic drawings, the movements being
impulsive and even "gesture-like".
In 1938 Lauretta Bender (20), impressed ty the experimental
studies of Wertheimer, Kohler, and Koffka in the Gestalt theory of
perception, published a monography in which she summarized her find
ings on the alterations of the "Gestalt function" in the copying of
geometrical designs by children and abnormal patients. Parts of this
monograph summarize her earlier papers on this subject (17, 18, 19),
which date back to 1932, while other sections are concerned with more
recent material not previously published. The Gestalt psychologists
had enunciated a number of laws which they believed to operate in
the perception or memory of visual forms— laws which are summarized
by Boring (27) under the headings of naturalness of form, figure and
ground, articulation, good and poor forms, strong and weak forms,
open and closed forms, dynamic basis of form, persistency of form,
constancy of form, symmetry of form, integration of similars and
adjacents, meaningfulness of forms, fusion of forms, and transposi
tion of form (pp. 253-255)— but their findings were essentially based
upon experimental investigations of normal adults. If it is true that
there is no necessary correspondence between the stimulus and the per
ceptual response because of the fact that dynamic principles operate
within the organism to structure the perception, is it also true, asks

- 16 -
Bender, that these same principles operate in pathological or im
mature states of the organism?
Bender's conception of the "gestalt function" and the rela
tionship of her theoretical position to the aims of her study are
stated by her as follows:
The gestalt function may be defined as that function

of the integrated organism whereby it responds to a
given constellation of stimuli as a whole; the res
ponse itself being a constellation, or pattern, or
gestalt. All integrative processes within the ner
vous system occur in constellations, or patterns, or
gestalten. Integration occurs not by summation or
subtraction or association but by differentiation, or
by increasing or decreasing the internal complexity
of the pattern in its setting. It appears that an
integrated organism never responds in any other way.
The whole setting of the stimulus and the whole in
tegrative state or the organism determine the pat
tern of the response. Starting with this thesis,
one may use the given stimulating constellation in
more or less similar setting and study the gestalt
function in various pathological integrative condi
tions in different organic and functional nervous
and mental disorders. Such has been the technique
that was employed in this study. Any pattern in
any sensory field may be regarded as a potential
stimulus. Visual motor patterns have proved most
satisfactory, because the visual field most readily
adapts itself to experimental study and especially
because the sick patient can usually cooperate with
the simple test of copying a few test forms, (pp. 3-4).
She develops her thesis toward the close of her introductory
chapter, at -the same time anticipating her findings:
The factor of becoming is present in the physical

world (Eddington). It accounts for the continuous
integrating physical processes and tendencies. It
is also present in the individual personality. There
it also accounts for tendencies-to action and drives.
In the final analysis the two are the same working
towards the necessity for completing gestalten in all
realms of nature.
- 17 -
Working with and against the tendency of becoming,

are the destructive forces whereby all gestalten or
simplified or destroyed. This tendency is seen in
the experimental work brought forth in this book when
the individual reacting to the stimulus is a deviate.
Even in these individuals, the drive to experience
complete gestalten and to contribute to the integra
tion of gestalten are always present. A gestalt is
always experienced, but a more primitive form tends
to emerge still whole in itself and still greater
than the sum of all its parts, (pp. 5-6).
The factors which determine the gestalt are listed by Bender
as follows:
1. The stimulating pattern in the physical world which

also must obey certain laws of gestalt.
2. The motility of the visual field which determines
spatial relationships.
3. The temporal factor determined by the motility and
sequential relationships which tend to become more
intricately integrated into the spatial relation
ships with maturation processes and are determined,
therefore, by the temporal factor of the life span
of tiie individual.
4. The motor reaction pattern of the individual, his
attitudes towards and actual participation in the
individually created experience.
5. The immediate tendency for each of these factors to
be non-separable from the others, (pp. 53-54).
Somewhat later, she adds:
The properties of the sensory field are determined by

the laws of its biologic nature which are limited by
the maturation level, the integrative integrity (lack
of lesion), the status of the personality, including
the emotional complexes, and the given situation, (p. 75).
The test itself consists of nine geometrical designs which were
selected from approximately thirty which Wertheimer (93) originally used
in his classical study of Gestalt principles. Each design is on a sepa
rate card, and the subject is asked to copy them one at a time. On oc
casion Bender administered the test with brief tachistoseepic exposures,
but the major portion of her research is based upon the procedure of
- 18 -
copying each figure without a time limit. In her monograph she re
ports the results of the administration of the test to children,
mental defectives, schizophrenics, manic-depressives, and patients
with organic involvement (sensory aphasia, dementia paralytica, al
coholic psychoses, and traumatic psychoses). She does not state the
actual number of cases upon which her conclusions are based so that
the question of the adequacy of her sample remains in doubt. In
general,, her report of her findings consists of the presentation
of a few cases illustrative of a given clinical entity followed by
a "summary" in which she generalizes about the population which she
has presumably sampled.
One exception to the above statement is her standardiza
tion of the Bender-Gestalt as a performance test for children based
on 800 children ranging in age from three to eleven and in education
from the pre-school level through the fifth grade. Bender was espe
cially interested in the maturation of the "visual motor gestalt
function" because of her conviction that many of the reproductions
obtained from psychopathological states represent a regression to
an earlier, and hence more primitive developmental level. She ob
serves that "It appears from the studies in sensory aphasia that the
visual motor gestalt function is a fundamental function associated
with language ability and closely associated with various functions
of intelligence such as visual perception, manual motor ability,
memory, temporal and spatial concepts, and organization or representa
tion." (p. 112).
Bender traces the genetic development of the visual motor

- 19 -
patterns in considerable detail. The drawings of children 2-1/2 to
4 years of age are "scribblings that represent motor play" and the
most primitive attempts at reproduction of visual form are whirling
or vortical movements perseverated in the horizontal direction.
Little by little the scribbling is controlled, and evidence of ge<-
stalt organization appears in the reproduction of forms as poorly
executed circles. Angulated forms, crossed forms, and parallel
lines present great difficulties. The gestalt principles enun
ciated by Wertheimer and others are of limited applicability to
the reproductions of children. Bender draws this conclusion in
her monograph and makes her position equally explicit in an arti
cle published eleven years later (24)s
An enclosed loop is the basis of all perceived form.

There is also a tendency to perseverate any one learned
(even if self-discovered) pattern wherevefr adaptable
to other perceived figures. Direction, especially
dextrad horizontal direction in the right-handed child,
is more important than distance or size. It is more
overwhelming than Wertheimer's principles of proximity
or similarity of parts. This predominance of the di
rectional factor is probably due in part to motofr fea
tures, and in part to the principle that the optic
field is organized on movement. Concepts such as se
ries and masses are more readily grasped by children
than absolute number of size. Wertheimer's principle
of continuity is important to the extent that it in
volves direction and series. His principles of 'gute
Gestalt' and natural geometrical figures are important
to the extent that they arise out of the primitive loop.
Otherwise, the principles of gestalt, as given by
Wertheimer, do not apply to the genesis of gestalt
in the maturation of the child's visual motor pat
tern. (p. 166).
Vertical rotations appear as a developmental phenomenon in
children of pre-school age but disappear by the age of seven or eight.
Differentiation of form, maturation of the motor apparatus, and capa
city for the representation of objects take place rapidly between the
- 20 -
ages of four and seven and, according to Bender (20), "all of the fig
ures are satisfactorily produced at the age of eleven years. Adults
add only a certain motor perfection or perfection in detail in sizes
and distances." (p. 112). In her later article (24), she adds: "The
standardization of gestalt maturation and function covers the ages
four to eleven, which is the age when language function, including
reading and writing, are developing. Like the Goodenough, it is not
valid after adolescence." (p. 165).
Although Bender (20) makes the statement in her monograph
that the "evolution of gestalten is a maturation process rather than
an educational or imitative one" (p. 26), she presents evidence in a
subsequent article (24) that suggests that maturation of the gestalt
function does not take place in a cultural vaccuum. She cites a study
by Dr. Alice Joseph and Dr. Veronica F. Murray in which they adminis
tered the Bender-Gestalt to 149 native children from the Chamorros and
Carolinians of Saipan in the Mariana Islands in which the drawings
differ markedly from the norms published by Bender for children of
comparable ages. Open figures are frequently closed, vertical ro
tation occurs at all age levels sampled, there is "more movement and
action", figures are influenced by earlier reproductions of other
figures, and in a number of instances, the figures are tied to each
other, the total reproduction tending to make an organized design.
Discussing the performance of mental defectives on the Bender-
Gestalt, Bender (20) observes that their drawings are much more varied
than those of normal children of comparable age and that they are re
produced with better motor control. The reproductions tend to be

- 21 -
small in size, the primitive loop is freely used, the patterns are
"rigid" rotations and perseveration occur, and difficulty with oblique
relations and angles are common. On the basis of the observation that
"some children rated as mentally defective by the standard tests and
social criteria are able to handle the visual motor gestalten in a nor
mal or superior fashion" (p. 138), she concludes that "there are multi
ple causes of mental defectiveness, wnich may be classified as (l) sim
ple retardation in maturation; (2) specific disabilities in the fields
of language; (3) dissociative phenomena which distort the whole person
ality; (4) impulse disturbances; (5) perceptual disturbances; (6) con-
fusional disturbances." (p. 149)*
Patients with organic brain diseases or injuries reproduce
the Bender-Gestalt figures in a grossly distorted manner. In cases
of localized brain damage resulting in sensory aphasia, for example,
the following features are evident; (1) reversion to primitive forms
seen in children and mental defectives, (2) perseveration, (3) use of
compact, inclosed, energy-saving responses and unit symbols, (A) trans
formation of dots into dashes or circles, (5) simplification and omis
sion of detail, (6) rotations or reversals, (7) separation of parts of
figures, and (8) fragmentation. Patients repeatedly tested after cer
ebral accident frequently show marked improvement in their reproduc
tions, the changes tending to follow a development sequence similar to
that noted in children, leading Bender to conclude:
These studies of desturbances in perceptual motor

gestalten in organic brain disease indicate that the
estalt principles are never fixed, but are the in-
fegrative response of the personality-as-a-whole in
any given situation; in disintegrating cerebral lesions
- 22 -
they tend to revert to more primitive levels, and,,

as the brain recovers from its insult, they tend to
follow the laws of developmental maturation in re
turning to the higher integrative responses. Ip. 7b).
Bender observes that "the disturbances in gestalt function does not run
absolutely parallel to the degree of sensory aphasia", however, and
seeks to explain tnis lack of perfect correlation in terms of tne cor
tical area affected. She believes that the gestalt function is in
creasingly affected the nearer the lesion comes to the occipital re
gion and concludes that "the area most probably involved in distur
bances of the visual motor gestalt function, as exemplified by these
copied test forms, is that between the temporal, parietal, and occi
pital lobes of the dominant hemisphere." Ip. 75).
In demential paralytica the outstanding features of the Bender-
Gestalt reproductions are (I) perseveration in the horizontal aextrad
direction, (2) substitution of letters or numbers for dots, (3) sim
plification, (4) fragmentation, (.5) rotation, (.b) transformation of
dots into dasnes, (7) difficulty with line crossing and (8) use of
primitive loops. Early cases often show a formalistic perfection.
Bender argues, however, that even in reproductions shoving marked de
viations from the stimulus figures the essential gestalt is preserved.
Patients suffering from alcoholic psychoses show (.1) a strokelike per
severation, which appears to be a perseveration of motor impulses rather
than of form, (2) substitution of dashes, loops, and jagged lines for
dots, 13) gross distortions, 14) angulation difficulties, 15) tenden
cies toward closure of open forms, (b) simplification, 17) splitting
of a figure into parts, (8) rotation, and IV) "hairy" lines. In the
- 23 -
traumatic psychoses one notes: (1) rotations and reversals, (2) sim
plification, (3) fragmentation, (4) gross distortions, (3) substitu
tion of numbers, crosses, dashes, and jagged lines for dots, (6) re
tracing, and (7) blunting of angles. In the fifth and final cat
egory of organic involvement discussed in the monograph, namely, acute
confusional states, the cardinal characteristics are 11) rotation,
(2 ) perseveration, (3) tendencies toward closure, (,4) mierographia,
[5) gross distortions, (.6) difficulties with overlapping and crossed
figures, (7) blunting of angles, and (8) simplification.
In schizophrenia gross destruction of the gestalt may be evi
dent, with a strong tendency toward splitting or dissociation of parts
of a figure which are normally joined together. Other noteworthy de
viations are Cl) rotations and reversals, (2) flattening of curves,
C3) simplification, (4) substitution of dashes and circles for dots,
(5) fragmentation, (fa) mierographia, (7) perseveration (within a fig
ure and from one figure to another), (8) line crossing difficulties,
and (9) ornamentation. In her 1949 paper Bender (24) characterizes
the test performance of adult schizophrenics in the following manner:
Gestalt drawings of typical schizophrenic patients are

easily recognized by the bizarreness of the configurations
produced, the frequent splitting in figures, not in such
a way as to make a 1gute gestalt1, and especially by show
ing an unusual cohesion between all the figures and an in
crease in movement in the figures, attempts to use these
figures as matrices for delusional ideations by ornament
ing them with connecting lines, destroying the original
gestalt and creating new figures, are usually noticeable.
In such instances, the gestalt test really becomes a pro
jective technique by allowing the person to be tested to
projact his own ideations into the test performance.' (pp.
169-170).
The reproductions of patients in the excited or depressive

-2k-
phases of a manic-depressive psychosis differ considerably, as one
would naturally expect. Manic performance is characterized by speed
of execution, whirligig drawings, free use of space, and strong ten
dencies toward ornamentation or embellishment of the figures without,
however, destroying the basic Gestalt. "Unlike the schizophrenic in
dividuals who make additions to the configuration, these do not de
stroy or distort the original figure but tend to embellish it or to
emphasize it with additions which are secondary to the original de
sign". (p. 110). While "bewildered depressions" exhibit many features
similar to those evident in acute confusional states and in some of the
alcoholic psychoses, except for the preservation of the integrity of
the gestalt, the milder depressions reproduce the figures with an ac
curacy, nicety, and precision suggestive of the meticulousness that
one might exnect to find in obsessive-compulsive neurotic. Such draw
ings hardly appear to be psychotic; if anything, they are "too normal".
In an effort to determine whether the Bender-Gestalt test
could detect malingerers, Bender asked a number of physicians, nurses,
and medical students to take the test as though they were mental de
fectives and in other instances gathered data from prisoners who pre
tended to be psychotic. Although she concludes that "none of these in
dividual was successful in neglecting the essential gestalt principles
which their maturation level would make it possible for them to experi
ence", inspection of the drawings might lead another investigator to
conclude otherwise.
Bender reports no data on the psychoneuroses, psychosomatic
disorders, or psychopathic personality. With regard to the neuroses,

she takes the position that disturbances in perception or in the visual
motor gestalt function are not to be expected.
Although the Bender-Gestalt has come into increasing use in
psychological clinics in the past decade, the published literature on
the test is surprisingly meager, luring the period between Bender's
(17) first published article on the test in 1932 and her monography in
1938, her husbnad, Paul Schilder, and his collaborators published three
papers relative to Bender-Gestalt performance in organics and in schizo
phrenics following shock treatment, and in the next six years two other
papers appeared on these subjects. In his 1934 paper Schilder (83) dis
cusses 35 selected cases of serious head injuries to some of whom he ad
ministered the Bender-Gestalt as well as several other tests. He states
that the following features are typical of the reproductions of such pa
tients in the acute confusional state: (1) the gestalt function is deep
ly impaired, (2) primitive structures appear, (3) angular figures are
changed into circles or ellipses, (4) dasnes are substituted for dots,
(5) changes in direction and conversions of angles into straight lines
occur, (b) perseveration and simplification are common, and (8) motor
execution is clumsy and awkward. Bender, Curran, and Schilder (25)
asked seven patients with Korsakoff's psychosis to draw some of the
Gestalt figures repeatedly. Part of the time the patient had the orig
inal before him, and part of the time he saw only his preceding copy.
Their results show evidences of (1) either sudden or progressive changes,
(2) closure, (3) simplification of curves, and (4) either contraction
of figures or expansion and elongation. They interpret these findings
as a reversion to a more primitive type of organization due to strong

- 2b -
"field forces" liberated in the patient by the organic disorder*
Orenstein and Schilder (73) administered the test to 19 schizo
phrenic patients as they came out of insulin shock and found that
the drawings show such disturbance in gestalt function as (1) per
severation, (2) substitution of circles and loops for points, (3)
substitution of curves for angles and of solid for dotted lines,
(4) change of angles into straight lines, (5) bringing of slanted
lines nearer to the vertical, (b) separation of gestalt units in
space, (7) transgression from one plane to another, (8) rotation of
90° and 180°, and (9) exaggeration of "primitive gestalt principles".
a year later Schilder (85) repeated the study on schizophrenics after
metrozol convulsion and obtained substantially similar results. In
1944 Stainbrook and Lowenbach (87) asked patients undergoing electro-
convulsive therapy to reproduce some of the Bender-Gestalt designs
both before and at various time intervals after the electrically in
duced convulsions. They note that even the earliest attempts at
copying the figures were reactions to the figures as a whole, a sim
ple loop standing for each element in the original, but the parts of
a figure are usually separated in space, angularity is the next fea
ture to be represented, although the parts are still out of propor
tion and dissociated. Perseveration is also noted, even in the ear
liest drawings. Later the elements are brought into their proper
spatial relationships and near the end of the post-convulsive recovery
period (21-25 minutes) the parts are connected to form the gestalt.
They conclude with a word of cautions
Perseveration and 'primitive visual motor gestalt pat-

- 27 -
terns' have been described by Orenstein and Schilder

during and immediately following insulin coma and have
been ascribed to a probability of lesions in temporal,
parietal, and occipital areas. Their descriptions and
illustrations of drawing bear certain resemblances to
our findings in the earlier stages of recovery, but a-
gain we believe that the temptation to correlate the
transitory postshock changes with similar and relative
ly persistent signs demonstrable in the behavior of pa
tients with known organic pathology should be resisted.
Even as late as 1944 references in the literature to the Bender-
Gestalt were few and far between. Mailer (64) in his review of personal
ity tests devotes one sentence to the test, while White (94), reviewing
the literature on projective tests, gives the Bender-Gestalt three lines.
Hunt and Cofer (54) > writing in the same compendium and reviewing the
literature on psychological deficit, are more generous, allocating nine
lines to the test. JJavid Wechsler (90) regards the test as helpful in
estimating the degree of mental deterioration in organic brain disease,
commenting as follows:
We have applied her test to patients variously diagnosed

as seniles, and found that the figures are reproduced by
these patients as primitive loops or segments of‘large
arcs, perseverated in a wavelight manner. This is pre
cisely the kind of thing that Bender found to be char
acteristic of children's efforts prior to maturation of
the psychomotor function (at M.A. level of approximately
2 years) and in particular of mentally defective children
of about this M.A. level, (p. 69).
In 1945 Fabian (38) reported a relationship between reading
disabilities and the incidence of vertical rotations on the Bender-Gestalt.
Testing 586 children ranging in age from 5 to 9 and in education from
kindergarten through the third grade, he found that rotation of the fig
ures from the horizontal to the vertical is a normal developmental phe
nomenon in children of pre-school and beginning school age but that it

- 28 -
virtually disappears by the age of 8 or 9• Over 50# of children aged
5 to 6 show rotations, compared with 20# at ages b£ to 7£ and Tfc At ages
7£ to 9. Using as experimental subjects a group of boys in the children's
ward Bellevue Psychiatric Hospital who ranged in age from 7 through 12,
had I.Q.'s above 80, and manifested severe reading disabilities, he found
that of 21 non-readers 16 (7fc#) vertically rotated one or more of the
Bender figures, and that of 25 boys showing a reading retardation of
from 50-80#, 15 (60#) showed vertical rotation. It should be noted, how
ever, that these boys were referred primarily for personality distur
bances and conduct disorders rather than for scholastic failure, and it
may be that the emotional factors were as significant as the reading
factors in contributing to the rotation phenomenon. Moreover, the size
of the experimental samples was small. In a second experiment figures
A and 3 (which were most frequently rotated) were presented in the verti
cal position to 23 children beginning the first grade. Under these cir
cumstances rotation did not occur, suggesting that rotation from the hori
zontal to the vertical is a tendency which is prepotent over rotation
from the vertical to the horizontal. Discussing this experiment in a
subsequent article, Bender (24) attributes verticalization to the fol
lowing factors: (1) the muscle mechanics of the arm, which favor it,
(2) movement tendencies in the gestalt function, (3) body image factors,
(4) upright posture, and (5) language development.
During World war II the Bender-Gestalt came into increasing
use, largely through the efforts of Army clinical psychologists who
found the test useful as a diagnostic supplement to such tests as the
Wechsler-Bellevue, Rorschach, TAT, Man-and-woman Drawings, Minnesota
Multiphasic Personality Inventory, and home-made sentence completion

- 29 -
tests. Several thousand Bender-Gestalt tests were administered at Mason
General Hospital whose psychological staff published a manual (10) in
1945* However, the most credit for popularizing the test deservedly
goes to Max Hutt who, as an instructor at the Adjutant General's School
for the training of officer clinical psychologists, devoted consider
able time to an exposition of the test and its interpretive significance.
A dynamic teacher, he encouraged his students to try out the test when
they returned to their own installations. His approach to the test 1b
outlined in a mimeographed manual entitled, "A Tentative Guide for the
Administration and Interpretation of the Bender Gestalt Test" (5b), which
he prepared for restricted distribution to officer clinical psycnologists
in June, 1945. This manual was subsequently reproduced by the psychology
sections of Veterans Administration hospitals and clinics and has become
more or less of a "Bible" to clinical psychologists who use the test.
Some of the interpretive statements are culled from Bender's monograph,
while others are based upon analogies with Rorschach elements and upon
Hutt's experience with the test, including "an intensive study of the
records of 130 psychoneurotics". He regards his data as "tentative and
suggestive" but also "meaningful and significant". He feels that the
test can be of "crucial value in analyzing the psychodynamics of the per
sonality in process", tvhile recognizing that "the principle of multiple
determinants affects this test just as it does any other", he is not re
luctant to assign psychodynamic meanings to the various test factors.
Because of the bearing that his approach has to the purpose of the pres
ent investigation, it will be well to consider Hutt's statements in de
tail. In the outline that follows, an attempt has been made to summarize
Hutts’ major scoring factors and their clinical correlates.
Scoring Factor Clinical or Psvchodvnamlc Correlate
1. Arrangement
I
a. Sequence
(1) Methodical Index of intellectual control and

adaptation to reality. Careful or
compulsive pre-planning.
(2) Logical Flexibility, spontaneity. Good

anticipatory reactions. Good
adjustment.
(3) Irregular Reflects inner needs rather than

good planning. Characteristic of
neurotics, occasionally seen in
psychotics.
(4) Confused Characteristic of psychotics,

occasionally seen in organics.
b. Cohesion
(1) Expansive Expansive, elated, or manic mood.

Oppositional or aggressive traits,
as in psychopathic personality.
(2) Compressive Depression, rigidity, constric

tion, withdrawal. Sense of per
sonal inadequacy. Perception of
the world as a hostile place. Hos
tility may be projected upon the
self (masochism) or upon the world
(paranoia)•
c. Margin (use of edge Need for an external aid as a sta

of paper as a guide) bilizing factor. Compensation for
loss or sensed loss of control.
2. Size
a. Expansion Overcompensation. Expansive mood,

manic excitement. Seen in some
neurotics and many hebephrenics.
- 31 -
Scoring Factor Clinical or Pavc^nri-ywmdc Correlate
b. Reduction Reduction in energy output or

drive. Constriction, rigidity,
stereotypy, withdrawal. Seen
in depressed or anxious pa
tients, occasionally in or-
ganics.
3. Form Modifications
a. Rotation of entire Distortion of reality. Bizarre

figure perception. Disorientation, dis
sociation. Almost never occurs
in psychosis.
b. Rotation of part Disorientation, dissociation. More

of figure frequent in organics than in psy
chotics.
c. Reversal (of left Disorientation, negativism resis

and right parts of tiveness. Present in psychopathic
figure) personality, severe neurosis, some
incipient psychotic states.
d. Regression (substitu Deterioration. Inaccurate percep

tion of loops, scrib tion of reality. Present in psy
bles, and dashes for chotics and organics.
dots, diamonds into
squares, etc.)
e. Angulation Poor motor control or faulty per

ception. If the latter, it is re
lated to emotional disturbance or
mental defect.
f. Simplification Immature emotional development

(as seen in the neurotic) or dis
turbed emotional functioning and
poverty of organized and channelized
drive (as seen in the organic).
g. Curvature
(1) Flattening Reduced affect, emotional blunting.
(2) Accentuation Increased affect, emotional re

activity
-(3) Irregularity Instability
(4) Connected Rigidity

straight lines
- 32 -
Scoring Factor Clinical or Psvchodvnamlc Correlate
h. Perseveration Emotional rigidity. Typical of

psychosis and Severe neurosis
(especially obsessive-compulsive).
i. Closure Neurotic sex comflict. Abulia

or inability to complete a task
(organic)
(1) Line crossing Psychasthenia

(figure t>)
j. Fragmentation Dissociation, loss of integra

tive capacity. Present in deteri
orated psychotic states and in brain
damage.
k. Overlapping (es Organic brain damage

pecially figure 7)
1. Elaboration Schizophrenia
("doodling")
m. Slope
(1) Downward Depressed mood
(2) Upward Elated mood
4. Line Quality
a. Sketching Insecurity (including organics)
(1) Repetitive strokes Masturbation
(2) Elongation and Sexual disturbance

repetitive sketch
ing of projections
of figures 7 and 8
(3) Motor Incoordina Physiological tensions, poor

tion muscle tension. Indicative of
intellectual or emotional dif
ficulty.
Following his presentation of the underlying psychodynamics
of the Bender-Gestalt drawings, Hutt proceeds to discuss "clinical syn
dromes" of drawing characteristics which, according to him, distinguish

- 33 -
between the neurotic, the schizophrenic, and the organic, (pp. 17-18).
Typical of the neurotic are (1) preservation of the gestalt, (2) ab
sence of rotation, (3) reduction in size, (4) phallic symbols, (5)
change in curvature values, (6) difficulty with crossings and closure,
(7) simplification, (8) reversal, (9) angulation, (10) occasional re
gression, perseveration, and fragmentation, (11) marked resistance to
the test or parts of it, (12) emotional exclamations, (13) mild dis
turbance in spatial relations, and (14) motor incoordination. Char
acteristic of schizophrenia are (1) rotation, (2) regression, (3) dis
sociation, (4) fragmentation, (5) elaboration, (6) destruction of the
gestalt, (7) chaotic or confused order, (8) gross misuse of space, (9)
marked condensation, and (10) occasional flattening, reversals, block
ing, and exaggeration of phallic parts. Most important in the brain
injured patient are (1) partial rotations, (2) vagueness and sketchiness,
(3) perseveration, (4) loss of detail, (5) fragmentation, (6) difficulty
with acute angles, (7) concreteness and over-simplification, (8) overlap
ping, and (9) exclamations and behavior involving impotency, perplexity,
and incompetence.
At about the same time that Hutt's manual was publiched,
Spiegel, Shor, and Fishman (86) used the Bender-Gestalt as part of a
battery of tests given to subjects at Mason General hospital. The
subjects were regressed to various age levels through hypnosis and
tested at these levels. Only two subjects were employed in the study,
however, and no specific data is given on the Bender-Gestalt. Irion,
Pascal, and Hobbs (11) administered the test to 33 psychiatric patients
and 23 enlisted assistant psychologists at an anqy psychological clinic

- 34 -
and had 10 of the 23 psycholgists grade the papers in terms of neatness
and accuracy of execution. This grading was correlated with emotional
disturbance, as attested by significant differences in the mean ranks
of the psychiatric patients and the group of psychologists*
Oddly enough, the test cards for the Bender-Gestalt were not
made available to the public until 194b, when they were published by
the American Orthopsychiatric Association in conjunction with a brief,
seven-page manual by Bender (22). Studies performed prior to this date
were conducted with "home-made" cards which contained minor inaccuracies,
those most frequently in use being a set of 100 cards prepared by an
amateur mechanical draftsman and distributed by Hutt at the Adjutant
General's School. Bender's manual is concerned mainly with instruc
tions for the administration of the test. It includes a plate showing
"typical" performance at age levels between three and eleven but con
tains no material on test interpretation.
Long interested in childhood schizophrenia, Bender (23 pub
lished a paper in 1947 in which she added this disease entity to the
pathological states whose gesta.lt performance she has studied. As
usual, she does not give the number of subjects upon which her observa
tions are based but describes the Bender-Gestalt record of "the" child
schizophrenic as follows:
The schizophrenic child copying these figures shows

many of the same problems which are shown in his
motility. There is a tendency to use old primitive
responses interlocked with the more mature capacities
which are expected from the maturational level of the
child. There is therefore an excessive use of the vor
tical movement even with good diamond forms. A series
of figures on a horizontal plane may be pulled around
into a vortical figure. The boundaries of circles are
- 35 -
uncertain and may be gone over several times. The

centers of circles are uncertain; there are no points
but many little circles, and for the same reason an
gular and crossed forms are fragmented. Action cannot
be readily controlled and figures are elaborated, en
larged, repeated. The total product makes a pattern
itself with a great deal of fluidity to it based upon
vortical movement. The perceptual patterns lose their
boundaries and therefore their relationship to the back
ground. (pp. 45-46).
In 1947, Wayne, Adams, and Rowe (89), matching overt homo
sexuals in an army disciplinary barracks with a control group of non
homosexuals, concluded that the Bender-Gestalt reveals sexual conflicts
in general but not with sufficient specificity to make it useful for
diagnosis of homosexual tendencies. In the same year Harrower (46)
gives brief mention to the test as part of a battery used in testing a
depressed child and Israel Wechsler (92) acknowledges its clinical value
in his Textbook of Clinical Neurology. Two years later Bender (24) wrote
an article on "Psychological Principles of the Visual Motor Gestalt Test"
which is essentially a restatement of her observations in previous publi
cations. In the same year Barkley (13) developed a "hapto-kinesthetic"
version of the test for use with organics. The Bender figures are re
produced in relief on nine plastic plates and the subject is asked to
feel the design and then draw it. The Bender-Gestalt test is then given
in the usual way, and the two performances are compared. That his find
ings are tentative and in need of more exhaustive experimentation is in
dicated by his conclusions:
Only a minor pilot study has been run to date, but there
appear to be marked differences between the performances
of the brain damaged and normal subjects. It has also
been found that many subjects who give a good reproduc
tion to the visual stimuli exhibit marked distortions on
the reproductions of the hapto-kinesthetic perceptions.
(p. 180).
- 36 -
In his chapter on the Bender-Gestalt in Abt and Beliak1s
book, Protective Psychology. Woltmann (102) briefly summarizes the
literature up to 1950, devoting the principal portion of his article
to a restatement of the material in Bender's monograph, frequently
quoting it verbatim without employing quotation marks but giving her
credit in a footnote, "I am indebted to Dr. Lauretta for a multitude
of suggestions concerning the writing of this chapter". He glosses
over Hutt's contribution, except to criticize his "overrefinement of
instruction" and to add that he introduced Hutt to the test. A few
suggestions are made relative to the clinical use of the test but in
general comparatively little new material is offered.
While this review concludes the known published references
to the Bender-Gestalt with the exception of two important studies that
will be discussed in detail in Chapter II, this historical summary re
view would not be complete without devoting a few paragraphs to other
visual-motor test3 which have failed to achieve the popularity of Ben
der' s test. One of these is the KLlis Visual Designs Test, which has
been applied primarily to children and adolescents and which, according
to Bell (16) "has proven of comparable value to the Bender test in the
diagnosis of organic conditions", (p. 345)• The test was standardized
in 1940 by i*ood and Shulman (103) on 1646 children ranging in age from
to 17£ years, and its diagnostic value with children having organic
brain diseases was reported in 1942 by Lord and Wood (62). Bach of
ten geometrical designs (two of which are identical with those used in
the Memory for Designs Test on the 1937 Stanford-Binet) are exposed for
five seconds and the subject is asked to copy them. Bach design is
- 37 -
given 1, £, or 0 credit, and a total score is obtained. It was found
that children with I.Q.'s of 85 or over and with known organic condi
tions frequently obtained low scores, leading the investigators to con
clude:
Extreme deviations from the norm in the visuo-motor

field seems to be a psychological entity of real sig
nificance for child guidance. In such cases of prob
able mild brain damage the usual neurological signs are
frequently absent. Consequently the psychological find
ings must be looked upon as chief aid to diagnosis.
(quoted from the original article by Lord and wood in
Bell (16), p. 346).
Wolff (98) describes a technique which he calls "graphometry"
in which subjects are required to draw six simple geometrical forms in
three sessions at three-day intervals under the following conditions:
eyes closed with right hand, eyes closed with left hand, eyes open with
right hand, eyes open with left hand. When asked in subsequent sessions
to alter the proportions of the figures which they had drawn, over half
of the subjects were unable to do so. Wolff implies that such consist
ency in graphic expression is related to personality variables and may
have diagnostic possibilities.
Appropriately classified as a visual-motor test but differing
from those previously described in that the copying of visual forms is
not required is the mirror drawing technique of Wechsler and Hartogs (91)•
Subjects are asked to join points successively by mirror drawing and
later by blind drawing (with a screen interposed between the subject and
the mirror). The performance is scored for several factors (time, dis
tance covered, segmentation, etc.), and suggestive data on a number of
subjects tend to indicate that disruption in performance is related to
the presence and possibly the degree of anxiety in the subject.

CHAPTER II
STATEMENT OF THE PROBLEM
While Bender was primarily interested in the Gestalt test as
an instrument for the exploration of the gestalt function, with emphasis
upon its maturational and pathological aspects, Hutt conceived of the
test as a projective personality technique, rich in interpretive sig
nificance and psychodynamic implications and capable of contributing
valuable clues to differential diagnosis. Clinical psychologists have
been quick to follow Hutt's lead rather than Bender's more prosaic
approach because they have long sought for a projective device less
time-consuming than the Rorschach or TAT which may be employed as a
brief supplement to the longer techniques in yielding new or confirma
tory personality data. There was a great practical need for a short
diagnostic test in the army mental hygiene clinic or neuropsychiatric
section of a station or general hospital, and Hutt's students were
therefore not reluctant to try out the Bender-Gastalt upon returning
to their own psychological installations. In a number of instances,
Hutt's manual was followed slavishly and uncritically, presumably on
the unverbalized assumption that the validity and reliability of the
test had been unequivocally established. The word, "tentative", in
the title of Hutt's guide was ostensibly forgotten or ignored, and un
warranted generalization took the place of cautious hypothesis. Deal
ing as they were with an essentially deviate population, there was

- 39 -
little danger of creating a high proportion of false positives by find
ing evidences on the test of anxiety, inferiority, insecurity, sexual
conflict, aggressive over-compensation, and emotional instability. Up
on separation from the service, many of the army psychologists incor
porated the test into diagnostic batteries in use at mental hygiene
units, mental hospitals, institutions for the feebleminded, and voca
tional guidance agencies. Before long, the Bender-Gestalt had come
into widespread use, despite the paucity of evidence regarding the
test's validity.
, There are many reasons for the test's increasing popularity.
The Bender-Gestalt taps two major aspects of the personality, the per
ceptual and the motor, can be used effectively with young children
and senile adults, and is fundamentally an emotionally neutral, in
nocuous, non-disrupting task which arouses little or no resistance on
the part of the subject. It acts as a buffer in establishing rapport
in the testing situation or as a "breather" between tests which are
more difficult or ego-involved. Failure is less apparent than on many
other graphic tests, since even patients with severe mental disorders
are usually able to respond in some fashion, thereby eliminating the
painful and embarrassing "I don't know" or "I can't" response. Woltmann
(102) recommends its use prior to the administration of the Draw-a-Man-
and-Woman test in order to reassure the negativlstic or reluctant subject
who says that he cannot draw. It possesses the advantages of short ad
ministration time and non-verbal content, which makes it widely appli
cable, regardless of language handicap or educational limitations.
Moreover, it has the advantage of being a "concealed" or indirect

- AO -
personality measure, as opposed to the usual personality question
naire, the answers to which may easily be "faked" by the average subject
who does not wish to reveal his shortcomings. It is in essence a pro
jective technique of the "partially structured" variety in which the
subject structures the stimulus situation in terms of his ego needs
and situational stresses. Hutt (55) > writing on the use of projective
methods in army medical installations, comments as follows regarding
the Bender-Gestalt:
Based as it is on fundamental gestalten. varying

somewhat in level and complexity, and requiring the
integration of visual perception with kinesthetic
control, this procedure offers at the purely non
verbal level data which are germane to another layer
of personality assayal. In addition to the obvious
distortions of the perceptual reality and to the
elaborations so characteristic of the schizophrenic,
it offers incisive evidence of obsessive or compul
sive trends of the neurotic, the individual suffer
ing from anxiety, severe reductions in size, confu
sion and retrogression of the organic brain-injtired
and the like. Observations of the general methods
of work, use of space relationships, meticulousness,
motor control, effort syndrome and the likes are al
so made possible. The writer has found this test
very valuable in military practice not only along
the lines indicated, but in differentiating emo
tional regression from the effects of organic brain
injury and in delineating special and primary fea
tures of the psychoneurotic. Improvement of the
patient is attended by improvement in the form and
quality of the gestalt reproductions.
This test presents the outstanding advantages of

simplicity of administration (especially in time),
simplicity of materials, and above all is valuable
in furnishing data on the integrative visual-motor
response pattern which is so often disturbed in the
so-called war neuroses and the traumatic and biol
ogical reactions to damage of the central nervous
system.
Surely a simple test that can accomplish all these psychological

- u -
feats must claim the attention of the enterprising clinical psychol
ogist! Hutt's claims are clearly more extravagant than those of
Bender, who was content to study the development, arrest, retardation,
and loss of function of visual-motor behavior in children and deviate
adults, but it is precisely because Hutt's approach offers greater
promise than it has achieved the greater acceptance. Moreover, Ben
der's monograph (20) is of rather limited usefulness to the clinician
because of its emphasis on extreme deviations which are the exception,
rather than the rule in practical clinical work. At times her writ
ings take on a diffuse, almost esoteric quality which make her pre
sentation difficult to follow. The specificity, systematic organiza
tion, control data, and operational definition of terms and concepts
which one is accustomed to see in the reporting of psychological re
search are largely conspicuous by their absence. The reader is gen
erally left in the dark concerning the actual number of cases upon
which she bases her conclusions, and quantitative data are almost en
tirely lacking. One gets the impression that there is a Bender-Gestalt
record "typical" of the schizophrenic or "typical" of the aphasic and
is then puzzled when he runs across a record in the clinic which fails
to fit the "classical" picture. The following statement, previously
quoted in another connection, is characteristic (20):
Gestalt drawings of typical schizophrenic patients

are easily recognized by the bizarreness of the con
figurations produced, the frequent splitting in
figures, not in such a way as to make a "gute ges
talt", and especially by showing an unusual cohesion
between all the figures and an increase in movement
in the figures, (pp. 169-170).
- A2-
Anatasi and Foley's (6) comment on Bender's work in their review of
experimental investigations of graphic productions in the abnormal is
also relevant:
It appears from these descriptions that there is

considerable overlapping in the performance of the
various clinical groups, (p. 204)*
Later in the same article, they voice two criticisms of experimental
work in the study of artistic productions which might readily be ap
plied to Bender's contributions:
Quantitative results are relatively rare, in many

studies only the author's conclusion being given,
with illustrative cases....Interpretations are
often made in terms of many Gestalt or psychoanal
ytic terms which are relatively vague, ill-defined,
complex, and mentalistic. (pp. 231-232).
But if Bender's articles tend to ignore individual psychody
namics and are insufficiently detailed to aid the clinician in prac
tical work with the test, the criticism might be leveled against Hutt
that he offers "too much too soon", that claims are made for the test
which have yet to be validated, and that he might have been wiser to
have delayed publication until more acceptable empirical data had
been made available. Although he states that the "clinical syndromes"
for the psychoneurotic, the schizophrenic, and the brain injured
patients are "substantiated by the author's research studies", he has
never published his original data in quantitative form, either by way
of reporting validity coefficients or of indicating the proportion of
false positives or false negatives to be anticipated. It is not
enough for him to say that his method of interpretation "works" well;
one would like to know "how well". Admittedly, his syndrome for the
neurotic is based on 130 "carefully selected psychoneurotic cases" but
only 40 controls, with no indication that the two groups were equated
for factors that might influence performance on the test, and he fails
to state the size of the samples employed for his "research" on schizo
phrenics and organics. Moreover, many of the "determinants", or scor
ing deviations, are inadequately defined, resulting in varying inter
pretations of them by clinicians who employ the same terms but assign
them different meanings. No actual illustrations of the form modica-
tions are given in the manual, so that one is hard pressed to ascer
tain whether an "angulation" or "closure" difficulty deviates suffi
ciently from the normal to be regarded as pathognomonic or how "ir
regular" the wave length in figure 6 must be before one may justifiably
posit instability in the subject.
Both Hutt and Bender are original, creative thinkers who are
not afraid to pioneer in unexplored psychological territory and who
occasionally are equally fearless in making the "inductive leap" from
data to conclusion while treading on uncertain ground. It may well
be that they are capable of drawing inferences from the test which
provide valuable insights into the patient with whom they are work
ing, but it may also be that they are relying heavily upon minimal
cues based upon long experience with the test which are generally
labeled as "clinical intuition" and which may be less readily commu
nicable to others. The fact of the matter is that the vast majority
of clinical psychologists using the test interpret it in a more or
less rough, subjective, global manner, drawing psychodynamic implica
tions on the basis of essentially unsystematized clinical observation.

- a -
If the test is to have real value to the clinician and if it is to be
given univocal interpretation by those who use it as a diagnostic or
prognostic instrument, both Bender’s and Hutt's provocative ideas
should be stated as hypotheses to be subjected to rigorous experi
mental or clinical standardization and validation.
The need for quantification and validation of all projective
tests has been recognized with increasing frequency in recent years,
despite the objections of some clinicians who seem to feel that any
attempt to apply statistics to projective devices would automatically
destroy their usefulness as instruments for assessing the "global"
personality. For them, there is no substitute for clinical experi
ence and insight, and any attempt at quantitative evaluation of the
elements or components of a test is regarded as "atomistic". It is
the "total configuration" which is important, since a scoring factor
cannot be considered in isolation but must be interpreted in the light
of its context. This point of view is epitomized in the following
quotation from Brown and Rapaport (29):
Usual standardization procedures of mental tests

have not been successfully applied to these (projec
tive) methods, as it is the configuration of factors
present rather than the independent quantity of each
factor that describes the personality. Broad experi
ence of the psychologist rather than statistically
reliable norms is the necessary prerequisite for
using these procedures, (p. 76).
White (94) recognizes the fact that a quantitative approach
may present problems to the research worker but believes that such an
approach is advisable:
From the nature of the material sought and from the

unavoidable indirectness of the seeking it is ob
vious that the interpretation of imaginative produc
tions will offer great scientific difficulties....
- 45 -
intuition can seize upon clues, but it is no easy

matter to establish reliability, observer agreement,
and validity, (p. 215).
It can be seen from this review that the psychol

ogical study of drawing and painting suffers from
no lack of ingenuity or stimulating ideas. The cry
ing need is for validation, (p. 24-6).
Bell (16), reviewing visual-motor tests, expresses himself in
a similar vein:
Their weakness, by no means insuperable, is in the

limited evidence of validity and reliability as yet
accumulated, (p. 348).
Sargent (81), in her comprehensive review of the origins,
theory, and application of projective methods in personality research,
also makes a strong case for quantification:
Quantitative method might profitably be applied

more extensively to the properties of the projec
tive tests themselves. For example, as Harrison
suggests, it would be valuable to have frequency
tables similar to those for the Kent Rosanoff word
association test, showing the relative frequency
of certain common phantasies produced by each of
the TAT pictures. If we think of these numerical
results as adding to the precision of the instru
ment itself, instead of reading into the figures
over-simplified generalizations about people, no
atomistic conception of personality is implied.
The clinician would not need to alter either his
theory or his interpretation of certain Rorschach
syndromes if he also knew more about the frequency
of the component determinants, both singly and in
constellations. Buch knowledge would, on the con
trary, serve both as an added support and as a
check on his conclusions, (pp. 280-281).
She goes on to point out that many of the scoring elements of the Ror
schach (e.g., color, shading, movement, common details, form, etc.)
have been intensively studied by Beck (14), Hertz (50, 51)» Hertzman
(52), Kisker (58), andothers, and it might also be noted that

- 46 -
Harrower-Erickson (48), Piotrowski (78, 79), and Monroe (68, 69, 70)
have used the "sign approach" with good result.
Cronbach (34) argues in effect that such studiesas the above
represent only a beginning, and that much remains tobe done along
these lines:
There ha3 been no systematic validation of the test

(Rorschach), score bv score or trait by trait. Ac
cumulated blind descriptions show that on the whole
the test does much better than chance; there is less
evidence that each specific determinant, such as tex
ture, is correctly interpreted. Until this evidence
is provided, probably the test will lead tosome in
valid inferences from scores, (p. 442).
And Gardner Murphy (71), who can hardly be accused of espous
ing an "atomistic" psychology of personality, emphatically remarks:
The impression has gotten abroad that there is an

antithesis between personality measurement and an
approach in terms of structure. Yet measurement
supports rather than negates emphasis upon whole
ness. (p. 668).
Commenting upon the commonly used technique of matching a blind inter
pretation of a projective test against a case history or independent
personality description as evidence of validity, Murphy cogently ob
serves:
Matching is done as a whole; hence it is impossible

to determine, so far as features can be isolated,
which ones lead to a sound judgment and which are
misleading. If it is maintained that the interde
pendent aspects of the method forbid any analysis
of separate components, it is hard to see how it
can be further developed....there is nothing to pre
vent rapid progress in the detection of good and
poor leads, and the development of weighting tech
niques to make the best use of the leads that prove
to be good. (p. 668).
The present investigator became interested in the Bender-Gestalt

- 47 -
when he attended Hutt's classes In clinical psychology at the Adjutant
General'3 School. Like many of his classmates, he was duly impressed
by the claims that were made for the test, and his enthusiasm grew as
a result of his subsequent supervised training at Mason General Hospi
tal, where the test was accepted as a useful diagnostic tool. On re
turning to his assignment as chief clinical psychologist of the neuro-
psychiatric section of the station hospital at Fort Bragg, North
Carolina, he added the Bender-Gestalt to the existing test battery,
which consisted of the Wechsler-Bellevue Intelligence Scale, Rorschach,
TAT, Man-and-Woman Drawings, Shipley-Hartford Scale, Minnesota Multi-
phasic Personality Inventory, and a sentence completion test and a
personality questionnaire, both of which were constructed by the author.
Many of the schizophrenics rotated the figures, and many of the neurotic
patients manifested angulation difficulties and tremulous lines, and in
general deviations from the stimulus figures were clearly discernible
in a high proportion of the cases. Generally speaking, moreover, the
deviations were greater in the psychotics than in the neurotics. No
controls were available, but it was assumed that normal subjects would
reproduce the figures essentially like the originals.
Separated from the service in February, 1946, the writer ob
tained a position a month later as senior clinical psychologist at the
New York University Testing and Advisement Center and began to adminis
ter the test routinely to both veteran and non-veteran clients seeking
educational or vocational guidance. Much to his surprise, he observed
many test records from presumably "normal" individuals that resembled
those which had been previously obtained on neurotic patients. The

- 48 -
disturbing questions began to present themselves with increasing in
sistence: Is it possible that most subjects, whether normal or ab
normal, will show "abnormal" deviations on the Bender-Gestalt which
will classify them as neurotic? Is Hutt's neurotic syndrome actually
a pattern which is characteristic of both neurotics and normals,
though differing from that of schizophrenics and organics? Is the
"success" of the test an artifact resulting from the fact that most
clinicians deal with abnormal people, whom the test will properly
characterize as deviant? Under such circumstances, false negatives are
unlikely and false positives cannot occur, thereby creating a spurious
impression of validity.
These questions assume even greater significance when it is
recalled that Bender was unsuccessful in finding any distinctive
Bender-Gestalt test performance for her neurotic patients. Writing
as late as 194-9, after she had had an opportunity to familiarize her
selfwith the military and the civilian use of her test, Bender(24)
concludes:
In my own studies I did not expect to find disturb

ances in perception or in the visual motor gestalt
function in the psychoneuroses. However, it was
often possible to evaluate severe anxiety and com
pulsive obsessional features, especially when they
occurred as defense mechanisms in individuals with
organic disorders or psychoses which threatened the
integrity of the organism.
During the war, the test was used in a number of

army installation centers to aid in differentiating
between emotional regression and the effects of or
ganic brain injury or schizophrenia. It was also
used in an effort to explore personality dynamics,
but the results of this research are still incon
clusive. (p. 170).
- 49 -
woltmann (102), who has worked in close collaboration with
Bender and who has used the test clinically since its inception, re
affirms Bender's position in his review of the literature:
Organic conditions and psychotic disturbances usual

ly produce a telltale pattern with obvious deviations
in the copying of the Gestalt figures. let it hap
pens very often that a clearly established diagnosis
of psychoneurosis from the test battery is accompanied
by very normal copies of the Gestalt figures. This
does not render the Gestalt test invalid, because
personality disturbances in the neurotic personality
seldom invade the visual-motor sphere, (p. 348).
He goes on to point out that observation of such features of test be
havior as questioning, need for reassurance, requests for help, and
carelessness may give the clinician insights into the subject's per
sonality, but adds that "these and innumerable other forms of test be
havior are found in every type of normal and abnormal personality pat
terns." (p. 349).
The general purpose of the present investigation is to deter
mine whether neurotics and normals can, in fact, be distinguished on
the basis of their Bender-Gestalt records. Most clinical psychologists
are agreed that psychotics and organics show gross deviations on the
test which are sufficiently bizarre or unusual to identify such cases
fairly readily without the necessity of developing an elaborate scor
ing system, but they disagree on the applicability of the test to
patients suffering from neurotic conditions. Before inquiring into
the validity of the psychodynamic interpretations of the individual
scoring factors as applied to neurotics, it would appear that the
logically prior step would be to ascertain whether the scoring factors
themselves differentiate between neurotic and normal subjects. If a

- 50 -
sign alleged to reveal anxiety or emotional immaturity occurs with
equal frequency in both groups, one might well question the validity
of such an interpretation. The specific aims of this investigation,
then, are to ascertain the discriminating power of the various scoring
elements, to isolate any which exhibit differential validity as re
gards normals and neurotics, and, if possible, to combine these dis
criminating factors into a total score which will effectively separ
ate the contrasting clinical groups. To accomplish these aims, it
is proposed to develop an objective scoring system, defining the scor
ing elements operationally rather than conceptually or interpretively,
and to subject the observed deviations of the reproductions from the
stimulus figures to a statistical appraisal similar to that employed
in item analysis techniques.
It should be emphasized that this is a study of empirical
validity; the effort is not to identify psychological processes or
psychodynamic correlates but rather to define the test's uses and
limitations in practical terms. The approach is similar to Zubin's
(104, 105, 106) attempt to revise the Rorschach scoring system to per
mit more exact quantification. Although what he (106) has to say con
cerns the Rorschach, it so clearly reflects the point of view of the
present study that it is quoted at length:
The primary emphasis in this new approach was psy-

chometrically rather than personalistically oriented,
the purpose being to present an objective frame of
reference for classifying Rorschach responses, rather
than a system for differentiating personality organi
zations of different individuals, (p. 277).
- 51 -
In order to make the Rorschach into a functional

test in this area (differential diagnosis), we need
to quantify or categorize the trends that have been
delineated in Rorschach responses and then compare
two contrasted groups or diagnostic categories on
the basis of these trends....After the differential
patterns characterizing many contrasted groups have
been developed, we may find a connecting link in
some hypothetical framework, but the proof of the
utility of the test will depend not on the elegance
of our theory, but on the precision of our differ
entiating and prediction, (p. 279).
Were it not for two important developments that have taken
place since 1947— the date of the inception of the present study— this
section of the thesis would now be concluded. In the past four years,
however, two serious attempts have been made to objectify and quantify
the Bender-Gestalt with almost completely antithetical findings— an
outcome which brings the present investigation into even sharper focus
and enhances the significance of its conclusions. The studies are
those of Billingslea (26), which appeared in January, 1948 and Pascal
and Suttell (76), published in June, 1951*
Billingslea (26) administered the Bender-Gestalt to 100 psy
choneurotic adult male patients, all soldiers at an Army hospital, and
to 50 normal adult male soldiers. The diagnoses for the neurotic group
were based on the judgments of four psychiatrists and Billingslea fol
lowing clinical interviews, examination of medical histories, and evalu
ation of the results of a battery of psychological tests. The control
group was judged to be "normal11 on the basis of having "no civilian or
Army history of crippling emotional disorders". It is stated that the
two groups were of "comparable age, intelligence, civilian and Army
occupational experience, Army combat experience, and home town populations, '
- 52 -
but no figures are given by way of substantiation. Since the standard
test cards were not then available, the cards distributed by Hutt were
vised, despite the fact that they admittedly contained minor inaccura
cies.
Billingslea developed an objective scoring system requiring
a special measuring scale consisting of a fine grid, a linear scale,
and a protractor, but its very precision makes it impractical for
clinical use, as attested to by the fact that it took him fifteen
hours to score a single record. The papers were originally scored
for 38 factors and 137 indices, but the scoring was so time-consuming
that he eventually scored all papers for 25 factors and 63 indices. A
factor is defined as a "certain rather clearly detectable test behav
ior observable in one or more of the figures", while indices "involve
the measurement on only one figure of such things as lengths of lines,
angles, areas, irregularities in shape, and rotation of the whole
figure or parts of the figure".
Billingslea rejected the test-retest method for obtaining
reliability on the grounds of contamination by the memory factor for a
short test-retest interval and of fluctuations in personality organiza
tion for a longer interval. He therefore attempted to utilize the
split-half technique by correlating added scores for indices from the
first five figures against those from similar indices from the last
four figures. The obtained coefficients were disappointingly low, forc
ing him to conclude that "the instrument is either unreliable as far
as the factors measured are concerned, or that the factors measured are
not consistently expressed by the individual with behavior patterns as

- 53 -
judged".
The findings on validity are equally disappointing. Billings
lea reports median scores for each index for the neurotic and normal
groups rather than the proportion of each group exhibiting each devia
tion making his data somewhat difficult to grasp, but in essence he
finds only 6 indices with median difference critical ratios of 3.00 or
greater (closure on figure A, shape distortion on figure 3, point angu
lation on figure 8, regression, sketching, and counting), noting that
in general "these do not form a particularly interrelated pattern and
certainly not a consistent one". When size differences are considered
as a factor rather than as an index, it is observed that "the trend is
for the psychoneurotic subjects to produce smaller figures than the
normal groups, especially in the cases of figures 1, 2, 5, and 6".
Comparing his results point by point with those of Hutt, he reports
the following areas of agreement: (1) non-occurrence of the distor
tion of the figure beyond recognition, (2) rarity of perseveration
and fragmentation in either group, (3) reduction in size of the figures
for the neurotic group, (4) more frequent occurrence of "curvature
values" among the neurotics, especially decrease in the size of the
sinusoidal waves in figure 6, (3) greater verbal resistance among the
neurotics, and (6) more frequent "indices indicative of sexual dis
turbance" in the neurotic group. Contrary to Hutt: (l) total and
part rotations occur frequently in both groups, (2) difficulty with
crossings is not present in either group, (3) regression occurs fre
quently in both groups, although the number of instances per record
is greater in the neurotic group, and (4) neither sloping angulation

- 54 -
nor reversal, is distinctive of neurosis.
Billingslea therefore concludes that his results support
those of Bender in demonstrating that "a clear syndrome for distinguish
ing the psychoneurotic test record cannot be established", but adds:
"There is nothing in the above findings that detracts from the use of
the instrument as a tool for intuitive observations of the client's
clinical behavior".
There are a number of criticisms that may be leveled against
Billingslea's study. (1) The size of his control sample is relatively
small, limited as it is to only 30 subjects. (2) Standard test cards
are not used, although it is doubtful that this factor is a major
source of error. (3) The split-half technique for determining reli
ability is technically unjustified, since this method gives mislead
ing results when the two halves of the test are not as equivalent as
parallel forms of the same test would be. The individual figures vary
in difficulty and in the particular problems they present to the sub
ject, as Bender (20) has shown. Sargent (81) has made the same point
with respect to the Rorschach: "...split-half correlations of Ror
schach factors mean little because the ten cards are admittedly un
even in the type and amount of response they produce, and because the
technique involves isolating factors from context." (p. 275)* (4)
No pictorial illustrations of the scoring factorB are given, although
it is evident to anyone trying to use the test that verbal descrip
tions by themselves convey different meanings to different examiners.
(5) The scoring system is too precise, finicky, cumbersome, and time-
consuming to have any practical utility. It may be that real differ
ences are obscured by the over-aeticulousness of the approach. An

- 55 -
example Is the high incidence of rotation in both the normal and
neurotic groups— a manifestation which Bender regards as "an acute
confusional feature", fairly frequent in psychotic8 and organics but
unusual in normal adults. Billingslea*s comment is typical: "It is
recognized that many of these subjects obtained scores that indicated
rotations of the figures to only ten degrees in either direction.
However, rotation, as defined, includes any declination from the nor
mal axis, and such behavior has been considered perceptually meaning
ful by proponents of the test." (p. 14). (6) Billingslea*s conclu
sion to the effect that "with our present knowledge, it is not possible
to establish a scoring system by which an insufficiently clinically
trained person might utilize the instrument" (p. 17) is a non seauitur.
The fact that his system is non-discriminating does not preclude the
development of a less involved and less rigid system which may be di-
agnostically meaningful. (7) Billingslea*s emphasis is upon considera
tion of the scoring factors in isolation rather than in combination.
It is true that he asserts that "when individual indices, representing
several different factors, are combined, the combination often demon
strates a stronger trend towards validity than do the individual in
dices" (pp. 18-19), but nowhere does he report any quantitative data
on such a combination.
A more recent attempt at quantification and validation has
been made by Pascal. Pascal was a classmate of the present investiga
tor's at the officer clinical psychology course at the Adjutant General's
School, and both students became interested in the Bender-Gestalt at
the same time, pursuing their investigations independently upon returning

- 56 -
to civilian life. He (75) published a preliminary report of his work
in 1949, but the details of his investigation appeared in book form
in 1951, written in collaboration with Barbara Suttell (76). They
developed a scoring system dependent upon inspection rather than the
precise measurement characteristic of Billingslea's approach and tried
out nearly 200 scoring deviations on normal and abnormal individuals,
retaining 105 on the basis of item analysis. Actual count indicates
that there are really about 40 different scoring elements which are
said to be discriminating, the discrepancy between the reported 105
and the actual 40 deviations being attributable to the fact that the
same deviation is counted as a separate unit each time it appears in
a figure. "Rotation", for example, is scored for all 8 figures
(figure A is not scored) and is regarded as 8 "signs" rather than 1.
Pascal differentiates his criterion groups on the basis of
whether they are psychiatric inpatients (primarily psychotic), psy
chiatric outpatients (primarily neurotic), or nonpatients (normal).
He feels that the patient-nonpatient breakdown is a more reliable,
though cruder criterion than specific diagnostic categories and adds
that he has deliberately attempted to avoid the problem of psychiatric
diagnoses. "Of chief importance, however, for our validity study," he
points out, "is the reliable fact that all our patients were patients
of psychiatrists, and it is upon this dichotongr, patient versus non
patient, that we primarily base our tests of validity." (p. 28).
Frequencies, weights, and phi coefficients are published in
the appendix (pp. 97-99) for each of the retained scoring items on
the basis of administration of the test to 260 nonpatients and 260

- 57 -
patients (110 of whoa were psychotics and 150 neurotics) matched for
age, education, and sex. Phi coefficients and weights were calculated
from abacs appearing in Guilford (43)• The item weights range from
1 to 8, depending upon their discriminating power. A scoring devia
tion which never appears in the nonpatient population but which does
appear 1 or more times in the patient population is given the max-timim
weight of 8. Differences significant at the 1$ level are reported
for each design in terms of the total number of deviations for that
figure exhibited by the two criterion groups, (p. 14).
Following the item analysis, the test was standardized on
474 adults (281 males and 193 females) from the Providence area rang
ing in age from 15 to 50 and in education from the 9th grade through
college. The normative data are admittedly weak with respect to the
age group, 40-50, and are not applicable to adults with only grammar
school education. Moreover, the raw scores of the high school group
are significantly higher (i.e., more unfavorable) than those of the
college subjects, resulting in a bimodal curve when the total distri
bution of raw scores is plotted. This finding is all the more surpris
ing in view of the fact that there is no significant correlation be
tween raw score and Otis l.Q. Since the raw scores are uncorrelated
with sex differences and with age (within the limits 15-50), the test
scores of male and female subjects are combined regardless of age,
but the raw scores of the high school and college samples are converted
separately into Z scores with a mean of 50 and a standard deviation of
10. The obtained raw score is transmuted into a Z score by the examiner
by reference to conversion tables which appear in the appendix (pp. 100-
101).
- 58 -
Validity data are reported on 323 psychiatric patients (187
neurotics, 136 psychotic) ranging in age from 15 to 50 and in educa
tion from high school to college. Compared with an average Z score
of 50 for the standardization population, the mean Z score of the
total patient population is 73*6, the mean scores of the psychotics
and neurotics being 81.1 and 68.2, respectively. Biserial correlation
coefficients are reported as follows: total patient vs. total non
patient population, .74; nonpatients vs. neurotics, .73; nonpatients
versus psychotics, .91; and neurotics vs. psychotics, .35- Discussing
the use of the test as a screening device, the authors note that a Z
score of 60 would screen out 82% of the total patient population (86%
psychotic and 79% neurotic) at a cost of only 20% of the nonpatient
group. Data are also presented to show that patients with low scores
on admission are more likely to be discharged as "improved" than pa
tients who obtain high scores.
About 100 pages of the book consist of a scoring manual in
which each of the scored deviations are defined and illustrated. Un
like the papers of Hutt and Billingslea, pictorial illustrations are
abundant, and the scoring instructions are sufficiently clear to make
the method communicable. As a supplement to the instructions, 45 test
records are presented and scored, providing the student with an oppor
tunity to check his scoring against that of the authors. Inter-scorer
reliability is high, according to Pascal and Suttell, who state that
a coefficient of .90 represents a fair estimate of the degree of agree
ment that can be attained between different examiners with practice.
They report that the two authors, scoring the records of 120 subjects
- 59 -
independently, obtained essentially similar biserial correlation co
efficients between normal and psychotic, normal and neurotic, and
neurotic and psychotic subjects, although there was a mean difference
in score between the two examiners of approximately four points. They
caution, however, that they "do not claim our scoring method to be en
tirely objective. It depends, as a matter of fact, a good deal upon
judgment; but judgment, which we have found can easily be selftaught
with the aid of the manual provided." (p. 12). Subjective certainty
is emphasized; when in doubt the examiner is urged not to score for
that particular deviation.
Test reliability is less satisfactory, however. The authors
tried out several odd-even splits, obtaining reliability coefficients
of about .50. They reject this method (which was first employed by
Billingslea) on the grounds that the subjects react differently to
each design and regard the test-retest technique as a superior method,
though not without its limitations. Administering the retest 24 hours
after the initial test to 44 subjects, they report a reliability co
efficient of .71. Retesting 23 normal subjects after 18 months, they
find the reliability to be .63. Practice effect is said to be negli
gible when mean scores for test-retest populations are compared, al
though some individual subjects show considerable variation. Pascal
and Suttell comment that "as users of the te6t we have not been pri
marily concerned with reliability as long as validity has been main
tained" (p. 17), and in an earlier article Pascal (75) argues:
These low coefficients in the fact of the higher

validity coefficients attest to the difficulty of
estimating the reliability of a projective technique;
- 60 -
for each individual tends to react differently to

each figure. That is, a high score in one figure
does not necessarily imply a high score in another.
It would seem that test-retest reliability would
give the best estimate of the test reliability but
this is complicated by the fact that people may
change in their approach to the test if the inter
val between the test and retest is too long. The
assumption upon which this statement is based is
that if the test is an estimate of adjustment then
adjustment may change over a period of time. It
would therefore seem that the best estimate of the
test reliability at the present time is the validi
ty coefficient.
Certain criticisms may justifiably be raised concerning Pas
cal and Suttell's investigation. (1) It is reported that nearly 200
deviations were tried but that only 105 were retained. It would have
been helpful to users of the test if data on all deviations had been
presented, so that they might check their own conclusions on scoring
elements which they may have been interpreting freely without statis
tical confirmation. (2) No attempt is made to relate the findings to
those of Hutt or Billingslea, both of whom have made challenging con
tributions to the same area of research. (3) It is assumed that re
productions made by psychiatric patients tend to "deviate more from
the stimuli than those of nonpatients, and the more psychologically
disturbed the patient the greater the number of deviations" (p. 5)«
It is true that all the deviations which the authors regard as scor-
able discriminate in the same direction, i.e., they occur with greater
frequency in the abnormal groups, but it is conceivable and, as a mat
ter of fact, highly probable, that some of the deviations may be more
typical of normal subjects, and one wonders why no such deviations
were found. The presence of a "normal" sign can be as significant as

- 61 -
the absence of an "abnormal" sign in connoting adequate emotional ad
justment. (4) In the item analysis tables, frequencies are given only
for the nonpatient and patient groups, i.e., the neurotic and psychotic
subjects are Imped together and treated as one criterion group. It
would have been instructive to the clinical psychologist if separate
frequencies had been reported for each scoring item for normals, neurot
ics, and psychotics, so that the relative discriminating power of the
item might be directly ascertained. Are there, in short, any signs
which differentiate normals from psychotics but which fail to distin
guish between normals and neurotics, and do any signs exist which
clearly discriminate between the neurotic and psychotic patients?
(5) The weighting system, particularly as it applies to zero frequen
cies, is questionable. Although most of the 105 scoring items have
modest weights (values of 2, 3, or 4), 38 items are given the maximum
weight of 8. One might expect that such items would have exceptional
ly high discriminating power, but the fact of the matter is that 31
of the 38 items are given the maximumweight merely because they oc
cur one or more times in the patient population and never occur in
the nonpatient group. Out of a total of 260 nonpatients and 260 pa
tients, for example, the following signs have zero frequencies in the
control group and frequencies of 2 or less in the patient group: ro
tation (figure 1), double row (figure 1), figure on 2 lines (figure 2),
second attempt (figures 2 and 3), distortion (figures 3, 5, and 6),
guide lines (figures 4 and 6), and design missing (figures 4 and 5).
No data are given for another 11 signs which are assigned maximum
weights. The argument that such minor differences may be ascribed to

- 62 -
accidents of sampling is advanced by Anastasi and Foley (8) who, in
a previously cited study, treated zero frequencies in their 92-item
check list for evaluating the drawings of 340 normals and 340insti
tutionalized patients quite differently:
Those items having consistently zero frequencies

in the normal group, furthermore, had very low fre
quencies (often two or three cases) in the abnormal
group. Because of the rare occurrence of such
characteristics the zero frequencies in the normal
group might be attributable to sampling error.
Since, therefore, their interpretation is doubtful,
no special consideration will be given to those
characteristics appearing only in the abnormal
drawings, (p. 188).
(6) Many of the items said to be discriminating actually lack validi
ty. The present investigator, converting the item frequencies for the
patient and nonpatient groups into proportions and applying the for
mula for the reliability of the difference between two proportions,^
found that about 20 of the 94 signs for which data are given are not
significant at the 5% level and that approximately another 20 do not
meet the 1% level of significance. It would appear that many of the
items in Pascal's scoring system constitute "dead weight" which un
necessarily complicate the task of the scorer. (7) Figure A is not
considered in the scoring of the test, but no reason is given for the
exclusion. In view of the presentation of 10 or more "significant”
deviations for each of the remaining eight figures, is it possible
that figure A is non-discriminating? (8) Pascal and Suttell repeated
ly refer to biserial correlation coefficients as evidence for the
validity of their scoring system, despite the fact that use of the
biserial r is statistically unwarranted for their dichotomous groups
1. The nomograph by Lawshe and Baker (60) simplified the computational

labor by permitting rapid graphical solutions of the differences
between proportions.
- 63 -
and gives the impression of an inflated validity which the test does
not actually possess. The point is made by Ellis and Conrad (37) in
their review of the use of personality tests in military practice:
In several of the studies involving biserial co

efficients of correlation, the coefficients were
calculated on the basis of an almost equal number
of "normals" and abnormals. Actually, since "nor
mals" are generally far more numerous than abnor
mals, biserial correlation coefficients should be
calculated only on groupings which approximate the
"normal"-abnormal ratio....Biserial coefficients
computed for such samples are spuriously high.
(pp. 404-405).
That Pascal and Suttell found differences between normals
and psychotics is not surprising, in view of the fact that institution
alized psychotics tend to differ significantly from normals on most
psychological tests, although the results would have been more mean
ingful if the authors had indicated the composition of their psychot
ic population; while they state that all psychotics used in the study
suffered from psychogenic disorders, they give no indication of the
proportion of schizophrenics and manic-depressives composing the sam
ple. Their finding that neurotics can be reliably differentiated from
normals on the basis of their scoring system is, however, of greater
interest, since it focuses upon the central problem of the present
study. Bender (20), Billingslea (26), and Woltmann (102) obtain nega
tive results with respect to neurotics, while Hutt (56) and Pascal
and Suttell (76) obtain strongly positive findings. As previously
noted, the present investigation was conducted completely indepently
of Pascal's study, which was unknown to the writer until its publica
tion in June, 1951, but Pascal's work does serve the very useful pur
pose of bringing the findings of the present research into bold relief.
- 64 -
Will the results, arrived at by an independent scoring method, resolve
the contradiction of previous studies and dissipate the bewilderment
of the clinician, who is told on the one hand that the test is of
limited value with neurotics and on the other that it possesses high
validity with that group, or will the findings be so ambiguous in
character as to defy definitive generalization?

CHAPTER 111
PROCEDURE
A* A t y— t Materials
The stlnmlus figures consist of nlno gecnetrical designs
originally used by Wertheimer (93) in hia classical investigation of
gootalt principles. They are reproduced below in Figure 1, which is
taken fron plates appearing in Bender's Monograph (20, p. 4) and Manu
al (22, p. 4).
Figure !• The Bender-Gestalt Test Figures

- 66 -
Figure A consists of a circle and a square oriented on its
vertex placed in contiguous relationship* According to Wertheimer
and Bender, this figure is experienced as two closed figures on a
background because each represents a "gute gestalt”, the principle of
continuity of geometrical or internal organization overruling the
principle of proximity of parts. Figure 1 consists of 12 dots the
distances between which are such that they are supposed to be per
ceived as a series of pairs, the governing principle being proximity
of parts. Actually, according to Bender, only about 25% of normal
adults perceive the pairing because of the fact that the differences
in the distances are so small. Figure 2 consists of three rows of
small circles so arranged that they are usually perceived as eleven
columns slanting from left above to right below because of the prin
ciple of proximity. Figure 3 is perceived as an arrowhead made up of
a series of obtuse angles of increasingly greater spread formed by
the arrangement of 1, 3, 5, and 7 dots, respectively. This figure
also illustrates the principle of proximity. Figure 4 illustrates
the principle of continuity, since it is generally perceived as an
open square with a bell-shaped figure tangential to the lower right-
hand comer. Figure 5, consisting of 19 dots in the semicircle and
7 dots in the tangential slanting line, is seen as an incomplete cir
cle touched by a slanting line, the principle of continuity being
prepotent over proximity. Figure 6, also illustrating continuity, is
composed of two sinusoidal lines, each consisting of 4 waves but dif
fering in wave length, crossing each other at a slant. Figures 7
and 8 consist of two configurations made up of the same units but

- 67 -
rarely perceived as such because in figure 8 continuity is the pre
vailing principle. Thus figure 7 is perceived as two hexagonal forms
with oblique overlap, while figure 8 is seen as an elongated horizon
tal hexagon in the center of which is a small diamond whose vertical
vertices touch the top and bottom of the hexagon at the center of
their longest sides.
According to Bender, all of the figures are "satisfactorily”
produced by the age of 11 years. Figures A, 1, 4, and 5 are produced
at 6 years, figure 8 at 7 years, figure 6 at 8 years, figures 2 and 7
at 10 years, and figure 3 at 11 years. Figure 2, taken from her mono
graph (p. 132) and her manual (p. 5) summarizes her data on the stand
ardization of the test with children from which she derived the develop
mental norms for the individual figures. It can be seen from this
figure that her criteria of "satisfactory” reproduction are rather
lenient because of her emphasis upon representation of the "funda
mental gestalten" rather than upon the finer nuances of more accurate
reproduction.
B. Adnrini st.ration of the Test
Although Bender did not provide any standard instructions
for administering the test until publication of her manual (22) in
1946, almost all users of the Bender-Gestalt have employed the same
basic method of administration. The stimulus cards, each of which
contains a test figure, are presented one at a time, and the subject
is asked to copy them while keeping the card in full view. Rulers or
other mechanical guides are not permitted, but otherwise the subject
- 68 -
rtfvi l rtaw* i Flfur* 1.flfur* 4 F t ( n r * 3 F lg u r* * H |m T •
q r w w mgr napr mgr

Mall.
90JC
9r*
• JTt
00
6 jr s
00OOfteo
CO
3 rn ■
Figure 2* S w a r y of Derelopnental Norms Obtained

from Bender*a Standardisation of the
Bender-Gestalt as a Performance Test
for Children
is free to proceed as he chooses* Questions about how the figures
are to be copied are usually answered in a noncommittal Banner or re
ferred back to the subject for decision* There is no tine limit, and
- 69 -
the reproductions themselves are not timed* Erasures, crossing out
of part of a figure, or making more than one attempt to reproduce a
figure are allowed. Bender and Hutt permit the subject to count the
cards or examine all of them before beginning with card A. Hutt,
Woltmann, and Pascal and Suttell frown upon card-turning but permit
it if the subject is insistent, and the same holds true of turning
the paper* All examiners except Woltmann permit the subject to use
as many sheets of paper as he wishes, Woltmann arguing that "such a
procedure overlooks and neglects the diagnostically important fact
that the final organization or distribution of the nine figures on
the same sheet constitutes an added Gestalt function" (p* 334)* Hutt,
on the other hand, feels that structuring the instructions in this
way actually contributes to the loss of valuable data concerning ar
rangement or sequence. Unlike other investigators, most of when re
gard sketching as an important diagnostic feature, Pascal and Suttell
prohibit it on the grounds that a skilled artist using this technique
can exactly reproduce the stimuli. Only Bender feels that it is well
to encourage the placing of figure A near the upper left hand corner
of the paper, other examiners preferring that the subject feel ab
solutely free with regard to placement of the figures*
In the present study three major innovations in administra
tion of the test were introduced* (1) Each reproduction was timed*
Although all previous investigators have ignored the time factor ex
cept to observe that extremely long or short times should be noted as
a qualitative datum, the present investigator felt that time might be
a significant factor in its own right and that its inclusion was as
-70 -
justifiable in connection with the Bender-Gestalt as it has been with
word association tests and the Rorschach. Furthermore, its suscep
tibility to objective measurement has much to commend it with respect
to a test for which completely objective measures are the exception
rather than the rule. (2) The administration of the initial test was
followed by a test of immediate recall. As soon as the subject had
copied all the figures, all stimulus cards were removed from sight and
he was asked to draw the figures from memory. This modification of
the basic procedure was introduced on the hypothesis that, if any dif
ferences in test performance are exhibited by normals and neurotics,
they will be accentuated in the absence of an objective stimulus. Nu
merous studies (e.g., 42, 77, 1, 39, 30, 44) on memory for fora show
pronounced alterations of the reproductions in normal adults, and it
is conceivable that certain latent tendencies, held in check during
the copying process, might reveal themselves as distinctive of neurot
ic adults under the pressure of an unexpected test of incidental
memory. Moreover, if emotional blocking is heightened under these
conditions, one would expect that the total number of figures re
called by the neurotic group would be significantly smaller than for
the control group. (3) An immediate retest followed the immediate re
call. As soon as the subject had completed his recall of the figures,
he was asked to copy the figures from the models in a procedure iden
tical with that employed for the initial test. This procedure per
mits a comparison of performance under conditions in which the test
materials and procedure are first unfamiliar and then familiar and
sheds light on the question of whether observed deviations occurring

- 71 -
during the first administration of the test persist or disappear un
der a second administration. Unlike previous studies, such as that
of Pascal and Suttell (76) in which retests were administered to a
comparatively small proportion of the total group tested merely for
purposes of estimating reliability, the retest was employed in the
present research as an integral part of the testing procedure with
implications for both validity and reliability* The retest subserves
the same function in this method of administration as it does in
finger-painting, where workers such as Kadis (57) and Napoli (72) re
quire from three to eight paintings per session, and in the Szondi
Test (35) in which repeated administrations are an essential part of
the procedure* Die importance of the retest in protective testing
is underscored by Sargent (81) who observes:
Tomkins, however, repeating the TAT daily with a

group of subjects found that 20 sessions were re
quired to bring out all the significant themas
for one person. This finding is enough to indi
cate that high reliability can hardly be expected
upon just one repetition of a test* (p* 275).
and by Harrower (47) who, referring to her own Multiple-Choice Ror
schach, notes:
The need to repeat the Multiple-Choice Test is

not only of practical importance; it also raises
the whole issue of whether or not the findings
of any projective technique are significant if
they occur at an isolated instance in time* (p* 162).
In the present study, the test was administered to the con
trol group by psychologists at the New York University Testing and Ad
visement Center and to the experimental group by clinical psychologists
at two Veterans Administration mental hygiene units in New York City.

While the writer administered a number of the teats himself, the great
majority of tests were given by others on the assumption that the re
sults would have more universal applicability if obtained under con
ditions approximating those under which the test would be administered
in actual clinical practice* This procedure would also reduce to a
minimum any unconscious bias on the part of the present investigator
to influence the outcome. The instructions were explained and illus
trated by the writer at staff meetings of the installations where
the test was given, and opportunities were provided for the giving
of "practice tests" to answer any questions that might arise.
The actual directions for administering the test follow:
Test Materials
1. White unruled paper, x 11"

2. No. 2 pencil with eraser and moderately sharp point
3. Bender-Gestalt test cards
4. Smooth writing surface
5* Stop watch
Teat Arim-SrH
I. Initial Test
Have all test materials in readiness. Be sure to gain

rapport before starting the test. Explain that you
have a short test that you would like subject to take.
Do not answer anv questions about the nature of the test.
Hold all the test cards in your left hand face down.
Place a blank sheet of paper lengthwise in front of
subject and say:
"I have nine cards here which I»m going to show

you, one at a time. On each card is a simple
line drawing. I*d like you to copy these figures
as W * ^ fta you cfrn."
Keep a pile of five or ten sheets of paper near by so

that the subject may feel free to use more than one
sheet if he wishes. Do not, however, suggest that he
may use additional sheets.
- 73 -
Present the test cards one at a time, beginning with

figure A. Be sure the card is correctly oriented.
Lay the card on the table at the top of the sheet of
paper* Remove the card when subject has completed his
reproduction and make appropriate entries on the Work
Methods Score Sheet. Try to make these entries as
quickly as possible so as to avoid unnecessary delay*
Present the next card in the indicated order* Each
card should be visible for the entire time that sub
ject is working on it. Expose only one card at a time.
Record the time (in seconds) required for each repro

duction, i.e., the time elapsing from presentation to
removal of the test card.
If subject turns the paper, return the paper to its

original position. Make note of such attempts on the
Work Methods Score Sheet (e.g., "attempted to turn
paper") for the relevant figure. If he turns the paper
again, let the paper remain as it is and note the
orientation of the paper on the Work Methods Score
Sheet (e.g., "paper horizontal"). Return the paper to
its normal orientation before presenting the next card.
If subject turns the test card, return the card to its

original position. Make note of such attempts on the
Worth Methods Score Sheet (e.g.,"attempted to turn
card”) for the relevant figure. If he turns thecard
again, let the card remain as it is and note theori
entation of the card on the Work Methods Score Sheet,
using the following notation: (normal orientation),
(turned 90 degrees clockwise), (card inverted),
and (turned 90 degrees counterclockwise).
Subject is not allowed to use any mechanical guides,

such as a ruler or a coin. If he attempts to do so,
explain that this is a free-hand drawing test.
Refer all questions about method (e.g., time, correc

tions, size, etc.) back to subject with such noncommit
tal remarks as:
"Do it the wav you think heat", or
"That*s up to you", or
"There are no rules about the wav you work".
If subject asks whether he may use the other side of the

paper or more than one sheet of paper, say: "That*a up
to you" or "Ton niAy use any amount you wish". Provide
additional sheets if requested.
- 74 -
II. Immediate R«can Ta«t
This test is to be given immediately following the

Initial Test.
When subject has completed the last reproduction of the

Initial Test, remove all test cards and answer sheets
from view. Place another blank sheet of paper length
wise before him and say:
"Di»yw of the figures as you can from

memory, in any order you wish."
Allow minute* for recall. If subject says he has

"run dry" before the allotted time, encourage him to
continue.
HE SURE TO OBSERVE THE ORDER IN WHICH THE FIGURES ABE

REPRODUCED SO THAT IT CAN BE INDICATED LATER. When sub
ject has completed the Immediate Recall Test, indicate
the order of recall by placing a small number "1" under
the first figure reproduced, a "2" under the second
figure, and so on.
Do not make any entries on the Work Methods Score Sheet.
If any unusual behavior is observed, make notes on

another sheet of paper.
III. TmwftfHate Retest
This test is to be given immediately following the Im

mediate Recall Test.
Remove the Immediate Recall answer sheet from view and

re-administer the entire test according to the direc
tions given for the Initial Test administration.
Place a blank sheet of paper lengthwise in front of

subject and say:
"Mow Ifd like you to take the original test over

again. Here is the first card."
Proceed as before. Record time for each reproduction. .

Make the usual entries on the Work Methods Score Sheet.
1. The term, "Work Methods Score Sheet", refers to a mimeographed

form on which the examiner records the "methods signs" (to be
discussed in Chapter IV) observed during the test performance.
- 75 -
C. The Subjects
The neurotic group consists of 108 World War II veterans
who, at the time they were tested, were undergoing treatment at •the
Mental hygiene Service of the Veterans Administration. The cases
were drawn from the mental hygiene clinics of the New York Regional
Office and the Brooklyn Regional Office, with the majority coming
from the latter installation. In-service diagnoses, on the basis of
which these subjects were receiving disability pensions, were not
used in view of the suspicion that many such diagnoses were more a
matter of practical expedience than of scientific accuracy. Rather,
the nosological classifications were predicated upon judgments ar
rived at by one or more staff psychiatrists on the basis of case
histories, clinical interviews, and the results of an extensive bat
tery of diagnostic psychological tests administered by staff clinical
psychologists— a battery which generally included the Wechsler-Bellevue,
Rorschach, TAT, and Man-and-Wotnan Drawings. The Bender-Gestalt was
incorporated into the battery for purposes of this study so that it
would be regarded by the subjects as part of the normal testing pro
cedure, but in no instance was the test used to establish the diag
nosis. If there was any doubt in the minds of the psychiatrists or
psychologists about the diagnosis of a given patient, the case was
not used. Cases with complicating features such as gunshot wounds,
ulcers, migraine, asthma, psychopathic personality, schizoid trends,
orthopedic disabilities, or neurological involvement were automatical
ly excluded, the effort being to obtain "pure” neurotic cases insofar
as possible. There was less concern for differential diagnosis within

- 76 -
the neurotic rubric than for differential diagnosis between the neurot
ic and non-neurotic psychiatric patients in view of the unreliability
of specific diagnostic categories, as Ash 112) and others have demon
strated. Italy native white male adult outpatients were included in
the experimental criterion group, i'he criteria for selecting this
group were so rigorously adhered to that over three years were required
to collect the data, administrative difficulties also slowed down the
process of acquiring data because of the fact that during peak periods
of case load the staff psychologists were frequently too hard pressed
to add another test to their batteries.
The control or "normal" group consists of 285 World War II
veterans availing themselves of aptitude testing and vocational coun
seling at the New York University Testing and Advisement Center under
the provisions of Public Law 34-6, generally referred to as "the G.I.
Bill of Rights". Only native white male non-disabled veterans re
ceiving no pension for physical or neuropsychiatric disability are in
cluded. For purposes of this study, a veteran is considered "normal"
provided he has no history of neuropathic traits or emotional malad
justment, gives no evidence of nervous mannerisms or emotional dis
turbance during two hour-long interviews or during a series of testing
sessions ranging in total time from three to eleven hours, has not
been given a neuropsychiatric diagnosis in service, and has not sub
sequently requested or received psychiatric treatment.
In order to refine the control group still further, the sub
jects in this group were delimited by the application of a psychometric
as well as a clinical criterion. The psychometric instrument used for

- 77 -
this purpose was the Minnesota Multiphasic Personality Inventory,
group form (49), which has been clinically standardized and validated.
It is believed that deliberate misrepresentation of responses— a pit
fall in employing any personality inventory— was reduced to a minimum
because of the fact that the test was administered in a vocational
guidance, rather than a personnel selection situation. Since a stand
ard score of 70 is generally regarded as the dividing line between nor
mal and abnormal scores on any one of the nine scales, the control
group was subdivided into three subgroups as follows:
1) Below 70 Group (N = 155). All MMPI scores fall below 70.
2) 1 above 70 Group (N * 68). Only one of the MMPI scale

scores equals or exceeds 70.
3) 2 above 70 Group (N = 62). Two or more of the MMPI

scale scores equal or exceed 70.
This breakdown into subgroups permits comparisons within the control
group as well as between the control group and the neurotic subjects
and increases the probability that at least one sizable portion of the
controls is sufficiently "normal" to meet the rigorous standards that
should be imposed upon a research study using the method of contrast
ing groups for purposes of establishing test validity.
Along with the MMPI, all control subjects took the Otis Gam
ma (74) so that the relationship between Bender-Gestalt test perform
ance and intelligence might subsequently be explored. Ideally, it
would have been desirable to have administered the MMPI and the Otis
to the neurotic subjects as well, but such additions to the regular
diagnostic test battery employed at the Veterans Administration mental

hygiene clinics from which the experimental subjects were drawn were
- 78 -
not feasible because the need to process cases would have made such
additional testing time prohibitive. Eighty-four of the 108 subjects
in the neurotic group were, however, given ffechsler-Bellevues, and the
relationship between Bender-Gestalt test performance and scores on
this test will be explored in a subsequent chapter.
The composition of the control and neurotic groups with
respect to age, education, marital status, and branch of service is
given in Tables I, II, III, and IV. Data for the control subjects are
broken down for the three subgroups and then combined for the "Total
Control" group.
The mean age of the control group is 24.22 years, while that
of the neurotic group is 29.76 years. The control group has a greater
proportion of subjects below the age of 22, while the neurotic group
has a heavier concentration above 30. As might be expected in view of
this age distribution, the groups also differ in marital status, with
the control group having a higher proportion of single men. With re
gard to education, the control group has an advantage of slightly more
than Ig years, the average subject having completed about one semester
of the high school senior year, in contrast with the average neurotic
subject, who has not quite reached the 11th grade. The principal rea
son for this difference in mean education is the relatively high pro
portion of neurotic subjects who have not entered high school. The
differences in age, education, and marital status are readily under
stood when it is recalled that subjects in the control group were seek
ing vocational guidance with a view toward continuing their education
and that many of them were planning to enter college under the "G.I.
- 79 -
TABLE I
Age Composition of the Criterion Groups
GrouD
Below 1 above 2 above Total Total
70 70 70 . Control Neurotic
Age f It f * f. % f % f 2
42-43 0 0.0 0 0.0 0 0.0 0 0.0 3 2.8
40-41 1 0.7 1 1.5 0 0.0 2 0.7 3 2.8
38-39 2 1.3 1 1.5 0 0.0 3 1.1 7 6.5
36-37 2 1.3 2 2.9 0 0.0 4 1.4 10 9.3
34-35 3 1.9 2 2.9 3 4.8 8 2.8 6 5.6
32-33 2 1.3 1 1.5 1 1.6 4 1.4 12 11.1
30-31 7 4.5 6 8.8 2 3.2 15 5.3 7 6.5
28-29 11 7.1 10 14.7 1 1.6 22 7.7 11 10.2
26-27 17 11.0 11 16.2 7 11.3 35 12.3 17 15.7
24-25 23 14.8 2 2.9 10 16.1 35 12.3 15 13.9
22-23 25 16.1 13 19.1 12 19.4 50 17.6 15 13.9
20-21 35 22.6 14 20.6 19 30.7 68 23.9 1 0.9
18-19 21 13.6 5 7.4 5 8.1 31 10.9 0 0.0
N.I.* 6 3.9 0 0.0 2 3.2 8 2.8 1 0.9
N 155 68 62 285 108
Mean 23.94 25. 50 23.46 24.22 29.76
S.D. 4.66 5. 16 3.98 4.70 5.74
* Not indicated
- 80 -
TABLE II
Educational Status of the Criterion Groups
Group
Highest Below 1 above 2 above Total Total
Grade . 70 70 70 Control Neurotic
Attained f * f % f % f * f %
Grad. Work* A 2.6 1 1.5 1 1.6 6 2.1 3 2.8
16 9 5.8 6 8.8 4 6.5 19 6.7 4 3.7
15 5 3.2 1 1.5 1 1.6 7 2.5 1 0.9
14 17 11.0 7 10.3 7 11.3 31 10.9 8 7.4
13 21 13.6 11 16.2 6 9.7 38 13.3 8 7.4
12 74 47.7 30 44.1 27 43.6 131 46 J) 27 25.0
11 6 3.9 6 8.8 8 12.9 20 7.0 6 5.6
10 11 7.1 2 2.9 3 4*8 16 5.6 15 13.9
9 6 3.9 4 5.9 3 4.8 13 4.6 10 9.3
8 1 0.7 0 0.0 0 0.0 1 0.4 16 14.8
7 0 0.0 0 0.0 0 0.0 0 0.0 3 2.8
6 0 0.0 0 0.0 0 0.0 0 0.0 2 1.9
5 0 0.0 0 0.0 0 0.0 0 0.0 3 2.8
4 0 0.0 0 0.0 0 0.0 0 0.0 2 1.9
N.I.** 1 0.7 0 0.0 2 3.2 3 1,1 0 0.0
N 155 68 62 285 108
Mean 12.49 12.51 12.35 12.47 10.74
S.D. 1.73 1.72 1.69 1.72 2.81
* Graduate Work
** Not indicated
- 81 -
TABLE III
Marital Status of the Criterion Groups
Grouo
Marital 70 70 70 Control Neurotic
Status f * . f. * f % f * f S
Single 111 71.6 39 57.4 46 74.2 196 68.7 40 37.0
Married 40 25.8 28 41.2 16 25.8 84 29.5 62 57.5
Divorced 0 0.0 1 1.5 0 0.0 1 0.4 1 0.9
Separated 0 0.0 0 0.0 0 0.0 0 0.0 4 3.7
N.I.* 4 2.6 0 0.0 0 0.0 4 1.4 1 0.0
N 155 68 62 285 108
* Not Indicated
TABLE IV
Military Service Affiliation of the Criterion Groups
GrouD
Branch of 70 70 ... 70 Control Neurotic
Service f % ___ f _ % f * f S f %
Army 110 71.0 45 66.2 44 71.0 199 69.7 91 84.3
Navy 38 24.5 16 23.5 14 22.6 68 23.9 16 14.8
Marines 4 2.6 5 7.4 2 3.2 11 3.9 1 0.9
Coast Guard 3 1.9 2 2.9 2 3.2 7 2.5 0 0.0
H 155 68 62 285 108

- 82 -
Billn. Under the circumstances they were more likely to be younger, un
married, and further advanced educationally than the neurotic veterans
whose problems were primarily emotional rather than educational or voca
tional. The fact that a higher proportion of the neurotic subjects had
served in the Army may be attributable in part to the fact that the Navy
tended to select recruits from the younger age and higher educational
brackets.
The question immediately arises as to how significant these
differences in composition of the criterion groups may be. The three
subgroups of the control group are remarkably similar with regard to
age, education, marital status, and branch of service, and it may be
noted that the majority of subjects from both the control and the neurot
ic groups are between 20 and 30, the modal grade attained is the 12th,
and the modal service affiliation is the Army. It is difficult to see
how either marital status or service affiliation could be related to
test performance on the Bender-Gestalt, but age and education might con
ceivably exhibit significant correlations. This problem was tackled by
initially performing an analysis of the scoring items for the Total Con
trol and Total Neurotic groups, isolating those items which showed sta
tistically significant validities, and then performing a second analysis
of these items on 138 control and neurotic subjects who were matched
pair by pair for age, education, and marital status. Only those items
which survived both analyses and which, by virtue of this procedure,
were unrelated to the variables under consideration were retained for
subsequent combination into a total score. The details of the proce
dure will be fully described in Chapter V.

CHAPTER IV
THE SCORING SYSTEM
A. Introduction
Bender did not develop any system of scoring for her test
because her concern with gross distortions and bizarre deviations of
the reproductions from the original stimuli made such a refinement un
necessary. Her paramount aim was to ascertain whether the gestalt was
maintained or destroyed in the drawings of children, mental defectives,
psychotics, and organics, and no systematic scoring device was needed
to note such obvious alterations as rotation, fragmentation, persevera
tion, primitivation, and dissociation. With this orientation, it is
not surprising that she failed to observe characteristics of test per
formance that were distinctive for neurotic patients. In the present
research the working hypothesis has been adopted that subtle differences
between normals and neurotics may in fact exist and that the way to test
this hypothesis is to develop a reasonably objective and communicable
method of scoring that will reflect those nuances of drawing which con
tribute to relatively minor modifications of the original stimuli.
Over-refinement and the undue precision of meticulous measure
ment, such as characterizes Billingslea's system, is, however, avoided,
since it is recognized that one of the virtues of the Bender-Gestalt is
its inherent simplicity and that any attempt to complicate it by evolving
a time-consuming method of scoring would almost automatically result in

- 84 -
its exclusion from the clinician's diagnostic armamentarium. Measure
ment is occasionally resorted to as, for example, in the case of size
deviations, but for the most part rulers, protractors, and metric grids
have given way to a simple inspection technique.
The scoring factors themselves have been selected on the basis
of (1) characteristics of the reproductions observed in the present study,
(2) findings on a variety of clinical groups cited in the literature (with
the exception of the studies of Billingslea and Pascal and Suttell, which
had not been published at the inception of the present research), and (3)
findings of the present investigator based upon administration of the
test to more than 1000 military neuropsychiatric cases (covering the
major diagnostic categories of neurosis, psychopathic personality, men
tal deficiency, simple adult maladjustment, alcoholism and drug addiction,
sexual aberrations, functional psychosis, and organic psychosis) and to
several hundred veteran and non-veteran "normal adults" requesting edu
cational and vocational counseling. A number of scoring items were orig
inally tried out but were eventually excluded from the statistical tabu
lations because of their complete absence from the 786 test records (a
test and a retest for each of the 393 normal and neurotic subjects) in
the present study. Many of these signs have been mentioned by Bender
and others in connection with the test performance of children, mental
defectives, psychotics, and.organics but, as might have been anticipated
from Bender's negative results with neurotic cases, were not observed
to occur with the normal and neurotic adults of the present 6tudy. Ex
amples of these discarded or unusable signs are scribbling, perseverated
vortical movement, substitution of numbers and letters for dots, stroke
like perseveration, closure of figure 4» ornamentation and embellishment,

- 85 -
and distortion of the gestalt beyond recognition.
A distinction has been made in the present investigation be
tween "graphic" signs and "methods" signs. The graphic signs are de
fined as those characteristics of the reproductions which are scorable
solely by inspection or measurement of the figures appearing on the
test record} they do not presuppose observation of the subject's test
behavior. Examples are size, contiguity, substitution, and asymmetry.
The methods signs are defined as those characteristics of the repro
ductions which are scorable only by observation of the subject's test
behavior. Examples are counting, paper rotation, direction (whether
a line is made in the dextrad or sinistrad direction), and initial
part (whether the left or the right part of a figure consisting of two
subfigures is made first). Hutt(56) refers to some of these methods
signs as being worthy of investigation but reports no relevant data.
Billingslea (26) notes the incidence of counting for his subjects but
otherwise ignores the manner in which the reproductions are made. Wolt
mann (102) gives due recognition to many of these signs, but reports
clinical impressions rather than quantitative date.. Pascal and Suttell
(76) suggest that notation be made of card turning or paper turning but
exclude all methods signs from statistical consideration. It might be
noted at this point that many other methods signs other than those ap
pearing in the present study might have been used. In the exploratory
work preceding the formal investigation a more comprehensive list was
employed but was discarded as impracticable because the examiners found
it impossible to make so many notations and still keep pace with the sub
ject. Those that have been retained have been observed to occur most
frequently.
- 86 -
The final list of all signs employed in the present study
consists of 82 scoring categories (70 individual figure and 12 global)
many of which are represented in more than one figure, the total number
of signs being 318. Each individual figure scoring category is num
bered, and an individual sign is designated by this number and by the
number of the figure on which it appears. "Guides", for example, is
scoring category number 9; should guide lines be used to reproduce
figure 5, the designation would be 9-5. Similarly, "initial part" is
category 65; should the diamond in figure A be made before the circle
is drawn, and should the bell of figure A be drawn before the open
square, the designations would be 65-A and 65-A, respectively. Certain
categories are represented in only one figure as, for example, "pairing"
(number 70) which occurs solely in figure 1 and is designated as 70-1.
Other categories, such as "total rotation" (number 48) may occur in all
figures and are designated accordingly as 48-A, A8-1, etc. "Global"
categories such as sequence, cohesion, and colliding refer to the in
terrelationship of all the figures and henee have no specific figure
designation.
The individual figure and global scoring categories are de
fined and, where necessary, illustrated, in the Scoring Manual below.
B. Scoring Manual
I. IndividualsFigure Signs
1. Length (all figures). Distance (in mm.) of the horizontal extent
of the reproduction. Measure from the extreme left to the extreme
right of the reproduction, using a transparent millimeter ruler.

- 87 -
Bm b p Io b *
_ - <j (j (1 o o o
p o o © o . . o c a
© o 0 0 ° 6 °0 u o /s
, 0 0 0 o O o o I o o
I
*-----------
Figure 2 Figure A
Figure 1 Figure U
2. Height (all figures). Distance (in mm.) of the vertical extent of
the reproduction. Measure from the extreme upper to the extreme
lower part of the reproduction, using a transparent millemeter
fuler. Examples;
o
O c O & c o 0
o o d o o c, o
Figure A Figure 2
Figure 8
5. Tremulous line (a. 4, 6, 7, 8). A shaky, quivering wavering line.
Figure 6 Figure U Figure 8

- 88 -
6. Erasure (any figure). All or part of a reproduction exhibits evi
dence of erasure*
7. Multiple attempt (any figure). Two or more reproductions of a given
figure appear on the test record. The initial reproduction may in
volve the entire figure or just part of it. The initial reproduc-
tion may be crossed out, partially erased, or left intact. Hot
scored if the initial reproduction 1b erased completely and the
second attempt is superimposed upon it.
Figure A Figure U
8. Figure A ftlfltttlVt (A). One or more reproductions appear above
Figure A on the test record, or Figure A is 3 inches or more from
the top of the page.
Examples:
i• . * 0 0• t*
CO
o 0 o o e • ®
c 0 O O 0 0 «
t) * *> ** 0 c
to o o oo
. o 0 0 * 0 0 °
C S V H O C
c
- 89 -
9 . Guides (any figure). Use of extraneous lines or points as an aid
to reproducing the figure.
Examples:
o— o q- - c o •&--- 1~— "6-- e~
^ o o— o c>— o — Ci -e * 6
0 u o o o -«*- fe e> ^ ‘ o
O C >
Figure A Figure 2
Figure 5 Figure 6
10. Overshooting lines (A, 4> 7, 8). Lines overshoot one another at
points of juncture.
Examples:
i
& >
Figure A
Figure 4 Figure 7
11. Retracing (A, 4, 6, 7, 8). k part of a line or one or more lines
are made with repetitive strokes which are more or less superimposed
on one smother. The line may be belabored or sketched.
Examples:
0 7
Figure A Figure 8
Figure 4
- 90 -
12. Projection retracing (7,8). One or more of the pointed projections
of figures 7 or 8 are retraced as defined in (11). Scored also for
Retracing (11). Not scored when entire reproduction is retraced.
Not scored if only the diamond in figure 8 is retraced.
14. Substitution: dot-dash (1, 3, 5). Dots are reproduced as hori
zontal or vertical dashes at least 1/16 inch in length. At least
four dots should be so reproduced to be scored.
f 1 I I i I I I I / /
Figure A
<■ N
, /
/ ' /
I /
Figure 5 Figure 3
13. Substitution: dot-circle (1, 3, 5)• Dots are reproduced as un
filed circles or loops. At .least four dots should be so reproduced
to be scored.
I
• 0 o
Figure 1 o
. . o
I
I
0© O O
0 °o Figure 3
3 •
o
Figure 5
- 91 -
16. Substitution: dot-ball (1, 3, 5). Dots are reproduced rs filied-
in circles or loops. At least four dots should be so reproduced
to be scored. Slightly enlarged dots are not scored.
Examples: ^
• • • # • • • • • • # „ » * * p
*
Figure 1 * * ,
• ° * *
•> » * Figure 3 '
Figure 5
Not scored:
17. Substitution: dot-scribble (1, 3, 5). Dots are reproduced as
"wiggly" lines, "butterfly" lines, or scribbles, at least four
dots should be so reproduced to be scored.
Examples: ,
\ ' v. -
i ( /
I
Figure 3 figure 5
18. Substitution: circle-dot (2). The circles in figure 2 are re
produced as dots or filled-in circles or loops, nt least four
circles should be so reproduced to be scored.

- 92 -
Examples:
« o
C> o
C 0 u
Figure 2 Figure 2
20. Numeration (1, 2, 3, 5, 6). The number of dots, circles, columns,
or waves in the reproduction differs from the actual number in
the stimulus figure. The correct number of elements is as follows
for the figures concerned{
Figure 1 12 dots
Figure 2 ' 11 columns or 33 circles
Figure 3 1, 3, 5, and 7 dots
Figure 3 19 dots in semicircle, 7dots in tangential line
Figure 6 4 wave crests in horizontal and 4 invertical
examples;
Figure 1
Figure 3
Figure 5
Figure 6
- 93 -
21. Ifava accentuation (6). Harked Increase in amplitude of hori
zontal or vertical wavy line. At least 2 of the waves should be
so reproduced to be scored.
Examples:
/
X.
Figure 6 Figure 6
22. Wave flattening (6). Harked decrease in amplitude of horizontal
or vertical wavy line. At least 2 of the waves should be so re
produced to be scored.
Examples;
Figure 6 Figure 6
23. Wave irregularity (6). Harked variation in wave lenght or ampli
tude of horizontal or vertical wavy line. At least 1 of the waves

should differ markedly from the others to be scored.
- 94 -
Figure 6 Figure 6
25. Curvature contraction (4, 5). Marked constriction of bell in
figure 4 and semicircle in figure 5.
Examples:
Figure 4 Figure 5
26. Curvature flattening (4, 5). -Marked flattening of bell in figure 4
and semicircle in figure 5.
£X§JHEl2&:
Figure 4 Figure 5
27. Asymmetry (A, 3, 4, 5, 7, 8). A symmetrical figure or a symmetri
cal part of a figure is reproduced asymmetrically. Lines equal in
length are made unequal, or corresponding angles equal in size are
made unequal, or regular contours are made irregular.

- 95 -
Brarolest
* *
o o
Figure A Figure 3 Figure A
< C Z 5 ~ >
Figure 5 Figure 7 Figure 8
28. Disproportion (A, At 5> 6, 7, 8). Distortion in the relative
size of the parts of a figure. Parts of equal length, height, or
area are made unequal.
figggdLSl1
Figure A Figure A
Figure 5 Figure 6
Figure 7 Figure 8
- 96 -
29. Displacement (A 3, 4, 5, 6, 7, 8). Shifting of the point of junc
ture of one part of a figure with the other, either to the left
or right or above or below. In the case of figure 5 > junction
of the tangential line at the center or left of the center of the
semicircle is scored, as well as junction at the extreme right.
Scoring for figure 6 is strict: the vertical line must cross the
horizontal at the middle of the third horizontal wave, or dis
placement is scored; displacement is also scored if the horizon
tal line crosses the vertical in the trough at either end of the
second (from the top) vertical wave. In the case of figure 7, dis
placement is scored when one hexagon is drawn sell below the other
or when the lower point of the left hexagon is not below the lower
point of the right hexagon.
RTfapplapi
Figure A Figure 3 # Figure 4
Figure 4 Figure 5
- 97 -
Figure 6 Figure 6
Figure 8
30. Contiguity (A, U, 5» 7, 8). Contiguous parts of a figure over
lap or are only partially contiguous, or overlapping parts are
made contiguous.
Examples:
Figure A
- 98 -
Figure A Figure 5 Figure 7
o~7< s— >
Figure 8
31. on (2» 3* 5» 7). Distortion of the angular direc
tion of a figure or of the angle formed by one part of a figure
with the other.
0 0 0 6 o o o
f t o o o o o do 0 o
t o o o o o 0 0 0 0 0 o a 0 * t 0
O O O O 6 0 0 0 0 0 0 0 t> <> 0 0 0
Figure 2 Figure 2
i I
4 •
i
» i
»
Figure 3 Figure 5 Figure A

- 99 -
32> Concentric arc (3). Two or more of the arrowheads in figure 3
are reproduced as arcs.
%
%
t •
33* Parallel Hwaa (4 , 7, 8). Parallel lines are reproduced as
converging or diverging.
£ssie1s£*
Figure A Figure 7
Figure 7 Figure 8
34. ealiMM (2). Slanting coluams of figure 2 are repro
duced as converging or diverging. Also scored for angulation (31-2).
ftgyU&gt
0 0 O 0 0 0 O 0
0 b 0 o 0 t> o C 0 0
o d d 0 ^ d O o 0 q
35* Parallel r o w (2). The parallel rove of figure 2 are reproduced
as converging or diverging.
y,Tn»nlan
^ e o o o o o o
O O O o o 0 o o o 0 o o 0 0 o O
o o O O ^ O O O O O O o o « e o O O O O O O O
„ O O o o 0.0 0 ° « 0 0 o 0 0 < > o o e

c O © o °
Figure 2
36. Horigontal iT-raonii^ (1 , 2). The horizontal progression of
dots or coluans is irregular and uneven rather than correctly
aligned or consistent in direction.
° o O 0 o O A t
O O o O. o c £ 0
& O c> to C O *
Figure 1 Figure 2
37. Irregular gracing (l, 2, 3). The distances between dots, circles,
rows, or coluans are markedly unequal.
£2S£E£s &:
c ° ° ® ° e o ©
o o O o o o 0 Q
............. 0 0 O O o 0 0 0
Figure 1 Figure 2
O O O o O O C O 0 ,
• • » #
•*. * *
\
t
o ° • « * . o c °0 ' : • * ■
o a O a 0 0
c o & 0 # 0
38. Perseveration (1, 2, 3) • Tendency to extend a figure indefinitely

- 101 -
There mast be at least 4 additional dots in figure 1, 4 additional
columns in figure 2, and 1 additional column in figure 3. Also
scored for numeration (20).
Examples:
Figure 1 , *« ',
. \ ’• 1
< S 0 6 « ° O O 0 6 O o o c o G . • * '
u t> 0 0 ° 0 0 c L. * C i. o I ‘ *
q 0 0 o S 0 v w 0 >■ o i. <-■ ‘ ’
Figure 2 Figure 3
39* Serial incompleteness (1, 2, 3). Tendency to reduce the number
of elements in figures 1, 2, and 3. Scored when there are 9 or
less dots in figure 1, 8 or less columns in figure 2, 3 or less
columns in figure 3* Also scored for numeration (20).
Examples:
o a e o>
4.0. Closed-open (A, 4* 7, 8). Enclosed figures or parts of figures
are not completely closed in the reproduction because of the presence
of gaps. One instance is sufficient to score.
Examples:
(also scored for Contiguity)
Figure A
- 102 -
L_ X
J 2
/-
> < (also scored for Contiguity)
Figure 8
41. Open-closed (4, 5). Open figures or parts of figures are closed
or nearly closed in the reproduction.
Examples:
(also scored for paralled lines)
Figure 4 Figure 4
(also scored for Curvature:

horseshoe)
« <
Figure 5 Figure 5
44* Splitting (A, 4» 7). The parts of a figure, joined in the stiau-
lu8 card, are reproduced so that they are separated by a narrow
space (less than £ inch).
Figure A
Figure 4 Figure 7
- 103 -
45. Dissociation (A, 4, 5» 7). The part* of a figure, joined in the
stimulus card, are reproduced so that they are separated by a space
of £ inch or nore.
L:
Figure A
Figure 7
» *i
Figure 5
46. Simplification (any figure). The reproduction is aade consider
ably less complex than the original, usually by substitution of
solid for dotted lines, straight lines for angles, loops or cursed
lines for angular figures, or straight lines for easy lines. Also
scored for Distortion (47).
o O
Figure 4 Figure 5 Figure 6 Figure 7
47. Distortion (any figure). Marked deviation of the reproduction from

- 104 -
the original, Including simplification, addition or onission of
angles, massing of dots, narked line irregularity, closure of
open figures, etc.
SBadftftt
a o Cl o ^ 0 * • i« i
. 5 » # J f a ?
n ^ o ° -
o 0 ^
<*«
m *
o * *
» .* •
* •
Figure 6
Figure 4 Figure 5
C3 C>
Figure 8
Figure 7
Figure 8
48. Total rotation (any figure). The entire reproduction is rotated
45° or nore from the orientation of the stimulus figure. Turning
of the stimulus card or of the paper with corresponding orienta
tion of the reproduction is not scored.
Figure 4 Figure 1 . Figure 3

- 105 -
49. Part rotation (A, 4» 5» 7, 8). Rotation of 45° or more of one. of
the parts of a figure. An exception is figure 5» whioh is scored
for a rotation of the semicircle of 30° or more.
Examples: '
— -i
Figure A Figure 4
Figure 4 Figure 5
r
s
N.
Figure 7 Figure 8
50. Reversal (A, 4> 7). The spatial position of the two parts of a
figure are reversed so that the left part is reproduced on the
right and the right part on the left.
Eframplgg.;
f
J
Figure A
\
Figure 4 Figure 7
51. Paper rotation (any figure). The paper is turned hy the subject
to a horizontal position while one or more figures are reproduced.
A reproduction so made is not scored for tptal rotation (48) if it
is correctly oriented with respect to the horizontal orientation of
the paper.
52. Card rotation (any figure). The stimulus card ie turned by the
subject at least 90° in any direction. A reproduction made under
these conditions is not scored for total rotation (48) if it is
correctly oriented with respect to the orientation of the card.
53. F ra g m e n ta tio n (any figure). Only an isolated part, or fragment,
of the total figure is reproduced. In the cases of figures 1 and

/
2, 4 or less dots and 3 or less columns are so scored; in these
two instances Serial incompleteness (39) would also be scored.

I
Examples:
Figure A Figure 1
54* Omission (A, 2, 3, 4, 5, 6, 7, 8). Omission of a major part of a
figure in the reproduction. More of the figure is present, how
ever, than would be true if fragmentation were scored.
Examples:
Figure A Figure 2 Figure 3 Figure 4
0 <
Figure 4 Figure 5 Figure 6 Figure 7 Figure 8
- 107 -
5$. Upward slope (A, 1, 2, 3, 6, 8). The reproduction Inclines up
ward. Scoring is strict; any perceptible upward slant is scored,
even if slight. The horizontal line of figure 6 is normally slanted
in the stimulus card but is scored if also slanted in the reproduc
tion, since the majority of subjects fail to observe the slant in
that figure. Scored also for counterclockwise rotation up to 45°*

B t j m w iI ftg 8
. * *

L.
7 o
o
®
o
°
e > o
i, 0
°0
»
o
a
0 O O
Figure 8 Figure 2
56. Downward slope (A, 1, 2, 3, 6, 8). The reproduction slants down
ward. Scoring is strict; any perceptible downward slope is scored,
even if slight. Scored also for clockwise rotation up to 45°•
JaSflElats
a o c b Q 0
o o O o © a a
° ° o ° = °c
° ° €>
Figure 3 Figure 8
Figure 6
- 108 -
57. Direction: slnlstrad (1, 2, 3, 4, 5» 6). Horizontally oriented
stimulus figures are reproduced from right to l'eft. In figure 4>
either the open square or the bell, or both, nay be made from
right to left. In figure 5, only the semicircle, and In figure
6, only the horizontal line are scored
Examples:
Figure 1
Figure 4 Figure 5
Figure 6
58. Direction* upward (2, 3, 6). A figure, part of a figure, or
a line is drawn in the upward direction, i.e., from below to
above. In figures 2 and 3> any one column so drawn is sufficient
for scoring. "

*
Examples: . *
o o o o o o O C o
o \ O C> t o o o o c ^ o c o
, to \ o O O o £> O O o ° a
Figure 2 Figure 3
Figure 6
59. Direction: centrifugal (A, 5). The bell of figure 4 or the

- 109 -
semicircle of figure 5 is made by starting at the crest of the
curve and drawing lines from the center of the curve to each end.
J6BSQlS£$
Figure 5
Figure 4
60. Counting: test figure (1, 2, 3, 5, 6). Subject overtly counts
the number of dots, circles, columns, rows, or waves in the stimu
lus figure, as indicated by moving pencil or finger over each
element of the figure, counting aloud, moving lips while counting
subvocally, etc.
61. Counting: reproduction (1, 2, 3, 5, 6). The same as (60), except
that the elements of the reproduction, rather than the stimulus
figure, are counted.
62. Recounting; test figure (1, 2, 3, 5, 6). Subject overtly counts
the elements of the stimulus figure, as defined in (60) two or more
times.
63. Recounting; reproduction (1, 2, 3, 5, 6}. The same as (62), ex
cept that the elements of the reproduction, rather than the stimu
lus figure, are overtly counted two or more times.
64. Rows (2). Figure 2 is reproduced row by row rather than column by
column. Making the top row only and then completing the columns
is not scored.
- 110 -
Ex&aPl»*
O O O a c d C G Q O
O O u a o o G GO
43 " o" O O O’ Q
Cs © e» o o
y
Figure 2
65. Initial part (A, 4» 5» 6, 7, 8). This refers to the pert of the
figure which is reproduced before the other part is aade. Scored
only if the part- indicated below is n&de-first:
Figure A - diamond Figure 6 - vertical line

Figure 4 - bell Figure 7 - left hexagon
Figure 5 - handle Figure 8 - inclosed diamond
66.' Angle differentiation (7). The two angles at the extremes of
either hexagon are made essentially alike rather than an acute
and obtuse.
Examples:
Figure 7
67. Leg aaortenlng (5). One "leg" of the semicircle is shorter than
the other. The angle formed between the horizontal and an imag
inary line connecting the two ends pf the legs should be at least
10°. An angle of 30° or more is scored for both leg shortening
and part rotation (49).

- Ill -
•'
0
10°
Figure 5
68. Side reduction (7* 8)* A six-sided figure is made with 5 or less
sides. Scored also for distortion (47)*
acsBEles:
figure y
70. Pairing (1). The dots of figure 1 are arranged in pairs. Three
or more pairs should be erident to be scored.
F.-ramnl m;
Figure 1
71* Time (all figures). Time (in seconds) required to reproduce each
figure. The stop watch is started when the stimulus card is pre
sented and stopped when the subject stops drawing or otherwise
signifies that he is finished.
72. Correction (all figures). An inaccuracy in the reproduction is
rectified by drawing another line without erasing the incorrect
one, by crossing out a part of the figure, or by partially re
tracing the incorrect part.

- 112 -
S b b Is s *
O o o ° o u 6
0-0 6 O 6 6 C S>
0 ^ 0 0 ■ O < ^ 0 o O O 0 O
Figure A Figure 2
Figure A
Figure 6
Figure 8
73* Multiple wtwalHng (A, U, 6, 7, 8)* The subject tskes two or sore
strokes to sake a line which is usually made by one continuous
stroke. Sketching is excluded fra* this definition, however.
EwpIqb:
o o
Figure A
Figure 4
Figure 8
Figure 7
- 113 -
74* Projection angulation (7, 8). The sides of the angles forming the
pointed projections in figures 7 or 8 are made unequal or lopeided
or asyanetrical, the points of the angles are blunted or otherwise
distorted or (in the case of figure 8 only), the corresponding
angles at the extremes of the hexagon are unequal* Scoring is
strict; slight deviations are scored*
Exanplet
Figure 7
Figure 8
75* Line sag (A, 4, 7, 8). Straight lines are reproduced as curved,
sagging, convex, or concave*
Figure A
Figure 4 Figure 7
d s i > > < z 2 n >

Figure 8
- 114 -
77* Curvature; horseshoe (5). The lege of the semicircle In figure
5 curve in toward each other, resembling a horseshoe.
* *
Figure 5
78. Continuation (1, 2). Figures 1 or 2 are made on two lines,
usually when the reproduction is incomplete at the right-hand
margin.
6 0 0 6 O o o e
O O o Q o e a c
0 O .0 0 O 0
% • • Figure 1
C o e
0 o O
y O Q
Figure 2
80. Cmpre squfrfrff (4 , 5). The bell of figure 4 or the semicircle
of figure 5 resemble an open square.
BFMPlf"!
Figure 4 Figure 5
- 115 -
II. Global Signs
Sequence. Order in which the reproductions are placed on the page.
Scored in terns of the nmber of sequential connections. A sequential
connection refers to any two reproductions which follow one another in
the sane order as the presentation of the stimulus cards. A reproduc
tion nay be directly above, below, to the right, or to the left of the
one which it follows. A connection exists only when no other figure
intervenes between the two successive reproductions. If a figure is
near the right-hand nargin and is followed by its successor near the
left-hand nargin below, a connection is scored. If a figure is at the
bottom of the left half of a page and is followed by its successor at
the top of the right half of the page, or vice versa, a connection is
scored. If a figure is at the bottom of a page and is followed by its
successor on the obverse side of the paper or on a new sheet, a connec
tion is scored.
BStikSfii
7 connections 6 connections 5 connections

- 116 -
Slgld Sequence. Each reproduction is placed directly below the pre
ceding one. Hie reproductions nay all be placed on one side of the page
or continued from the front side to the obTerse side or to a new sheet
of paper. Not scored, however, if the reproductions are placed on one
side of the page in two colunns. Also scored for 8 sequential con
nections.
Exajglesj
CO
, o e « 0 * » t> ® e »
oo
„ . • e « • « . # . «
^ 08 a &
*
•»• i* »»
Front Back
Cohesion. The degree of cosipression or expansion of all the reproduc
tions, expressed in terms of theamount of space filled, as follows:
Cohesion: 1/3. All of the figures are reproduced in the upper,

or lower third of one page.
Cohesion: 1/2. All of the figures are reproduced in the upper

or lower half of one page.
Cohesion: 2/3* All of the figures reproduced in the upper or

lower 2/3 of one page.
Cohesion: 1 All of the reproductions occupy a full page.

- 117 -
Cohesion: 1-1/3. Reproductions occupy all or pert of one page

plus 1/3 of the back of the page or 1/3 of a
new sheet*
Cohesion: 1-1/2* Reproductions oceupy all or part of one page

new sheet*
Cohesion: 1-2/3* Reproductions occupy all or part of one page

new sheet*
Cohesion: 2 Reproductions fully occupy the front and back

of a page or one side of one sheet and one
side of a second sheet*
Cohesion: 2 / More than 2 pages (front and back of one sheet

or one side of each of two sheets) used to re
produce the figures*
Ezaaples:
Front Back
C O ........
C O
•o o o o
a * • 0 0 ® 9l•,Q.
9 0 .4•
i \
% % %
1 /
/
Cohesion: 1/2
Second sheet. Use of aore then one sheet of paper on which to place
the reproductions. Refers to the use of two or aore separate sheets,
not to the front and back of a single page.
Collision. Two or aore reproductions overlap or run Into one another*

- 118 -
C O
Figure A and 1
Figures U and .*
Figures 5 and 6
gJBbSring* Q»e or aore of the reproductions is numbered by the subject.
£SHBil£:
0£> V ..
* * » ■* . «'
* ' * I I •*
V7
< z n ~ z ,
Ccapartafota. Che or aore lines are dram to separate the reproductions.

- 119 -
Total Time. Sum (in seconds) of the times required to complete each
of the reproductions*
(a) Time: 139. Scored when the total time is less than 140
seconds.
Short figures. Presence of one or more reproductions equaling or
falling below the 10th percentile of the Total Control group in length,
for the Initial Test and the Retest, the critical lengths are given in
the table below.
Long figures. Presence of one or more reproductions equaling or ex
ceeding the 90th percentile of the Total Control group in length. For
the Initial Test and the Retest, the critical lengths are given in the
table below.
Flat figures. Presence of one or more reproductions equaling or falling
below the 10th percentile of the Total Control group in height. For
the Initial Test and the Retest, the critical heights are given in the
table below.
Tall figures. Presence of one or more reproductions equaling or ex
ceeding the 90th percentile of the Total Control group in height. For
the Initial Test and the Retest, the critical heights are given in the
table below.
- 120 -
TABLE V
Criteria for Scoring Short, Long, Flat, and Tall Figures
Short figures Long figures Flat figures T»n figures

initial Retest Initial Retest. Initial Retest Initial RetesT
Oil
A 27 29 58 55 15 16 32 31
1 69 70 150 146 — — — —
2 86 82 171 168 13 13 26 25
3 25 25 53 54 22 22 46 47
4 33 31 55 52 35 32 57 54
5 30 31 57 56 32 31 60 58
6 86 82 156 147 55 52 96 92
7 30 30 55 53 39 36 68 63
8 63 58 108 104 13 12 24 21
CHAPTER V
RESULTS
As a first step in the analysis of the results, separate
treatment was accorded the global and the individual figure scoring
signs, and comparisons were made of the differences between the cri
terion groups which were statistically reliable or insignificant on
the Initial Test and the Retest. Because of its uniqueness the Recall
Test was regarded as a separate unit and will be considered at length
in a subsequent section of this chapter after presentation of the find
ings on the Initial Test and the Retest.
A. Global Signs on the Initial Test and the Retest
Tables VI and VII show the means and standard deviations of
the lengths of each of the figure reproductions on the Initial Test
and the Retest for the criterion groups and the critical ratios of the
differences between the means of the Total Control and Total Neurotic
groups. It is evident that the three criterion sub-groups (Below 70,
1 Above 70, and 2 Above 70) are much more nearly alike with respect
to the mean length of each reproduction than are any one of these groups
and the Total Neurotic group. Comparing the Total Control and Total
Neurotic groups, a consistent tendency is noted on both the Initial
Test and the Retest for the neurotic subjects to make their reproduc
tions shorter than those of the controls. Although only two of the
- 122 -
rod * o
O O IT
•H *H 00 Oi vO cn
«P -P I
00
•
*A rH
•
$ 00 00 $ 35
•
CM
h ^ -4l
o
rH
♦ i +
o
I
rH
I
o
♦
o
♦
o
I
s 3
O -rl
•H
«P u
o ©
0 -P
• • cn
• cv• 0
• 10
• O'• O'• cv• 0
• *8 3
o rH
CM to 00 00 O' 0 M3 0 cv Or &
u to CO cn cn p —t CV
1 p cv ®
u h m
r-l+3 '
5 • ^ g CV cv cv CT\ t-
ag '—' 0 3
eh 5 a 00 rH O' NO
<n a— cn Q cn i 9 rH 00 P a
-
3>
©
S4^ H 4h ®
rH
5 3 ° ©
h g
4-4 o 5J
O 0] 00 M3 00 cv CO CO
® n o • • 4h „
Ao & cn rH cv CV 00 c^ c- 0 -P
rH IT CO—9 cn cn rH cv 3 r—4 cd
H O tO
s 55 55
5 ^55" 00
CH h fl O »n 'O cv to• cv vn c ca
0 £ o8 a
" e b
a O ®® PP M3 00 to Ht H P
$} 3 CV
rH cn 9 a 00 at
® * O <0 o
5 JQ P
u>5 | S
,s ®
s fl
1-94^
oh m_* c*- O' 00• O rH O' cv 00 M3
® P Q •
M3 H H 0 'O c^ 00 1 Pa
1 pm §5
® O
® h E-4
sO coH cn cn a 3 p cv w ca
®
Eh ® &
c*\1 p 11 p oa
* Cm 'O •—> aJ t> «r•
» to• cn cn C"• cr•
\ vn CO ® 3
1 H 4-1 O • » O rH
cd P cd CM * o P Pr
5
§
5°p
P ® O
p s .3
4
3 p
cn 45 %
o
O' td
3 *a{
a) H P P
M o •
P
► cn cv p g,
B 4h O
® 3 0 0
o • cv• cv
• O' v> CO cn P 3
O ® o
o ® CO •
—I 0
i 0 CV p M3 rH M> 'P fH M
> CV <n 1
—1 cv 3 00
3 O a! I
—I
■E O -P
P P c\i3a? 5-
O II a) o
5 O op
5
o
CO Po o
cd
cv O'•
O'
o
p
v0•
'O
cv
p
cn
•
00
cn 4
ON
4
NT
a
cv
O'
9
4
0
00
p p
P o
£o a®
TJ 3 P .*
OflP
n) Oi H Or
o• cn O' p cv ir» 3 3
« P«0
®«
£
in
H
Q
CO'
lf\
rH
•
cv
cn
cn
<n P
•
cn O O
a
'
cv p g« f H
u>
4-1 ®
55-3
H H°£„ C*- o HE cv t- CV vn cv
«
ip
3
a
•
a
• •
00
cn 9 s ia •
cv
00
3 3P
U P 3
P
mho
®5
o
35
o
EH
■an*
P CV cn VO 00 v£>
•
® 1 cv p in cn
—I O' o 00 O ON a<
o 00 CQ — ' P «n cn i cv rH ©
h ©
ft ft ft •H
©
»A 3 2 II
• o• cv« P cn cn O' 00 cn
h n O 3 • •
d Eh © a] * i 00 00 m p p O' I oo
© (-4 a '— cn O' o cn 5- P P a
5 ft§
p ■* 3 Vi ©
p
CQ d ©
Q< © ©
• * 3J
°g vn 00 c- VT\ r a p O'
©I Q
♦ Vi
3 o
3S O on cn to o m o OP
(3 a p 'o
CQ>ft c cn 3 I—I cv ©
«H ft © ft
a u
Q 3 p 33
O h © p cn bo
® g cv cv cn a ©
© ft * § © ©
© d
o
3
00 cn o p p
3
© ©
«n 00
3-d ft © 3 o
opo a
© © ©
n 33
„a
p p
^© 3 a 3 • • o• cn 00 ra cn 00 ©
© o O
c- m
ft
> 3g S H
Vi
cm Vi TJ
«
►
cv CO'
nO
O rH
cn cv a
00 o
cv
ON
a 3 a
•rt
© ©
©
3 O ft SI §
U ^ ;fJ8c " UT\ cn p cv 00 P O' C-, i>
-P n
© 3
3 » © O rH
a -p
o ©
ft 04 I I
cv -5
4 O'
3
ra
3
00
cn
* 3
O'
cn
o
00
p o.
^ *©
-p a p
© © Vi O
O O o •
ft ft
i—i CM ra ra p p p
a a
33
ft • « • •
r-
• • © d
o
o o c~ ir \ cn p On O' £*• a u
p o « co p cv cn rH C
M P oo
*2 a 58 p
© C
O * © © © o
■B
p rH ft
CV ft O II OP
a
p © P
cd o• cv cn rH O
• -vf
• p p
©
-P O O
» • • •
O P
P o
CM
CQ d ft Vl 51 cn 't 00
P o
ft § 3 o ©
•al (H ® *
© ft o f t p *
9
© PS t- CM U*\ cn ra o• O O' s 3§*
a at • • • • • • 9 0 0
© o irv cn o oo o 00 cv ra h tih
© *r ra- p CM cn rH p CV p P 0 00
a cr
H
»H ©V
Vi
33'
3i Q g n 00
5 3
ra
c*-
p
>a- p
55 3 4 fc
00
p
a' b'!
u p
n h o
©
p
a
o
CO P P
1 S3
a ton
< CV P m ra 00
p
ft *
- 121 -
reproductions on the Initial Test (figures 1 and 2) and four of the
reproductions on the Retest (figures A, 1, 2, and 6) result in differ
ences significant at the 5% level, the fact remains that the mean
length of the figures reproduced by the normals exceeds the mean length
of the neurotic reproductions in six of the nine figures on the Initial
Test and eight of the nine figures on the Retest. Only the differences
for figures 1 and 2 are, however, statistically significant at the %
level on both tests.
If the comparison is made not between the reproductions of
the normals and neurotics but between the reproductions of either
group and the actual lengths of the stimulus figures, some interest
ing facts emerge. As Table VII and Figure 3 indicate, only three of
the nine reproductions (figures 5, 7, and 8) are made longer than the
stimulus figures by both groups on the Initial Test, and this trend
persists on the Retest except for figure 8, which is made equal in
length to the stimulus figure by the Neurotics. Figure 1 has the
greatest absolute decrease (20 mm. and 24 mm. for the controls on the
two tests and 31 mm. and 37 mm. for the neurotics), but in relative
terms figures A, 1, and 3 are decreased by nearly equal amounts (rough
ly, about one-fourth of the original length). By way of contrast,
figure 5 is consistently made longer by both groups on both tests, the
relative increase ranging from 3056 to 3656. Figure 4, on the other
hand, is made almost identical in length with the stimulus figure by
both the normals and the neurotics on the Initial Test and Retest.
There is also a tendency for both groups to make the Retest reproduc
tions somewhat shorter than the Initial Test reproductions.

- 125 -
O m o
-28
O
♦33
e •H cv cv H
a H -P i 1 1 i ♦
9)
®
u ®
O oi EH ® II cn
1
-37
-27
*n
•17 -11
O
♦11
a a
m ® as
rH
1 i ♦
<*H Cl
•20
o ® Os o CO
o 9 cn rH
I* 5
Comparison of the Actual Lengths of the Stimulus Figures with the Mean Lengths
A
p O
rH
CO cv Os sO cn
09 I I H> ♦
®
p
of the Figure Reproductions on the Initial Test and Retest for the
®
c*S
H P ■
Ig
EH ® '
SS
CO
cn
W
Os
CO
O
rH
m
cn %
CO
o
rH
OS
cn £
o.
u CO m co Q O
p O CV cn ->* to
o a
Total Control and Total Neurotic Groups
rH rH
eh o •
o
•&« m cn cn in cv sO \D
«9
CV cv i—I rH cn 9Hi
0 rH P I i i I l ^ I Ht
a) 58
®
Uu oiO H 2 cn rH c» H CV C- CO CO
as rH cn S
a
M ®
a) I i i I . 1 , * Hi
fH
9-1 O
o o sO c- o cn o (4
rH
O rH in rH CV rH
p U I i i I l ♦ I
a 1° 5
O "S a
e5 H O
O CO O Os CO CO CO ®
CV r—t
I I I I ♦ + 0
o ■6
43
H
Os ca
52 CO rH c- Vf\ sO
$ *O ©
cn O cn a C
% h ©
d * a a
SS §
®
i-4rt Or.
CV sO CO Os v\
1
a a cn rH
rH
a CO o
p
o
g
5
ttf>
3 © ®
5 cv m cn cn ~4-
V\ cn cn cn Oj c^ a
sr*
® LM
P Cn
rH rH
a
CO 5
SP
I CV cn <r\ CO
- 126 -
'' ;tt
A A A A A ' / A ' V /• V / A / . A A A A ' / A A A / A / / / / / / / / / / / A A A ' / / / / A / . A

a a a / a / . a a a a a a a a ' / / / A a a a a a a ' / / / / a ' A a a ' / / / / / / / . a a a ///
00
T T
■i 1111111111111111111111111111111111111111111111111111111111*11111111111111111
I m i 11111111111111111111i i 1111111111
i i i i i i i m r r t i i i r i i \\i i CO
i r 1-r
T T • / A ' / A A A '/. A V .A / . A / . A A A / . ' / /,' /
/.A /.'//. A A A /.A A '/ / / / / , ' / / / / / / / c0)
TOR TOTOK i-l
3
I I I I !I I ,1 Ii i i i : i I II II I i,i, r^
T O i i m I I iiiiilll
I T "I vm I I * I 1 0)
A
03
Q
■ ■ ■ ■ ■ A A A V A 7 A / A /. V A V A A A A A / A / / / A V A A A A A / V A A A A A A A A A A V A A A A A / / / // A A / A / A /
■ ■ ■ ■ ■ A / A A A A A /, V A • /A / A A A A A A '/ A A A / A /. A A ,A , / A A A A A A '/ A / A A A A A A V A ' / A ' / A A A A V A p S o
»H»::»::::s:sss::::::s::s:s::ss::::as::s::ssss«Bs 'Se P o
10 <H
co © P
I i i i i r i t i , i ; 11111 111 nill 111111111111111
II i i > , : 1iii:iiiiiiiiiiiiiiiiiiiiiiim iiiiii
iIIII i■I '",11111 I!IIIIIIIIIIIIIIIIIIIIII I||| II|||i|hi; |
* 1111111** 111111111 n 11111111111><11i i i i i 1111 i t i m 111111111
rs
— ■MwwmW W W ! §
■ I M M
id •H
p rH
z
I / / A A A A A A ' - / , A A A A A A V A, A A A A
' A A A A A A A A A A A / A A AA. A A A /. A 2
JS H P
4
. i n .... o
I I I III:94879349
Us
C
OT S
I ’I I'll!!
© ° H
*15 O P
' A A A / . A A A A A A A A A A A A A A A, A
'. A / , A A A A A A / . A A A / , A A A A A A A *‘ 35 §
1 1 1 I I 1 1 1 1 1 1 1 1 ■ 1 1 1 1 1 1 1 1 1 1 1 1 , I , . I ! 11
^ 5. c«sxao
nini111;im 111111111r,111111ii: ii © © ft
©
H © J3
© h P
■ A A / , A A A A A A A A A A A A A A. A A
■ A A A A A A A A A A A A A A / . A A, A / , 5 5 l fn
! O H O
<0 «h
© © P
xi x : m
i • M 111; | i j H I ! I p p ©
p
l iiilliir!i 9h 9h ©
o om
c
i m n A A A A A A /A / A / . A A A / / / A A A / ' / / . A A A A / A / A A A AA A A A A A A A /.'/A A A A A A A / . A / . ' / / / / A / A A A A o©
■ ■ ■ I A / A / A A A A A A ' / A A A A A / , ' , ' / / / A A A A A A .' / A A A A A A A / . ' / A ' / /,'/A A A '/. A A. A A A A ' / A ' / A A. A / , A A ,' / A, •H
X4T- Ph
©
£
s
I I I i i t i i i i , ) i I I I I I i I I I i I I I I I i : i I 1: 11111 >i 11 i i i n 11111 i m M 1111, 1111 n i i i l l I i n II l i l i i l l l i i i i i i I I I I i l l l l l l I o
■I i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i 11, i i i n i i 11 i i 111111 i i i i m 111111 n 11111! 11 m u i m u 11; i i i 1111 i n h u h
I 1 I | I I 1 I I | I I I I 1 I I I I I I I I I I I 11 I I I I 1 I I k I I I [ |•| \ I |_illllll I i l l
cn
! ■ A A A A A A A ' A A A A A A ' A. A / , A A, A A A A ' / A A A A A A ' / A A /AA. A A A A A A A A A A A. A A A A, A / / / A ©
■ A / . A A A A / A A A / . A / . , A ' / A A A / A A A / . A A ' / A ' / A A A / . A / . ' / A A A / A A A A ' / A ' / A A A A. A A A
t— it -fc t s.
id
m i l m i n i i m u m m 11 m m i i i m m 11 i i m : i i i i 11111 i 111 i i i i I i i i i : i i i i i i iii;iii,iiim ii,iiiim m m im m
<1111111111111111it1111M111111111II1111m 11111m iii11111i i i i i i I i i i ; i i i i i i ii:I:i11111111m m 111:111111(111111111
| I | I |-L|-| !-i—i-j-1 Lt i i i t i i ii
i ii i ** i i i i i \ i i i i i t i i i i I i ii
Just as the reproductions of the controls tend to be longer
than those of the neurotics, so do they tend to exceed the mean height,
as Tables IX and X attest. This tendency is less emphatic for height
than for length, however, since differences in favor of the controls
are found on only five of eight figures-*- (figures 1, 3, 4, 6, 7), al
though the differences are consistent in direction on Initial Test and
Retest. That some of these differences are attributable to chance is,
however, suggested by the fact that only one figure (figure 6) shows a
significant difference at the 5# level on the Initial Test, and only
two figures (figures 3 and 6) discriminate reliably on the Retest. On
ly the differences for figure 6 are statistically significant on both
tests.
Comparing the size of the reproductions with that of the
stimulus figures, it is evident from Table XI and figure 4 that most
of the reproductions exceed the originals in mean height on both tests
and for both criterion groups. This finding is in decided contrast to
that obtained with reference to the variable of length, where the orig
inals exceeded the reproductions for the majority of figures. The re
productions which exhibit the greatest relative increase in height are
figures 2, 3, 5, and 8, but the most striking result is the huge in
crease in figure 2, whose reproduction is three times as high as the
stimulus figure. The reproductions of figures 4, 6, and 7 most nearly
approximate the originals in height, while only figure A is consistently
1. Figure 1 is regarded as unidimensional and hence is not scored for

the height variable.
- 128 -
d * ■
W O ifN
H *H
m cn CM
sO
o cn cn rH
cn
cn
cn -4 O'• 00• CM • O'
■P -P I • • • • •
•H cd o rH O rH o CM o H
h (2 ^
o
* 1 * ♦ 1 ♦ * 1
-4 O rH cn m cn
* ♦ • • •
O 00 O' c- cn O' rH vO Oi m
Ml O H rH rH
rH -P rH
. e
cn 38 II
oo 'O rH ^ cn -4
* H » S5 • • ft •
f^4 » 5 Z CM O' cm cn -p o 00
CO CM cn -4 c=- ir\
4> P. Cm
58 M «
°
<n
o M S 00 o cn oo cn cn
u gig.
a) P 9 S
H oo
H O N
cn C- ITV O' 00
a c-
3
-4
W h
35 o Q 3 -p ii
•H ■— O fl cn cn
O h d •H
O
3
O 9
-P
9 fc O
* 8 * cn
CM a cn
cn 4 £
•
—i
cn
e'
js
Xa9 ?s zB
O 9 vO cn o 00 00 -4
a o -PCm 9d9 H-Pcd 9
O'
9
-4 rH O'
•
vO O'
5 9O
cm
9 CM rH rH
CQ > vO
I9 O 9 <£IMH -TEhJ
Cm
5 1 P" — 00 CM nO r- O'
♦
“i ?§ a 4
OB 3 g ! CM ■* ■4 c-
fl -p cn l> cn I—I
0
•H
39 O
1 9 Cm O r- O iA NO IfN 00
1| O O vO -4 r- O
•
o
♦
o
•
cn
Tf d o
CQ rH
cd
9
> s H Oi rH
6 O H -P +> O CM 3 O II
■—■ cd C-- <n cn oo O' O' to* -4
■a
3
CO
•H rH
H 3 CM
CM
en
cn
c^
tt
d mi
*825
Total Control group
cd
0,0
sO cn c- o 04 SO 4> 00
9
9
P<
W
—* o o
m
cn
H
r- O' O'
a »n
rH
a
H H M I
m cn to vO rH
Z
cn
CM 9 CM
cn $ 4 pi cn
c-
ii
CM cn cn sO oo
(*,
- 129 -
rH
© * __ ®
O O to O' cn o
P P to
*
CV £ to
•
-t
•
to to 55
P P 1 • 4
P at H rH CM rH rH CM o o a a
o
X a ♦ I ♦ + I ♦ * I O P
■ •rH
P IH
O ©
3 ■**
• • vO sf to o to• cn p- •-8 §
Q | • • • • • • •
© sO SO a <n 't Q. &
p 8 CQ'— 3 rH o
® rH p p
H 1 ,-*«S o •r|
fl P ITS 5 u II ©
& P o fl P CM to O'
p Ih EH © a • • • • • • • • •
Oh O a X | CM ON o rH to 'O 5b no
CM rH cn V* $ p •rl P
® CtH O
p
5 1 o © ^
Or •*3 2
p ©
o © fl © • • CM
• CO
« to• p
4 '*
4 NO
• p
4 *5
Or cd Q. a a •
43 fl © fl •1 v£> o O' p to o O -P
O o a O p to CQ >- p p rH Ctf
cd X X i—1 o CV
£) c3 © C3 u 55
sr 5 p fl
X-H« to
P fl 5 o o fl to tn CM CM CM
4 to CM •M «
o O p EH o 35 • • • • • • • 4
P fl p o a | m CO <n cn cn P MD 55
m h © o CM H cn '4- 't t» $ P ©
p ® © u ® O
X! P * fl jd P
hop P © 53 TJ
P M a
« o 5 Q • • p 5
a p fl °• § to• rH
• CM
•
CM
• -#
•
c-
4 rH
• CM
4 ©
© © O §
5 o o 5 a
5 ©
XI e X CM CQ'— to -* O' CM '£>
o o tfl ► vO p P P rH •H
5 fl EH CA CD
3 © cn o II ©
p o Pi -fl © X- •P CD
9 o p © g a x—H £*- m P- o• sO cn NO ■s4
i 0) 9
P © CM S • • • • 4 • 0H
w p p <n CO cn 'O
fl CO p rH CM rH cn £ S3 p
o ®n o
P p Pi
1 *
p « « P
cd a fl • ♦ o •
P 5 oO P Q,
► © o a
•0 cn
• -<*■
• to
• cn
• '£>
• p to
4 P fl
® jo p • •
Q p o p 2 CQ — M3 o. to O' o cn
X $ p rH H 2 Si
T) a © 5 Q i
—i
b o o o CV £ p tt 3) O
cd P Eh © i> x-** O P
"O ra P X ♦ P rH to O' P- cn ON 'i- p p
fl fl © p a | * • • 4 • • • 4 p o
o a O cn
5 p 5 £ CV cn S3 S3 3 5 !3
CQ p p o ®
© cd _a
fl o ©
■a P
© 8 • •
Pi x**"*la a
•Q
CM
•
cv
4
cn
•
CM
4 'O
•
to
•
O' CM
•
s $
w Or 4 ® o t
fl © CQ w NO *A O' O' rH o h Eh X
at a * rH rH p O no
® oo cm «
S rH rH > 1
© •x-N
•
43 3
a 25 H
• on O'
•
'0
•
N0
• • O' O
cn CM O' o> sal
* J cv £ cn cn 3 vO S3 £ P.o
mho
®
3P H
.S 5
as
cm m to sO to
fa
- JJU -
■a
Ji5
w fl S
VHP fl
V / . V 6 • / '/ / . V / . ' / A / -V / ■ •./-./ ■ft

/.V/.'/- /■•/' //.V/.
V/,V /,'/<'/'//.V/-'//,'//>’
i I I I I I I II I I I I I I
■ -K=acr » r ■■
■■■E=ak-M^K^HTBI
I I I I ^ H t : K n i 3 n « l l l ii11111,11111111 m iii!ti11f■■1111!i;;111:11-11 ,
■ ■■R - »■ ■ T M ■ ■ IiIIIIIIIIIIIIIIIIIIIIIIIII M II I
i I II,I I
II I 1 I . I II
.
©CP
55J
V t Oh ©
rw'-ni* o oa
■■crfiar
§
S.
u0)
I III I a
111:ml'11111
i uy •H
fa
'/'/.'//.'//,V A V //V //V ///'t’

/.'/■/.'//.v /.V
V / , '/ / , '/ /'//O /.'//■/.'/ ////.• / / / 7
fhH±hhtttrrtiiJr\ x©
-19
♦217
♦11
o
♦21
1 % 1
«\ c^v o orv
♦ 9
-<*
♦217 ♦13
1 ♦ 1 1 ♦
CM
-11
♦22
CM o CM
CM
1 ♦ ♦ ♦
nO rH A0 o H CM
n
J3
i ♦ ♦ 1 ♦ ♦ ♦
22
19
30
17
4 3
0> c^\
e*\
«> r^N On CM rH vO O^
H **>
I C M ■3 I ♦ ♦ ♦ *
♦
't V\ H O H c»> ~4-

rH
I ♦ I ♦ + + +
S in
rH
t- CM CM K\
CM
C*- AO H
CM ®
w w
I
5 » * ♦ ♦ ♦ ♦
ra fl
-P o -P
j3-h a
oo-P o
1
■h g o
-<t C*\ V© H On vtn -t Cf\ ®
rH
® p I ’+ ♦ ♦ ♦ ♦ ♦ ♦ 0
W 'O H
O eg ■6
H h -P
« a. o ■P
3£ ** 01
CM
O CM <
D
C
M
<z$ cd
o>
® p 0
d>
,
tM l*« 5
o ® On <Q ><\ jf\ CM C-
CM C^ u-\ h
0
a tp a
•P o ®
fn >
'•H
1 V
00
o C- V© D- ®
o
I i CM CM cf\ a
ra
-p
■§»
rH
®
XI
CM fr\ 3
AO *a|
&H ♦
- 132 -
made flatter than the model on both tests. The discrepancies between
Initial and Retest are somewhat less for height than for length (see
figures 3 and 4), and figure U Is most like the model with respect to
both variables.
Since many of the differences In length and height of the
individual figures were consistent in direction on the Initial Test
and the Retest for the Total Control and Total Neurotic groups, an ef
fort was made to take into account individual consistencies in size
on a given test record by treating size as a global sign. According
ly, the 10th and 90th percentiles for the Total Control group were
computed for each figure with respect to the distributions for length
and height. With regard to length, any reproduction equalling or fall
ing below the 10th percentile of the norm group (the Total Control)
was designated as a "short" figure, and any reproduction equalling or
exceeding the 90th percentile was designated as a "long" figure. Simi
larly, a reproduction at or below the 10th percentile in height was
called a "flat" figure, and one at or above the 90th percentile was re
ferred to as a "tall" figure. Tabulations were then made of the num
ber of short, long, flat, and tall reproductions appearing in the Ini
tial and Retest records of each subject, and frequency distributions
were prepared for the three control sub-groups, the Total Control
group, and the Total Neurotic group. These distributions appear in
Tables XII, XIII, XIV, XV, XVI, XVII, XVIII, and XIX.
Inspection of these tables suggested that the major considera
tion was not the number of times a deviant reproduction occurred but
- 133 -
TABLE XII
Percentage Distribution of Short Figures (Reproductions at or Belov

the 10th Percentile of the Total Control Group in Length)
for the Criterion Groups on the Initial Test
Group
(1) (2) (3) (4) (5)
Belov 1 above 2 above Total Total
No. of 70 70 70 Control Neurotic
Short . (N = 155) (N = 68) (N = 62) (N = 285) (N = 108)
Figures i 1 * t i
8 0.7 0.0 0.0 0.4 0.0
7 0.0 0.0 1.6 0.A 0.9
6 1.3 0.0 0.0 0.7 2.8
5 1.3 1.5 1.6 1.4 1.9
4 3.9 4.4 1.6 3.5 7.4
3 2.6 2.9 8.1 3.9 2.8
2 7.7 11.8 14.5 10.2 13.0
1 23.2 22.1 9.7 20.0 29.6
0 59.3 57.4 62.9 59.7 41.7
Total 100.0 100.0 100.0 100.0 100.0
TABLE XIII
Percentage Distribution of Short Figures (Reproductions at or Belov

for the Criterion Groups on the Retest
Group
(1) (2) (3) (4) (5)
Short (N - 155) (N = 68) (N » 62) (N - 285) (N - 108)
Figures % % i % %
9 0.7 0.0 0.0 0.4 0.9

8 0.7 0.0 1.6 0.7 0.0
7 0.0 1.5 0.0 0.4 0.9
6 1.3 0.0 3.2 1.4 4.6
5 2.6 4.4 1.6 2.8 5.6
4 1.9 1.5 6.5 2.8 5.6
3 3.9 0.0 3.2 2.8 9.3
2 14.8 7.4 9.7 11.9 16.7
1 18.7 29.4 21.0 21.8 22.2
0 55.5 55.9 53.2. 55.1 34.3
Total 100.0 100.0 100.0 100.0 100.0
- 134 -
TABLE XIV
Percentage Distribution of Long Figures (Reproductions at or Above

Group
(1) (2) (3) (4) (5)
Long (N - 155) (» = 68) (N = 62) (N = 285) (N = 108)
Figures % . % . % % i
7 0.0 0.0 1.6 0.4 0.0

6 1.3 0.0 1.6 1.1 0.9
5 1.9 0.0 0.0 1.1 3.7
4 2.6 1.5 4.8 2.8 4.6
3 5.2 7.4 9.7 6.7 9.3
2 9.0 10.3 12,9 10.2 9.3
1 23.2. 28.0 22.6 24.2. 22.2
0 56.8 53.0 46.8 53.7 50.0
Total 100.0 100.0 100.0 100.0 100.0
TABLE XV
Percentage Distribution of Long Figures (Reproductions at or Above

(1) (2) (3) (4) (5)

S?
ii
Long (N = 155) (N = 62) (N = 285) (N = 108)

Figures % % % . % .... % ....
7 0.7 0.0 0.0 0.4 0.0

6 0.0 0.0 1.6 0.4 0.0
5 0.7 0.0 3.2 1.1 1.9
4 3.2 2.9 3.2 3.2 0.9
3 2.6 5.9 4.8 3.9 7.4
2 12.3 16.2 9.7 12.6 14.8
1 25.2 22.1 32.3 26.0 19.5
0 55.5 53.0 45.2 52.7 55.6
Total 100.0 100.0 100.0 100.0 100.0
- 135 -
TABLE XVI
Percentage Distribution of Flat Figures (Reproductions at or Below

the 10th Percentile of the Total Control Group in Height)
Group
(1) (2) (3) U) (5)
Flat _ (N_= 155) (N = 68) (N = 62) (N = 285) (N = 108)
Fieures % . __ % .... . % % .. %
8 1.3 0.0 0.0 0.7 0.0
7 0.0 0.0 0.0 0.0 0.0
6 0.7 0.0 0.0 0.4 2.8
5 2.6 0.0 1.6 1.8 4.6
4 3.9 1.5 1.6 2.8 5.6
3 6.5 1.5 9.7 6.0 4.6
2 12.9 11.8 3.2 10.5 8.3
1 27.7 23.5 21.0 25.3 30.6
0 44.5 61.8 62.9 52.7 43.5
Total 100.0 100.0 100.0 100.0 100.0
TABLE XVII
Percentage Distribution of Flat Figures (Reproductions at or Below

Group
(1) (2) (3) (4) (5)
Flat (N * 155) (N - 68) (N - 62) (N = 285) (N = 108)
Figures % _ %. ... % % % ...
8 0.7 0.0 0.0 0.4 0.0

7 0.0 1.5 1.6 0.7 0.0
6 0.7 0.0 0.0 0.4 1.9
5 1.9 0.0 3.2 1.8 4.6
4 5.2 0.0 1.6 3.2 5.6
3 5.8 4.4 11.3 6.7 7.4
2 12.3 4«4 8.1 9.5 9.3
1 23.9 22.1 22.6 23.2 26.9
0 49.7 67.7 51.6 54.4 44.5
Total 100.0 100.0 100.0 100.0 100.0
- 136 -
TABLE XVIII
Percentage Distribution of Tall Figures (Reproductions at or Above

Group
(1) (2) (3) (4) (5)
Tall (N - 155) .__(« - 68) (N * 62) (N - 285) (N - 108)
Figures % % * %. %
7 0.0 0.0 0.0 0.0 0.9
6 0.0 1.5 0.0 0.4 0.0
5 1.3 0.0 1.6 1.1 0.9
4 3.2 0.0 1.6 2.1 3.7
3 5.8 5.9 11.3 7.0 5.6
2 13.6 13.2 8.1 12.3 13.0
1 18.7 22.1 29.0 21.8 25.0
0 57.4 57.4 48*4 55.5 50.9
Total 100.0 100.0 100.0 100.0 100.0
TABLE XIX
Percentage Distribution of Tall Figures (Reproductions at or Above

Group
(1) (2) (3) (4) (5)
Tall (N - 155) (N = 68) (N = 62) (N » 285) (N = 108)
Figures % % t i %
5 0.7 1.5 0.0 0.7 0.0

4 1.3 0.0 1.6 1.1 2.8
3 4.5 4*4 9.7 5.6 11.1
2 11.0 10.3 17.7 12.3 11.1
1 25.8 28.0 9.7 22.8 18.5
0 56.8 55.9 61.3 57.5 56.5
Total 100.0 100.0 100.0 100.0 100.0
- 137 -
rather the mere presence of a single Instance In the test record of a
short, long, flat, or tall reproduction.On the Initial Test, for ex
ample, 58.3$ of the Total Neurotic group had test records with one or
more short figures as contrasted with 40.3$ of the Total Control group;
on the Retest, the percentages are 65.7$and44.9$> respectively. The
critical ratios of these differences are3.21 and 3.85 for the Initial
and Retest, respectively. Only the differences for the short figures
are significant, however, the differences for the long, flat, and tall
figures failing to reach the 5$ confidence level. These findings are
summarized in Tables XXVII and XXVIII, in which the percentages of sub
jects in each of the criterion groups having at least one short, long,
flat, or tall reproduction are given.
Normal and neurotic subjects cannot be distinguished on the
basis of sequence, as Tables XX and XXI clearly demonstrate. About
three-fifths of either group reproduce the figures in a logical, sys
tematic order parallelling the order of the presentation of the stimu
lus cards on the Initial Test (8 sequential connections), and this
proportion rises to four-fifths on the Retest. Most subjects profit
from their initial experience in placing the reproductions on the page
and produce an arrangement on the second trial which is more nearly
in accord with the demands of the objective situation. Of interest
also is the relative scarcity of test records with the loose or con
fused sequence said to be characteristic of psychotic records. If a
confused sequence is defined as the presence of 4 or less connections,
only 4.3$ of the Total Control and 1.9$ of the Total Neurotic group
present records which would be so classified on the Initial Test, and

- 138 -
TABLE XX
Percentage Distribution of the Number of Sequential Connections

Group
(1) (2) (3) (4) (5)
Number 70 70 70 Control Neurotic
of (N = 155) (N = 68) (N = 62) (N - 285). . (N = 108)
Connections % . % . % % *
8 60.0 58.8 66.1 61.1 61.1

7 23.9 25.0 12.9 21.8 21.3
6 7.1 10.3 9.7 8.4 9.3
5 3.2 4.4 8.1 4.6 6.5
4 4.5 0.0 1.6 2.8 1.9
3 0.7 0.0 0.0 0.4 0.0
2 0.7 1.5 0.0 0.7 0.0
1 0.0 0.0 1.6 0.4 0.0
Total 100.0 100.0 100.0 100 JO 100.0
TABLE XXI
Percentage Distribution of the Number of Sequential Connections

Group
(1) (2) (3) (4) (5)
Number 70 70 70 Control Neurotic
of (N = 155) (N = 68) (N = 62) (N = 285) (N - 108).
Connections % * % « ■I
8 76.1 73.6 83.9 77.2. 81.5

7 19.4 22.1 11.3 18.3 12.0
6 3.9 2.9 3.2. 3.5 3.7
5 0.0 1.5 0.0 0.4 0.0
4 0.7 0.0 0.0 0.4 0.9
3 0.0 0.0 0.0 0.0 0.9
2 0.0 0.0 1.6 0.4 0.9
1 0.0 0.0 0.0 0.0 0.0
Total 100.0 100.0 100.0 100.0 • 100.0
- 139 -
this percentage drops to 0.8$ and 2.7$, respectively, for the Retest.
If this definition is broadened to include records with 6 or less con
nections, less than 20$ of the criterion groups on the Initial Test
and 5$ on the Retest would be so classified. If loose sequence rep
resents a "confusional feature0, as some writers attest, one can only
conclude that neurotics are no more "confused" than normals.
The average time required to reproduce each figure on the Ini
tial Test and Retest is given for all the criterion groups in Tables
XXII and XXIII. All of the differences are small, 17 of the 18 dif
ferences between the Total Control and Total Neurotic groups being in
significant at the 5% level. The one difference that is significant
at the 5% level (that for figure 5 on the Initial Test) fails to hold
up on the Retest. The variabilities of the time distributions are con
siderable, the standard deviations being almost as large as the medians
themselves. On the average, less time is required to reproduce each
figure on the second trial, the Retest medians in all instances being
smaller than the Initial Test medians for both neurotics and controls,
as figure 5 shows.
If total time to reproduce all the test figures rather than the
time per reproduction is the focal point, it can be seen from Table
XXIV that the neurotics take 22 seconds more than the normals, on the
average, to accomplish the Initial Test and 8 seconds more to complete
the Retest. The difference between the Total Control and Total Neurot
ic groups on the Initial Test is almost significant at the 5$ level,
but the difference is palpably insignificant for the Retest. If only

. 140 -
O
•P
-P
in
0
o H -* r*
m 3 o vO 3 <n 8 H
? 1 A
o H H o CM H O H 55
a> 8
Medians of the Time Required to Reproduce Each Figure on the Initial Test
&
and Total Neurotic Groups and Critical Ratios of the Differences between
for the Criterion Groups plus Standard Deviations for the Total Control
=s iH
b O
O t.-
.
i
Cm -P
o
•P od
r- m ■v?m 'O O' h O
•—» a! o H
-P o
H•
vO o o m t- o r~ o ®O
CM 00 H•CM• CM• CM•
Medians of the Total Control and Total Neurotic Groups
O' O' O' m in i- i H o 00

in -P
—' o 003
• •
in cn
» •
rn m
• • •
CM rP o
•
O' O'
•
I
• •
—II—I 'O o ah
a> m h CM H iH CM H CM H CM CM S 3
* a o
to £-i
J5
H ‘ f-i
r-t O 00 o m vp in in ->*m o m m m \D in r~ m H O o
^ q} h CM oo in £- oo r - in CM O CM -t CM CM ^ O' CM O O rP <0 tp
-P -P • • • • • • • • • • • •
o a (i O' on H m o cn rP I> cn oo O' cn cm in h 5
H OO a H cn h CM H rH CM H CM CM ®
-p 00
s> e
o
to to
® «P
Oi -P
> sO • CM Oi & O o 00 m m (0 »
O o O' O' if\ 00 CM 'O
1 cn % r- it © • • * • P5
f-i a) CO rr- r- CM o O' m -P
C3 55 w H m CM 3 !— I CM
CM ®
5
4* -P
®
05
> • vrv in tn CM >• M5 O cn 00 5
O O• O O' CM rH O' in r- on
c\i S P © • • • • • • • • 3 to
a) QQ O' o O' i
—I CM O' tn irv o ®
3 tn —I
1 rH CM r4 CM rH •p -p
-P d
n-l o
U <H
°9 1<H
W in
—» O O H
h
w
h
®
r»
||
O'
c*\
o
vO
in
o
Si
O'
r>
O'
CM
e*-
CM
O'
c-
in *®In-WaP
m I
—1 cn CM CM c\t O
a «H (0
e O
43 <p
a.
•a.
H
s® 3°. »aa «• 'aoo• T
£ a 5 Q 3 Q.
CO
a co a co a co a co aJ co• a co a co •a^
a co a co CQ
O
CM m 'O oo
(*l *
- 141 -
* A minus sign before the critical ratio indicates that the time is longer for the Neurotic
1—1
a * un x© x© x© X© Hf
o
•H
o
*H 1
IN-
• 3. IP dN UN O CO 2
group. A plus sign indicates that the time is longer for the Total Control group.
-P O o O i
—1 o o
H ♦ * 1 ♦ 1 1
h « ■
—*
-P o o
o .
CQ, 1-1 P o O O' O dN O' d\ Qxx© O' x© O' c- O CO O' rH
UNCO UN CU CNi CO ON CU C- H rH UN O iH CNi
ir\ 5 g H • • ■ •
5-t a * C-sO O CO Ox CO CU
E ® £ 3 34 44 CU iH 34
st as
rH UN
i-l O CO H O' Ox CM CO -T tO rH O dN O' UN O' cu O' UN
CU vO O- CO UN O', cr\ CNi CO dN x© (—i CO rH
• • » • . • • •
dN O £4
CO x© OrH xO x© dN xO Ox t" > dN O CU
H O CM H i—1rH 3 4 CU H
O SB
4> Oi
> X© O' O UN dN x© o ox dN
o O d\ O O' o cu• dx
^ oA R
dN ©
CO H f- CO rH CM
rH CNI CU
cu 25
s s d\ O "sr cu CO ■HT O' dN

O'' o o dx o O' x© CO x©.
oT J§ ^ * • • •
*—" aj
25
rH
H
xO
CU
xO
ri
t-
I
—I
UN
—I
1 iOx
—I CU
i
—i
x© O ITS ow VN
* u> 31 <? CNi
•
U*\
•
IT\
• •s•r iC\
•
Oi
•
U“\
•
CO \0
04
O r- IfN O oi
'— ID
«
3 4 rH rH Oi rH
E-«Q «
VC «
3 S3
C Q C Q R Q flo flO S3 Pi R « S O RQ
<H HI <H •n • >n • T3 *
s co a co a co
iq • in •
a co a co a
*
co
Xt
a
•
co
id • T) •
a co a co
ojo
CO CO CQ
§ g -3
•H P -P
CM dN UN x© CO
1ft
a
- 14* -
r ■ ”-1— "” “ “ ■■ - "■ n “■n — inn n nnn pPnnn p p p n * p ~p _
““
pi ■■■
■-
■
■ ■"
-----—_ _—_—-___ __ _ ___ _ _ __ __ __ ■n__ __ __ _r— _—, ____ ___
- — -at *- F— 2r——- ——————“-------- -------------■--- --F --------- --- ----
-----=I- | _ F — —__ _ — _ _ ___ —_ __ _ ___ _ _ _ _ _
t3 -*r ——n
““— “ B—— til -—--—— — _--------- -- -- ---
-JS jF on H r“ ~ ■““”“
i 9
J ■ sr z 1
— n
|VU n T 2 r—n=n n
— _ U L r - - LC
Figure 5 • Comparison of the Median Times Required to Reproduce Each Stimulus

“j b ---.r t " h-H;
j.-------- j------ TTr 1 nI 1
□Ed —iQ d sd d t
■■■■■■■■■■
i i i r i i ^ i r i K i■■■■■■■■■■in
i l'/y/y
■■rk»KrBn*BP~H l/.y/.yyv"
Figure on the Initial Test and the Retest for the

iiE=ap*ap>nai ■

■■■■I
■■■■I ■■■■■■■
MM'/-'. //.yyvy ’/y
■/. /////, /
/. y
■ ■ ■ / / / / V ////- /,//'//,
■■■■■■■■■■a
■■■■■■■■■■■i■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■
■ I
■râia m/'-a ■■■■■■■■■ ■■■■■■■■■■■■■■■■■■■■■■— ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ —
■rnEaa ■”./■ ■■■■■■■■■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ a
■r a n a//a ■■■■■■■■■
■ ■■■■»■■■■•■«■■■■■■■■■
■■■■■■■■■■■■■■■■■■■■■■a ■ ■ ■ ■ ■ ■ ■ ■ '/'/ ' / / '/ / /,'//,'/ /,'/ /■■',/ '/ / '/ /. '//,V /.'//,//. ■/ y ■/ yV
■■■■■■■■■■■■■■■■■■■■■■■ ■ ■ ■ ■ ■ ■ ■ a v/.yy vy yy y y / y /' / //'//•/////,'// y y /.'//,'//.
!■■■■
■■■■■■■■■■■■■■■■■a
!■■/'/Yv/'/ ''/v/ v/'y /''■/'/•/ ''//''/‘v/-•//'/
•o
r r
ri~r
!■■'/'' / y ' / y / y / - ' / y-'/y / y y y y y / y / y /.'//.y y v y ' / y v y / y / . y /.y y y y / y / y / . y / . y y y

! ■ ■ v /. y / . y / , y y v y ' / y ' / y / y / y / y / . y y y y ' / y y y / y /. y /, y / . y y / y / y / y / . y / . y y y y y y y
(SPM0390 O II
- 143 -
Medians of the Total Time Required to Reproduce All the Figures on the Initial Test Alone
P
o O O
ON 00
p P I
■h
fn O §
I S3
O
and the Retest Alone for the Criterion Groups plus Standard Deviations for the
Total Control and Total Neurotic Groups and Critical Ratios of the Differences
©
5
U
o
o
Between Medians of the Total Control and Total Neurotic Groups
•H 00
-P O ON O «
0 r-i o• o • If>rr>
8P
O CM o
1 II o> 00 H
SB »
xa
a>
H •S
p
O 00 • ONsO O' --t
U CM O
0)
00 CM
• •
nOU*\
• •
«
-P
O CQ 00 CM O cfs
g " 4 00 2"
O 55 rH -P
cd
CO
© CM
Q>
•P
t> nO • 05
— o p O o
C'N P t - II ©
n—' aj CQ
23 *
CM •—
•p
cd
—
©
>
o
•H flO
cm <a c -
' ■ a
II
00
NO
o
•
Q>
CQ
0
1 NO
CP\
rH
Cd fH
O 00
•rl
r
H — P H
•H O
U h
° S3
© O
i*. Ip 00
,0R K lf\
’<D II 23
o o
23 ■P Eh
JS ©
S
3 4*
£T3 Q• £■xj<=•
* •H P
a o
S CO a co 0h
©
■3 p
SS
ca
p £ ©
ra P
© ©
EH (U *
- 144 -
very short total times are considered, approximately one-third (33*7%)
of the neurotics require less than 140 seconds to complete the Initial
Test; for the Retest the figures are 49*1$ and 45*4%, respectively.
The variable of cohesion was analyzed by tabulating the num
ber of subjects in each criterion group using varying amounts of space
to reproduce the figures. These frequency distributions, expressed
in percentages, appear in Tables XXV and XXVI for the Initial Test
and the Retest. Inspection of these distributions indicates that the
maximum discrimination can be obtained by establishing a dichotomy be
tween subjects requiring half a page or less and those utilizing more
-than a half page. On the Initial Test, for example, 8.3% of the Total
Neurotic group crowd their reproductions into half the page as con
trasted with one-fourth that proportion (2.1%) for the Total Control
group; on the Retest, the figures are 13.0% and 5«0%, respectively.
Both of these differences are significant at the 5% level.
Tables XXVII and XXVIII summarize the findings for the glob
al signs on the two tests. "Connections" is redefined as the pres
ence of 7 or 8 connections; "Cohesion: refers to the use of half
the page or less to represent the reproductions; "Time: 139" indicates
that the total time to reproduce the figures is less than 140 seconds;
"Short Figures" refers to the presence of one or more reproductions
equalling or falling below the 10th percentile of the Total Control
group in length; "Long Figures" refers to the presence of one or more
reproductions at or above the 90th percentile of this norm group in
length; and "Flat Figures" and "Tall Figures" refer to the presence
of at least one reproduction at or beyond the limits of the 10th and

- U5 -
TABLE XXV
Percentage Distribution of the Amount of Space Filled (Cohesion)

Group
(1) (2) (3) (4) (5)
Pages (N » 155) (N - 68) (N = 62) (N = 285) (N = 108)
Used % . % % i %
Over 2 0.0 1.5 0.0 0.4 0.0

2 4.5 4*4 3.2 4.2. 4.6
1-2/3 4.5 4*4 9.7 5.6 12.0
1-1/2 12.3 13.2 19.4 14.0 7.4
1-1/3 14.2 13.2 11.3 13.3 8.3
1 51.0 48.5 37.1 47.4 51.9
2/3 11.6 13.2 16.1 13.0 7.4
1/2 1.9 1.5 3.2 2.1 8.3
1/3 0.0 0.0 0.0 0.0 0.0
Total 100.0 100.0 100.0 100.0 100.0
TABLE XXVI
Percentage Distribution of the Amount of Space Filled (Cohesion)

Group
(1) (2) (3) (4) (5)
Pages (N = 155) (N = 68) (N « 62) (N = 285) (N = 108)
Used %... % _ % % %
Over 2 0.0 0.0 0.0 0.0 0.0
2 0.7 0.0 4.8 1.4 3.7
1-2/3 4.5 2.9 4*8 4.2 2.8
1-1/2 5.8 7.4 11.3 7.4 1.9
1-1/3 7.1 7.4 4.8 6.7 6.5
1 56.8 58.8 46.8 55.1 50.9
2/3 20.6 17.7 22.6 20.4 21.3
1/2 3.9 5.9 4*8 4.6 11.1
1/3 0.7 0.0 0.0 0.4 1.9
Total 100.0 100.0 100.0 100.0 100.0
- 146 -
1o *o ir>
O H
H CM
• • CM
• •
8
33:53 CM <r\ CM C\
o
I I I *>
o
•H Cd
)
—I-P O
— > fid 0 H sr«lff'OCÔM)HvO(J'Cf(in
w\ -P • • • • • • • • • • • •
o N'j<^CM40O'OO"tOt0 0'
Eh 1 ■OOr-HHHM'vvAirv-Mt’ H
a a
CQ
H IT
H O tO
-5 fid fn CM O'
"t -P -P • • CM• • • • •\ 1ACO
• • •ri rl >
• •
— 'O g N CMC*>q^^^Cj-^CMHCM«n
a) H H O CO rH i—I C*\
•H
-P ®
O 2d
>O CM
vO Of'C'C'rlCMrH'-O'CVCMCM'T
o o • ••• •«••••••
> i ■as
— ' fid C'- H cîocniA cm
m m CM ^
iCQ t 58 tXiCMC^'4'ÔOCMsOif\Ovr\'<r
• • • • • < • • • • • •
CM ,2 M C*\vO 3 H N P- tO N H O r| !M
COHH 'Tvpff\'4- m
. «>
ITi O c n C » \ O M M in\D O'P-!>C0
u .2 P"
H rl SH • • • • • • •
CAO 3 ^ Q ^ i A O J H O H s O
• • • • a
®
CQ
H to r l r l ->t ^ VA ^«T fA
iCJJ a
JQVO
5>H d 4h
•H cd O O
m -P o
o ®
m Eh H O
T1 *rl
sr 3 * -P
d ® S
a
id En &h m H H H a o *
q> «h T3 fl -P g
o H O h bO-P
rH UOOflgj <d Oh
o •H *41 O H g i
ffiCQCOPh eh a « 4
a o o h
- H7 -
-
© X®!
H P
in O
OT• CM
•
m CM ©
I I
o
CT* W
©
Pi ©
<Sh
a
O— ©
5, H tO Pi an
H P O
r-* S O r l
if\0'Ottc>
• • » • «,sfirtioCk
• ! • • •O'0'a'
• • o O
a
«\P 6 m m on CM >f\M iriCMT; rl m m ©
^ o 3 a — H H
O' vO "T irvnJ- H '-i‘
1 m o
© Eh (D S3
S5 25 9
o &
o 3
o o
©
H vr H Pi
HOOT i n -<}■ CM C" O ' on -X> « O i - M O H WJ
» • • • • • • • • • • •
a ii l f \ TO O 0 ' * > l A N n 0 « M J ' © ©
O' H -T •'3’ 'J
H O
o a "’ g,
P
© ©
M P ©
M
CM
©
» P
vO CMmmCMOTOT'tC-OTOOTOT © ©
5 in
I cn PQ© £ n U > H ' 0 m ' 0 - 4 ’O T « ' 4 ,O s f ' 0
P
©
O
H
Pi O' h sr in -4- m -t O T3
CO
O
CM
H fl
TJ H
C _
•* a
O H©
H
© P
> $ f-mtcoHomHinoa'O © ©
cm j? R
ai
n W4 Vf\OHOsfCÔJ^tHOvnc*\
0s rH rH Vf\ «H ft
H
a ctJ
O c oa
§
H
Pi Qt
• • • • • • • • • • • • ©
B8 inv0r-jO^|ÔC^'*O'4C4 3
a .8
’© O' 3*8
CP © H O
as U P & OT
O O W>
Ch Pj Pi
© p .. O
POO
©
a pfjc*-
bOH © <P
•H d O
8 tQ-P o
•H
CO
o ©
9^ ©
C © -P 1
CD
o
5 £
C3 o o
o o pci CO CQ eh a
$
- 148 -
90th percentiles, respectively, in height. The remaining entries are
defined in the scoring manual in Chapter IV. Only four of the twelve
signs (Second Sheet, Short Figures, Cohesion: -g, and Time: (139) are
significant at the 5% level on the Initial Test, and this number is
reduced to two (Short Figures, Cohesion: g) on the Retest. The or
der of placement of the reproductions on the page (Connections, Col
lision, Rigid Sequence), numbering the reproductions, placing the re
productions in compartments, variations in height, and excessive length
all turn out to be non-discriminating variables. Most consistently
significant are spatial compression and shortness of length of the re
productions.
B. Individual Figure Signs on the Initial Test and the Retest.
The percentage incidence of the individual figure scoring
signs on the Initial Test and the Retest is given in Tables XXIX and XXX,
but for purposes of the immediately ensuing discussion, which will focus
on the general characteristics of the reproductions common to both nor
mal and neurotic subjects rather than upon differentiating features,
the data are summarized in more convenient form in Tables XXIX-A and
XXX-A, which present an analysis of the 67 scoring categories into in
dividual figure scoring signs occurring in 20$ or more and 5$ or less
of the cases in both the Total Control and Total Neurotic groups on the
initial and final tests. Among those scoring categories which occur
very infrequently are tremulous lines, erasures, more than one attempt
at reproducing a figure, use of guide points or lines, retracing, pro
jection retracing, curvature flattening in figures 4 and 5, perseveration,

- 149 -
closure of open figures, dissociation, simplification, distortions
(except for figures 7 and 8), total rotation, part rotation, reversals ,
paper rotations, card rotations, fragmentations, omissions, recounting
of test figures, recounting of reproductions, curvature distortion in
figure 5, and continuations. Scoring categories which are frequently
represented (20$ or more of the cases) are overshooting lines, substi
tutions (especially dots into dashes or filled-in loops), numeration,
wave irregularities in figure 6, asymmetry, displacements, angulation
difficulties, horizontal irregularities in figure 2, opening of closed
figures (figures A, 7, and 8), counting of test figures and reproduc
tions, leg shortening in figure 5, projection angulation, and sagging
lines. Disproportions are common in figures A and 7 but rare in figures
6 and 8. Contiguity difficulties are frequent in figures A and 8 but
rare in 7. Corrections are rare in figures 1, 2, 3, and 5 but common
in 7 and 8. Distortions are rare in all figures except 7 and 8, where
they attain an incidence of 10-15$. îth regard to individual figure
signs, those which occur most frequently (roughtly, 40$ or more of the
cases) are overshooting lines in figures A, 7, and 8; incorrect number
of dots in figure 5; wave irregularities in figure 6j asymmetry in
figures 4, 7, and 8; displacement in figures 6 and 7j angulation dif
ficulties in figure 2; opening of inclosed spaces in figure 7, count
ing of test figure in figure 2; making the left hexagon first in
figure 7} leg shortening in figure 5; projection angulation in figures
7 and 8; and line sag in figures A, 7, and 8.

- 150 -
1
•rl
I
—I d
P © in VTi v0 o
01 u % CM O « i o
« © •rl ■H
P
I
r^v
•
CM
•H I I o
rH Vi O Vi
flj d O O «
33 S P
a ©©
a p p
M © fl O *£ &
■© OS •H <0
© O rH P O
ri O '
mc^-^vOt~ir»t-'i,\oa)0'0'«rvoeocôc,«a'<J'C^vOflo bp ©
jfl © *H d
P * «H SS £ •s«! vOnt-tj 3 CM m t~s0 O NO o C M l A O m O O m M C M •H ©
C*> CM CM Vi ©
d d EH © O P
o © ho Z Z O ©
H *H © O
© © CO •H
© T3 •
6)v! ©
■H -H H
CO Q ©
i
—1m
UO © S3 o to l*\©©
• • • J•"•* C
• ^•N•© •
© O• C•t•© I
•M ^•>•C V•!•O r
• l•' f
• OI^ *
© 0*' 3&S,
•dH J
Pd PU V. Q 3
Vi CM
P ' O M t M q q i n ^ i n o r i M H o Q (< \o h o o h c m m 3 WH
H ,-d © w O
EH § '
cn CM CM O
O *H ,S P © © Vi
o o P o z ®dP
co © © P rH d
« O.0Q © PiC5
S
© o d O
,*H O H •H <J
)+ > h O
© CM 'S H
> vO CM©CM©'t©©^0© 4 -s O v O O O C O O O 'O O m C M C M
• • • • • • • • • • • • I 3 .$
?
O O
mJd c- l O OsfOOrlOMX^t^ ^ 0 .o
O
H a) P © ■— © H -3- cr, cr\ CM H •H O
P H ©
33 2 J CM * g
t
•s
H*■*H» ©
Pm
o
► H as rl *rl d
■h o © ©P P
•d <H 3 o o
d © © © to
M rHd PO -P >
—* o o
3 sag,
■rl © .H
« 0.64 © CM £> C- I
w r oj
(M C M iA » N - t © N ir \0 'O s f O O O H O H O O O H Z «
cm c^v CM m
5 m rl
Q. d H *
© bO
ch a © © jtj d
o o 3*3
Vi H
©O O © • *5
H O°
H ©
d dP O I A ( *' \I t ^ H H C -C T 'O tO rr'iO O O 'C '-C T .C M O C A O O O O H < 2 & ©
3 *o
o £ i* O'strid'OSWrjl'NifHinHOHlciO 333
o PS cq
H Cd rH H CM ©
d d a <h
S tfo
© O '3O3 H ©
H Vi
© H ©
HO H Cm O
© d
I I a © ©
d, H TJ
« d* ■don
P
Vi O
h
« <H O >rS
s d ^ «h|^ i^rid ^ ^ ^ <d <ri to o t>to<>o6în3tStocr>
----------- ■* “
^ ^ dl
■ O' o H cm m 3 m
ae o
c
a, O CO CMCMCMrr,-^~T'^’'4-'t'J »© m m m m m 3
CO *
- 151 -
ontinued)
CM O' C- CO o
• CM
• vO• <r\C~-
• •
CM CM CM >n cm
I I I l l
o
•H co
H 4» O. C^vO 0'0'0'0''4’if\0'CC\0'0'0'0'0'0 O' CO O' '•4'O' O' O' CO C*\O'
■o • • • • • • • • • • • • • • • • • • • • • • • • • •
vr\ H TiO O O O 0 '0 '0 0 ' W M f \ O O r l O O C ' I H O r f\ H c f\0' O r l
IH §<D CM lf\ CM CM CM CM CM rl M5 C^
3 S3
H »T
H O CQ CMC0rlI>CCHOiff\C0'^tClrlMHiAM'OC'QrlCMsfC'C^'0(N!
■s* CT CM riOrlHm'OlACO(MryOCMC|N9 0 3 0 a ' rH-iO fc^
|M CM
f\Mm
-4- CM rH CM H
§
O 3
CM
'O I'Mft^'fiNOOHHHMOOCMWOO'OOrlHOO'OHCO
C*\ J8£ H ' O O H ! r> O r l C O ' O O O C s H O C ri<,\ D O r H O \ 0 \0 0 7 ' O i a ) M
'aJ H -<)• CM H CM CM H H CM CM
CM 3
© CO
t> MO MO''0 0 0 0'HO'C"tN+O Q i f t M O O O O « N O M O M M
CM JS ^ | wi t'CMlf\OO(MCMlft't>(MinOH'tOOOOC0l'\OCMif\sf't
CM H FA CM H C^CM
(continued)
-— , 0 0 0'cr\'>ff-0't-ir>C0HOE^-4C-0'CMC'-OC*-Oir>-'i-I>Hcr\CM0'
• • • • • • • • • • • • • • • • • • • • • • • • I ©
H H C~
—•• o
■&e n H O ' O H O ' t i c w O O ' d ' C O O r l ^ O O O O'OCOO'OsOFNH
CM C\ CM H CM C^CM
m
TABLE XXIX
Is
o >h
t> CO
CMCtNJAmJi £> O' H
C" C~ t-
"!f H
»A
H r l r l H r l H H r l H H H H H H n n n
rl rl CM 'JlTlKMCMA'O'O'O
cm m
-O
CO
- 152 -
(continued)
ctf \c CO sO O' H o
o o C^M CM CM Q CM
3 3 1 • • • •
CM CM c*\rr\ CM
♦ ♦ I I I
to A ' *
^• O 0
•
'• O•O 0• '• ^ •v 0• O• r l
•
O• C •r\•C M• f•f "• C•0 '• 0•' O• O•r !• ©•O O• O•t 'a O *P M* O
S I ■br C O d o b O H f - O ' COMrlHACr ^ CfMO'Osl-OHOO
lrl^ CQ rH sO UA i
H »r
H O 00
'f l{ h CM »TvCOC^'4--^C'CM»rvO'sOCOCOvO'4-C"i,
• • • • • • • • • • • • • • • •
C^s}-C^C,
• • • •
'0'NfsfCT'0'OsOt»\
• • • • • • • a
-P -P
^ O fl II ■sr ' O r l O r l O O M O ' O C ' l S ' © I ' O ’f i O O O O O ' r l O O >tvO '•O ~-t IT\
rH CM KMrl Nrl Hr-i 'TCf>
H O
o a
©
> H c M ' 0 \ O O O C M C M C ' O t t i f \ 0 ' O O O v O O O C s H O O C r\ K M ( M \ O K
§
o ■OR v f l ^ H r i O O n s t ^ H M | n ^ O O O H O O p C O O O O Ll ! \ n H M
s CM
CM >f\ CM ■sfC«A
©
> ©irioooiAOO')'Ocf\c)of-a)if\ooooiAcfNt'OOH©©s!-st
CM ,8 R aR H H O oorirag'oqMff'rjHcnooo rr\ -sf O O -
st(*>00 'tsf
N inh CM H
(continued)
o* P i
nr C^rr
•
Mt~0'C^t^vO'OC~Cf\>CCMO'Of-C^r~C^tr'OCf\C-t-rHC~COCOCO
• • • • • • • • • • • • • • • • • • • • • • • • • a *
MjR C Q H O H O O C M H C ^ q c M H H O C J 'O O O O c n C 'jO O C rO O M M A irv
• © tn'JrlHCMCM •'i’C'A
«
TABLE XXIX
§p
•rl rH rH rH
I I I CMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCMCM
8
O CO
sAci-O'C0d)H-^»ANAc^C0C^CQr!|C'lAtf\'ici-C0Qr^C\ICC\'<f
CO rH cm cr\c»Mc'\c*,vc^c'Nc''iHrirnrnr\«MnvrM>rv'OsO'0'£>sO
- 153 -
(continued)
rH
a)
•oH *oH I
. $ o
-P -P ^ rc\ NO - 4-
+ I I
O
CTNOC'-ONOOCVCMCOCMC'-Ot'-CMC'yHOOC-NOONOOOC'-NONOONî-tiO
■BS H O ( f \ O ( \ WCN
I ^ Cf\J' r-4
i f\J O i«—
fI i O1i-4/rr\M f \r4
C 0 H O O < n - ^ O O O O m u M A O i r \^ l >
H vr>
H O CO C^-<'?C,
~'i'HCMMNO‘r\'4,cîACC\0'4,0'00'<}''^H'i'CN-'»i,ONCMCOCMH
aft O O O O H C O t - C M N O N p O C M C O O N N O O O H H O H O O O C^ -<J-CM rl O'
CM rl CM H H CM H H
o *
® cm
> N0 'OOQOvO^CVC'HOONOrlOO'NOO'OCMOOOKsO^NOCMHn
§ ^ Oo pI> II
2 1— • aJ 'A HOOOH^sf s f ' t O j O f f fl
H CM CM
mOjÔOrlnOOO^Hrlrl^'OH
f'Vrl
C3 CM *
£ O O O itcO it\CMnO O tf\ CM CM CO O n O O -<*■On O KMfl O O « O ' O «0 «!

CM .Eg O O O rl CM C°\■ ' O O C ^ C O r l . O r j t O l A O O N t C M O r l r l O O f i O C ' I O t O O O
CM rl CM CM rl -«T
(continued)
^ B C~ E'-C- C N C N i n W ' O W O ' O I^M) H \ r l t N O t ,O t N ( f\ O O O v O ( 0 ( M < \ ' #

H H<D g
^ 'e* O O O O O m ^ N l f \ O j O ^ O \ D > O O O O O H O O O ( N | K M ^ O ( 0
CQ
TABLE XXIX
o .
s CM CM
CMcr\mcnor\rr\r^c^c<>rcv (r \ ( r \ ( ' > c 9\r < - \r f> c r ^ r \c N N \c ^ c r \ c n c '> c ^ \ c < ^ c ^ c < ^ ( r \
gc/Nûk-digogaSp^cMC^co-.
CTn r l CM - tiO rIiA^cnuîrvirilrNv^SNd
flo CT» g lC M c A ^ iA ^ g tO ^ rl
O CO H r l rl rl CM C M C
CMMft\
C ^(A C^\ (A ( A s j
C^C^C^C^^ A 1A 1A l A I A L A I A l A vO vft
CO
- 154 -
(continued)
3o o ir> «A CM C-
•H £ O• lAvO
+3 -p i • •
cm CM C- c^
£ ^4 I I 4* +
o
•H CO
•
H -P O
— >ij O H (00'0'if\M)0'0"0'0C'OC'iO(f
« • • • • • • • • • • • • •\CMt*t^0'OffvOC'0'f'0"0'0'tC'
• • • • • • • • • • » • » • •
»fN -P U WJ ( M O O ' O l f \ r l O K M A ' 0 O f “ N " ^ O t t ) ' 0 O O ( ^ O 0 p\Htr>Hlf\'t^'O
— HO 3
® a i
—t ^ rt CO H CM rl r
—I CM
a 53
H ir
H O CO H s f f - O t C r l O-^-rH O C O t n O C O ' - O C M W t O ' O s t f - ' J W M ^ ' t i A s O
-tj h N .
Ht -P -P H O O t ,' C M H O E ^ C O C rNCMCy' CMOOs C q H s f c M C M O O O C r\ OC M m c » > C O
o a i H rf\H N H H H H H
H o
o a
CM
g > sO \0 0'Of“CMOOtCPM>N'Qr'C'(ri\OcnCMM5rl OOOlTkvOC-HOCM
3 o P
o c^>& t> I
h '
—a)
OrlO'nOO'iH t ' ^ H O ' t ' r i o i r i n H t t O O O ' O H0
H H i n H H C M H
O
*^'
— HO O ^ CM
a a
CM
2 $1 i A O O O ' ' J O O ' ? ' t N O N « O H ( rls(COÔ'COiftOiftO'0'tcy!^C\!

cm 5 R i r I O O N M O 0 ^ s t M ) 0 r | r l ( M 0 O' rH CM rl rl 3 H C M O C O t n £ > C < ^
cd H'Tr-ICMr-ICMr-i rH (<\rl rl
5*5
(continued)
to IT C»C^C~C^CMT'OC^Hj-t>OCMvOOOCÔCMCMCOOrÔO'C,*Cf\rlCMC>-
r~T <2 9 H
■—l r<dt r * a 1* O O O C^ H rl O O to Oc^f^fr\o O W r l ^ c ^ ^ O H O H O O v O n t Q
H CNi H rl rn rH rH
CO
a
TABLE XXIX
tan
Is
O *r C M C ^ p J m s O C ^ O i H lf\ '■_____
£> C~7 CO CJ^ O .HC
. d* >
SQr
O ir- l«s<ff il of CÔQCT
t S C Q C >iH C~- 0s
“ >T\CM‘
O CO CM CM CM CMCMC<^r<'vr<'\vt~'T-'T-^'<t-'3''-*ir\U’\ t-
CO
- 155 -
(continued)
(0 CM OTl
< O'
o o H• &• £• c«-\ 3• F*•
•H iH •
+> +> >* CM CM CM CM CM cr\
gs-1 I ♦ I I I I I
C M f " i O '0 < O t t O ' > r » N i f M T 'O r H O r - f t '( T ' < r i ( J > O f f ' s^C|' ( l f \ '4 ’sO sO d \ o
ITi Q n > H O C M ^ IC J ^ O C J O O ® H O O c*>
lH i0) H ir> CM - t CM CM sO
as
H
H
O tO CÔ'nCM t ' l f i ' O C ' i O t ' s H O i f i C ' t - v O ( \ i « J ( y . C ^ s f O t t r l l C \ O C O t ' C O O
C f S j H 6ft • • • • • • • • • • • • • • • • • • • • • • • » • • ♦ • • .
U NC Ôr H^ ^ O O iO sJ Hf HqH^Fn- ^ H NOOi r^l rHl ^ O OrHO O O H t O ^ C4
^ CH M O t CN*ir\
O^
-g g »
o a
CM ^ (Js
vO
I
o
^QP
c«
CM
tO (vM A O '^ ( N i O 'r * * 0 ,» ( > ^ 0 \ O H O ,* H s O O H nO O O
•
*40 vO
sr n
"4
vrtQ |— | (V^ (>| rt\ A) r \) r r \ r» u vArw vA t r» m _ i r \ r \
O'OjNÔrlvO
HrlCO H
vO • t f IT\ <N sO UN.-I
t(V l i t * __i >#i
i _j .#•*
CMH
' * c f \ H wvs^
to
t
c*>>r\
58 COrr\Oi't>A(J'0(ri('ttlOC^sl'OrlC'(<'>-<i'OCOOOOlC\CO>C\^'CilOst>J
o n O O O O v O C ' H N O O O v O ( O t " t O ^ ' t O s t ’0(0
i 0 0 OHtOvO -d-H H O -4-
**> H CM H H vO H n H CM. H CM t n
as
(Continued)
O'O'OO ® O O'C'fwC rl (’ ■lOi f-tt f'C' O OOO c f\'0 CMO (0

.ggS
tH t> •aft rri(^'*010rlC'0(AxO(^H(r \(ntr\Olf\0('iOOr'-,’ ''-l'
O r tin r'"'
H tn t^ O lftO
•C®Q N CMr i CMr I rH r l CMrH rH rH CMr l CM IA
2!
TABLE XXIX
bo
si- IT\WN VTV
^SJ I I I I J
o cp\ u> O vO C- O' TI' 1£\m!>t'OUN^C'tOO'HC'O'rlC'J ^C^CT'c!>rHCMcAlAt>-
O CQ ^ £> 00 HHr-5r-iCMCMCMCMCMCMcr\-^--^vf\Vf\»rN>f\vcS'®s0s0v0'0'0
tQ
- 156 -
(continued)
oj _ in| ON t- o NO
o o P P• ONNO Vf\ cr\ CM
•h *h i • • • • • •
p P U*\ CM - 't CM CM m C M
•H d *4 I I I+ +• I I
5^'
o.
•rl fiO
rH P O ON<7'NOsi-C‘-ONONOGOOmCOCMONrHC'-ONtOONOu-\OON\Ox|-CMEN-ONNOOGO
OH
v r\P H
^ EoH 3 li HHmC^cnHOCMCMNOC^£r\OsJ-lT\OCMHO(T»CMHmOOCÔt'-4,C
m C M t ' - C ' H H rH CM H rH rH
M
^ rH•
©
as as
H u>|
O 00
*-* cd h CM ^ N O f » M ) O ^ M < ,inOMrs'^NtÔiH'^NftNO'Mcr> > N ' t O ' 0 0
-*P P TS
<» rHCMCNjmr^vOO CMCMf»^rg>rHOJt>OCMrH 0 ^ f ' H ( r lv O tr\O O C O ^ -^
■
— o
Eh § "
O 5ZJ
© CM
> nO NONOcrjCMCMNONONOC'-O'O'O'CMNOrrxO O nO N O r l » 'O f f > f f > C M O O r i O CM
o$ fr\P
• • » * • • • • • • • • • • • • • • • • « • • • • • • « • •
H H r j ^ m r l H H O N ^ N m Q N O O H HN^--tHCNjC^fôONOrH^-
u aS
C3 CM
£ W> i f \
• • •
s f O ' O If ! IA f \ (J- Ol I f ! O NO
• • • • • • • • • • • • •
O O Ifv O >
a * • •
^
*
O CO s i - 4 O O CM (<■> C^
• • • • • « • « *
CM5 O H rH O N fin O rH H jA C M t^ O J O l^ C ^ O C M rl O -^sf O » M M O O ^
'— aS O-
(continued)
vr>
£ or\\0 rr»rH CM C*- O CM CM tf\rH CM cm nO £*- O nO CM O CM CM On UM CM nO C*- O nO On CM
^ o
70
1
• • • • • • « • • • • • • • • • • • • • • • • « • • • * • •
n ssa
iH r-i iH OJ N f- <r\o O CMrH v O s O O r H r H O O O C M r H O CM O H UMTlCM O O O N O
w a> rH CM H t> ir\H H rH H CM H H
CQ
as
TABLE XXIX
no
UMAIAvD\DsOvO'^)vO'nO'^)ÔnO 'O vO vO \O>0nOn0>OvOs0nOnOsOnOnO

£fl| i I i i n IT I li I • J, 1 J J, CM • itvn
I OI EJ"I~ t O QIHIC MIC Mli II
o C M > O in '0 > 0 'H O H C M (^ tO O rH N O C 'rH lf\CMC\
O CQ
CQ
C" C- 00 rH CM CM CM CMCM CMC M ^Û M UMMMAUM UM1 ® nO nO nO nO C* f -
- 157 -
(continued)
at
o O
uN O CM
•rl *H
W»
•
cf
«
\ m• Si•-
P P CM CM CM cr\
I I * I
O
O
•H 00
P O M fo CT> a - C^X H SO IT\ CT> O' sO 00 O 'n O C T ' O C ^ C T ' C T ' 0 s cr \cr\C<‘» 0 0 < YA \ 0 C<\ C*-
-
— • at O H
in P ts« Cs'Cf\Hrl(J'H'M(OiAO'^C^(,MfiOOM30r(HOO<r\CO(T'(T'XOCr>Cci
O H IT\ H vO CM l f \ CM H m H in rr\ \Q m
V
25 53
(
—Iin
r-l O 00
•'-* at U CM C‘*
*iC»
•-»t
•N-X
•iT<•‘
icnCMCMC'-C'-00mCM'tfO000\C'--Mj'rHOv©s00‘
• • • • • • • • • • • • • • • • • • • 'C'-«>^C‘
• • • • -00
•
-4-P P U\M) O O KM rt COM) f- Q C M O 'P 'O Q O ( M N O H C M C M " 4 C M M ) '0 0 CM
^ o fl H - 't ITl CMM O CM rH ' J H C- CM rH O m
H O
O 25
OONOOOOri'O^ff'ÔiC'-inO'OOrlONCM'OOff't^Nu’ iOMftCM
• • • • • • • • • • • • • • • • ■ • • • • • • • l i t * *
•4tnOO'tOOrlrlcn-4rOO-<tr)OOÔrOnA10'Ntt)0''jrlvO(n
•—■cd l f\ vO C^sO cnH H S cM ricO nO M3
CM
®
(7"40lftiC«if>sl,C'iiC\Sfv(MT>H^'Off'OOOMMfiO>C\^’tO(T'HH-4lf\
CM ,2 R etfj O i C ^ O H t J Q HC^Cvm^
QvvOQ-^4^
-HH O
OÔ'C CSMOOO i t O O H H j n2t-
c»t^»-l M j CM^fi A W
O’O ^ H
■
— at 5 cm in cn rH -<■ CM rH m \© vO
(continued)
m
m
—• * o o rH lArl
• • • O• O• 00
• O'C*
• • to
• CT
• 'n•O • •-Mi-•m •O •O' •m •<n• •O'•C i•CM•sO• f00tm * CM*E-*CM
rH rH c- pcnr-jC^mmmt'-cnl
M)r'rlO3W\rH0'lf\rHOI00t'l*\OfOCMiHOrlpc^rjt'«MAIf\>Cr
•— ® 0 -t CM M3 CM rH m H >
c*-( cm
M rico
H cn m
m- t
CQ
as
TABLE XXIX
00
• H fl j
Ih hi
O *H
O CQ
i•n'1 J J J rH
I. J. J I J 1 I. I i I ^ J L J I. ! i I J. ! I. I ! 1
CM 00 O n O rH cn O VJnO 00 OK rH invA 00 CM cn -MtuS in n£>
r) rl N CM CM C ' t n n ' t M ’M ' t lft'O'5'5 f-CN S S
CQ
- 158 -
rH
(0 IT On V\
O O ON o £•
*H *H 1 • •
-P «P CA CM CM
•H cd ^4 I I I
u p3
a
o .
H -P C§
-— ■- a t O H
t'-ONC'-PNO'HOCOO'flCNff'tN^ff'OOt-tlOîiN'TUMfi
tft-P f ^ O ' O tO H ^ O 'itO t ^ Q O ^ A O O O N f ^ O i r l r l C ^ O'sO
-'r0 \Q H c ^ H n ’ H rH ( ^ n O i a
H 5ID 1
a a
rH IP
1
—t O <0 H^COC^rHHHC^C^»Avr%^r«*%HCsi^C^\Dî*rHHs£>rHO
• • • • • • • • • • » * « • • • • • • • • • • •
b CNi
3 1 -p V! HOHlNH'd'H'inNO'OÔNHAO'OifNrlNHsl'WNt^
n£
>
~ v-n rH ON H ON CM CfNVN
EH g "
o a
O CNi
O O C M I > - O c A O I > C ^ H i H i - C M >T\n£> O G O r - i t f \ \ O C M C O H C O < ' r\
HN r2 ^ I Vi O O C N O N O r l O M l O O ' C N <r" O H s f O t t ) ' O H t n u N W ' * r l
s—^ C0 ITV nO rH C N H (Ni CM ICNnO
1
CNi 55
£ 3 ( A l A C O - t O ' C A O C M t - C M f ' - m O O O ■ ^■ O s r «
• l « • • • • • • • • • • • -
O «N C \i l i N S f O i
cm 2 ^ Vi , *^~>~'0 \OC^crNC'^rHOOO'iJ-OC,-OOOrH'>qrHC'-vO
00 C'- CM Q
■— cd CCN mD CCNnO
«I
(concluded)
IT
ft IT c*-yO'4’
• • • ^•t •“'•
4’O>O't-00'^’
• • • • • t^C>c<’
* • •\O'f-000NO«O'Cr
* • • • • • • \•cfN•CNi
♦ON
Vi
nO -5 H CM CA H CM ir\ -V
■3 *
a
TABLE XXIX
00
S ooopoocpoocoopooooooopopoôotyepooooopopco 00 00 00 00 00 00 00 00
»-■ 6 I III
o A O' O r-l CM A 0Q A O A o nO i> 00 i-j CM VfNsO A tO COM
MCA ^ JC N
O co H r-i H CM CMCMCrNCA-Hi-'<f'i-'T>AIAcfMA'l)'2>C' C*-C*-C“-C-
CO
- 159 -
1
P
CO O O IO X O l l l l l l l l l l l K O XII P
d
o
or less; - - not relevant

o o ox O I I I I I I I I I I I X X X O X I
SO o o o IO I I I I I Ix XIIIOXI I
IT I o o O I I I IX I I IO O X O I
I I I I I I I I I I ox X
CO I o o OIIIX X I I I I I IO I IX
incidence of 20$ or more; o » incidence of %

CM I o o O l l l l l l l X I I I I I I I I IX I
i o o i o i i i x IX I I I I I I I I I I I
ox I I I I I I I I I I I IX X XII
m
•35
E-t
(not scored for that figure).
® ■s4O3
P
P O P T)
m b rl fH I
aJ P gl o ®
b„pX_I O_ .P in H
bjf
01 5 J. J> ^ J. 2
a O O O OP
§ s fl t) "O O
$CS &
e I
® Is %
o § ® o ® a
P a) O P
a
P P SP d
a) Pi p § >* d V
P P bO O P P P P P ^ a p o p
h ® «s5 O a -rl pa pa pd pa pa pp 0 ® P P b
o ®H OP P 01 o d P p
o CU ® (0 P O O ••HI cd O a5 bo a) a
01 CM CQ . P
m p
®aJ ® P P P P P h r l P H tt
O 4O 4P d o
•H aJ H ®P OP P P P P m o d §p d R
pi d 5»®inddddd p p o a o
faÔOSSPkCQCQCQCOcas; q o o <3 o X
iA'OMlOÔrjNînvOl>«OOHCM(«MQ\Otw « a ' O r l N
H rlrlrtrlrlrH rlC M C M tM C M C M C M C M C M tM (*\ffM r\
- 160 -
(continued)
eo I I I X I I I O O O I O O O O | I I | | | | | o
I I I X I O O O O O O O O O O I I I I I I I I I I X
NO I I I I I I I I I O O O I I O O O O O I O O I
IT I I I I I I O I O O O O I O O O O I I I X O O I X
I I I O O O O O O O O O O O O I I I I I I I I
Cu
co I I O O I I I I O O O I I O O O O O O I O O I I
(N X O I I I I O O O I I O O O O O O I X X
I I I O O O I I I I O O O I I O O O I O I I X X O i l
I I I I I I I X I O O O O O O O O O O O O I I I I I I I I O
d
® o
H a
i? cd® o
•H © ^3
p CQ 3 fl -ii rT 3
® u
•rl
o
hi
®
p
CQ 1
h0
d
d
®
p
-p 'p
CQ u pP Chi f
l -p P
o © cu
2 0 _ H
H ®
©H3 p p © ®
p cd 01 1
® 3 © p at d
d d § © CU -P P
13
o
o
CHS
o
d •H
d o d d cu © o © «
p
cu o s' o Ti O O p p
o hi •H O O 01 *H o d o p •H P H *H S.Q p
fl HOP p o ® ® •rl a) -p aJ -P -P m ©
•H P a) d cu ra h O P o H d cu
p c-
d P o o d as 3 d w -o o o 60 h0 *rl H
o O ID C f Idr-1®
l -p o d O P H H ■ d p P
o O p t> P o p p ® rl TJ cd P P •S h d d
CO i—i H H N 5o ® I -P o - © P * o o
cd a} as •H ® ©
u u u
d h ©
© H ©
as p
n -P p
© od a a ® d "d o o m P
s s S op®
CU CU IP W H CU
CU C U H
O CO o C
o a)
OQ H CU L B.I33
O D P O Q P
° O 88
O O 04 « « H
rr\'Ct’crcvOt^tOO'OH'4’>OvO^<C)CJ'OHOiCA->tirisOC^tOCXOHOJcr\-^-ir»
r^c^c«%c»>crvcncri'4''4-'T>i,Xf^t'^-^-uMf\vf\iAcf\ircirv»nvnif\vOvO'OvOvO'0
- 161 -
co I IO IX XXIII
XI IX XXI
vO I I I I I
in IO I Io
X I
IO I I I I I
oi IO I I IO I
O I I I IO I
I I I IX I I I
(concluded)
O'
M
a>
+>
aj
O
o
TABLE XXIX-A
o
CO
do DO13
1-3 CO
vO O N O
- 162 -
o
nt
•rl t H
P -P
C*1
•
cr\
CM
vO
•
I
P
8o
CM
I I
o
©
p© a
®
« © P
d u
© P as •rl «0
o r-l P O vO CO CT't* O I"- IC1vO rr\ <7'<r\ (j\ cj\ oo os cj\
5 * •H r-v crt O H ........... I I 4 > • •
tf\CMOC'\i-lcr\sOO-4,eO«OC*\('f>rlO CM O rl O
to st;E-4 ©
£ if\ r*1 rr\ fTj <r\
©
o 6 ! 3 S5
d •rl
© CO
U
© a>
CO
hOP cd H «rt
a q xi r-l O tO tOiAst r l s f r l \ O W ( f\Ov C J ' S f C M O C ' © O r l W
•rl 0
•H U
. (d Sh cm
p p
• • • • • • • • • • • • • • • • • • •
N N O C M O tf\ir » -M '0 's ts f> -< o O r H O H H
O o d II a* cr\ rr, CM CM
O fr4 O
CO O S5
01 ©
a 9*®
o u
I 5-4 o
CM
E«4 P ° H 2 vO C J v O O CM O P l t O
o a> • • • • • • • • • • i•C x•t t•M •
O vO vO O O vO
• * • ♦ •
H (2 £> S 2
•— ' as
C \ H O 0 1 O Q ' * i f » r c H i f \ i f t ^ O H r l O H H
-T CTN cr\ f r \ CM
J H
T3 OS CM
3$
•H P
65 «
©
«
H O © to
0) M 3O *-P >
O O
sO Wi l f \ O CT'MltOCT''4CMCOtO O l s l O o m o o o
6-* cd CM .Q C- II ■b* r lO ir c H 'O N C M r ) M ® if t 't O O r-l O O CM
as rr\ or\ cn fr*
2!
Cm 04 5
O ra a)
04
® a
o o o
a h
®o IT|
■n "S •O
•h a o i
—I CM CM t- O O cr\H ITN'O « M f \ l f l O ' 0 ! M J ' 0 ( rvrf\
q o o — N. O O aO fl
d -m
M fn H H C" R Wl
-— ©
f ^ C ^ O O O O i > i A O > t ‘r > ^ ^ O O H O r l H
<*\ -4- H (V
© +> 5 co a
U)>H o
M H
5
d °
© «>
25 Jf g <X$ <si <sl «al «3l <aj <d <3l <£ <d <ca2 <sl «3J
©
04 o Jx vi cJ-tfc * ri. ri.tl <4 4 3 1 2 3 1 3 I
O CO H r-l CMCMCMCrv'd’-T-<t-SJ''5-'^»fMA
CO
- 163 -
(continued)
vO CA ir \ o CA
S o vO O O CA NO
• • • • •
CM CM CMC M CM
1 1 1 1
O
i-ttO
H -P O
. (d O H
V\ 4> H vO CM C M O C M C M O O v O s O r H r H C A C A O r i r l O ' ^ O C O CM >fr H C-
w EO-t 3
®
» rH tf\ rlHmH^rl rl ri CM CM
as as
rH W H
rH o to (f \ r l ' O l C \ 0 ' O t * f - s f r l i C \ f * ( O C M O < O s f C M C f \ - * '■ 0 vO C - E" CM
IH C
M • • • • • • • • • • • • • • • • • • • • • • • I *
-P >ACMCMCMEÔOOOO'CMrr\CMi-IOCMOHCMOtOHOO'©
O a ii rH CA CM CM rH CM rH rl rl rl rl rH
Eh o
o as
CM
§ > vO '0 '0 ^ (M ® O O O O rl® rlrf\l» \O ® O K l'tO s f(< \® O
• • • • « • • • • • • • • • • • • • • • • • • • t*-
•
o cr\ ,Q rlrlrl^nOOOrlMOItOrlrlOHOMO'OC'rlrlOC'
h •— td H ~>t CM CM rH rH rH iH CM H H
C3 CM
® to
j
» eg O* V\
• CA
• '■•i’lA
• O• i•r\ U\
• M\
• CA
• CM
• H• vO
• C-
• O• • O
• CA
• •to •O •rH •OA•m •«\•H
CM ^ ^ ‘O r l O M H O r l H ^ O ' O O ' O M O M O O C OOO' OrlrlO'
— ' cd H VA CM rI rl rl CM rn r—I i
— IrH rH
ts
(continued)
VT
IT
H lAvO C M C ^ f ' O f ' t ' f ' O C M v O M C ' O ' O C ' ^ O C ' t t i ’N O O " CM
h '®5 P
• v D O irHs f HOJO ' O O O WH O t AO(I Hr\flQO'OC\iOOHO*^CNiOO
Ol iHp-irHrH rH
PQ
TABLE XXX
00
d
•H <a$<J«s}<j^|rHrHrHrH rH rH rH rH rH rH i—| i—I
h §1 I I I I I J I I I I till
O -rl \sO £*•O v\ v A ^ c U to o<cq1 a'j,
s __ rH CM CA O
O CO
CO
rH rH rH rH (V lA a vo \o vo r~
- 164 -
(continued)
at IT
o O
I
•H *ri
-P -P ^
I
O
O
•H «0
H -P O O' O O' Cl C^ O' If\©
^ II O r • • • • • • • • • • 0• '•© •C '• O•O •> •
O O f \ N © C ' H O t - O
• • • • * • • • •
W-P h W O O rlC M « 0 © M H iA O lf\H O O ffll*\O n M fiM c ^ rlO m O
o 3 it I—I CM ITv cr\ CM H cn C*'* H
Eh 0)
SB SB]
H >T
rl Ott f"'M40<0'Olf\ri>OrlOC,
• • • • • • • ’• • • • •N'^sfrlO'0'sfOOCMCM'0'^Ntsf
• • • • • • • • • # • ! • •
O fl II OOrlO'Me)fflt''0^0'OOOrlOr|Ot'C'(^'JHOOO
CM -~t CM H H CM CM H
Eh O
o as
£ Si © O C M ( ,l O ( ' I H H O U \ O C M O O ' 0 H i A O f f ' W O t t 0 ' O O O

1 J ^ it t t t » H M O n O O r l © ' 0 O n i n O M N o o o
H O n H H( MCMfsf
fn * cd CM 0"\ CM H
C3 CM *
© t#j
> <0 OOU\«0Hv0t0«0OCMOs|-OO0"<|-t-OOOU'\W\CMir\OO
CM ,8 R O 0 H H t M l 0 ® r H t t ( r\ O C » O O ( M ' 4 ^ O ® W r l r | r p\ H O O
w at rl CM M •—IC\i — 1 ‘ •" CM CM ■
t*
(continued)
>r
ir
* l-H C‘*C‘
~cr\'4'CT'0'E-tt)HC\10HC‘
• • • • • « • * • • • • •-C‘•-0\0'OC'-0'f-ICMCMC'\OC-C-
• • • • • • » • • « » •
H0 £ » O O H W C \ r H > l f t t ' l f t O C ' O O O H H O ( ,M f ' l C \ i A O O O O
•® CM -T CM r-l H CM CM H
PP
TABLE XXX
CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM (*>
t 81 1M J II
» • cIi II
r U AI,VvOC^«gO'COCMm»A'£)C'Qr-|CMC<NHtCMsOC~
1 J I I. I I I I 1 J
NO „ _ -.
C-- O' CO
.
O 03
CO
iH CMOAcr\c^c»>fr>(^rfM-4-uMr»»r\«rMrc'3)\OsOvONOl>-
- 165 -
(continued)
Critical
m CO O '
Ratio
c? H cn
l • • •
w cn c\i
i i l
surotic
- 108)
rotal
0 ' m o ' i l > - * t C ‘~ 0 ' r H t ~ C M C M O ' ' O v G O O O O O ' M ' O m t O O

(5)
• • • • • • • • • • • • • • • • • • • • • f t # #
o c 5 £ i S B c^ , f > 0 r 1 c2 . Q 0 0 m o o o o h i * > cc\ o « q > * o
H CM C\| H r-i'-tH H rH r-l
* *
H m
H O CO to o O ' t - > o -sf t o co i f \ p » t ' i n i H > t t > - s | - - s f ' ^ - t o c \ t f - o o c -
? * * J * 1 • •• • • • ♦ •• • •• • •• • • • »
*R H S d 9 C' 1 O f f l ^ O , ^ o c , % r H O O O O H H ^ O ' O ' 0 O
§ CM rH CM rH m i—1
^ w»
o 5B
(N - 62)
2 above
nO O ' l n s O CM H O C" r - C- O CMvO O O O O nO CO nO O M M T in O

70
§
(3)
o V* rH OJ ' J £ { C^lvO O O ' O M ^ O ^ H O O O O r l ' t H O ' O ' O H

h rl rl u H iH
<3
1 - 68)
above
- t CM c n t - C O H O •sJ-'O O ' O ' O ' t U I O O i n O i m C M l t O O ^ K \

70
® ® ® ® ® ® # ^ # e ® # e e e e e e # e % e # # 0
(2)
w O ^ H O > O C ^ i n i n i n O - ^ H O O H O H H H O m I > H
rl rl H rl rl -ifr
_ .
i—i
»
TABLE XXX (continued)
m
in
» H f-m v O 'O i'-o r - o o i> c ^ c n c M t~ i> c n o t> c n c ''im c "ic o c M O
o O
H H M O ^ H C M r-rH O C r'rH C T 'C '-rH c n O O rH O O rH rH -^ -rH in m O
'—' o
® 5
bd
rrvrrvcncncncnrncncncrvcncrycnrncnccvcrvcncncncncncrim rn
•28
O *H 0 ' ' J t f \ ' i I > c i l > t j ' rH CM £ > 0 0 C ^ - C Q H C M A - V l A ' i ^ - C O i i r ^ C M
O CO rl rl rl rl (M C M C M c n r r \ C r \ r r \ ' < J - H j m W M n m M \ m i r \ m ' O \0 vO
CO
- 166 -
(continued)
o o «0
•rl *rl
i—I
•
C"~o
-P -P CM
UO S ' * I
■
— v a) O rl CÔvOO'OsO(nOtOtf\C^rlCVr-ir^CÔxO'CÔNC^r>OOvOC^
ir\ -P H O O ' 4 ,H O « ,
\tt>C\jCMN£l£\^’OjH|’\0rmOcmcmOCmcmOO'4’Cm
—' O 9 H
Eh a>
a a
iH in
rl O tO
rj d h IN • • • • « « • • • • • • • • • • • • # • • • • • • •
O O rl m o ~-tvO O 't'O m O O'r) mt'I'O O O O H O O O r H
O -fg
lI.^ H rnirl CM rlri m n
EH O
o a
Ô'ONOmrlmtOOIrltONvO'JloOHOOO'OOOf'UV
cm J8 ° « -as
I
o
1* Cj
(V
rlOrlmOvOCOrlsflACOmmCMOvOOtOOOOrlOOO'j
H ■Sfr CM CM rl cm H
I % m o O umu\~4-^}-rm-<*t'-tio l> o O' O O O *r»0 C- to

CM % ^ H H O O H H > ' J O H' t (cm
M » lH > sCM
t m > s r o m O O O H O H N -t
tO
W' a!
H 5
er
Vf>
* rH 0 0 0 ' 0 ' 0 ' O H f |M n c W O > r l O O ' t f M N O m O O ' f - O t O ' O
oO
rH E' II O O r l m O N P " O M ( M t k Off'>OHrlMlf\OrlOHOOlArl
a> H cm H CM H rl CM rl
PQ
5
bO
V ^ I" *’ * 1 • • ' I 1J i i i * i l l L » J I * • • J I
O *rl cm cm um-vO C T ' O i - l u m ' s O C - t O O ' O r l c m O r i ' - t i r N C ' - t J Q O ' . H C M r ^ C T '
UC O NO C- H H CM' cm cm cm cm cm cm cm nr -c st n jn fr—
-■*■ «m •tr
-\—urn—m
CO
- 167 -
(c o n tin u e d )
rH '- 'j
at
o o
vrv 00 vO e'
en o cn >n O' -*
cm
• • 3 3 'O 'O cm
•
-4-
•
2 3 ' m on CM CM CM - 4 CM CM CM
I I* I I I I
u
o.
•H 00
H -P Q vO 00 cn in 00 O O' 00 -<tvO m CM O O O' mt*- 4 cc\ O ' O to O ' O 4 rl
*—■» cd O rl
m -P h in C~- O' m CM O O CM O t ' - O ' C n m O O H ' O C M O O O O
o 3 ■ CM -t CM rl rl CM C- cn rl cn
E -t a>
55 SS
i
—Iin
Cn m m r l O in sO 00 O cn m r-l C~ on 00 'O o >4 oo •M’ ^ i n
) • • • • • • •
US NC\iqin>HOCVt'0''t(MvpHO(Mrtr| O O O O O r lC I
s Hi sO r-l rH CM rH H H l> HH rl
o a
£ nO O C- -4- O' C" o o o o O ' i n n n s t o o M O ' m \0 o o m o v o o r i

• •••••
IfHw- aj i o c-o'no'o O'4-Oj'Orjr-jtÔOOJOjp^HOO'OOHHOO
rl rl m
C3
cm 3
a> oo
t> sO M C M s t’ r l O ' O O ' t f ' O O O O C M H O ' mvO C M O ' i n O O O M O O m m
oo
N > Iw HM) ' t C M m O O ' t M H C O n M C M O c n C M rlO O c n O O rlrl
>— ' a t rl CM H rl rl C- CM r l rl
ss
m
in
* rl O' t " CM r l O' O C - ' O c n - ' l ' C N r i r ' - C ' - C ' - r l CMC'-O t > C- O C*NvO
^ o o
H rH R —I C O O ' c n C - r l O O C > q I > C M g ) 0 O t t t t - ^ O O O O ' O O r l C M
(J) CM CM
cq
■5
c?
8
s -t 't'+'-jt’-'tinininininminininmininmininmuMninmin
I I
m CM - ______
m cn i n o - i c^- ck -st- i n 'A t » i i n ' i c ^ o o c ) ' H ' i c ^ 5 ' r i ' 4 , c^cf'
OCO vO r- c— c- oo J H H r l cm cm cm cm cm cm c n - s f - « t - 4 m m m m
CO
- 168 -
(continued)
O O £• 0
vr\ 3•$ • ^3
3 2 ' CM •sj- C
M irv CM
£ (S'41 I 1 I I I I
o
o
•H CO
•P O 0s O O ' O ' K N O 0 " 0 -M vO 0'0>A'0 0'iAO"tt''OsfO'0'0
« o
iftC\lrlHO'(r\ O l f t t ' W \ r l O O O P M ' O t t H D O M t ' O O O
w O gl H CM rH iH vO -t CM P- H
h »
2! 2;
H ir>
H 0(0 OC^CO>tOrHP-rlC'\\OCArl'M'4,« C M s f ( M ' t t X ) ' 0 0 « ) O H INf
! • • • • • • • •
CrlPHOlAlf\Orl^'tlC\rlOHn'frO ' P'«'vnjJNri_lo_l
a v O H H P O r l O O
rH rH 1A CM CA --T rH
'£ § "
o a
0) cm
> vO J> CM CM O s t O O C ^ v O C U O O n t M C M ift'i'O A ' t O O O s O M )
ca JB ^ r-rOcrvCArHCÔOrHi-H^tdOrH^-rAvdcji ICM O' O O O
' aj CM CA H
CM
E s C A O ' O O CM O O O' O' O' ■'*O m O ' C O Nf c*\O i a P P O U N O vrvO

cm'J> It wi OiAOOCACAOCMCMCMP-OHCMCA-sfOcAHCMC'-OHOHO
'd rH rH lf\ r l 1ft M rl
rH ■*
CP
*■ lf'' A P O ' O t O C- C\>0'lftift0'0 C\ C*> CM -Mf'O CA CM CO O M> O £> O
•® ■«* N P H O ^ « H O n \ 0 M H O H O n « O H u w O O C M O O O
rH ir\ CA MO M f r H
m 2;
bo
VMClUMrtlAlftlfMCMft'OvO'O'O'O vQ \© nO vO vO sO vO sD sO sQ \© \0
I I I I I J I I I I I J II.
O -rl O r l N C M A t ' t M t ' O m s O t ' O ' r rl
i O rHCMCACOO'r-l'Ac'-CQrHCM
OC O 'OM)sOvO\OM)t'P-CO rH CM CM CM CM CMCM CA ~-t'H’--J IA >A
CO
- 169 -
(continued)
nO cn CM O' nO
o o O O' CO
• • •
O'
• o
•
33 ' CM CM CM 1
—I CM
3 (S * I I I I
a
o
•H co
H 43 O O O C 0 '0 CM(’>C't'<T\'M,P%0'0 '(r\lCi0'O(Mf\lf\v£)C0 C0 0'0' 0' 0' 0' 0'
oid
m 43 So :^
CM CM CM Ifio CO 1
AM) O't'O'rl rH 'O'O H N W ' O ' O O ' S t C M O O n H O O
— o 3 ii ^ H H r-l rl rl '4‘ M) CM if! CM ih IA rH m
Eh ®
» a
—iHg d
i IP
CM O'-^'O-i’t-m'OCOrH '4,'tC^t^l'ncn>3’CM«0'4-C'- CM(ÔCOtf\rlst!>
.. cm
43 w H H r H r H O O O ^ i n C M C O O O C J ' msO O rl > r j ^ O H O ' H O
•— o a I I r-l i-l cn M) rl M 1 N r l s f S
Eh o
O 25
® CM
> nO l > m O C 'C O CM n n t ' C M t ' M ) O U M f t n P * CMvOO O ' O ' O O ' O O ' O O C*-
cn 5 > 'tO C 's fc r \r-jO O O C rl O 'H O 'Tv'O r l O ' M H O l ’ M M O O r l C M O O O '
>-h cd rH H cn <n r l ICl (M m c n rH Iin in
CM
> $ c o c n m c o o o o 'c o m mco o mco O 'H fcM 'O H m cm O ' o m -•$ O ' m e *
■
«P.'
CM 43
• a)
rjOr-lr—l o o m r —l O H C O O rH cn CM -4-sO o ^ t > > o c n i n o H t > c M H ^
i—I rH i
—I rH rH cn CM -M CM rH in vO
(continued)
IT
VT
'-t'OO'OC-ocncocnO'Cô c*-O'Co cMHO"OC~cot"'OoO'OC-ocn
p^Hl
iH Ha>
^ ■a« C O rH H O 'O O C M -*C M rH C > -O O c n in in C O C M iH O m O 'O O rH O iO O O '
r-i H H H sf in H i n cm vf m
GQ
TABLE XXX
bo
a
•H
8 t-C't-C'-P-C-C^C'-C'-C'-C'-C^C-
f - CO O' O r H n o i f t• 1 J i q> i >o
J I J I I I I I ..........................
O T« invAc^coorHincMpk ini t> ct'OHcm
O CO
CO
m m m n o 'O c- h h h cm cm cm c n c n cn --t -4-'0
- 170 -
'si »r H O O P~ CM C*>
-2.49
o o C" CM CM O■ c^vO
• I • ■ »
P P ' -4-CM CM CM CM CM
I* I I I I
to g , ' *
o
•H CO
64.8
P O i-ltOir\OCT'HsOOCT'CMC~I>l>C'-rc\CMCMO''QOCr\'OOC'0"0
64.8
^ P. P
g H. * O'CMcnmcfNO'lfvHO rîC^(n> tfMvO (M>-OCrlOO"ÔOif\'t
-4’ '•3'rH vO *■$ O \0 H •4 CM ^4 H CM
H ®
S3 Z
67. A
r—I VO
51.2
V\C'-Otf\U'*\OW,\tO-4»f\C-H-4CMtOO'NOrHV'\HC~H'4CMC<'\0\
CMf1O O' cc\ if\(MrlHfr\ ' O C M ^ ' O ' ^ r d C M H 0 ' H ^ C 0 H ' d O ' c ^
CJ o ■§ - * CM rr\ sO -st vO Brlmrlm H
H O
O Z
® CM
66.1
75.8
► vO 0-4t>ir\C-vOCM'OOI>i-H\Oa>HC*>CÔ C ' ^ v O r l r l B M ' m
I
u cd
oo
rr» JOt~ R O
_ 0>
_ O_> -4 O' H c*SH O 0 CO H JA\0 Q S O' cn C" H to to H O' -4\0
H CM CM
CM *
69.1
55.9
►
--N O O 3
CM XI >
H 0 '« 0 0 'H < M O 't0 'C C i> d > d f-1 1 0 0 'M 0 in s fo a t0 if\'d H 0 '
• • • • • • • • • • ■ • • • • • • • • • • • • • • A
CMir\\0ir\O'U'H-4CMQ'44'''tHM)r';sOr-t'4OrHr-|H'4CMCM
rr\ M) '~
if\
cd \5 B ri I*\H cr\ 1—I 1—i CM
(concluded)
vo
VO
43.2
64.5
* ( M J ' « ) O v O > B C ' m ' d r l C r\ C ' l r l B H ( n C '^P^tO IftCPvff'rl CM

rl(f\ W C ' C M C O ( M O r l B f ‘ r l K M 1 0 H f " C M O t t ) H l f \ ' O H r l B c f>
■B r\l cm Ml re\ '■O M) H (p| o' r-l
ffl z
TABLE XXX
M
5u g »>-t'i>c^t^(ototototototototorototototototototato to
74-8
75-8
o 11 1 1 L i l l i 1 1 7 7 1 7 1 i l j i 1 I 1 1 1 I
\0«X>CMC«'V'4vri»f\'sO<T'0 H C M > O C O C O O v O C'-rH >T»nO MNtO. CM
O CO M3 MO c- I> C~- I> Hr-(HCMC\ICC\CO'4'd'd'VMAlO M3-'-2 t> P
CO
- 171 -
%
$
Analysis of the Scoring Categories into Individual Figure Scoring Signs Occurring
00 O O I O X o i l I I I I I I I I I X O X I I I
§
O
in 2056 or More and % or Less of the Cases in Both the Total Control and
O I O X I I I I I I I I I I I X X X I I
®
M
sO O I O I O I I I I I I X O X I I I O X I I I I
P
§
IO O IO I I I X I I 1 0 0 O III
OOOIO I I I I I I I I I OX X I I mM
Total Neurotic Groups on the Retest^
H
IO O IO I I X I I I I I I O I I X I I M
O
ITt
cm IO O IO I I I I I I X I I I I I I I I I X I I
P
0
IO O IO I I X I X I I I I I I I I I I I I I
0
1
O O O O X I I I I I I I I I I I I X X X I I I I
w
ST £
a> M Go
O MH
® dp ap
H iO O
4 O r H H U
m M r-i M I S'S
a) P at O ® d
Mts U jl OilrH o bfl
IO 13
o p tn PM P CM M
w , 13 O O O O O P S3 0 , O
® +J © § alT3 *a xi t3 o O £ P p
p & a P
at a
® o® o ca
H4 W
> a)
o « it m • • •# i t 1 d ® ’S
M
o
S3
as a<
O ® H
® «aj
p
o
to M
§ P P P P P O oO Pat ®M o «
a d a pp O H M
0 •H P P3 P3 ■PP+J
add
o o o
•rl p P d l ^ f
_
0 p
O >•
n
S2a ^ 9
•p
p o n
S P P
o d p p o
M H
:a d
0 o
tJ n
•n
« p
o
o a. ® ca J3 O O P iH p P trl at at p P I I ® o at So at - a O
C
O
1a ^ 01 P P P P P M H P rH P d
®p -§ M co at m 01 01 o ® ® ® PtP at
® at P O
I s Is
w d _ .. .. M 11w
M M 1 « £ ® m 5 5^ i l l ! P O a o a S at
£h W a h o O « CM co co CQCQCQ Ss ts pS E* O O <4 Q Q O p U Ot do X
irv\Of-®(T'OHCN('Jif\vQf'()qOHNCr\>f\'OC'CpO'OHN(r\'t
HHHHHHHHCMCMCMCVCM(MC\iCVCMC^C^cr\cr\rr\
- 172 -
( c o n tin u e d )
to I I I I I X I I I O O O I O O O O I I I I I I I I
I I I X I O O O O O O O O O O I I I I I I I I I I
SO I I I I I t I O O I I O O O O O I O O I
vo I I I I O I O O O O I O O O O I I I O O I
I I I O O O O O O O O O O O O I I | I I I I I
•H
ft co O I I I I O I I O O O O O O O O l O O I
CM X o I I I I O O O I I O O O O O O l X X O O
I O O O I I I I O O O I I O O O l o i l O O I
I I I I I X I O O O O O O O O O O O O I I I I I I I I
„ d
® O
tH
& 0
(0
01
®
■s I* i p 1-1 §
Ih M O I h t J
o -p
1
txt
®
p
cd
1
ID
b
•H
0
a
C
®
p
H
®
d ®
CD & U f t 13 P Ih
a) P O 01 0 ,
d p u ® ®
® 01 ftp Ih
•H -H d O
a CO Sh cd d o d d d a, n d o ® ®
"S -rl a. o o o o ® o_ P Ih
01 i-4 I
1-1
o u d ts p ■H • H *H P 4rH
o d 2H p ® ® cd p •p -p o ® •• •• •• bO bo
•H Ih cd
H -2 Id f i
d ft cn M P o H
m t3 do do do &0 •• •• 0 S3
h o o d cd 1-1 * s s d bO*H dH
-a} o (D fl I •H *H p o o d o Ih iH -H iH q d -p -p
.I a ■o P o •H h Ih u p p p
CO § ^ * ®
a So ®
® 01 a 01
2 Ih £ © o ® I-I CD
J.
d -H m° p
Ih
.’pr _ 8 S | Do UoO® 1 ^ 1 1
fe
o« u h £ ti § § o o
cd o ® r-l a, p,-h a s cd §* cd 2 Jj? 5, O o o ® ®
w pH 53 M PL, CO o O C Q O CO Q ft Ph O h O D Q Q Q Q O O P 3 P h Ph
CQ
lOvsO C ^ t O O ' O H s f v O » ' Q t t t q C 7 ' O H C M C O ' + m v O f-«OOÔr-ICMCO-<t
*5 coc^cf\focOHt-4--4-Ht-^-v^-^--<|wMonr\«r\>f\vf\iov>ov«Mo»vOvOvOvOvO
- 173 -
to O I I O I O X X I I I
X X I X X X I
vO I I
o I Io
CO O I I I I I
CM O I I IO I
o I I IO I
O I I I I O I X I I I
a
o
•H
(concluded)
s ■P
aj
•H
3 -P
at
O
a® 3
tv M ® -p
a] -c?
h0Td
t IP ® -H t3
M <j 1-3 CO a-, o a a.
2
irwo > eg O
vO vO '•O so
- 174 -
Discriminating Power of Individual Figure Signs

on the Initial Test and the Retest
Tables XXIX and XXX show the incidence, expressed in per
centages, of the individual figure scoring signs on the Initial
Test and the Retest for the three control sub-groups, the Total
Control group, and the Total Neurotic group as well as critical
ratios of the differences between the Total Control and Total
Neurotic groups which are significant at the 5% level or better
(i.e., critical ratios of 1.96 or higher). On the Initial Test
40 out of 248 signs (16.1$ of the total) are discriminating com
pared with 41 out of 230 signs (17.8$ of the total) on the Retest.
Thirty-one of the 40 Initial Test discriminating signs and 35 of
the 41 Retest discriminating signs are negative, that is, they oc
cur more frequently among neurotics than among normals. The total
number of signs is somewhat less for the Retest than for the Ini
tial Test because only signs with frequencies above zero are en
tered in the table and certain signs having such frequencies
on the Initial Test were absent on the Retest. By way of recapit
ulation, Table XXXI presents the number of signs that was tabu
lated for each figure on each of the two tests, and Table XXXII
shows the number of discriminating signs obtained for each of the
figures on the initial and final tests.

- 175 -
TABLE XXXI
Breakdown of the Total Number of Signs Tabulated for the Initial Test
and the Retest on Each of the Bender-Gestalt Figures
Number of Tabulated Signs

Figure
Test 4 1 2 ? _ 4_ 5 6 7 8 Total
Initial 26 26 27 30 29 30 27 27 26 248
Retest 24 20 24 29 29 30 26 26 22 230
TABLE XXXII
Breakdown of the Total Number of Signs Significant at the % Level on

the Initial Test and the Retest for Each of the Bender-Gestalt Figures
Number of Significant Signs

Figure
Test A 1 2 3 4 5 6 7 $ Total
Initial 2 6 4 3 5 6 7 4 3 40
Retest 3 4 1 3 6 7 8 5 4 a
Selection of Signs for Inclusion in the Final Scoring System
In order to develop a scoring system which would be intrin
sically stable and reliable rather than dependent upon the summation
of chance differences, a sign was required to satisfy either of the
following criteria before being selected for inclusion in the final
scoring scheme; (1) discrimination between the Total Control and
Total Neurotic groups at the % level (critical ratio of 1.96) on
either the Initial Test or the Retest and at the 10$ level (critical
- 176 -
ratio of 1.65) level on the other or (2) discrimination at the 1$
level (critical ratio of 2.58) on either the Initial Test or the
Retest and consistency in the direction of the difference on the
other. The 31 individual scoring signs and the 3 global signs which
are consistent with these criteria are presented in Table XXXIII
together with their corresponding critical ratios.
Since, however, the two major criterion groups differ some
what with respect to age composition, educational background, and
marital status even though they are essentially alike with regard
to nationality, race, sex, and military service, the possibility
cannot be ignored that differences in these three variables may con
ceivably account for some of the obtained differences. In order to
give due consideration to this possibility, an attempt was made to
match as many of the neurotic subjects as possible with an equal
number of subjects from the Below 70 group with respect to the three
variables under discussion. The Below 70 group was selected as the
contrasting group because it satisfies both the clinical and psycho
metric criteria of normality adopted for this investigation. A
total of 138 subjects was so matched— 69 neurotics and 69 controls—
and will be referred to in the remainder of this paper as the Neurot
ic Matched Criterion group and the Below 70 Matched Criterion group.
While perfect matching could not be achieved because of the require
ment that each case be matched with respect to three variables simul
taneously, the success of the matching technique is evident from
Table XXXIV. Perfect matching was achieved for educational status,

- 177 -
■s
8
frequently in the Total Neurotic group. A plus sign before the critical ratio in
8% !£»g'CnspO'OCr\0'-st-ONtJO\OCMCMO'tt)r-l
*H 'rl \OxOC'~CT'OcnvOt'-CMir\rHC'\t'-CT'CMcn'4-\OCMi>tO
* A minus sign before the critical ratio indicates that the scoring sign occurs more
P P • • • • • • • • • • • • • • • • » # * # #
I I I I I I + | | | | | ++ | | | | | | |
o
O CO
•P •H O
in H -P H O' CO VT\ If\f-CM 'H’C"~ OnCM VTyCOvOt>CO lfMT\-Hf'■i’ O' CM
ID
■p I I* r\> «J r*v <*oro N.
COCMrHO'COCMt^COmifNCO^t rr\ pr\ [v fT^ j—| ^ [*>.
dicates a greater incidence of the sign in the Total Control group.

(D u O M f M ^ r l CMIM CM in m rH rH CM~3" m m CMC"-
03 Eh © a
2!'-'
<C\
CO O ' O ' m C M ' O v O C M C T ' O O O O O ' v O c n c n <r\vO s f O t O
CM
US 'JMMH«H\00>>\OvOOH N lO f f 'r lH I ^ C M
CMC^CMi—Ir H r - ii—IrHCMCM m H H CM H H rH vO
©
o o
•rl *r| crs^ 0it t0P«0^'Od*C''I>0Jt'cr,0iv0'0,-irHQ
V v C '- C n 'O m r - c n O 'C M O 'O O I f l 'O b - r H O ' C '- i - i t - O
-P -P • « . « . • • . . « « • • •
H r l CM CM IACM N O rl (M N N < ^l^\
I I I I I I + I I I I I + + I I I I I I I
S3
-p
m
ID O CO
EH •H O
H -P H
H © O
© •g 5 11 ^ CÔ'OOCÔCO-^mcnu,
5C^>A'tvOcrvsO'4-l>C»>
CM i f\ CM CMvO 0T\ CMvO m C M CM m m CM
eh © a
a —*
H
ICO rH m
H O W O r H C O i —IC'-mmC'-CT'O'CMr—|(T\-<i''£>Cr'C'--'OCOm''4'
© El CM
■P-P 0«J O H lA IM ^ ifw O O' sO rH O' CM ir\ OQ O O r-l rH m C M
o a n^ CM rH m H rH m C MrH CM in
Eh O
o a
(e
o
rH
J©3
CO
bp-p
a © <<<!HHHrHiHC'ICMCM<,M <\'*'tsf'^«MAif\lf\©
O mvA m O rH O m Q rH Q rH tAcT'CM m C^-O' C^-d> CM
m C- rH Ifl'O vO > i f \ \ 0 © ' 0 \ 0 i l M O C ' > t M C M if\vO
- 178 -
a)
O O O -^•tOCAsOcACM'J’rHCMO'A
•H r-1 A O CM'-tOO'OOC-'t'-tACMtO
•P -P • • • • • • • • • • • •
CM-4CMACMCMCMrH-~*CMCMCA
O I I I I+ I I+ I I I I
O 00
■p •H O
to r l - P H A O' A '-t'-O CM CA O' rH CM O -t
ID qi O • • • • • • • • • • • » •
-P CA'OOOQAOOOOO'CMCAAA
I & -4- CM \3 O H A ~4 CM rH nQ ~4
<3 H ® SB
a '—
H A
rl O «)
IS *4 CM
#)sf(Mt0'0't>>iA0'O0'rl
• • • • • • • • • • • • •
-p -p ^ O^DO sO r) H H O O CA r-j IT\ ^ ^
O fl II ^ cv lA < f r l
Eh O
O a
H
cd
o CACAO'O'C'-O'OCMO-tCMHO
•H .. C ^C -O O O 'A tA C M A A E -C M C M O
-P -P • • • • • • • • • • * « «
A C M H
h J-CMCACMCMcA H C M C A C A
I I I I♦ I I+ I I I I I
O 00
•H O
H -P rH t0«t'lrl'04C'l(fl«>'0(t\ct\iA
cd O • • • • • • • • • • • ■ a
aS CMC'-CA'4AOOtOCAC-OOtOO'
•H IA CM O' > N rl 1 A - 4 H A H
-P eh ©a
a '—
rH A
rH O 00 rHC'-AvtTAC'-CMOsOAHCAC-
Cd U CM
-P -P CM ~4 CA00 (An-O CA CM 4 0 N 0 I A
O fl II N r i® • 4 r l C>- CM H -4 CA
EH O
o a
h |N
.
.iCA O'
8 J CM
ICIAO'
J tOII
oo a
© M) vO'O vOvO NO C**
I I I IO
Q H AMD (A -rl
to 8 CM CM CM A v'
vO 'O 'O \D ca <n -P ..
© U ©
•s2 3
o CO EH
- 179 -
O
•H O'
+a xO
O CM cr\ r l <»>
C\ c*'!
©a
a —'
O'
gO S O
o c~- M V O O
II «H c^\m
©
CP a
CQ X3 ©
© -P
© © O ©
rH *rl U U
DO Ch 2 S
•3 d Ch * Sf
CO © ©
CQ 3 CP CQ
o -—
•H O'
+3 vO
o O' C- irv
jj II Oh C'i'4’HC''iS0C'-rr\O''OrH lf\ o
CM
C
M C
M
3
aS
a0 O'
+>
£ 'O
•rl O' t- m
fnHH to C- c^ 0"0 H rH tf\ o
SH R I
O Ii CM • • •
© C
M CM C
M
cp a H rH
1
o>
T3
a©
t'- \ 0 m-^c^CMrH O O' tO Ci •H •
g r lr lr ln r lr lr lr l © X3 Q
o © © •
$
CO
s s C
Q
o -—
•rl O'
+> vO '■* O -*
o O CMC'^CMmt'-tOirvCM to O cn
b ii < o
«
l>
Is CM C
M
O'
I* 'O 'O to p
o o rHCMCMmCM'OC-CM'©C»'vir\ CM o C"
rH £> II <M rH rH i
—I • • •
© f- vO
cp a CM CM
d
H O ' l > « n c f\ r H 0 ' C ' - i f \ C ^ r H ©
©
bi
"■T 0“\ O'* cr\ C^N CM CM CM CM CM a
©
•H
•o o
•
itO'i^CMOtO'i'VcMO © © •
^ ( ' r'vc*'\cf\ « ' f\ r < > CMCMCMCMCM 3 3 CO
- 180 -
the median grade attained being the high school senior year with a
standard deviation of approximately two grade levels. The closeness
of the matching for age is demonstrated by means and medians which
are less than one year apart and by standard deviations which are
less than one-half year apart. The number of single and married per
sons is also quite similar for the two groups.
The percentage incidence of the tentatively selected scor
ing signs and the critical ratios of the differences between the
percentages of the two matched criterion groups are presented in
Table XXXV. Since the number of cases is smaller in the matched
groups than in the original groups, the critical ratios tend to be
smaller even when the absolute differences are identical. To take
just one example: a percentage difference of 17 points on the Ini
tial Test for the original groups on sign 61-2 (36.956 for the Total
Control and 53•!% for the Total Neurotic group) results in a critical
ratio of 3.01, whereas a difference of 22 points for the matched
groups (34*856 and 56.656 for the Below 70 and Neurotic groups, respec
tively) results in a critical ratio of only 2.63. As a basis for in
clusion of the signs in the final scoring system, the criterion set
for this second weeding out process was somewhat more lenient, namely,
discrimination at the 5/6 level on either the Initial Test or the Re
test and consistency in the direction of the difference on the other,
as Table XXXV discloses, 27 of the 31 tentatively selected individual
signs and all 3 of the global signs survive both screening procedures.
- 181 -
The 30 signs (16 graphic and 14 methods signs) which con
stitute the final scoring system may be grouped as follows:
Description Sign
Numeration (number of wave crests in figure 6 20-6
Wave flattening in figure 6 22-6
Asymmetry in figure 5 27-5
Displacement in figures 5 and 6 29-5

29-6
Contiguity in figure A 30-A
Parallel lines in figure 8 33-8
Upward slope in figures 1 and 2 55-1

55-2
Sinistrad direction in figures 4 and 5 57-4*

57-5
Upward direction in figure 6 58-6+
Counting of test figure in figures 1, 2, 3, 5, 60-1

and 6 60-2
60-3
60-5
60-6
Counting of reproduction in figures 1,2, 3, 61-1

and 6 61-2
61-3
61-6
Initial part (left hexagon made first) in 65-7*
figure 7
Angle differentiation in figure 7 66-7
Pairing of dots in figure 1 70-1*
Correction in figure 4 72-4
Sagging line in figures A and 4 75-A

75-4
- 182
cd *
o o t'0'U\OrlrlO'W(T'MOt'ff"£lHt*f'00
P p NrllÔrl«rlO'U>MO>lf\OPMTi'0 0'n
Percentage Incidence and Critical Ratios of the Differences between Percentages
-P -P • • • • • • • • • • • • • • • • • • a
H O O ^ c y O H H r \H O « (* M M H H O IH
I+ I I I I I I I I | 4. | 4. 4. | | CM
I
for the Matched Criterion Groups (Below 70 and Neurotic) of the Tentatively
o^
p P O'
CQ
«
P SO
O UM>r-CMCMHOO>\OtOU>i-ieoCOr-'Cj--^-C'AO
« • • • • • • • « • • « * • « • « • »
Selected Individual Figure Scoring Signs and Global Signs
P £ H r y ^ r ' C f M O v O O Q O H sfM) UMT\® < C*\M3 M)

C- CM C*'!CM so CM CM -•* CM H CM rl
<3 ® as
a '—
O'
it vO s|H«)t0tt(^PM'MAO|if»>vO O O M 5
oo • • • • • • • • • • • • • • • • • • a
H t- || rMv0^1AOOOff'stO't«>O\0rl-JH ^
M3 CM 0^1 inCMOI^rlHH CM rl rj rH CM
m as
cd
o o tX)CMO'ir\0'OCMCÔvf\»r\vOCMCMC^NOî,0'0
•rl p
P P
(c\\0^®OKNOtt-MC>O'3Oc,
* • • • • • • • • • • • • •\S'J'TC'l>
• » • * •
H O N O O C M O I O O H C M r l ^ N C M O C M m ^
I I I I I I I I I I I 4 > I * + | | I
u p ja
O P P
P
CQ o. H
0) O
E-t P O'
P nO ~-^CMU'\ClOCMC'\C‘---4'E‘~O0'C»\c<-\O‘,\C'~OOCOvO d O
0 • • • • « • « • « • • • • • • • * • * P u
1 11*1 CM('rv(r\r-l«0\0-s3-CM-t\OTiOOC'-C'~tOir\crvOO
®s sOCM^CMsOfnCMvOCMPPCM vOvOVN
as —
a
H
O'
» vO tOO'flOsfCM'^'vOHCOCOO'NOIsfOH £•- V\ UN C~
o p
Ht'
®
HW( OtqHC^tQC^HmHW^CÔcnOÔtOcncotO
VT\ H N H rH rj W\ CVJ *— ~* rH CV >fr
pp a
i?
»
o
®
x>
rH ® x!
vO I H I nO nO vOI rllMsflftvO M j H CM rr, 3o
s +2® *
J I i iT i i I i i d ® ® cn
CNl|vi»|AWMrc a cr t>o So p p
p
•aj CO
%
- 183 -
<>
cd
o o OJ Oi l A ^C O H O f f l H ' f
•H *rl >0 (M CM O' tt O -«4-cn cv cQ in m vO DO cn
P P • > • • • • • * • • « • • • •
cncnoiHHcnc\iv}-c\icnc\<C'ic\coio
i l i l l l + l + i i i i i *
O —v
•P •H ON
10 P vO ( C n C > r l r l ' 0 O t ' ( M 0 ' 0 ' ( r\ H < 1 0 C \
O
9e | ■ Wl cn i-t CM -cf H r-I
OS <d as
as'—
O'
* sO o i o ^ ' O s f o
• • • • • • • • •
o iv o o
• • • •
m 'O
• a
cm
Sfc « ■e* o o c - t> ^ o i r \ 0 ' 4- o r ^ O ' H O ( M
® CM ' O Oi N H CM OJ
oq ss
cd
o tf>NCq<^C'inOOifMAt0H«l'0O
C'-0'r-t'00'C‘-inCMC''eO'50in^'0'
P P > * • • • • • • • • ! • • • •
stNCVCMNHC^lÔOOf^NNH
o llllll+l + lllll*
o^
•H O'
P vO <n cn m v O sO CM P \ 0 ' 0 ( M C M H > f ' « )
O • • • • • • • • • • • • • • •
aS g N W| 0' O<n \0C -O 0 ' t n rH c n CM i n t0 c n rH
•H '4-cM~-tmcMH'4’-4-HCMir\‘n m cm
P ® 25
•H a •
a
O'
» sO if'^tiHco r- O' O' c»\o st’s t m o cn
op • • • • • • • • • • • • • • •
H C~ N 'J-^vOst(ON®0'0>'OOOM'0
TABLE XXXV (concluded)
CD H cm cn cm H H --d-cn cn cn
m s
H|C\1
m \ O H cm cnso
cn
^ a •H L-
.
.iO'
6 i i i, i, i i i 1 i i i i o
•H OQr-lrlHr-lin'OpCVjin in p
CO \Q \Q \Q \Q <0 vO 'O N C- C"-C- 01 p ••
e M ©
■B
o S-SCO E-h
- 184 -
Description Sign
Spatial compression Cohesion: \
Presence of 1 or more reproductions Short Figures

of abbreviated length
Total time to make the reproductions Time: 139*

less than 140 seconds
(Signs followed by a plus occur more frequently among normals than
among neurotics; all other signs discriminate in the opposite direc
tion, i.e., are more characteristic of neurotics.)
C. The Recall Test
Tables XXXVI, XXXVII, XXXVIII, XXXIX, and XL present a de
tailed breakdown of the frequency of recall of each of the stimulus
figures by the subjects in each of the criterion groups on the Imme
diate Recall Test. These tables give, for each figure, the recall
position (the number of times the figure was the first to be recalled,
the second to be recalled, etc.), the total frequency of recall (ex
pressed as an absolute number and as a percentage of the total number
of cases in the group), and the relative rank of that figure with re
gard to its recall frequency. Table XXXVI, for example, is read as
follows: Figure A is recalled by 114 of the 155 subjects in the Be
low 70 group, or by 73-5% of the total number of cases. With respect
to the order of recall, 40 subjects recalled Figure A before any of
the other figures, 3 subjects reproduced it as the second figure, 6
as the third figure on the Recall Test record, etc. Figure A ranks
sixth with regard to recall frequency, since five other figures are
recalled more often. Tables XLI and XLII list the recall positions
- I85 -
as percentages rather than absolute frequencies for the Total Control
and Total Neurotic groups. Table XLI, for example, is read as follows:
90.256 of the 285 subjects in the Total Control group reproduce figure 1
on the Recall Test. Twenty per cent of the 285 cases recall figure 1
before any of the other figures, 22.8% recall it second, 9.856 recall
it in third position, etc.
Table XLIII presents a comparison of the criterion groups with
respect to the relative frequency of recall of each of the figures, ex
pressed as ranks. Figure 8 is recalled more frequently than any of
the other figures by four of the five groups and is tied for first posi
tion in the remaining group, while figure 4 is least often recalled by
all of the groups. For the entire group of controls, the relative fre
quency of recall (from most to least) of each of the figures is as fol
lows: figures 8, 6, 1, 7, 2, A, 5, 3, and 4. The tanks of the figures
for the neurotic subjects vary only slightly from those of the "norm"
group, as one can observe from the close correspondence of the two
curves in figure 6. The reliability of this rank order is evident
from Table XLIV, which gives the intercorrelations of the rank orders
of all the criterion groups. The lowest of these intercorrelations is
.921 and the highest is .996. The correlation of .950 between the
Total Control and Total Neurotic groups militates against the use of
relative recall frequency as a discriminating variable. The findings
on the relative frequency of recall of the various figures agree in
general with the vast majority of studies on serial learning which in
dicates that the first and last responses are the easiest to learn and
- 186 -
TABLE XXXVI
Frequency of Recall and Order of Recall of Each of the Test Figures on

the Immediate Recall Test for the Below 70 Group
Recall Position
Figure 1 2 3 A 5 6 ,7 8 9 f Rank
A 40 3 6 8 17 17 14 8 1 114 73.5 6
1 28 A2 15 14 18 13 7 2 0 139 89.7 3
2 7 20 37 18 12 13 13 0 0 120 77.4 5
3 7 8 12 13 3 12 11 1 1 68 43.9 8
4 3 3 2 9 10 11 10 6 0 5A 34.8 9
' 5 16 21 19 11 8 10 13 5 1 104 67.1 7
6 7 21 26 A2 30 15 6 1 1 149 96.1 2
7 A 12 11 22 30 31 18 6 0 134 86.4 4
8 A3 25 27 18 18 U 6 2 0 153 98.7 1
TABLE XXXVII

the Immediate Recall Test for the 1 Above 70 Group
Recall Position
rare 1 2 3 4 5 6 7 8 9 f ... % Rank
A 14 3 2 5 4 10 7 1 0 46 67.7 6
1 17 11 4 6 12 6 2 2 0 60 88.3 3
2 1 14 13 10 7 5 3 1 0 54 79.4 5
3 2 4 8 7 3 0 1 0 0 25 36.8 8
4 1 1 0 1 5 6 2 1 1 18 26.5 9
5 6 9 8 5 3 6 4 0 0 41 60.3 7
6 7 12 10 12 13 5 5 0 0 64 94.1 1.5
7 2 4 9 12 10 12 8 2 0 59 86.8 4
8 18 10 14 9 5 5 2 1 0 64 94.1 1.5
- 187 -
TABLE XXXVIII
the Immediate Recall Test for the 2 Above 70 Group
Recall Position
Figure 1 2 3 4 5 6 7 8 9 f i Rank
A 6 6 4 4 9 12 5 3 0 49 79.0 4
1 12 12 9 7 10 4 2 2 0 58 93.6 2
2 5 8 11 4 5 4 5 1 0 43 69.4 6
3 4 2 2 9 2 1 3 1 0 24 38.7 8
4 1 0 1 2 1 5 2 4 0 16 25.8 9
5 8 5 7 9 5 6 1 0 0 41 66.1 7
6 4 11 17 10 9 5 1 0 0 57 91.9 3
7 1 5 5 12 10 8 6 0 0 47 75.8 5
8 21 13 6 5 3 4 6 1 0 59 95-2 1
TABLE XXXIX

the Immediate Recall Test for the Total Control Group
Recall Position
rure 1 2 3 4 5 6 7 8 9 f % Ran)!
A 60 12 12 17 30 39 26 12 1 209 73.4 6
1 57 65 28 27 40 23 11 6 0 257 90.2 3
2 13 42 61 32 24 22 21 2 0 217 76.2 5
3 13 14 22 29 8 13 15 2 1 117 41.1 8
4 5 4 3 12 16 22 14 11 1 88 30.9 9
5 30 35 34 25 16 22 18 5 1 186 65.3 7
6 18 44 53 64 52 25 12 1 1 270 94.8 2
7 7 21 25 46 50 51 32 8 0 240 84.2 4
8 82 48 47 32 26 23 14 4 0 276 96.9 1
- 188 -
TABLE XL
Frequency Of Recall and Order of Recall of Each of the Test Figures on

the Immediate Recall Test for the Total Neurotic Group
Recall Position
Figure 1 2 3 A 5 6 7 8 9 f _ % Rank
A 26 A 6 6 6 6 9 3 0 66 61.1 7
1 30 25 13 6 7 9 2 1 0 93 86.1 3
2 A 18 23 13 5 5 3 1 0 72 66.7 6
3 7 7 10 11 2 2 2 3 0 AA 40.7 8
4 3 6 3 9 7 3 6 A 0 a 38.0 9
5 12 17 10 10 13 7 5 2 2 78 72.2 5
6 A 7 10 20 22 23 6 2 0 94 87.0 2
7 1 7 13 11 21 15 13 A 0 85 78.7 4
8 21 16 15 17 10 9 8 3 1 100 92.6 1
TABLE XLI
Frequency of Recall and Order of Recall, Expressed as Percentages,

of Each of the Test Figures on the Immediate Recall Test for the
Total Control Group
________________ Recall Position

Figure 1 2 3 4 5 6 7____ §___ 2__ Xfitel
A 21.1 4.2 4.2 6.0 10.5 13.7 9.1 4.2 0.4 73.4
1 20.0 22.8 9.8 9.5 14.0 8.1 3.9 2.1 0.0 90.2
2 4.6 14*7 21.4 11.2 8.4 7.7 7.4 0.7 0.0 76.1
3 4.6 4.9 7.7 10.2 2.8 4.6 5.3 0.7 0.4 41.2
4 1.8 1.4 1.1 4.2 5.6 7.7 4.9 3.9 0.4 31.0
5 10.5 12.3 11.9 8.8 5.6 7.7 6.3 1.8 0.4 65.3
6 6.3 15.4 18.6 22.5 18.3 8.8 4.2 0.4 0.4 94.9
7 2.5 7.4 8.8 16.2 17.6 17.9 11.2 2.8 0.0 84*4
8 28.8 16.9 16.5 11.2 9.1 8.1 4.9 1.4 0.0 96.9
- 189 -
TABLE XLII
Frequency of Recall and Order of Recall, Expressed as Percentages,
of Each of the Test Figures on the Immediate Recall Test for the
Total Neurotic Group
Recall Position
Fieure 1 2 3 4 5. 6 7 8 ? Total
A 24.1 3.7 5.6 5.6 5.6 5.6 8.3 2.8 0.0 61.3
1 27.8 23.2 12.0 5.6 6.5 8.3 1.9 0.9 0.0 86.2
2 3.7 16.7 21.3 12.0 4.6 4.6 2.8 0.9 0.0 66.6
3 6.5 6.5 9.3 10.2 1.9 1.9 1.9 2.8 0.0 41.0
4 2.8 5.6 2.8 8.3 6.5 2.8 5.6 3.7 0.0 38.1
5 11.1 15.7 9.3 9.3 12.0 6.5 4.6 1.9 1.9 72.3
6 3.7 6.5 9.3 18.5 20.4 21.3 5.6 1.9 0.0 87.2
7 0.9 6.5 12.0 10.2 19.5 13.9 12.0 3.7 0.0 78.7
8 19.5 14.8 13.9 15.7 9.3 8.3 7.4 2.8 0.9 92.6
TABLE XLIII
Frequency of Total Recall, Expressed as Ranks, of Each of the Test
Figures on the Immediate Recall Test for the Below 70, 1 Above 70,
2 Above 70, Total Control, and Total Neurotic Criterion Groups
GrouD
Below 1 Above 2 Above Total Total
70 70 70 Control Neurotic
(N - 155) _ (N - 68) (N - 62). . (N = 285) (N - 108)
Fieure Rank Rank Rank Rank Rank
A 6 6 4 6 7
1 3 3 2 3 3
2 5 5 6 5 6
3 8 8 8 8 8
4 9 9 9 9 9
5 7 7 7 7 5
6 2 1.5 3 2 2
7 4 4 5 4 4
8 1 1.5 1 1 1
TABLE XLIV
Intercorrelations of Frequencies of Total Recall (Expressed as Ranks)
of the Test Figures on the Immediate Recall Test for the Criterion Groups
rho
1 Above 70 2 Above 70 Total Neurotic
Below 70 .996 .933
1 Above 70 .921
Total Control .950
- 190 -
retain and the middle ones most difficult* Unlike Hovland (53), how
ever, who found that primacy was stronger than recency in the learning
of nonsense syllable lists, the present investigation reveals that the
last three figures, considered as a group, are more readily recalled
than the first three.
The relative frequency of recall in first position, that is,
the number of times each figure was recalled before any of the others,
is presented in Table XLV, in which the rank order of the figures is
given for each of the criterion groups, and in figure 7, in which the
percentages for the Total Control and Total Neurotic groups are plotted*
For the entire control group, the relative frequency of recall in first
position, from most to least, is as follows: Figures 8, A, 1, 5, 6,
2, 3, 7, and 4* The close correspondence of the ranks of the figures
for the neurotic subjects is evident from figure 7 and from the inter
correlations in Table XLVI, the correlation between the control and
neurotic groups being .88. Table XLVII permits a comparison of the
frequency of total recall and the frequency of recall in first posi
tion of all the figures for all of the criterion groups. Of particu
lar interest are the disparities in rank of the two recall variables
for figures A, 6, and 7, Although figure A ranks sixth with respect
to total recall, it ranks second with regard to frequency of recall in
first position for the Total Control group. Figures 6 and 7, on the
other hand, move downward in rank, with shifts from second to fifth
position and fourth to eighth position, respectively. Similar shifts
in position are noted for the neurotic subjects. It is interesting to

- 191 -
Total Control, N-J85

Total Neurotic. N»I06
LOT JXE.4444
T e s t ¥\ q ore
figure 6. frequency of Recall of £acb of the Test figures

on the lenediate Recall Test for the Total
Control and Total Neurotic Groups
- 192 -
TABLE XLV
Frequency of Recall in First Position, Expressed as Ranks, of Each of

the Test Figures on the Immediate Recall Test for the Below 70,
1 Above 70, 2 Above 70, Total Control, and Total Neurotic Criterion Groups
Grouo
(N * 155) (N = 68) (N = 62) (N = 285) (N = 108)
Figure Rank Rank Rank Rank Rank
A 2 3 4 2 2
1 3 2 2 3 1
2 6 8.5 5 6.5 6.5
3 6 6.5 6.5 6.5 5
4 9 8.5 8.5 9 8
5 4 5 3 4 4
6 6 4 6.5 5 6.5
7 8 6.5 8.5 8 9
8 1 1 1 1 3
TABLE XLVI
Intercorrelations of Frequencies (Expressed as Ranks) of the

Test Figures in First Position on the Immediate Recall
Test for the Criterion Groups
rho
1 Above 70 2 Above 70 Total Neurotic
Below 70 .867 933

1 Above 70 771
Total Control .879
- 193 -
Total Control. N*285

Total Neuroric,N*IOo
p t it '- :
figure 7. Frequency of Recall in First Position of Each of

The Test Figures on the Inaediate Recall Test
for the Total Control and Total Hieurotic Groups
- 194 -
TABLE XLVII
Frequency of Total Recall and Frequency of Recall in First Position,

Expressed as Ranks, of Each of the Test Figures on the Immediate
Recall Test for the Below 70, 1 Above 70, 2 Above 70, Total
Control and Total Neurotic Criterion Groups
Group
(N = 155) (N * 68) (N - 62) _ . (N = 285) _ (N * 108)
Total Total Total Total Total
Figure Recall First Recall First Recall First Recall First Recall First
A 6 2 6 3 4 4 6 2 7 2
1 3 3 3 2 2 2 3 3 3 1
2 5 6 5 8.5 6 5 5 6.5 6 6.5
3 8 6 8 6.5 8 6.5 8 6.5 8 5
4 9 9 9 8.5 9 8.5 9 9 9 8
5 7 4 7 5 7 3 7 4 5 4
6 2 6 1.5 4 3 6.5 2 5 2 6.5
7 4 8 4 6.5 5 8.5 4 8 4 9
8 1 1 1.5 1 1 1 1 1 1 3
TABLE XLVIII
Correlations of Frequency of Total Recall and Frequency of Recall

in First Position for the Below 70, 1 Above 70, 2 Above
70, Total Control and Total Neurotic Criterion Groups
GrouD rho
Below 70 .483
1 Above 70 .654
2 Above 70 .633
Total Control .546
Total Neurotic .254
- 195 -
note that, although figure 6 is recalled by 95$ of the Total Control
group, it is recalled in first position hy only 6$ of that group and
that for figure 7 the percentages are 84$ and 3$, respectively, while
for figure A 21$ recall it in first position even though only 73$ of
the entire group recalled it at all. The correlations between total
recall and priority of recall are given in Table XLVIII, the correla
tion for the Total Control group approximating *55 as contrasted with
.25 for the neurotic cases. Inspection of the rank differences in
Table XLVII suggests that the correlation coefficients would be more
nearly alike if the Total Neurotic group contained a larger number of
cases.
If one thinks not in terms of the interrelationships of the
test figures but rather in terms of the total number of figures re
called, can the normal and neurotic subjects be distinguished on this
basis? That the answer to this question is essentially in the nega
tive can be seen from Table XLIX, which gives the frequency distribu
tions, means, and standard deviations for the criterion groups. None
of the differences are statistically significant, the average number
of figures recalled clustering around six for the entire group of
393 cases, although the neurotics have the lowest mean of all the
groups. This is a surprisingly high figure, considering the fact
that the nine stimulus figures were presented during the Initial Test
without any instructions to the effect that a subsequent recall would
be required. The variability is somewhat greater for the neurotics
than for the controls, with a tendency for the neurotic group to contain
\
- 196 -
TABLE XLIX
Distribution of the Total Number of Figures Recalled, Means, and Standard

Deviations on the Iimnediate Recall Test for the Criterion Croups
Group
(d (2) (3) (4) (5)
Figures (N = 155) (N = 68) (N « 62) (N = 285) (N - 108)
Recalled* f f f f f
9 2 1 3 1
8* 2 0 2 2
8 13 3 7 23 12
7* 11 4 5 20 7
7 35 16 11 62 15
t>k 21 8 6 35 13
6 41 14 14 69 18
5k 9 8 4 21 7
5, 10 6 5 21 9
4i 1 2 2 5 7
4 8 4 7 19 4
3k 2 1 0 3 4
3 1 1 2 5
2k 3
2 0
lk 0
1 1
N 155 68 62 285 108

Mean 6.40 6.13 6.14 6.28 5.93
S.D. 1.08 1.16 1.26 1.15 1.60
Critical
Ratio *1.79
(m 4 - m 5)
* A reproduction received a score of £ when only 1 part of a figure con

taining 2 parts was given (e.g., reproducing only the circle of
figure A or the open square of figure 4). If a reproduction con
sisted of 2 parts taken from two different figures, it was also
scored as £ (e.g., reproducing a figure consisting of the circle of
figure A and the bell-shaped curve of figure 4)*
- 197 -
more instances of extreme paucity of recall* Thus 13 of the 108
neurotics (12.0$) recall 3i figures or less compared with 5 of the
285 controls (1.8$).
If one focuses on the frequency of recall of each of the
figures rather than on the total number of figures recalled, and if
the difference in incidence of recall of each of the figures is ex
amined, it will be seen from Table L that seven of the nine figures
show no reliable differences between normal and neurotic subjects
but that two figures (A and 6) discriminate at the 5% level. Analy
sis of Tables XLI and XLII indicated that this discrimination could
be sharpened by considering the recall position of figures A and 6.
If the number of neurotics and normals recalling figure A in the fifth
through the ninth positions is compared, the percentages are 41.1$
and 23*2$ for the Total Control and Total Neurotic groups, respective
ly— a difference giving rise to a critical ratio of 3*58* Similarly,
the percentages of controls and neurotics recalling figure 6 in the
first four positions are 62.1$ and 38.1$, respectively, the critical
ratio of the difference being 4.24. These data are given in Table LV,
in which the recall of figure A following the recall of four or more
other figures is designated as A (5-9), and the recall of figure 6
among the first four figures to be remembered is designated as 6 (1-4).
Up to this point the discussion of the Recall Test has been
limited to the number of individual or total figures recalled and the
order of recall. But, one may well ask, what about the characteristics
of the reproductions themselves? Are there any spatial or formal modi
fications which are peculiar to the Recall Test or which will reliably
- 198 -
Frequency of Recall, Expressed in Percentages, of Each of the Test Figures on the
in H
Immediate Recall Test for the Below 70, 1 Above 70, 2 Above 70, Total Control,
O' P- to
o o &
• o• to o• cn cn & ol in
• • •
and Total Neurotic Criterion Groups plus Critical Ratios of the Differences
•H *H I
CM CM
+ n a T* T1 * ■j
8
Ha rH p- t- CM
co CM P- to CM
3 to i i Cn P- to O'
rH m
-4- cm
•
CM O'
•
cn co cm O'
j5
o
o
>o
r-
o irv -4 «
£-• O
O 25
cn <o O' * O'
o• <3• -4
• p-
• to rH O' to CM
• • • •
I O' cn O' to m <3 H m m
p- O' o cn CM O O' P- O'
<0 to
> 3 cn en m o'* to H
• •
CM .8£u W c-
•o o >o <o -4 -O -4
i e' cn CM 2 O'
rH a
w
to o
if>
*
v\ O' to
• •
p s c
l>
cn
-4
-4
cn
>o
O' eg
CO
CM cn m to
- 199 -
differentiate the normals from the neurotics? The initial effort to
answer these questions was to compare the two matched criterion groups
(Below 70 and Neurotic) with regard to the incidence on the Recall
Test of those signs selected for inclusion in the final scoring sys
tem for the Initial Test and the Retest. Signs that had to do with
the line direction of the figures (57-4, 57—5# and 58-6), counting
(60-1, 60-2, 60-3, 60-5, 60-6, 61-1, 61-2, 61-3, 61-6), time (time:
139), and initial part (65-7) were necessarily omitted, since observa
tion of these nmethods signs” would have immeasurably complicated the
task of the examiner. Those signs for which comparisons were made
were 30-A, 75-A, 55-1, 70-1, 55-2, 72-4, 75-4, 27-5, 29-5, 20-6, 22-6,
29-6, 66-7, 33-8, and Cohesion: 5. Consideration was also given to

all of those signs mentioned in the Scoring Manual but not included
in the final scoring system for the Initial Test and the Retest by
tabulating their frequency of occurrence in random samples of 25 each
drawn from the total number of cases in the Total. Control and Total
Neurotic groups. Since no trend toward significance was noted for
these additional signs, tabulations for the entire group of subjects
were not made and do not appear in the accompanying tables.
Inspection of a number of Recall Test records suggested that
certain modifications were present when the figures were reproduced
from memory which either infrequently or never occurred during the
initial and final tests (when the reproductions were copied from
models). These changes are classified as distortions, confusions,
half figures, closures, and rotations. Rotation has been defined in
the Scoring Manual, but the remaining signs peculiar to the Recall
o
CKJ CO Figure a
o
0
O
O
O
to
O
o
0
0 o
o
e> o
o
Figure 2
0
o
o
O &
O O O & 0 o o O c G
O o o
O O 0 o c c a o 0
0 & o
0 o 6 O £
0 o 0 o 0 0
0 c
Figure 2 6
\ % t a 4
% 0 %.
• 1 \ \
' • 1 * \ e «
< : ; #
>
: •: • €
e / / A • >
Figure 3 / / e> A
/ Figure 3
$
*«' ; 1 ■
V. '
/ / ; / ”r
/
/
J
Figure 3
i
i
i
Figure 4
Figure 5
Figure 6
Figure 8. Exaaples of Distortion Taken from the Immediate Recall Test.

Figure 7
Figure 7
Figure 7
Figure 8
figure 8 (continued). Exaaples of Distortion Token fron the I— ortiato

Recall Tent.
cn cu rPoo
Figures A and 4
Figures A and 4
Figures A and 8 (or 7) Figures A and 8 Figures A and 5
Figures 1 and 2
Figures 4 and 6
Figures A, 4, and
• • • b 6 O O O • • >
Figures 1 and 5
Figures 4 and 7
Figures 7 and 8
Figure 9. Examples of Confusion (Intraserial Assimilation) Taken from

the Immediate Recall Test.
- 200 -
Test require definition and illustration.
Distortion is given a more inclusive definition than its
counterpart in the Scoring Manual, namely, a marked deviation of the
reproduction from the original, including simplification, addition
or omission of angles, marked line irregularity, reversal of parts,
or perseveration. Examples of recall distortion, taken from both
normal and neurotic records, are given in figure S. Distortions oc
cur very frequently regardless of the group to which the subject be
longs, 57% of the matched Below 70 group and 6k% of the matched neurot
ic group having one or more distorted reproductions in their recall
records. These figures are considerably higher than the 15% to 20%
incidence of distortion in the Initial Test and Retest and the distor
tions are more extreme. From Table LI, which presents a breakdown
TABLE LI
Percentage Incidence of Distortions of Each of the Bender-

Gestalt Figures on the Immediate Recall Test for the
Matched Criterion Groups (Below 70 and Neurotic)
Group
Below
70 Neurotic
is.-621
Figure I
A 2.9 7.2
1 1.4 2.9
2 4.3 5.8
3 5.8 5.8
4 11.6 15.9
5 0.0 1.4
6 10.1 15.9
7 43.5 37.7
8 2.9 4.3
- 201 -
figure by figure, It Is seen that the reproduction of figure 7 in the
Recall Test is much more likely than the other figures to deviate
grossly from the model, approximately two out of every five subjects
having difficulty with this figure. The actual proportion of dis
tortions in subjects attempting this figure is even greater, since
the percentages in Table LI are based on all the subjects in each
group, including several who did not recall figure 7 at all. Repre
sentations of angulation, size, and the number of sides present
especial difficulties. On the other hand, figure 8, which consists
of the same two hexagons, though rarely perceived as such, is only
infrequently distorted. Figures 4 and 6 show distortions in from 10-
15% of the cases, while distortions in the remaining figures are

quite rare. The proportion of distortions of figure 4 among those
actually attempting it is considerable, for it will be recalled from
Table XXXIX that figure 4 is recalled less often than any of the
other figures. None of the differences in Table LI is significant,
however, although one might have expected on an a priori basis that
the neurotic cases would have had much more difficulty with the re
productions than the control subjects.
Half Figures is a term referring to the reproduction of
only one part of a figure which consists of two parts in the original.
Examples in which Half Figure would be scored are: reproducing either
the circle or the diamond of Figure A but not both, making an open
or closed square but omitting the bell-shaped curve in figure 4, re
producing only the semi-circle of dots in figure 5, making only one

- 202 -
wavy line in figure 6, making only one hexagon in figure 7, and omit
ting the inclosed diamond in figure 8. Slightly more of the neurotic
than the control subjects manifest this sign (17.4$ and 11.6$, re
spectively), but the difference is not statistically reliable.
Closure-4 refers to reproducing the open square in figure 4
as a closed square. It is scored whether both square and bell-shaped
curve appear, whether the square appears alone (also scored for Half
Figure), or whether the square appears with a part of another figure
(also scored for Confusion). Closure occurs more often among normal
subjects, the percentages being 17*4$ and 7*3$ for the normals and
neurotics, respectively.
Confusion (intraserial assimilation) is scored when a re
production consists of two parts taken from two different figures
and combined. Examples of confusions appear in figure 9, and a
breakdown of the incidence of the sundry combinations is given in
Table LII. Figures A and 4 are most frequently confused, combina
tions of the two figures appearing in 17*4$ of the controls and 13*0$
of the neurotics— an insignificant difference. Combinations of
other figures occur relatively infrequently. If one considers the
number of cases in which one or more instances of confusion are pres
ent, the proportion of normals and neurotics is identical (24.7$)— a
result in marked contrast to its complete non-occurrence on the Ini
tial Test and the Retest.
Table LIII lists all the signs for which the Recall Test
was scored, the incidence of these signs in the matched criterion

- 203 -
TABLE LII
Percentage Incidence of Confusions for All Combinations of Figures

Appearing on the Immediate Recall Test Shown by Subjects in
the Matched Criterion Groups (Below 70 and Neurotic)
Below
70 Neurotic
Figure (N = 69) (N * 69)
Combined $ %
A and 4 17.4 13.0
A and 5 1.4 1.4
A and 8 1.4 1.4
1 and 2 1.4 1.4
1 and 5 1.4 1.4
4 and 6 2.9 5.8
4 and 7 0.0 1.4
7 and 8 0.0 1.4
A and 4 and 6 0.0 1.4
groups, and the critical ratios of those differences between percen
tages which are significant at the 10$ level or better. The only
sign not previously discussed is "Short Figures (4)”, which refers
to the presence of 4 or more reproductions at or below the 10th per
centile of the Initial Test reproductions of the Total Control group
in length. Of the 24 signs, 11 signs showing significant differences
at the 10$ level were selected for further analysis. All the Recall
Test records of all the cases in the Below 70 group (N = 155) and the
Total Neurotic group (N * 108) were scored for these tentatively
selected signs, and those signs still significant at the 10$ level
were selected for the final weeding-out process. Seven of the 11
signs survived this second screening, as Table LIV shows. The records
of all cases in all of the criterion groups were then scored for these
- 204. -
TAHLE LIII
Percentage Incidence of the Tentative Signs on the Immediate Recall Test

for the Matched Criterion Groups (Below 70 and Neurotic) plus Critical
Ratios of the Differences between the Groups Which are Significant
at the 10/6 Level or Better
Groun
Below
70 Neurotic
_(N = 69) (N - 691 Critical
SiRn i i . ... Ratio*
30-A 16.0 23.3

75-A 24.7 23.2
55-1 37.7 24.7 *1.66
70-1 7.3 7.3
55-2 29.0 17.4 *1.63
72-4 2.9 13.1 -2.25
75-4 16.0 18.9
27-5 8.7 8.7
29-5 11.6 13.1
20-6 65.3 65.3
22-6 14.5 13.1
29-6 47.9 63.8 -1.90
66-7 60.9 69.6
33-8 14.5 18.9
Cohesion: £ 14.5 27.6 -1.91
Distortion 56.6 63.8
Rotation 1.5 11.6 -2.44
Confusion 24.7 24.7
Half Figure 11.6 17.4
Closure-4 17.4 7.3 *1.82
Short Figures (4) 8.7 27.6 -2.97
A (5-9) 36.2 23.2 *1.69
6 (1-4) 59.5 40.6 *2.26
3jj or less 1.5 8.7 -1.95
* A minus sign before the critical ratio indicates that the scoring sign
occurs more frequently in the Neurotic group. A plus sign indicates
a greater incidence of the scoring sign in the Below 70 group.
- 205 -
TABLE LIV

for the Below 70 and Total Neurotic Groups plus Critical Ratios of the
Differences between the Groups Which are Significant at the 10$ Level or Better
Group
Below Total
70 Neurotic
(N * 155) ... (N * 108) Critical
Sign . . $ % Ratio*
55-1 26.5 25.0

55-2 18.1 15.7
72-4 8.4 8.3
29-6 49.0 59.3 -1.6b
Cohesion: £ 18.1 31.5 -2.46
Rotation 7.1 11.1
Closure-4 17.4 9.3 *1.96
Short Figures (4) 12.3 24.1 -2.41
A (5-9) 38.1 23.2 *2.65
6 (1-4) 60.6 38.9 *3.55
3h °r 1®8S 1.3 12.0 -3.28
* A minus sign before the critical ratio indicates that the scoring sign
occurs more frequently in the Total Neurotic group* A plus sign in
dicates a greater incidence of the scoring sign in the Below 70 group.
TABLE LV
for the Criterion Groups plus Critical Ratios of the Differences between
the Total Control and Total Neurotic Groups Which are Significant at the
5$ Level or Better
Group
(1) (2) (3)' ' (4) (5)
70 70 70 Control Neurotic (4 - 5)
_(N = 155) (N - 68) (N - 62) (N * 285) (N = 108) Critical
Sign % . _ _ % i i ____ Ratio*
29-6 49.0 58.8 53.2 52.3 59.3
Cohesion: £ 18.1 28.0 19.4 20.7 31.5 -2.13
Closure-4 17.4 20.6 12.9 17.2 9.3 *2.21
Short Figures (4) 12.3 10.3 12.9 11.9 24.1 -2.69
A (5-9) 38.1 28.0 46.8 41.1 23.2 ♦3.58
6 (1-4) 60*6 60.3 67.8 62.1 38.9 *4*24
3£ or less 1.3 2.9 1.6 1.8 12.0 -3.16
* A minus sign before the critical ratio indicates that the scoring sign occurs
more frequently in the Total Neurotic group. A plus sign indicates a greater
incidence of the scoring sign in the Total Control group.
- 206 -
7 signs, and those signs resulting in differences between the Total
Control and Total Neurotic groups that were significant at the 5$
level were retained for inclusion in the final scoring system of the
Recall Test. The 6 signs which survived all three screening procedures,
their incidence, and their critical ratios are shown in Table LV.
D. Scoring of the Test Records of the Matched Criterion

Groups with the Selected Signs
After having isolated those signs on the Initial Test, Re
call Test, and Retest which individually discriminated reliably between
the normal and neurotic subjects according to the item validity cri
teria specified for this investigation, the next step was to rescore
all of the records for these nselected signs", add the signs algebra
ically, and tabulate the distributions of these summated scores. The
actual sign-by-sign scoring of the Initial Test, Recall Test, and Re
test records for each of the 138 cases in the matched criterion groups
appears in Tables LVI, LVII, LV1I1, LVIX, LX, and LXI in the Appendix.
In these tables, the presence of a sign which occurs more frequently
among neurotic subjects is indicated by a minus (-) and the presence
of a sign occurring more often among the controls is designated by a
plus (+)• The sums of these signs, designated as Initial Score, Re
call Score, and Retest Score, appear without sign when the number of
minus signs outweighsthe plus signs and appear with a plus sign when
the number of plus signs outweighsthe negative signs. If, for example,
a given test record has six negative signs and three positive signs,
the sum is designated as 3; should the record contain four positive

- 207 -
signs and three negative signs, however, the sum would be designated
as *1. In all of the tables that follow, and in the ensuing discus
sion, a 8core which is not prefixed by a sign is understood to be
negative.
Frequency and cumulative percentage distributions of Ini
tial and Retest scores for the Below 70 and Neurotic Matched Cri
terion groups are tabulated in Table LXII and plotted in Figures 10,
11, 12, and 13* On the Initial Test the means for the control and
neurotic subjects are 3»07 and 8.06, respectively— a difference yield
ing a critical ratio of 9*78. Although it is true that the frequency
distributions overlap to some extent, the fact remains that almost
one-third (30.4$) of the control group score below (i.e., make more
favorable scores) the lowest score of the neurotic cases. Maximal
discrimination is obtained in the score range from 4 to 6. A score
of 5 or above, for example, would screen out 88.4$ of the neurotics
at the expense of only 26.1$ false positives; a score of 7 or above
would reduce the number of false positives to 10.1$ but would identify
only 72.5$ of the neurotics. On the Retest, the means are lower for
both groups— 2.10 for the controls and 6.46 for the neurotics— but
the difference is also highly significant (critical ratio of 8.90).
The score range of 3 to 5 yields the maximal discrimination, indicat
ing that the critical score for the Retest should probably be set
about one point below the cutting score for the Initial Test. A Re
test score of 4 or above would screen out 81.2$ of the neurotics at
the expense of 23*2$ false positives; a score of 6 or above would

- 208 -
TABLE LXII
Distribution of Scores, Cumulative Percentages, Means, and

Standard Deviations on the Initial Test and the Retest for
the Matched Criterion Groups (Below 70 and Neurotic)
Initial Test Retest

Group Group
Below 70 Neurotic Below 70 Neurotic
(N 69) _ (N - 69)_ ... (N = 69) (N ** 69)
Score f cf* f of* f ct% f
16 2 100.0
15 97.1
14 1 97.1
13 2 95.7 1 100.0
12 5 92.8 4 98.6
11 1 100.0 4 85. 5 5 92.8
10 98.6 10 79.7 2 85.5
9 1 98.6 4 65. 2 7 82.6
8 3 97.1 9 59.4 1 100.0 6 72.5
7 2 92.8 13 46.4 3 98.6 7 63.8
6 6 89.9 3 27.5 3 94.2 8 53.6
5 5 81.2 8 23. 2 3 89.9 9 42.0
4 16 73.9 3 11.6 6 82.8 7 29.0
3 7 50.7 1 7.2 10 76.8 5 16.8
2 7 40.6 4 5.8 16 62.3 6 11.6
1 6 30.4 10 39.1 2 2.9
0 9 21.7 9 24.6
♦ 1 2 8.7 4 11.6
* 2 3 5.8 3 5.8
♦ 3 1.4 1 1.4
«■ 4 1 1.4
N 69 69 69 69
Mean 3.07 8.06 2.10 6.46
S.D. 2.86 3.16 2.33 3.31
- 209 -
telowM M a<challN*6?
Neurotic (Matched), N* 69
( • /• )
Trequency
£h
In itia l Test Score
Figure 10. Frequency Distributions of Initial Test Scores for

the Below 70 and Neurotic liatcned Criterion Groups
- 210 -
below 7o (Matched^ Nr 6y
Neurotic (Matched), H*
¥
I
m.
+H +3 4Z +1 * 3 * * 4 7 6
Xvntial Test
Figure 11. Cumulative Percentage Distributions of Initial Test

Scores for the below 70 and Neurotic Matched
Criterion Groups
- 211 -
U 4I4U! !|' 1" I" 1ill 1I1IIII111I[I111111U-UlLUlUlU-lgT

£r±m Lffl fffi
EE -
5elow To (Matched^ Ns67
Eli u:i
Neurotic (Matched), N* 67
3533
"Frequency (*/*)
T,ete-s +
Figure 12. Frequency Distributions of Retest Scores for the

Below 70 and Neurotic Matched Criterion Groups
- 212 -
below 7o (Matched), Nr 6Y
Neurotic (Matched), N* t>9
(_%)
too
Freûenc^
li
1:
a
ulafive
iW
Kete. 5 t S core.
Figure 13. Cumulative Percentage Distributions of netest Scores for

the below 70 and Neurotic Matched Criterion Groups
- 213 -
reduce the number of false positives to 10.1$ but would identify on
ly 58$ of the neurotics; a score of 3 or above would pick up more of
the neurotics (88.4$) but would increase the number of false positives
to 37.7$« At the lower tail of the distribution 24*6$ of the controls
score below the lowest score of the neurotics.
Both the Initial Test and the Retest scores significantly
differentiate the control subjects from the neurotics, with the Ini
tial Test showing slightly higher discriminating power. Bi-serial
correlation coefficients between these scores and the criterion are
not reported because the equal number of subjects in the matched
groups does not approximate the normal-neurotic ratio in the general
population. (See Chapter II for an elaboration of this point.) Since,
however, each of the two scores is significantly related to the cri
terion and since the correlation of the Initial and Retest scores is
only moderately high^, there was a strong possibility that a combined
Initial and Retest score would possess greater discriminating power
than either test score alone. Accordingly, the Initial Test and Re
test scores of each subject in the matched criterion groups were
summed into what will hereafter be referred to as a Combined score.
The Initial, Retest, and Combined scores for each of the 138 cases
in the matched groups appear in Tables LXIII and LXIV of the Appendix;
frequency and cumulative percentage distributions are tabulated in
Table UCV and plotted in Figures 14 and 15* The means are 5*17 and
______________________________
1. The correlation between Initial and Retest scores for the entire
group of 393 normal and neurotic subjects is .67* The signifi
cance of this correlation in estimating test reliability will
be discussed in a subsequent section of this paper.
- 214 -
14*52 for the Below 70 and Neurotic Matched Criterion groups, respec
tively— a difference yielding a critical ratio of 11*00. As with the
Initial and Retest scores, approximately one-third of the normals
(31*9$) obtain scores below the lowest score made by the neurotics.
Discrimination at the upper end of the distribution is, however, con
siderably better: 43*5$ of the neurotics exceed the highest score of
the controls in contrast to proportions of 14*5$ and 27*5$ for the
Initial and Retest scores, respectively. This lesser degree of over
lapping is reflected in generally better differentiation at all points
of the distribution with maximal discrimination in the score range
from 7 through 10. A score of 8 or above, for instance, would screen
out 89.9$ of the neurotics at the expense of 23.2$ of the controls.
These figures may be compared with false positives of 26.1$ and 37*7$
for an 88.4$ pick-up on the Initial Test and Retest, respectively.
The frequency and cumulative percentage distributions for
the Recall scores are given in Table LXVI, and the cumulative per
centages for the matched criterion groups are plotted in figure 16.
The means for the Below 70 and Neurotic groups are +.89 and +.07, re
spectively— a difference yielding a critical ratio of 4*82. The close
contiguity of the cumulative frequency curves is in marked contrast
to the greater separations achieved by the Initial and Retest scores,
and the smaller number of signs for which the Recall Test is scored
also raises some question as to the stability of the obtained scores.
Certainly the Recall Test does nothing that the Initial Test and Re
test cannot do better, but the question remains as to whether its

- 215 -
TABLE LXV
Distribution of Combined Initial and Retest Scores, Cumulative

Percentages, Means, and Standard Deviations for the Matched
Criterion Groups (Below 70 and Neurotic)
Combined Gtoud
Initial Below
and 70 Neurotic
Retest (N = 69) . (N « 69)
Score f cf% f cf*
26 1 100.0
25 2 98.6
24 1 95.7
23 2 94.2
22 2 91.3
21 4 88.4
20 2 82.6
19 3 79.7
18 3 75.4
17 5 71.0
16 5 63.8
15 1 100.0 6 56.5
14 4 98.6 3 47.8
13 1 92.8 4 43.5
12 1 91.3 2 37.7
11 4 89.9 7 34.8
10 0 84.1 1 24.6
9 1 84.1 6 23.2
8 4 82.6 3 14.5
7 11 76.8 3 10.1
6 10 60.9 1 5.8
5 3 46.4 1 4.3
4 5 42.0 1 2.9
3 2 34.8 1 1.4
2 7 31.9
1 3 21.7
0 5 17.4
♦1 2 10.1
4-2 2 7.2
*3 1 4.3
♦4 1 2.9
4-5 0 1.4
4-6 1 1.4
N 69 69
Mean 5.17 14.52
S.D. 4.47 5.49
- 216 -
* |
$
EE
^ Q « > 4 O <0 « >

ft SI ft 3 3
- 217 -
131rni
S. Sl S & & O
r» W| <o5
( % ) A ^ u a n b j j i ^ 9/\i.4P| n i * i n )
- 218 -
TABLE L O T
Distribution of Scores, Cumulative Percentages, Means, and

Standard Deviations on the Recall Test for the Matched
Grouo
Below
70 Neurotic
Recall (N = 69) (N = 69)
Score _ . _ _ f cf* f cf*
-2 A 100.0
-1 6 100.0 16 94.2
0 19 91.3 24 71.0
*1 25 63.8 21 36.2
*2 15 27.5 A 5.8
*3 4 5.8
N 69 69
Mean ♦ .89 ♦ .07
S.D. 1.03 1.00
- 219 -
fcelow Jo(Mafched4Ns6 $ j &

i l j i i i i i ; Neurotic (Matched), N - f m 4
| | | iTT n t i S M
(•/•)
too
7«
C u rviula'H ve. T v c c |u c v \c y
<jo
85
8®
IS
7o
65
6o
SS
SO
45
4°
35
30
25
to
IS
to
S
♦4 *3 ♦* ♦ f o i X
K e c a ll Score
Figure 16. Cumulative Percentage Distributions of Recall Test Scores

for the Below 70 and Neurotic Hatched Criterion Groups
- 220 -
intercorrelations with these tests are such as to add anything to the
discriminating power of the Bender-Gestalt when it is combined with
the Initial and Retest scores*
In order to check upon this possibility, the Initial, Re
test, and Recall scores of each subject in the matched criterion
groups were summed into what will hereafter be designated as the
Total score* The Initial, Retest, Recall, and Total scores for each
of the 138 cases in the matched groups appear in Tables LXVII and
LXVIII of the Appendix; frequency and cumulative percentage distribu
tions are tabulated in Table L H X and plotted in Figures 17 and 18*
The means for the Below 70 and Neurotic Hatched Criterion groups are
4*29 and 14*45, respectively— a difference yielding a critical ratio
of 11*42* Of the neurotics, 43*5$ exceed the highest score of the
controls— a figure identical with the corresponding separation made
possible by the Combined Initial and Retest score* At the lower tail
of the distribution, 27*5$ of the controls fall below the lowest
score obtained by the neurotics compared with a corresponding figure
for the Combined score of 31*9$* Maximal differentiation is present
in the score range of 6 through 9* A score of 8 or above would screen
out 87% of the neurotics at the expense of 24*6/S of the normals*
These figures may be compared with a Combined score of 8 or above,
which results in an 88.1$ pick-up and a false positive rate of 23*2$.
This analysis indicates that the addition of the Recall score to the
Combined Initial and Retest score adds little, if anything, to the
discriminating power of the Bender-Gestalt as scored in the present
investigation•
- 221 -
TABLE U d X
Distribution of Total Scores (Initial plus Retest plus Recall),

Cumulative percentages, Means, and Standard Deviations
for the Matched Criterion Groups (Below 70 and Neurotic)
Group
Below
70 Neurotic
Total ____ .O L E >2)__________________
Score f cf% f ct%
26 2 100.0
25 2 97.1
24 94.2
23 94.2
22 3 94.2
21 6 89.9
20 2 81.2
19 3 78.3
18 4 73.9
17 4 68.1
16 4 62.3
15 1 100.0 6 56.5
14 2 98.6 5 47.8
13 2 95.7 3 40.6
12 2 92.8 1 36.2
11 1 89.9 4 34.8
10 2 88.4 3 29.0
9 3 85.5 5 24.6
8 4 81.2 3 17.4
7 2 75.4 4 13.0
6 8 72.5 1 7.2
5 4 60.9 1 5.8
4 9 55.1 1 4.3
3 4 42.0 1 2.9
2 6 36.2 1 1.4
1 3 27.5
0 3 23.2
+1 6 18.8
<-2 3 10.1
*3 1 5.8
♦4 1 4.3
*5 1 2.9
*6 1.4
t-7 1.4
+8 1 1.4
N 69 69
Mean 4.29 14.45
S.D. 4.62 5.78
- 222 -
« O
V <H
•« ;
Z [
♦+
( •/ ^ A t li■S n
- 223 -
V\
to
(•/.) h o x A a o b v A * aAUcinwun")
- 224 -
TABLE LXX
Distribution of Combined Initial and Recall Scores, Cumulative

Combined Group
Initial Below
and 70 Neurotic
Recall (N - 69) (N - 69)
Score f Ct% f cf£
17 1 100.0
16 1 98.6
15 0 97.1
14 1 97.1
13 3 95.7
12 3 91.3
11 4 87.0
10 1 100.0 14 81.2
9 2 98.6 6 60.9
6 1 95.7 7 52.2
7 2 94.2 5 42.0
6 3 91.3 6 34.8
5 9 87.0 6 26.1
4 5 73.9 5 17.4
3 6 66.7 3 10.1
2 11 58.0 1 5.8
1 8 42.0 3 4.3
0 7 30.4
*1 7 20.3
+2 3 10.1
*3 1 5.8
♦4 2 4.3
*5 0 1.4
t-6 1 1.4
N 69 69
Mean 2.30 7.99
S.D. 3.05 3.45
- 225 -
TABLE Lffl
Distribution of Combined Retest and Recall Scores, Cumulative

Criterion Groups (Below 70 and Neurotic
Combined Group
Retest Below
and 70 Neurotic
Recall (N = 69). (N = 69)
Score f cS% f cf*
14 1 100.0
13 2 98.6
12 3 95.7
11 2 91.3
10 2 88.4
9 12 85.5
8 1 100.0 8 68.1
7 1 98.6 4 56.5
6 3 97.1 5 50.7
5 4 92.8 7 43.5
4 5 87.0 • 8 33.3
3 5 79.7 3 21.7
2 8 72.5 7 17.4
1 12 60.9 3 7.2
0 9 43.5 2 2.9
i-l 14 30.4
♦2 4 10.1
*.3 2 4.3
*4 1 1.4
N 69 69
Mean 1.22 6.39
S.D. 2.56 3.43
- 226 -
Other combinations of scores are of course possible*
Table LXX presents the frequency and cumulative percentage distribu
tions of a combined Initial and Recall score, and Table T.rai shows
the distributions of a Combined Retest and Recall score. The criti
cal ratios of the differences between the means of the Below 70 and
Neurotic Matched Criterion groups are 10.16 and 9.94, respectively,
for these two sets of scores. The critical ratios are only slightly
higher than those obtained for either the Initial Test or the Retest
alone on the same group of subjects but lower than that yielded by
the Combined Initial and Retest score. Since the Combined Initial
and Retest score is more discriminating than either the Initial Test
or Retest score alone and since it discriminates as well as the more
cumbersome and time-consuming Total score, the conclusion is warranted
that it is the method of choice where maximal differentiation with a
minimum of time expenditure is the desideratum.
E. Scoring of the Test Records of the Unmatched Criterion Groups.
Can the unmatched criterion groups be differentiated as suc
cessfully as the matched groups? This question was answered by re
scoring the test records of the remaining 255 normal and neurotic
subjects for the selected signs used in the final scoring system.
Table LXXII shows the frequency and cumulative percentage distribu
tions on the Initial Test and the RetestHfor the 86 subjects in the
unmatched Below 70 group and the 39 subjects in the unmatched Neurot
ic group. For the Initial Test, the means of the two groups are 3*56
and 8*59, respectively, which may be compared with the means of 3*07
- 227 -
TABLE LXHI
- Distribution of Scores, Cumulative Percentages, Means and Standard

Deviations on the Initial Test and the Eetest for the
Unmatched Criterion Groups (Below 70 and Neurotic)
Initial 1Test Retest

Group Group
Below Below
70 Neurotic 70 Neurotic
(N = 86) (N - 39) . (N - 86)____ (N -
Score f cf% f cf% f cfSG f cf£
16 2 100.0
15 94.7
14 2 94.7
13 3 89.6 2 100.0
12 2 81.9 94.7
11 1 100.0 3 76.8 1 94.7
10 98.6 4 69.1 4 92.2
9 98.6 4 58.9 1 100.0 5 81.9
8 4 98.6 4 48.7 1 98.6 6 69.1
7 5 94.0 3 38.4 3 97.5 1 53.8
6 10 88.2 3 30.7 6 94.0 5 51.2
5 11 76.6 2 23.1 6 87.0 7 38.4
4 11 63.8 3 17.9 11 80.1 2 20.5
3 . 13 51.1 3 10.3 12 67.3 4 15.4
2 14 36.0 2.6 18 53.4 5.1
1 7 19.7 1 2.6 7 32.5 2 5.1
0 7 11.6 15 24.4
*1 1 3.5 4 7.0
*2 2 2.3 2 2.3
N 86 39 • 86 39
Mean 3.56 8.59 2.55 6.77
S.D. 2.49 3.74 2.35 2.93
- 228 -
and 8.06 obtained for the matched Below 70 and Neurotic groups.
These differences are statistically insignificant, as are the differ
ences between the standard deviations of 2.49 and 3.74 for the un
matched groups and 2.86 and 3*16 for the matched groups. For the Re
test scores, the means are 2.55 and 6.77 for the unmatched groups
and 2.10 and 6.46 for the matched groups; the standard deviations
are 2.35 and 2.93 for the former and 2.33 and 3*31 for the latter.
The dose correspondence of the distributions of the matched and un
matched groups is graphically portrayed in Figures 19 and 20, where
percentile curves of Initial and Retest scores are given.
Similar findings obtain with regard to the Combined Ini
tial and Retest score. The Initial, Retest, and Combined scores for
each of the 125 cases in the unmatched Below 70 and Neurotic groups
appear in Tables LXXIII and LXXIV of the Appendix; frequency and cumu
lative percentage distributions of the Combined scores are tabulated
in Table LXXV. The means of the matched and unmatched Below 70 cases
are 5.17 and 6.12, and the standard deviations are 4*47 and 3*90, re
spectively; the means of the matched and unmatched Neurotic cases
are 14*52 and 15*36, and the standard deviations are 5*49 and 6.09,
respectively. All of the differences between the two normal groups
may be ascribed to random fluctuations, and the same holds true for
the matched and unmatched neurotics. This is clearly seen in Figure 21,
which depicts the percentile distributions of the four groups.
The next step was to rescore all the test records in the 1
Above 70 and 2 Above 70 groups with respect to Initial, Retest, and

- 229 -
rrmr
below 7o (Matched)
below 7o (Unmatched)
Neurotic (Matched)
Neurotic(Unvnatched)
(•/•)
frequency
I
Cum ulative
Hi
1 1
1 tou/xa it
J n i t i a l T e s t ^core.
Figure 19. Comparison of toe Cumulative Percentage Distributions of

Initial Test Scores for the Matched and Unmatched
Below 70 and Neurotic Criterion Groups
below 7o (Matched)
below 7o (Unmatched)
Neurotic (Matched)
NeuroticCUnvnatched)
I oo
("/O
as
frequency
Cumulative
6e
as
Figure 20. Comparison of the Cumulative Percentage Distributions of

Retest Scores for the Hatched and Unmatched Eeloe 70
cnt Neurotic Criterion Groups
- 231 -
TABLE LXXV
Distribution of Combined Initial and Retest Scores, Cumulative

Percentages, Means, and Standard Deviations for the
Unmatched Criterion Groups (Below 70 and Neurotic)
Combined Groun
Initial Below
and 70 Neurotic
Retest . (N = 86) (N » 39)
Score f cf* X cf*
29 1 100.0
28 0 97.3
27 1 97.3
26 1 94.7
25 0 92.2
24 1 92.2
23 0 89.6
22 4 89.6
21 1 79.4
20 2 76.8
19 0 71.7
18 4 71.7
17 0 61.4
16 2 100.0 2 61.4
15 1 97.5 6 56.3
14 1 96.3 0 41.0
13 3 95.1 1 41.0
12 2 91.6 3 38.4
11 3 89.3 1 30.7
10 3 85.8 3 28.2
9 5 82.4 2 20.5
8 7 76.6 3 15.4
7 6 68.4 2 7.7
6 14 61.5 0 2.6
5 13 45.2 0 2.6
4 5 30.2 1 2.6
3 5 24.4
2 6 18.6
1 3 11.6
0 4 8.1
♦1 3 3.5
N 86 39
Mean 6-12 15.36
S.D. 3.90 6.09
- 232
4*
+>
s i 1 1
s i a!
si
4 1
|S
if
i?
P
33
ti
%*
o
l«
si
s?
H X3
a u
Is
• •
JG £
4» 4>
*4 h
o ■
■aH •
h
h O
| ft fv j3 i;
5 3: * 8 g V
V tt w %
W» Q Vft « a o 'K>
j »K>o w
•o ^ « R ♦ r*» Po* <—# o—
(•/.) AoMsnba.^ 9vn}P|nvMn}
- 233 -
Combined Scores and then to cross-compare all the criterion sub-groups.
The Recall Test scores were omitted in the comparison of the unmatched
groups because of their negligible contribution to the total variance
of scores in the matched group comparisons. Table LXXVI presents the
distributions of scores, means, and standard deviations on the Initial
Test and the Retest for the three control subgroups (Below 70, 1 Above
70, 2 Above 70), the Total Control group, and the Total Neurotic group.
Comparable data for the Combined Initial and Retest score appear in
Table LXXVII, and the Initial, Retest, and Combined scores for each of
the 130 cases in the 1 Above 70 and 2 Above 70 groups are given in
Tables LXXVIII and LXXIX of the Appendix.
The frequency distributions of Initial Test scores are
plotted for the Total Control and Total Neurotic groups in Figure 22.
The degree of overlapping of the frequency polygons is somewhat
greater for these groups than for the Below 70 and Neurotic matched
groups— a result consistent with the interpretation that the matched
cases are members of more sharply contrasting groups. Thus it is
seen that only 14$ of the Total Control cases fall below the lowest
score of the Total Neurotic group, whereas the figure for the matched
groups was 30.4$. Similarly, only 6.5$ of the Total Neurotic group
exceed the highest score of the Total Control cases, whereas the figure
was 14.5$ for the matched groups. Disregarding these minor differences,
the fact remains that the controls and neurotics are reasonably well
separated on the basis of Initial Test scores, as the percentile curves
in Figure 23 patently indicate. The fundamental congruence of the Be
low 70, 1 Above 70, and 2 Above 70 curves and their clear-cut separation
- 234. -
oJ
to O ' CM
mo
® t ^ ' t ® © C ^ N O ' C | \ 0 0 tOvO -vt •
>0m
3 S
® sts
b
O
85'
a rH m
+> g , NO
o 0
® o aO f0i »II NHHtnHmOrlO'nHWNO'H
r-1rHCM'4,'4'Cr>rr\C^r—I m CM CM
49 h
ffla to
6-t O CM
pS
a
U S3
® o
$? ® CM m ~4
® 4* I* vO CM --t
-g 43 0
0 *H ® SRl. CM m CM
0 h 4» NO
o ®
+> PS CM ■*
© o
® *rl
m m
P- ■'*
jo ° I CMCMm'^rHmcHC^C'-CMm to CM CM
•H 0 rl i
— I1—I vO
»
: .*
ur> m m
® o tr>
H
«© ■ CMvOO'O'C-CM^fE'-'ttOmi m CM CM
H CM CO H CM m
0 m pH
O
® H
0 O
0 14 o to nQ CO
•H -P •h o cm m
+3 0 3 +> H
0 O <•4 Sf m m P - f - -4 tO Cf\ P - m O vO -4 t 1 to to m
O 1—I rlrl pH o
t H 8 -
® as
Q as'
T3
f4 H m to CM
0 H O W m p-
£ h CM • •
Cm pH C M ' ^ « M ^ r \ ( V O ' t W O t v\t0lft'0 rH m cn cm
O 0 ■ pHpHCM-*^3-C«>-tCMCM to
1 49 E4 O CM
CO 0 n o as
®
S H
a
0 S © CM O' O
► vO tO v O
■CM i
to O CMCr\Cr\CMCMC\jHC^tO''0'tr-|rH CM m CM
0 sO
£ £
0
^
CM
*
a
0
*> • to CM O
to o > P'O t o to
0 43 • •
CMHf^'t'*CM©HH'ttOH C*\ CM
O H
r pH rH pH 58
CO
«H O
o t>- * m 't P-
mvO
0 *
o o <♦4 CM pHC^C'-sONOE-OrHC<\\Or'Mn m m cm
H • I pH rH CM CM CM pH pH m
« m
m 55
43
® *4 0
O \Q m ' + m c M p H O O'ttOC-'O m ^ t m C M r H O H C M ^ M 0 Q
u pH pH rH pH pH rH pH + + * ♦ ® •
CO as a co
- 235 -
TABLE LXXVII
Distribution of Combined Initial and Retest Scores, Means, and

Standard Deviations for the Below 70, 1 Above 70, 2 Above
70, Total Control and Total Neurotic Criterion Groups
Combined GrouD
Initial Below 1 Above 2 Above Total Total
and 70 70 70 Control Neurotic
Retest (N - 155) (N « 68) (N - 62) (N « 285) (N - 108)
Score f f f f f
29 1
28 0
27 1
26 2
25 2
24 2
23 2
22 6
21 5
20 1 1 4
19 1 1 3
18 1 1 2 7
17 1 2 3 5
16 2 2 0 4 7
15 2 0 1 3 12
14 5 0 2 7 4
13 4 2 0 6 5
12 3 3 2 8 4
11 7 1 1 9 8
10 3 6 3 12 4
9 6 4 9 19 8
8 11 7 9 27 7
7 17 3 3 23 4
6 24 5 3 32 1
5 16 6 5 27 1
4 10 8 5 23 2
3 7 3 7 17 1
2 13 6 3 22
1 6 4 0 10
0 9 4 3 16
♦1 5 0 1 6
♦2 2 1 1 4
♦3 1 1
♦4 1 1
♦5 0 0
*6 1 1
N 155 68 62 285 108
Mean 5.70 6.57 7.11 6.26 14.85
S.D. 4.16 4.66 4*64 4.48 5.71
- 236 -
from the Total Neurotic curve leave little doubt that these cases
were drawn from a larger population whose constitution differs radi
cally from that of the neurotic group.
The findings for the Retest scores are similar. There is
more overlapping for the Total Control and Total Neurotic groups than
for the matched criterion groups, according to Figure 24. Of the
Total Control cases, 20.4$ fall below the lowest score of the Total
Neurotic group, as compared with 24*6$ for the matched groups; only
6.5$ of the Total Neurotic cases exceed the highest score of the Total
Control group, as compared with 27.5$ for the matched groups. Never
theless, the percentile curve for the Total Neurotic group is signifi
cantly set apart from the curves of the three control groups, as
Figure 25 plainly shows.
As with the Initial and the Retest scores, the frequency
polygons of the Combined Initial and Retest scores for the Total Con
trol and Total Neurotic groups in Figure 26 overlap to a somewhat
greater degree than the corresponding curves for the Below 70 and
Neurotic matched groups. Only 21.4$ of the Total Control cases fall
below the lowest score of the Total Neurotic group, as contrasted
with 31.9$ for the matched groups; only 19.4$ of the Total Neurotic
cases exceed the highest score of the Total Control group, as con
trasted with 43*5$ for the matched groups. But again, the most sig
nificant finding is the isolation of the Total Neurotic percentile
curve from the three nearly congruent curves of the control sub-groups
in Figure 27.
- 237 -
Total Control, N*285

Total Neurotic,N*I08
c%>
30
frequency
10
♦6-
+s
I n i t i a l Test Score
Figure Figure <U. Frequency Distributions of Initial lest scores for the ie
- 238 -
11:1ililliiiil!U U ;J111!!!11:i|;|!i1
i &elow7o( Nsl55
1Above 70, N=68
-iit‘Hi
2Above7o,N*62
Total Neurotic, N* 108
10o
C0/°)
F rc^ju cd ty
rtftfen
C u m u lative
9±tism
t» +x +i o i a s 4 5 * r 8 /o // iz is i t is is
I n i t i a l Tes+ Score
Figure 23. Cumulative Percentage Distributions of Initial Test Scores

for the Below 70, 1 Above 70*, 2 Above 70, and Total
Neurotic Groups
- 239 -
Total Control, N*285

Total NeuroNc,N*IO& .j..
30
£%)
22
Frequency
20
12
IO
X __
IO II- /*+-
15 15
Figure 'at*. Fr»-MUoncy distributions of Retest bcores for tne i'ot’

-i
uontrol 'mo Total Neurotic Groups
t
I
240 -
5elow70, N-155
1 Above 70, Ns6 8
2 Abovc7o,Nl 6 2
Tota.1 Neurotic,N«IOS
/0O
V bo
U- «/5
35
30
*5
20
15
/O
Ke+esi" Score
Figure 25. Cumulative Percentage Distributions of Retest ocores

for the Below 70, 1 Above 70, 2 Above 70, and
Total Neurotic Groups
- <ti*± -
aI
i>o.~0
LLli mi m*
(%) A ? u 3 n b » J i
2 3
h
£i
« 0
- 242 -
A3M9nb9-l+
J A U e i n w O
Figure 27* Cumulative Percentage Distributions of Combined Initial and Retest Scores for
the Below 70, 1 Above 70, 2 Above 70, and Total Neurotic Groups
- 243 -
Can the three criterion sub-groups be differentiated from
one another on the basis of the test scores? As Tables LXXVI and
LXXVII indicate, there is a progressive increase in the mean Bender-
Gestalt score which parallels the degree of "psychometric abnormality"
(as defined in terms of MMPI scores) of the three groups. ' For the Ini
tial Test, the mean scores of the Below 70, 1 Above 70, and 2 Above 70
groups are 3-34, 3.82, and 3-89, respectively; for the Retest, the
mean scores are 2.35, 2.75, and 3.23, respectively; and the Combined
Initial and Retest scores are 5-70, 6.57, and 7.11, respectively.
While the trend is in accord with expectation, the question immediate
ly arises as to whether any of these differences are reliable. Table
LXXX, which gives the means, differences between means, and critical
ratios of the differences between means of the Initial, Retest, and
Combined scores for the matched and unmatched criterion groups and
for the total criterion groups, provides the answer. None of the dif
ferences between the Below 70 and 1 Above 70 groups for any of the
three kinds of scores is statistically significant; the same holds
true for differences between the 1 Above 70 and 2 Above 70 groups.
When the Below 70 and 2 Above 70 groups are compared, however, differ
ences significant at the 5% level are obtained for the Retest and Com
bined scores. None of the differences in any of these control sub
group comparisons meet the more rigorous demands of the 1% confidence
level, and when these small differences are contrasted with the great
disparities in score between the normal and neurotic subjects, one is
compelled to conclude either that the Bender-Gestalt Test is not

- 244 -
rH
0}
O o CM p
$9
•H . o pi CA O p o
Means, Differences between Means, and Critical Ratios of the Differences between Means of the
P P
to C
M
t §
Initial Test Scores, Retest Scores, and Combined Initial and Retest Scores for the Matched
® O
h
O
o
CO
a
® v\ O'
s ® CA CM CO IA CA
& _ •
p a O' O' to
®
p
and Unmatched Criterion Groups and for the Total Criterion Groups
d
as
C" CM
rH IA
CMp o c*- r~ rH P IA
H CA r- «
a CA H CMCO
o • • « • • • • • « • t •
a IA MS’ P ia IA P IA C*- p I" P
rH
as
o o
iH *H
P o
P P
O' o•
•H a] to CM C
M
h a!
® a
8 ®
o
COa d m
® ® a P CM CO NO
p h ® as CA C
M 00 O'
09 © £ ©
®
P Ch ®
a CA
a •H p
Q
O p IA IA IA CA CA m <A (A O '
H -sj- iat- CA C*- CA CM t> CM P IA
® • * • t • • • • • • • •
a C
Mp C
Mp CMC
M CM CA CM CA CMP
a! „
o O
rH rH CM O to IA
P P
P P CM CA 8
O'
O S)
®
O
o d <&
CO ® a o C«\ 00
rH
® as
£ ® ON o CA 58
a) IA Nf
® 3
P
d C^p -
C4O
-to
oi
P O' -'tO' CM O' COP
cS
®
o« o• ia
t •
va
• •
CA tO
• •
to to
• •
IA CM
• •
a CA to CA tO cr\ CA CA CA CA CAtO
O' O' P O' IfN CO IA CM
NOp CO CA Vf\
rH
«A P
rH
58 PCM IA CO
CO O
CM rH
■93tS
® ©
P P
O o e>
p p rH *H
OP
i
e» c» o
u
C
JJ
P PP
o 3
p. P ® P © o a
BI BI«jj
£ O £ O
§! OP
© ©
rH
©
*ai rH <!
®
3o 3o
m a n rH m cm H C
M
- 245 -
sufficiently sensitive to make such fine distinctions within the nor
mal population or that the groupings arrived at on the basis of Minnesota
Multiphasic scores are artificial because the MMPI itself is of dubious
validity or limited sensitivity.
Table LXXX also recapitulates the findings of paramount in
terest in the present study, namely, the comparisons between the Below
70 and Neurotic matched groups and between the Total Control and Total
Neurotic groups. All the differences for the Initial, Retest, and Com
bined scores are highly significant. Apparently, either the Initial
score or the Retest score may be used with equal effectiveness, but
the Combined score is the most discriminating and the most reliable
of all.
F. Effect of Education. Age, and Intelligence upon Test Scores
The similarity of the results for matched and unmatched
groups suggests that such factors as education and age bear little or
no relationship to the Bender-Gestalt test scores of adult subjects.
One would therefore expect that intelligence would be uncorrelated
with Bender-Gestalt test performance. These expectations are confirmed
by Table LXXXI, which summarizes the correlations of these variables
with the Initial and Retest scores of the Total Control and Total
Neurotic groups. The number of cases upon which these correlations
are based is not always equal to the total number of cases in the
group because data are missing in some instances. The correlation be
tween years of schooling and Initial score is -.11 for the Total Con
trol group and -.12 for the Total Neurotic group; for the Retest scores,
- 246 -
the correlations are -.10 and -.06, respectively. There is thus a
very slight tendency for the better educated subjects to make lower
(more favorable) scores, but not too much should be made of this since
the correlations are not statistically reliable.With respect to age,
the correlations are .06 and .04 for the Initial and Retest scores of
the controls and .20 and .16 for the neurotics, but none of these is
significant either. As with education, the correlations between Otis
I.Q. and test score are negative but small, namely, -.15 for the Ini
tial Test and -.17 for the Retest. Wechsler-Bellevues rather than
Otis Gammas were administered to the neurotic group, but the results
are similar. For the 84 neurotics on whom Wechsler-Bellevue scores
were available, the correlations with Initial and Retest scores are
-.10 and -.15, respectively.
TABLE LXXXI
Correlations of Grade, Age, and Intelligence Test Scores with Scores

on the Initial Test and the Retest for the Total Control and
Total Neurotic Criterion Groups
Group
Total Control Total Neurotic
N Initial Retest N Initial Retest
Grade 282 -.11 ±. .04 -.10 * .04 108 -.12 * .06 -.06 ± .07
A e ra 277 .06 * .04 .04 A .04 107 .20 * .06 .16 a .06
Otis I.Q 285 -.15 ± .04 -.17 t .04
W-B I.Q. 84 -.10 * .07 -.15 ± .07
- 247 -
G. Scoring of the Initial Test Records of the Matched

Criterion Groups with Pascal and Suttell»s Signs
As noted in Chapter II, Pascal and Suttell (76) are the only
investigators other than the present writer who have combined Bender-
Gestalt signs into a total score which, it is claimed, will differenti
ate normals from abnormals with a reasonably high degree of accuracy.
Pascal and Suttell claim that their scores will distinguish normals
from neurotics, normals from psychotics, and neurotics from psychotics,
although the latter differentiation is admittedly more subject to error.
From the standpoint of the subject of the present paper, the main in
terest centers around the relative effectiveness of Pascal and Suttell1s
signs in distinguishing between the normal and neurotic cases employed
in the present investigation. This question assumes significance in
view of the fact that many of the signs used by Pascal and his colla
borator were similar to signs tentatively tried out in the present
study but rejected for the final scoring system because of lack of
statistical validity. If their scoring scheme is valid, it should
hold up with the present group of controls and neurotics, and the suc
cess of this cross-validation would add immeasurably to the general
usefulness and applicability of their system. The Below 70 and Neurot
ic Matched Criterion groups were selected for this purpose because
they are sharply contrasting groups and because such factors as race,
nationality, sex, age, education, and marital status were held con
stant, eliminated, or neutralized.
Accordingly, the present writer learned Pascal and Suttell*s
scoring scheme and scored several of the test records which appear in
- 248 -
their book until his total scores closely approximated those given in
their scoring manual. All 138 cases in the matched criterion groups
were then scored in accord with this method. Tables LXXXII and
LXXXIII of the Appendix present the detailed scoring of each record
by giving the signs and thejLr assigned weights found in each of the
figures and in the "configuration", the total of these weights (which
constitutes the raw score), and the Z score corresponding to the
weighted total. The raw scores were then tabulated as in Table
LXXXIV, which gives the frequency distributions for the Below 70 and
Neurotic matched groups for the groups as a whole and also by educa
tional breakdown within each group and in addition shows the means,
standard deviations, and Z-score equivalents for each of the sub
groups. Figure 28 graphically depicts the raw score distributions
for two matched groups as a whole. The raw score means, differences
between means, and critical ratios of the differences between means
are given in Table LXXXV.
Education is unrelated to the size of the score of either
the controls or the neurotics, as is evident from the fact that the
grade '8-11, grade 12, and college trained subjects obtain almost iden
tical mean scores within the criterion group of which they are a mem
ber. Pascal and Suttell claim that education is a significant factor
and present separate norm tables for college and non-college trained
individuals. Of much greater significance is the failure of their
scoring scheme to distinguish between any of the normal and neurotic
groupings. For the normal and neurotic subjects having less than
- 249 -
CAOCM'OcACAr-Ô'flOi^flOCM ft
sO
■g “
o m
-tcA
HOOHWWvO^^HNH CA O' CM
u
■H CH e o CM CM H s
-P vO
O
S"
®a r n in
Z' VArH
o
*^3
CM CM I 'O
3 mcn
<M HON(MOONr\(M^NOH 0 > t0 vO CA
rH CM rH vO
o
§
s IT\lf>
U"•\lf•>
H (^(^00 O (M O' uaO .
rH rH rH 'O CM H 3
«a £ -
'itC-
H O ' 4 ,'0CMlA''i'rH CAC-tO vO
o CM CM sO
t'- O'!
NO
O M
rH
£
m •85 O O
cm m
• •
C- '■*o 'O
CM CM H UA
MACA
H CO
• •
«H O' MA CM CO
rH CM i—I IA
I
• rH
^ d)<H
® •>i'0"i'p'^0''^0''4-
sO 0''4'
>A UA ~-fr~T CA (A CM CM ■H O'
o ► o .
I ° I I I I l . ilt i i i T 3 « 53*9
® •
13 eg & m «r»-♦ ■'$■caA CA CM CM lAQino
mao u * o « rPl
\ O m O
a a co ci S' ® Z
- 250 -
TABLE LXXXV
Raw Score Means, Differences between the Means, and Critical Ratios of
the Differences between Means on the Initial Test Record for the
Matched Criterion Qroups (Below 70 and Neurotic) Scored
According to Pascal and Suttellrs System
i •
Groun
Below
70 Neurotic Difference
,_(N - 621... . (N -69) between Critical
Grade Mean Mean Means Ratio
8-11 25.15 28.60 3.45 .72

12 24.20 27.55 3.35 .99
College 27.45 29.40 1.95 .62
Total
Group 25.55 28.45 2.90 1.38
four years of high school the critical ratio of the mean difference
is .72; for those having four years of high school, the critical ratio
is .99; and for subjects with one or more years of college the ratio
is .62. For the groups as a whole the raw score means of the control
and neurotic cases are 25.55 and 28.45, respectively, yielding a
critical ratio of only 1.38. The Z-score equivalent of the raw score
mean for the control group is 60 and for the neurotic group, 63. The
Z-score mean is higher for the normals than it should be if Pascal
and Suttell*s standardization is accurate, while it is below expecta
tion for the neurotics.
Pascal and Suttell*s administration of the Bender-Gestalt
differs from that employed in the present study in two respects:
(1) the time required to reproduce the figures is not recorded, and
- 251 -
Below* (Matched),
Neurotic (Matched! Ns6?
40 4s so ■& so ts jr©
I n i + i a l Test- S c o r e £ P a s c a l a n d S u ttell's . S i^ n s
Figure 28. Frequency Distributions of Initial Test Scores for the

Below 70 and Neurotic Criterion Groups,
Using Pascal and Suttell1s Signs
- 252 -
(2) sketching is prohibited. However, sketching occurs relatively
infrequently in the matched group cases, and it is difficult to see
how the single factor of recording time could wipe out the presumed
differences in test performance which are implicit in their method
of scoring. Moreover, this one factor would not account for the
relatively high score of the controls (Z-score of 60) and the rela
tively low score of the neurotics (Z-score of 63). Pascal and Suttell
report Z score means of 50 and 68 for their control and neurotic groups,
respectively, the mean for their entire patient group of neurotics and
psychotics being 74. They go on to suggest that "one standard devia
tion above the mean, or a Z score of 60 would seem to be a reasonable
cutoff score since theoretically that should include 84 per cent of
the nonpatient population, which seems about right in view of what iB
known about the incidence of psychiatric illness in the so-called
normal population" (76, page 35). Because their nonpatient distribu
tion was somewhat skewed, a cutoff score of 60 actually screened out
79/6 of their neurotics and 2056 of their normals. If a cutoff score
of 60 were applied to the matched groups in the present study, it
would screen out only 59% of the neurotics at the high cost of 4656 of
the controls.
H. Rasu^ts of the Cross-Validation Study
If Pascal and Suttell’s scoring scheme holds up so poorly
under the strain of cross-validation, one may well speculate as to
the fate of the scoring method developed in the present investigation
under similar circumstances. The procedure of scoring the test records

- 253 -
of the very subjects upon whoa the item analysis is based is open to
the objection that the summation of chance differences may result in
a spuriously high differential that cannot subsequently be duplicated
when applied to other samples drawn from the same population. In per
sonnel selection situations the multiple correlation validity coeffi
cient resulting from the most favorable combination of weighted test
scores obtained by the group upon which the zero order coefficients
were calculated is almost always larger than that found when the
selected test battery is tried out on a new group of applicants. What
will be the extent of this shrinkage in validity in the present in
stance under comparable conditions? Will the validity be reduced al
most to zero, as with Pascal and Suttell's signs, or will the validity
be sufficiently high to warrant the use of the scoring scheme in prac
tical situations encountered in the clinic and the guidance laboratory?
Ideally, it would have been desirable to have obtained sam
ples of normal and neurotic veterans drawn from the same populations
upon which the item analysis was performed, but this procedure was
• not feasible because of the virtually insurmountable difficulty of
obtaining additional neurotic cases from the Veterans Administration
mental health clinics in New York City. With an increased case load
and a retrenchment in staff and with many staff psychologists carry
ing on research of their own with patients available for this pur
pose, it seemed unwise to press the chief psychologists, who had al
ready been exceedingly cooperative, for more cases. The thought then
occurred that the cross-validation study might profitably be conducted
with non-veteran male adults differing markedly in personal adjustment,

- 254 -
for if the test held up under these conditions the conclusions concern
ing validity would not have to be restricted to the veteran population
but would have a wider applicability to the male population at large.
It was decided to select the cases from the files of the
New York University Testing and Advisement Center— a testing and voca
tional guidance agency which serves the metropolitan area and suburban
regions. The vast majority of clients 18 years of age and over who
avail themselves of the Center's services desire educational or voca
tional guidance, but a certain proportion of these individuals are
subsequently found to have personality problems sufficiently severe to
warrant the recommendation that they undergo psychotherapy as an ini
tial step in the solution of their vocational problems. The testing
and counseling procedure consists of (l) an initial interview, in
which personal history data regarding education, employment experience,
health, family background, marital adjustment, interests, hobbies,
special talents, and personal and social adjustment are obtained, (2)
10 to 16 hours of group and individual testing, including the Wechsler-
Bellevue, educational achievement tests, aptitude tests, interest
tests, personality inventories, and projective personality tests

(Bender-Gestalt, Man and Woman Drawings, a sentence completion test;
less frequently, the Thematic Apperception Test, Rorschach, Szondi,
and Rosenzweig Picture Frustration Test), (3) one or more counseling
conferences in which the findings are presented to the client and re
commendations made, and (4) a detailed written report which integrates
biographical and test data and recapitulates the major results of the
entire procedure. Since comparatively few of these clients had

- 255 -
educations below the high school level, the cross-validation cases
were selected from a file of approximately 1,000 clients ranging in
age from 18 to 50 and in education from four years of high school
through college. Each of these cases was evaluated from the stand
point of whether the subject was effecting an unusually good or an
exceptionally poor emotional and social adjustment. Only when per
sonal history data, Wechsler-Bellevue scatters, and personality tests
pointed consistently in one direction or the other and only when these
data were consistent with the impressions of the psychologist who
originally handled the case was the subject selected for inclusion in
the cross-validation groups. The Bender-Gestalt test records were
removed from the folders prior to this appraisal so that they would
not influence the final evaluation. The Bender-Gestalt tests were
in all instances administered with the same instructions that had
previously been used for the veteran subjects but they had not of
course been scored, since no scoring scheme was available at the time.
Most of the poorly adjusted subjects who were selected would probably
have been diagnosed as neurotic by a psychiatrist (no cases in which
there was any Suspicion of psychosis were included), but the question
of diagnosis was avoided by merely designating the cases as satisfac
torily or unsatisfactorily adjusted.

A total of 118 cases which satisfied the above criteria were
finally selected. Of these, 54 (39 males and 15 females) were desig
nated as the TAC Total Satisfactory Adjustment group, and 64 (43 males
and 21 females) as the TAC Total Unsatisfactory Adjustment group. Ori
ginally, it had been intended that only males would be included, but a
- 256
smaller number of female records was added In order to shed some light
on the question of whether the method of scoring is valid for both
sexes. These groups were subdivided into several sub-groups which are
described in Table LXXXVI with regard to size, education, age, marital
status, and sex.
As a first step, the male subjects were matched, pair by
pair, for educational background, age, and marital status. Matching
was possible for 58 of the 82 male cases, and the two groups so
matched were designated as the TAC Male Satisfactory Adjustment group
(N - 29) and the TAC Male Unsatisfactory Adjustment group (N • 29).
The subjects in these two groups range in education from the 12th
grade through four years of college, with a mean grade of 14*41 in
each instance. The age range is 18 through 49, the mean ages of the
satisfactory and unsatisfactory matched groups being 25*56 and 25.68,
respectively. Sixty-nine per cent of the satisfactory matched group
are single compared with 79.356 for the unsatisfactory group.
The sign-by-sign scoring of the Initial Test and Retest
records of all the males in the cross-validation groups appear in
Tables LXXXVII through XC of the Appendix. Frequency and emulative

percentage distributions of Initial and Retest scores for the matched
male groups are tabulated in Table XCI and ogives are plotted in
Figure 29. On the Initial Test the means are 5*45 and 7.73 for the
TAC Male Satisfactory Adjustment and TAC Male Unsatisfactory Adjust
ment groups, respectively— a difference yielding a critical ratio of
2.96. Discrimination is about equally good in the score range of 4
through 8, reaching a maximum at 7. A score of 8 or above would screen

- 257 -
Sh oa c- t* to
o "4 NO -4
o
Sh
O O
rH 0 P p O' p
fl
O
•H
*
bpW
3 ~
CO
O'
nO
ft
P
•
£ P d i
•H -P
« |
Ta Jm
s
*rl
in p o O' CO nO
•
O *n
& *5 o o O p O' UN
O <sJ CO CM CO CM CM p CM CM
3 tr
«H O *4 NO CM
O a) 31 3 co 51 P
• to 5 O'
*h
m a CO <0 to to to C~ to
d *H
43 -p
q) as
-P 09
CO d vO NO o o to
U\ Cl UN
ft NO rH
• 58 pft
H & § •
U>
a _ IT c- NO P O'
33 a H CM CV CM CM P cQ CM
IT C* C* ^4 \n O'
sO sO 5a P *P NO
■p
d CO «H
d
0 09
•H JS vO 58 C" o O'
p O' oft o to
3013«Sj S3
p p
It
ss
o
£S
o O' O'
H
UN
P
•*4
H
a> a) CM CM cp H C
M u>
u>
0 m
•>+»
3 at -P -P
3 to
-P
33
CO H
rH
3
3
9 l 3 I 3
I
1
3
i
3
•f
3 HI
9
*r“!
3
3
1*n ® ^
3
•r
p
3
1 •o •d
•H ® <s 3 <1
§
3fc
£I 3Io
o
1 i .. S3 Eu & •p
I <s
® u
o
■p
o
<ri
• Vi
3s
o
<d
«p
3 u
3
4»
o
td 4> a 3 !3
o
<2
CQ
as
O -p
■ssa
33
a
§
+!
di
eh &
O tH
 0 ' © f ' t t 0 ' 0 '
• • • • « • • • • ♦ • • * • » •
0'0'0(r\c\<J'©iftO'Wl>,C'0 (•SvO'fl
o • O' OO'O'O'O'tOtOC'-'OsrmcMCMH
5 "S 53^
31 33
•p •»-»ad
*g- C*A ^
• •
S3
a ^
I H O O rlH H ^ N v O C M ^ C M N N O N O ' C- CA
CM
as
o o
•H <0 O s O ' O ' O H O ' O CM i n C- CA -xf
-P <H
d ta
fcv • ••••ftftftftftftftft
O 'O ' O sO fA UA O' i n -tf- O O OA C*A
■H *rl
► -P
© 01
a*! O O 'ff'O 'O 'P 'O in fA C M H
a co Vi m
O <n P in o
-SfrUA
II <g* H O O r i W O I -t'O M C I O I O H
&
ft
-* CM
ft
•p
® §
01 f j
aa
es
3
•<-•
a >g
•> o «j
a SS
*h
o -p « o \© 'O V© vO H to 0 " G t O O t - c o
• < • * • < • * • • • • • • •
«0 O '
-■3 3
a > Vi
■p
o
d H
© cr
8 vOsO'0\prr\CMUACO'^HOC*AC*A'0
0'0'0'00'tDt>m~3-C*ACMHrH
© l a>
b0n<rl
3fl h i
to -p
. o m 3
O -H § "
£• •H
© o d
u W <aj O ' C*- m
h
©
„ D
■p s CM
(U
©
t>
II to
9
1■p 3
0
■p
01 t- O 'O 'O CM CAsO OA xfr r-l m O ' O '
a e & -P • » • • • > « • • • » •
O 'O 'O 'O O 'D O flQ r-i'4 -O 'O 'O
o © 33Soj O O ' O ' CO cS UA xfr CM rH
•> 5 1 S3*1
Vi w It
© M O w P O'
h O
O <H • ♦
O HOdACM'OcrACMVA^tHOCM O' in cm
CO -P CM
n
S3 ©
d Pi
o
•H 0
I5
31
m ^{£}^JI}3p33<,'t0Cws0'A '>!*<r'c't H O '+ 9
© Q•
a a co
- 259 -
CtaatlatiTa Percentage Distributions of Initial Taat and Bateat Scores

for the Matched Male Cross-Validation Groups
■■ ■ ■ e a a a s s a a a s s a a a s ■
g a a a e a a a s io a a a a n a a s i
ifia s a e e a s e e a e e a e a a e S
SSSSS5S5SS5S2S18*221
rifora 29*
8 5R &
l % 3 9 A i f e | n u i r Q
- 260 -
out 55*2% of the maladjusted subjects at the expense of 20.7$ false
positives; for a score of 5 and above, the corresponding figures would
be 86.2$ and 58.6$, respectively. On the Retest, the means are 4.45
and 7.34 for the adjusted and maladjusted cases, resulting in a criti
cal ratio of 3.66. Maximal discrimination is present in the score
range of 5 through 7. A score of 8 or above screens out only 6.9$
of the controls compared with 51.7$ of the disturbed cases, but a
score of 4 or above would pick up 65.5$ of the controls as well as
86.2$ of true positives.
The Initial, Retest, and Combined scores for all the males
in the cross-validation groups appear in Tables XCII and XCIII of the
Appendix; frequency and cumulative percentage distributions for the
matched male groups are tabulated in Table XCIV, and ogives are
plotted in Figure 30. The means of 9.90 and 15.07 for the adjusted
and maladjusted male subjects yield a critical ratio of 3.64, and maxi
mal differentiation occurs in the score range of 13 through 16. A
score of 14 or above weeds out 62.1$ of the maladjusted subjects at
the cost of only 17.2$ false positives; 86.2$ of the poorly adjusted
and 55*2$ of the well adjusted cases are screened out by a score of
10 or above.
The distribution of Recall Test scores for -the matched male
groups is given in Table XCV. Although the direction of the differ
ence is in accord with expectation, the difference itself is statis
tically insignificant, as the critical ratio of 1.31 attests.

The findings for the entire group of males are similar to
- 261 -
TABLE XCIV
Distribution of Scores, Cumulative Percentages, Means, and Standard

Deviations of the Combined Initial and Retest Scores for
the Matched Male Cross-Validation Groups (TAC Satisfactory
Adjustment and TAC Unsatisfactory Adjustment
Group
TAC Male TAC Hale
Satisfactory Unsatisfactory
Adjustment Adjustment
(N - 29) (N - 29)
Score f cf$ f cf$
32 1 100.0
• • 96.6
• ••
24 1 96.6
23 0 93.1
22 1 100.0 2 93.1
21 0 96.6 0 86.2
20 0 96.6 2 86.2
19 0 96.6 2 79.3
18 0 96.6 1 72.4
17 0 96.6 4 69.0
16 3 96.6 1 55.2
15 86.2 2 51.7
14 1 86.2 2 44.8
13 2 82.8 0 37.9
12 3 75.9 2 37.9
11 3 65*5 1 31.0
10 3 55.2 4 27.6
9 1 44*8 1 13.8
8 4 41.4 0 10.3
7 1 27.6 1 10.3
6 24.1 0 6.9
5 1 13.8 0 6.9
4 1 10.3 2 6.9
3 6.9
2 1 6.9
1 1 3.4
N 29 29
Mean 9.90 15.07
S.D. 4.55 6.12
- ZbZ -
^ n ^ s 3 $ « s * a
(% ) A o u 9 n b 9 j ± “9 A » ^ c |n v M n ^
- 263 -
TABLE XCV

Deviations on the Recall Test for the Matched Male Cross-Valida
tion Groups (TAC Satisfactory Adjustment and TAC Unsatisfactory Adjustment
Group
TAC TAC
Recall (N_ - 29) (N - 29)
Score f cf$ f cf*
-2 2 100.0
-1 3 93.1
0 13 100.0 10 82.8
♦1 7 55-2 8 48.3
+2 8 31.0 4 20.7
*3 1 3.4 2 6.9
N 29 29
Mean *.90 ♦ .52
S.D. .92 1.25
those for the matched groups, as Tables XCVI and XCVII, which give the
frequency and cumulative percentage distributions of the Initial, Re
test, and Combined scores, and Figures 31 and 32, which present the
corresponding ogives, indicate. On the Initial Test, the mean of
the TAC Total Male Satisfactory Adjustment group is 5*13, and the mean
of the TAC Total Male Unsatisfactory Adjustment group is 7.44, the
critical ratio of the difference being 3*80. The means of the Retest
scores are 3*92 and 6.74, respectively, and the means of the Combined
scores are 9.05 and 14*30, the critical ratios of the differences be
ing 4.15 and 4.53* As with the matched male groups, maximal differ
entiation occurs in the score range of 4 through 8 for the Initial

- 264 -
o• c*
• c-
• c- • o• • • • • •
• on cv •
cm cm \o
• « • • • •
Distribution of Scores, Cumulative Percentages, Means, and Standard Deviations
O ® cn OC^r-E'-tf'K*\er\vpWV\->i-f*0«00'0"4-C\l
O O'O'O'O'O'tOC'tÛN'TONONH
on the Initial Test and the Retest for the Total. Male Cross-Validation
Groups (TAC Satisfactory Adjustment and TAC Unsatisfactory Adjustment)
•38
■P •<-»
’ r-vO
a} T3 ^
U <4 ONvO ON
-T
•p
ta £
9©
C3
-P ,
tt!L '+'0 ON CM CM CM
-3 © 0 ^ §
8-^
© i
as g
f)■rjtB $«3•
•P 73
r ) O O r ( « N N > A > UN00 s t O H O' ON CM
% C*N
O C '- C '- C ^ C '- O N ^ i '- ' t t Q O " 0 '0 ONvO C"

-3d *3
o © c«S 0 'O 'O 'O 'O '0 0 0 0 U N ''T O N C 'iH r -l
.
•P
3Si
• •
to
a HOOOH(OnH'0'*m>*0!(OOI ON t"- CM
3 rH
E-«
3
V S|
O (O C N 'O N (< M 7 H H
• • • • • • • • « • • •
O t ' > 0 ' r l © [ f \ ' 0 ( 0 0 UN UN
O O' O' 00 00 nO UN ON H
ss
Vi ra
a 0 ONnO
•H f? H UN
■P T3 ■ • »
VH iH O O N O N ' O U N O N U N O ' C M O C M O' UN CM
ca ON
o ^ H r ^ H ? i r l H C,'COC^VOU'''*rr'f'i H O r +
o m®
CO SB 3 CO
- 265 -
Qu 9>
I-
l ^H o
nJ A 4O ^ O UOt W~'*m) Q U *>*V 0 K) ^ o *>
(*L) A ‘3 M 3 n b ^ ^ 4 -aAueinvurO
- 266 -
TABLE XCVII

Deviations of the Combined Initial and Retest Scores for the
Total Male Cross-Validation Groups (TAC Satisfactory
Adjustment and TAC Unsatisfactory Adjustment)
Group
Combined TAC Total Male TAC Total Male
Initial Satisfactory Unsatisfactory
and Adjustment Adjustment
Retest (N - 39) _(N = 43)
Score f cfi f cfi
32 1 100.0
• •
• •
e
24 1 97.7
23 0 95.3
22 1 100.0 2 . 95.3
21 0 97.3 0 90.7
20 0 97.3 4 90.7
19 0 97.3 3 81.4
IS 0 97.3 2 74.4
17 0 97.3 4 69.8
16 3 97.3 3 60.5
15 0 89.6 2 53.5
14 1 89.6 2 48.8
13 2 87.2 3 44.2
12 3 81.9 2 37.2
11 A 74.4 1 32.6
10 5 64.1 4 30.2
9 1 51.3 2 20.9
8 A 48.7 0 16.3
7 1 38.5 3 16.3
6 5 35.9 0 9.3
5 5 23.1 1 9.3
4 1 10.3 2 7.0
3 1 7.7 0 2.3
2 1 5.1 1 2.3
1 1 2.6
N 39 43
- Mean 9.05 14.30
S.D. 4-33 6.09
!
- 267 -
(®/.T A ^ U S n b a j a
aAHPinuin^
v e o < o o x i o ( « o
Figure 32. Cumulative Percentage Distributions of Combined Initial and Retest Scores
for All the Male Cases in the Cross-Validation Study
- 268 -
Test and 5 through 7 for the Retest; for the Combined score, the range
is 11 through 16— somewhat wider than previously. The Recall Test
means are +.85 and +.72, respectively, a difference which falls far
below the accepted standards of statistical reliability.
There were not enough cases in the female groups to employ
the matching technique successfully, and there was every indication
from similar findings on the matched and unmatched veterans' groups
in the original validation and on the matched and unmatched male cross-
validation groups that matching was essentially superfluous, since
neither age, education, nor marital status are significantly related
to the test scores. Actually, the satisfactory and unsatisfactory
female groups are similar with regard to education (mean grades of
13.07 and 13.71, respectively) but differ slightly in age (mean ages
of 26.60 and 33.10, respectively) and marital status (73.3% and 61.9%
unmarried, respectively). The frequency and cumulative percentage
distributions of Initial, Retest, and Combined scores for the TAC
Female Satisfactory Adjustment and TAC Female Unsatisfactory Adjust

ment groups are given in Tables XCVIII and XCIX, and the correspond
ing ogives are plotted in Figures 33 and 34-• The Initial score means
of 4.60 and 7.14 and the Combined score means of 6.93 and 11.67 yield
statistically significant differences (critical ratios of 2.82 and
2.56, respectively), but the Retest score means of 3.27 and 4«52 do
not (critical ratio of 1.25), although the difference is in the ex
pected direction. The Recall Test means of +.47 and +.43 are palpably
insignificant. A score of 8 or above on the Initial Test would screen
out 42.9% of the maladjusted females at the cost of only 6.7% false
- 269 -
O tf\E" -♦£*■H -'tf’vO cnto tf\00

u O (T> 00 C'- \5
© 4©h I d
„
•H I «
t o CO
4.52
3.48
O -P 27*1
< ©
H « % <H CMHl^HNHHrnoiNNH iH
CM
■p
£
a
3
«r «
0«nC«-p~<*>C^C»NOC^t'C^
•3 § tr
o 8 nOvOPNt^fÔ r^vOvO
O'«0 00C">r>crv0UH
- © -9
2.55
3.27
fe <CDP
27*
»g
< rlrlO ITl
O i T i O O O r ) ' O H C rlif\«
M I! • •
> O r l r l H M ' ® ' J O ' ,4
Jsfl
8 O COtoNO ITV'^cnrH 7.14
2.92
1 m a 9
© -p
a
I ■p CMCNJO'+pHC\JC'JlT\rIrliH H
Oi
a
e-«
.-I
©
« O cr\ c^sO (c\cc*vO c»\(*\

OO'O'O'O'MMOtMrtH
3
2.45
4.60
HOOOHOI(t\'4CMOC'l CO
©
h OlriOO'«>NOlft'0(,
M V H O H♦ Nr
S.D.
Mean
o H H H
o
CO X
- 270 -
C%) A?uanJ>94 4 dMieiniMn*)

a
Figure 33 • Cumulative Percentage Distributions of Initial and Retest Scores for All
the Female Cases in the Cross-Validation Study
- 271 -
TABLE XCIX

Deviations of the Combined Initial and Retest Scores for the Total
Female Cross-Validation Groups (TAC Satisfactory Adjustment and
TAC Unsatisfactory Adjustment)
Group
Combined TAC Female TAC Female
Initial Satisfactory Unsatisfactory
and Adjustment Adjustment
Retest (H - 15) (H - 21)
Score f cfi f cf£
21 1 100.0
20 1 95.2
19 1 100.0 1 90.5
18 0 93.3 2 85.7
17 0 93.3 1 76.2
16 0 93.3 0 71.4
15 0 93.3 2 71.4
14 0 93-3 0 61.9
13 0 93.3 2 61.9
12 1 93.3 0 52.4
11 1 86.7 1 52.4
10 0 80.0 0 47.6
9 4 80.0 1 47.6
8 2 53.3 1 42.9
7 0 40.0 2 38.1
6 0 40.0 4 28.6
5 1 40.0 2 9.5
4 1 33.3
3 0 26.7
2 1 26.7
1 1 20.0
0 1 13.3
♦1 0 6.7
♦2 1 6.7
N 15 21
Mean 6.93 11.67
S.D. 5.39 5.60
- 272 -
TAC female Satisfactory A d ju s tm e n t,^ 15

TAC female Unsatisfactory Adjustment, N=2I
100
rt:
Frequency (*/»)
60
Cumulative
20
10
C o m b i n e d Initial and Tê-feat Score
Figure 34. Cumulative Percentage Distributions of Combined Initial

and Retest Scores for all the Female Cases
in the Cross-Validation Study
- 273 -
positives; a Combined score of 13 or above would label as maladjusted
47.6$ of the poorly adjusted and 6.7$ of the well-adjusted females.
In the case of the Retest, discrimination is fairly good at the upper
end of the score range but poor at the lower end. In view of the
fact that two of the three kinds of scores shown to be significant for
the males are also significant for the females, it is probable that
the scoring method developed in this study is applicable to either
sex and that the lack of completely parallel findings is attributable
to the small size of the female samples.
The final comparison concerns all 118 subjects in the cross-
validation study, namely, the 54 subjects in the TAC Total Satisfac
tory Adjustment group and the 64 subjects in the TAC Total Unsatis
factory Adjustment group. The similarity of these total groups In
educational background, age, and marital status is evident in Table
LXXX.VI. The mean education of the satisfactory group is 14.00 grades
with a standard deviation of 1.55; for the unsatisfactory group the
figures are 13.89 and 1.69, respectively. The mean ages of the two
groups are 25.74 and 29.38, and the standard deviations are 7.66 and
8.94* respectively; 68.5$ of the satisfactory cases and 70.3$ of the
unsatisfactory cases are unmarried. With regard to sex distribution,
71.2$ of the well-adjusted and 67.2$ of the poorly adjusted subjects
are males.
The frequency and cumulative percentage distributions of
Initial, Retest, and Combined Scores for all the subjects are given
in Tables C and Cl, and corresponding frequency polygons and ogives
are shown in Figures 35 * 36, 37, 38, and 39. If the frequency polygons
- 274 -
fr O't'tsKMriH P\vD0"0®iArtMH«)H
• • ......a.. ...*.«.
15: OCOflOtpÔ l A O r l v O O HCAE-COC'--<i'I>-CA
31
O Vl
EH TO
o oo'O'O'^TOai >^5 m ' t m o i H H
•pi
3.68
O P
■a) at
O
H O O H r l ' t i n ^ O v O m ~-tvO f- CM -t CA CM •sfrvO
**! vO
p
OHHrl'+TO^frlTOtt'OtÔ'
H
4SS
O TO
O
|-4 O to 00 CO '4’C*>p\vO tO CA -sfrif\CAH
H Vl n
ID 3
2.42
^ 4* »I Sf
<t HOOCMvOCMC-OflOOinHHH
i-1 rl V\
O ONCAto O' C- \0 o c -
I ,d
3. p
3 o to-4
H
H« g »
u p f»as
2.87
■si TO to! St
H to <; HOOO(<MCiTOir\>'Oll\0'TO'JTO
p
IB
«
EH
P
H d
TO
H h
o
P CS
•H
d
H
OCf><nC^CMCMC«^Hir»H-'tt~
d OvOvOO»fvCMO'<flHr|>TO
31
5 .«i>
6h V
| *
OO'O'O'toSlD'TTOrl
o TO 3 "
O *H sa tX)
2.49
C\lr|OCM»C-vOinst<^CMHOHCM
s
SJD.
o
o H H H H H H H H» ♦
CO » a
- 275 -
11mirnim nn n i-t-uiu m h ih ih-w i th h iiii1111n 1111111i-TOTf

T AC Total Satisfactory Adjustment, N* 54
TAC Total llnsafis fa c t^ Ad u)tm^t.N»64
#)'$*%)&$+#*&'*%&&**%+&'*%+&'*%)&**%(
cvC/«)
F ir e , q \ j e n
Xm +ial Test" Score
Figure 35• Frequency Distributions of Initial Test Scores for

All the Hale and Female Cases
in the Cross-Validation Study
- 276 -
4 44.
!!!!
O-Ju
>a©
10
♦f IO
♦3
Ke+est* Scor e
Figure 36. Frequency Distributions of Retest Scores for All the

Male and Female Cases in the Cross-Validation Study
- 277 -
h
o
I f
CO
S e
US
I
1 3
3*7
s l
3^
31
*.s
°.
11
il
a ta
?!
r
&0
a. +>
IA
— h,
8 ?-S U 2 . 3 3 «l U i U 5 3 *5
( % ) A ? i A 9 n b » A ± aA.i+einiMn^
- 278 -
TABLE Cl

Deviations of the Combined Initial and Betest Scores for the Total
Male and Female Cross-Validation Groups (TAC Satisfactory Adjustment
and TAC Unsatisfactory Adjustment)
Combined Groun
Initial TAC Total TAC Total
and Satisfactory Unsatisfactory
Retest Adjustment Adjustment
Score (H - 54) (N = 64).
f cf£ f ef£
32 1 100.0
e •
• •
24 1 98.4
23 0 96.9
22 1 100.0 2 96.9
21 0 98.1 1 93.8
20 0 98.1 5 92.2
19 1 98.1 4 84.4
18 0 96.3 4 78.1
17 0 96.3 5 71.9
16 3 96.3 3 64.1
15 0 90.7 4 59.4
14 1 90.7 2 53.1
13 2 88.9 5 50.0
12 4 85.2 2 42.2
11 5 77.8 2 39.1
10 5 68.5 4 35.9
9 5 59.3 3 29.7
8 6 50.0 1 25.0
7 1 38.9 5 23.4
6 5 37.0 4 15.6
5 6 27.8 3 9.4
4 2 16.7 2 4.7
3 1 13.0 0 1.6
2 2 11.1 1 1.6
1 2 7.4
0 1 3.7
♦1 0 1.9
♦2 1 1.9
N 54 64
Mean 8.46 13.44
S.D. 4.68 5.97
ri <1 H
>0
'
>
O-
i)
93
N-
N.
>4)
>o
- 28U -
V)
>
c % ) A o u a n b ^ j 9/\i4e|nu»rQ
»0
M)
«
Figure 39. Cumulative Percentage Distributions of Combined Initial and Retest Scores
for All the Male and. Female Cases in the Cross-Validation Stud/
- 281 -
for the cross-validation groups are compared with those of the veteran
groups (see Figures 25 and 26), it will be noted that they are similar
in shape, particularly with regard to the Retest and Combined scores.
The degree of overlapping is consistently greater for the cross-
validation than for the original groups. The differences between the
mean Initial Test scores of 4.98 and 7.42 for the satisfactory and un
satisfactory groups, respectively, yield a critical ratio of 4.93} for
the Retest score means of 3.74 and 6.02 the critical ratio is 4*03}
and for the Combined score means of 8.46 and 13.44 the critical ratio
is 5*08. For the Initial, Retest, and Combined scores, the score
range resulting in maximal differentiation are 4 through 8, 5 through
7, and 11 through 14, respectively. The Initial and Combined scores
are roughtly equivalent in discriminating power, and both are slight
ly superior to the Retest. An Initial score of 6 or above and a Com
bined score of 10 or above each screen out 70.3$ of the poorly ad
justed cases at the cost of 40.7$ false positives} a Retest score of
4 or above, on the other hand, has a false positive rate of 51.9$
for a pick-up of 71.9$.
The frequency and cumulative percentage distributions of

Recall scores for the total satisfactory and unsatisfactory groups
are shown in Table CII. As in all previous comparisons of the cross-
validation groups, the Recall Test fails completely to achieve any

reliable differentiation, the critical ratio of the difference between
the means of +.74 and +.62 being .64.
Table CIII recapitulates the findings in the cross-validation
study with respect to the Initial, Retest, and Combined scores. This
- 282 -
TABLE CII

Deviations on the Recall Test for the Total Male and Female Cross-
Validation Groups (TAC Satisfactory Adjustment and TAC
Unsatisfactory Adjustment)
Groun
TAC Total TAC Total
Recall (H = 54) .. (N = 64)
Score f cf* f cf*
-2 3 100.0
-1 2 100.0 4 95.3
0 23 96.3 24 89.1
♦1 17 53.7 19 51.6
♦2 11 22.2 11 21.9
♦3 1 1.9 3 4.7
N 54 64
Mean ♦.74 ♦.62
S.D. .88 1.13
table may be compared with Table LXXX, which presents analogous results
for the original veteran groups. Shrinkage in the size of the differ
ences between means (and hence in validity) is evident in all instances
as we proceed from the veteran cases to the cross-validation groups,
but the very significant finding emerges that the differences for three
of the four kinds of scores utilized in the present study, namely, the
Initial, Retest, and Combined scores, are statistically reliable.
There is also a trend in the two tables for the three kinds of scores
to be ranked as follows with regard to discriminating power (most to
least); Combined score, Initial score, Retest score.

- 283 -
■3
on 'O
•8 3 3 UN UN 8
p p
CO Oi
2o ga
<8
®
® 5
39J
UN
H CM ss
-P _ irv UN
® Sri
£>
u
ON
& » a O' S
O' 'O CO
a d
UN ON
•8 ° $ » CM O
p p
o\
® S 3
u
o 9
o O
CO (3 a
9 9 O' CM UN
p
a
h ® to CO
•
CM 8
■3
cm CM 02
(S % * *
CM CM CM
M
M
$ O'
•
3 CM IT if O
0 ON vO ON ON vO
1
.9
vO
O'
O
to
CM
CO
CO
O'
P P • •
CM ON CM
a
h
o ®
<s o a
M M
O a Sf
S
•3
o
§ «
ON UN
•
«H P ®
- r
02 CM CM CM
3
% £ *
•a a
H
s i5 55 3 2 a 8 51
£ UN UN r~
O' O' O'
CM CM ON 5 s CM V\
"S
fl ®
3 s 1 "S
I I •3 T-
3 *«-» 3
3 •o
T» a
T3
“**
o ô
3
2 ..
&
i &
3
3
8
H
£ lii
P £ C h m
I? 3
<p H O H <.P
8 5. S
H O H Vl
5 a a a a a
p p *h
o a o
H 3 9
o a o P
3333
o a o p
3 Sp S E-c•H H h p
p
bh a
a
e-i*
h h a
P a
o a oa *_ ts O 3 u a o c o a o J
a CO
H 3 ra a O a
H
co a
E-t
» a co a P
Eh EH
- 284 -
While it is evident that the scoring scheme developed in
this study achieves a reasonably satisfactory separation of the well
and poorly adjusted subjects though not nearly so satisfactory as the
separation of the normals and neurotics in the original criterion
groupsi one may well inquire as to whether all of the signs hold up
under the exigencies of cross-validation. The data bearing on this
point are presented in Table CIV, which gives the percentage incidence,
for the entire satisfactory and unsatisfactory male adjustment groups,
of the signs used in the final scoring scheme. No tests of signifi
cance are employed, the main interest centering around the question
of the consistency of the direction of the differences on the Initial
Test and the Retest. Seventeen of the 30 signs discriminate consist
ently on both tests in the expected direction; these are 20-6, 22-6,
29-5, 30-A, 55-2, 57-5, 58-6, 60-1, 60-2, 60-3, 60-6, 61-2, 61-3, 61-6,66-7,
65-7, and 72-4. Three signs discriminate in the expected direction on
the Initial Test but show a reversal on the Retest, namely 61-1, 70-1,
and 75-A. Four signs discriminate in the expected direction on the
Retest but show a reversal on the Initial Test, namely, 33-8, 55-1,
75-4, and Short Figures, while 6 signs show negligible differences,
namely, 27-5, 29-6, 57-4, 60-5, Cohesions 2, and Time: 139. None
of the signs show any large reversals of direction on both the Ini
tial Test and the Retest. Were the size of the samples as large as
those employed in the initial validation, some of the reversals in
direction might have disappeared, but in general title results accord
with expectation.
- 285 -
TABLE CIV
Percentage Incidence of the Signs Used in the Final Scoring System for the
TAC Total Male Satisfactory Adjustment and TAC Total Male Unsatisfactory
Adjustment Groups on the Initial Test and the Retest
Initial Test . Retest

Grout) Grout) __
TAC Total Male TAC Total Male TAC Total Male TAC Total Male
Satisfactory Unsatisfactory Satisfactory Unsatisfactory
Adjustment Adjustment Adjustment Adjustment
(H - 39) (H - A3) (N • 39) (H - A3)
. Sign * % t t
20-6 15.4 27.9 20.5 32.6
22-6 7.7 23.3 17.9 18.6
27-5 15.4 16.3 17.9 20.9
29-5 5.1 11.6 7.7 14.0
29-6 53.8 55.8 61.4 62.8
30-A 33.3 53.5 41.0 51.2
33-8 17.9 4.7 28.2 32.6
55-1 17.9 11.6 10.3 16.3
55-2 17.9 23.3 12.8 23.3
57-4* 38.4 39.5 35.9 39.5
57-5 2.6 7.0 0.0 7.0
58-6* 20.5 18.6 17.9 9.3
60-1 76.8 83.7 43.6 48.8
60-2 69.1 88*4 38.4 65.1
60-3 28.2 48.8 7.7 44.2
60-5 56.3 55.8 35.9 34.9
60-6 15.4 41.9 7.7 39.5
61-1 35.9 58.1 28.2 25.6
61-2 61.4 86.0 46.2 67.4
61-3 20.5 37.2 12.8 44.2
61-6 2.6 34.9 7.7 25.6
65-7* 74.4 62.8 71.7 58.1
66-7 7.7 11.6 5.1 9.3
70-1* 17.9 11.6 7.7 14.0
72-4 20.5 37.2 17.9 32.6
75-A 33.3 39.5 33.3 23.3
75-4 28.2 16.3 17.9 32.6
Cohesion: \ 0.0 0.0 2.6 4.7
Time: 139* 12.8 11.6 33.3 34.9
Short Figures 33.3 25.6 35.9 37.2
CHAPTER VI
DISCUSSION
Perhaps the most significant finding of the present investi
gation is that normals and neurotics are much more alike than different
with regard to perceptual-motor responses to simple geometric designs,
to the extent that such responses are measured by the reproduction of
the Bender-Gestalt figures from a model or from memory. Only about
one-sixth of the deviations for which the reproductions are scored dis
criminate between the normal and neurotic subjects in the initial val
idation study, even when the relatively lenient criterion of differen
tiation at the 5% level of confidence is invoked, and the number of
discriminating signs shrinks considerably if a more rigorous test of
statistically significance is applied. This result is in essential
agreement with that reported by Billinglea (26), who found only 12 out
of 63 indices (195&) significant at the 5% confidence level. It is in
partial agreement with Bender (20) and Woltmann (102) who, without re
porting quantitative data, conclude that the Bender-Gestalt is of lim
ited value in distinguishing normal from neurotic individuals. It is
marked disagreement with Hutt (56), who speaks of a "neurotic syndrome"
on the Bender-Gestalt without, however, offering any statistical sup
port for his conclusion, and with Pascal and Suttell (76), who report
that 105 of their 200 signs (52.5%) differentiate between their patient
and non-patient groups. It is only fair to point out that in the latter
- 287 -
instance a sign is regarded significant if it distinguishes normals
from a combined neurotic and psychotic group of patients, and that
their figure would be somewhat smaller if the psychotics were excluded
from the comparisons.
Modifications of the stimulus figures are frequent in both
neurotic and normal subjects, according to our findings, but these
deviations very rarely approach the point where they can be described
as gross destructions of the gestalt, except when the reproductions
are drawn from memory in the absence of an objective stimulus, and
even under such circumstances the incidence of distortions is approxi
mately equal for the two contrasting criterion groups. We would agree
with Billingslea and Pascal and Suttell that gross deviations are non
discriminating and that, if any separation is to be effected, more
subtle modifications must be defined, observed, and isolated. We would
further agree that a feasible scoring scheme should take account of
these minute deviations, but we would take exception with Billingslea
in his belief that such a method of scoring should be accomplished by
equally precise physical measurements. Billingslea took as long as fif
teen hours to score a single record— a procedure which might be justified
in the research laboratory but not in the psychological clinic. We agree
with Pascal and Suttell that scoring should be accomplished by inspection
rather than measurement but that the scoring criteria should be so care
fully described that no great loss is experienced in going from quanti
tative measurement to qualitative observation. Moreover, this qualita
tive inspection should be supported by quantitative normative data rather
than by loose generalization based on clinical experience with a predomi

- 288 -
nantly abnormal population. Hutt's statements, for example, should be
set forth as a series of hypotheses to be subjected to more rigorous
validation rather than as conclusions to be uncritically accepted and
applied. Interpretation of such tests as the Man and ttooan Drawings
and even the Rorschach would probably be less extravagant in some quar
ters if their users had more widespread experience with nonpatient popu
lations and if they were willing to accept the strong possibility that
these tests do not have validity coefficients of 1.00.
In the present investigation, none of the "abnormal" signs
occurs exclusively in the abnormal subjects, and none of the "normal"
signs is found solely in the control subjects— a finding similar to
that reported by Anastasi and Foley (8) in their experimental study of
drawings but differing from the results of Pascal and Suttell, who re
port and give maximum weight to signs having zero frequencies in their
control group and frequencies of one or two in their patient group. It
might also be noted in this connection that one might have entertained
any one of four hypotheses prior to the present study: (1) deviations
of the reproductions from the stimulus figures are more frequent among
abnormals than normals, i.e., normals tend to conform more strictly
to the stimuli and the greater the number of deviations, the greater
the disturbance; (2) deviations are more frequent among normals, i.e.,
abnormals tend to conform more rigidly to the stimuli, and the fewer the
deviations the greater the disturbance; (3) deviations are equally char
acteristic of both normals and abnormals; and (4) certain deviations are
more characteristic of normals and others more characteristic of abnor
mals. Most clinicians implicitly assume hypothesis #1, a corollary of

- 289 -
which is that, the more nearly the reproductions approximate the originalB
the more likely the examinee is to be normal and that the more the record
deviates from the original the greater the severity of the disturbance.
This hypothesis is ostensibly assumed by Pascal and Suttell, since all
105 of their original 200 signs discriminate in the same direction, that
is, occur more often in their patient populations. One might argue that
their results lend strong support to the hypothesis, but the question re
mains as to whether they found signs which did discriminate in the oppo
site direction, i.e., had negative validity, but discarded them because
they were not in accord with their own tacit working hypothesis. Pascal
and Suttell do not raise this issue, nor do they present data on the in
cidence of the signs which they discarded. That they favor hypothesis #1
is evident in their conclusion that increasing scores are paralleled by
increasing degrees of psychological disturbance. The hypothesis that de
viations are more characteristic of abnormals is also supported by the
fact that in the present investigation 31 of the 40 Initial Test signs
(77.5%) and 35 of the 41 Retest signs (85.4%) which discriminate at the
5% level are negative in sign, i.e., occur more frequently among neurotics.
Hypotheses #2 and #3 are rejected on this basis, while hypothesis #4 is
also given support. Billingslea also found signs which discriminated in
both directions, and such signs are implicit in Hutt's formulations. The
two relevant hypotheses may therefore be combined and restated as a con
clusion, namely, that deviations of the reproductions from the originals
occur more often in the records of maladjusted individuals but that cer
tain deviations, though fewer in number, are more characteristic of rela
tively well-adjusted individuals, and that both types of deviations should

- 290 -
be considered in evaluating Bender-Gestalt test performance. Whether
the degree of psychological disturbance progressively parallels an in
crease in the excess of "abnormal” over "normal" signs cannot be ascer
tained from the present study, since the subjects were not graded ac
cording to severity of disturbance and since, unlike Pascal and Suttell's
study, psychotic subjects were not tested. A high correlation between
size of score and severity of disturbance seems unlikely, however, in
view of the high degree of overlapping of the control and experimental
groups— an overlapping which is especially pronounced in the cross-
validation study. Cut-off scores may be set which will separate normals
and abnormals in varying proportions, the normals tending to make more
favorable (lower) scores and the abnormals more unfavorable (higher)
scores, but as the critical score is raised there is a steady increase
in the number of false negatives. The occurrence of high-scoring "nor
mal" records and low-scoring "neurotic" records would also be difficult
to explain. Although it is likely that psychotics will tend to make
higher scores than neurotics and normals, the interpretation of increasing
severity of disturbance with increasing score within the neurotic or mal
adjusted groups in the present investigation is not given strong support
by the data. The presence of three or four signs which occur very rarely
in the standardization group (e.g., total or part rotation, dissociation,
fragmentation, gross distortion, compartments, numbering, perseveration,
confused sequence) may be much more significant than the occurrence of
eight or nine deviations that are found more often in both groups. It
should also be noted that the scoring scheme developed in the present
study is intended to differentiate adult normals from adult neurotics,

- 291 -
or relatively well-adjusted from relatively poorly adjusted individuals,
and that its applicability to psychotics, organics, children, mental
defectives, etc. is not known. There is no reason why the method could
not be extended to such nosological groups by appropriate item-analysis
techniques, however.
The hypothesis that differences in test performance between
normals and neurotics will be accentuated in the absence of an objective
stimulus during recall is not given strong support by our findings. Only
six discriminating signs could be isolated in the initial validation study,
and the differences between the means of the criterion groups is consider
ably less than that obtained for Initial and Retest scores. When applied
to the cross-validation groups, the differences on the Recall Test disap
pear almost completely. Moreover, even in the initial validation study,
the Recall score adds little to the differentiating power of the other
scores, warranting the conclusion that the scoring of the Recall Test is
superfluous for the populations sampled in this study. If, however, a
Retest score or a Combined score is desired, it is necessary to adminis
ter the Recall Test, since the original data were obtained under these con
ditions. It seems likely that the Retest and Combined scores will not
differ markedly from those obtained in the present investigation, even if
the Recall Test is omitted, but this is a problem for future research, and
until the results of such research are forthcoming, it is recommended that
the present method of administration be adhered to. If time is limited, the
difficulty may be surmounted by giving only the Initial Test, for the results
obtained under these conditions are in essential agreement with those found
with the longer procedure, except for the fact that greater confidence can
- 292 -
be attached to the findings when Initial, Retest, and Combined scores
are consistent in their diagnostic implications.
The hypothesis that emotional blocking in the neurotics
would cause them to reproduce fewer figures in the Recall Test is also
given only meager substantiation. The controls do recall slightly more
figures than the neurotics, but the difference is not significant except
for the recall of 3£ figures or less, where the differences are small but
significant and in a direction that is in accord with the hypothesis.
Perhaps the most striking observation is the comparatively large number
of figures recalled by both normals and neurotics; of the 9 stimulus
figures, 6 are recalled on the average, despite the fact that the reten
tion test is one of incidental memory. This high figure attests to the
simplicity, familiarity, and meaningfulness of the stimulus figures,
since it is highly probable that the mean of nonsense figures would be
much lower. Fehrer (39), for example, gave subjects repeated brief ex
posures of geometric forms and determined the number of exposures re
quired for a correct drawing. It was found that among the easiest were
(1) known geometric forms (circles, semicircles, squares, diamonds, ob
longs, triangles, etc.) and (2) the above with a simple correction (ob
long with a gap in one of the sides, small square placed askew a larger
square,etc.). Most of the Bender-Gestalt figures would appear to fall
within the "easy" category.
The absence of significant differences for distortion, half
figure, and confusion supports the view that neurotics cannot be dis
tinguished on the basis of gross modifications of the gestalt, even in
Recall, where such differences should be exaggerated. Distortions are

- 293 -
found in three out of every five records, figure 7 alone accounting
for two of each three records on which distortions appear— eloquent
testimony to its difficulty* Most of the distortions take the form of
a simplification of the stimulus figure (e.g., fewer waves in figure 6,
fewer arrows in figure 3, fewer sides in figures 7 and 8). Reproducing
only one part of a figure which originally contained two parts (half
figure) may also be regarded as a form of simplification of the gestalt.
Intraserial assimilation (confusion of two figures) is present in one-
fourth of the recalls, particularly for figures A and 4-, but there are
no instances of object assimilation (reproduction made like an object
which the original resembles). Emphasis (exaggeration of unusual char
acteristics) is rare, while closure of the open square in figure A oc
curs fairly often, though closure of the bell in figure A and the semi
circle in figure 5 is most uncommon. Most of the changes predicted by
the Gestalt theory that the memory trace tends not toward vagueness but
toward "better figure" and that the trace, instead of weakening, becomes
more stable by yielding to its internal stresses are conspicuous by their
absence, probably because the Bender-Gestalt figures are "gute gestalt"
to start with, have few internal stresses (because balanced and symmetri
cal), and do not resemble objects* Normalization (object assimilation),
regularization (autonomous structural change), and emphasis (exaggeration
of an unusual characteristic)— changes which Gestalt psychologists find
so frequently in their studies of memory for form— are given little oppor
tunity to appear in the Bender-Gestalt stimuli. Moreover, any such changes
that might occur would be more likely to manifest themselves over a long
retention interval rather than in immediate recall.
Under any circumstances, it would be taking a rather limited

- 294 -
view of things if one were to ascribe all the difference between the
original and the reproduction to memory factors, since such factors as
an originally incomplete perception, past experience and present set,
and the exigencies of drawing also play a part. Vagueness cannot be
represented, as Woodworth (104) points out, forcing the subject to
draw a figure which he knows must differ from the original, and Kuhlmann
(59) perspicaciously observes: "Recall is very largely not recall at
all and can never be described even half correctly by calling it re
production. It is rather a construction, not a reconstruction a con
struction of a certain result that is accepted in place of the original,
and far from a reconstruction of a past perception."
With regard to the frequency and priority of recall of the
various figures, recency is somewhat stronger than primacy as far as
total recall is concerned, and the middle figures in the series are
least often recalled. In marked contrast to figure 8, which is re
called by 97% of the control group, figure 4 is recalled by only 31%.
Oddly enough, figure A, which has the greatest advantage with respect
to primacy, ranks sixth in total frequency of recall. With regard to
recall in first position, however, figure A ranks only behind figure 8,
while figure 7 drops from fourth to eighth rank. The correlation be
tween total frequency of recall and priority of recall is positive,
but only moderately high (r = .55). Similar results for a projective
test were obtained by Loeblowitz-Lennard and Riessman (61), who showed
25 college students 8 TAT pictures, each for 5 minutes with the usual
instructions. They were then asked to recall them by describing each

in one or two sentences and numbering them as they came to mind. Pic
- 295 -
ture 1 ranked 7.5 with respect to total frequency of recall, whereas the
last three pictures held second, third, and fourth ranks, when order of
recall was considered, however, Picture 1 ranked fourth, while Picture 7
moved down from second rank to seventh. Why the figures or pictures re
called first are not necessarily the ones recalled most often is a ques
tion worthy of more systematic investigation. It is interesting to note
also that the ranks of the Bender-Gestalt figures for both frequency and
priority of recall are very similar for normals and neurotics alike. The
recall processes of neurotics, in so far as the Bender-Gestalt figures
are concerned, differ but little from those of normal subjects with re
spect to such variables as extent of recall, order in which the recall
takes place, and deviations of the recalled material from the original.
To what extent do the findings of the present study agree with
those of other investigators? The strongly positive results of Pascal
and Suttell are in marked contradiction to those reported by Billingslea,
who in turn found little to support Hutt's "neurotic syndrome". The
conflicting findings of Billingslea and Pascal and Suttell are of inter
est because both attempted a quantification of the scoring categories,
and one wonders how two investigators, both of whom applied presumably
objective methods to a test which had previously been interpreted by
clinicians in a highly subjective manner, could arrive at such opposite
conclusions. Hutt's formulations merit special attention because the
majority of clinical psychologists have followed his lead in assuming
psychodynamic correlates of specified deviations in the reproductions
and in applying his criteria for the differential diagnosis of normals
from neurotics. Do the present findings justify the cautious reser-

- 296 -
vations of Billingslea, the extravagant enthusiasm of Hutt, or the
sanguine optimism of Pascal and Suttell? A detailed comparison of
our findings with those of these investigators should be of help in
answering this question.
Let us first consider Hutt's claims. (1) With regard to
sequence, Hutt (56) says: "An irregular order is revealed by place
ment of the drawings on the paper out of sequence, more in accord with
the inner needs of the subject than with the task and more in accord
ance with space allotments than with the requirements of good planning.
It occurs characteristically in psychoneurotic records and less fre
quently in psychotic records." (p. 11). Our findings show a logical
or regular sequence infrom three-fifths to four-fifths of all subjects
in the initial validation study with no significant difference between
the normal and neurotic subjects in this respect. (2) Regarding regres
sion, he states: "Regression is a manifestation of deterioration; at
the adult level it is present in the records of psychotics and organic
brain damage. It represents an attempt at simplification of the task
and inaccurate perception of reality. It is usually 'irreversible' in
the clinical sense." (pp. 13-14). Actually, regression (or substitution,
as we have called it) occurs quite frequently in the records of both
neurotics and normals, whom one would hardly classify as "deteriorated”.
This sign is not discriminating in our study, although there is a trend
for the neurotics to show a higher incidence. Pascal and Suttell use
this sign but do not indicate whether the greater incidence in their
patient group was due to the contribution of their neurotic or psychotic
subjects. Billingslea finds that regression is frequent in both his

- 297 -
controls and neurotics but that it occurs more often in the latter
group. It would appear that Hutt's interpretation or regression as
a manifestation of deterioration or inaccurate perception of reality
is rather questionable. Is far as its "irreversible'* nature is con
cerned, there are several instances in the present study in which dots
are converted into dashes or circles on the Initial Test but repro
duced as dots on the Retest. Some subjects, when subsequently ques
tioned, said that they knew that there were dots in the stimulus fig
ures but that they reproduced them as dashes, circles, scribbles, or
filled-in circles because "it's easier that way" or "it takes less
time". Apparently, their "perception of reality" was accurate even if

*
their motor execution failed to give a literal representation of that
perception. (3) Concerning sketching, Hutt asserts: "...insecure
patients...will frequently sketch portions of the figure with repeti
tive strokes, often revealing a masturbatory quality in such work.
Elongation and repetitive sketching of the prolongations of figures
(the projections) is a frequent correlate of sexual disturbance.
Special attention should be given to figures 7 and 8 in this regard."
(p. 16). In the present study sketching (or retracing) does not sig
nificantly differentiate normals from neurotics on any figure. Pro
jection retracing on figures 7 and 8 was scored as a separate sign
in order to check upon Hutt's interpretation, but the differences are
negligible. Moreover, the length of figures 7 and 8 does not provide
differentiating data. Hutt also implies— and has stated in his lec
tures— that difficulty with the projections in figures 7 and 8 in gener
al is indicative of sexual disturbance or conflict. If this is true,

- 298 -
then 60%-8($ of our subjects have such disturbances, in view of the
incidence of such signs as 74-7, 74-8, 12-7, and 12-8. Many psychol
ogists might argue that this may be a reasonable figure, but they
would hardly contend that the incidence of such conflicts would be
the same for neurotic and normal subjects. (4) Hutt has also claimed
that closure of open figures is indicative of neurotic sex conflict,
but in the present investigation not a single instance of closure was
observed in either the control or the neurotic subjects, except in
the Recall Test, where closure occurred more frequently in the nor
mals. (5) Hutt also refers to a special instance of closure in fig
ure 6 which is said to be pathognomonic of psychasthenia. "A special
form of closure difficulty occurs among psychasthenics (statistically
significant): they have difficulty with the crossing of curved lines
(figure 6). They often draw one figure (the horizontal sinusoidal
curve) accurately, but leave a gap where the vertical figure inter
sects the other one." (p. 15). Not a single instance of this type of
distortion was observed in any of the subjects in either the initial or
the cross-validation study. Pascal and Suttell report only one instance
of distortion in figure 6 among their 260 nonpatients and 260 patients
(neurotics and psychotics), and it is not certain whether even this
one instance involves the line-crossing difficulty. Billingslea found
no instances of this deviation among his 100 neurotic subjects. One
must either conclude that this sign is of negligible diagnostic signifi
cance or that the neurotic cases in the cited studies did not include
one instance of psychasthenia. (6) Of interest, too, is Hutt's refer
ence to the "depressed” and "elated" slopes of figures 1 and 2. Accord-

- 299 -
ing to Hutt, an upward slope of either or both of these figures con
notes elation, while a downward slope is characteristic of depression.
If an upward slope is indicative of an elated mood, the results of the
present study would force one to the inescapable conclusion that ela
tion is more typical of neurosis than of normality, since the signs,
55-1 and 55-2, both occur with significantly higher frequency among the
neurotic subjects. As far as downward slope is concerned, one would
have to infer that normals are as depressed as neurotics, since the
differences between the criterion groups are negligible. (7) Hutt
also contends that angulation difficulties are characteristic of
neurosis, but this sign (scoring category 31) fails to differentiate
the criterion groups on any figure. Difficulties with angulation are
quite common on figure 2, occurring in approximately half of the 393
cases in the initial validation study, but the differential incidence
is statistically unreliable. (8) Another sign which Hutt regards as
characteristic of neurosis is condensation and simplification. This
deviation is very uncommon in our cases and occurs no more often in
neurotics than in normals. (9) Reversal is also said to occur more
frequently in neurotics, but not one instance of this deviation was noted
in the present study among either neurotics or normals. (10) Modifica
tions of curvature, particularly with regard to figure 6, are claimed
by Hutt to constitute part of the neurotic syndrome. Irregularity of
the sinusoidal curves, said to indicate emotional instability, is non
discriminating for our cases (the sign is 23-6). Translation of the

curves into a series of straight lines, said to indicate rigidity,
never occurs. Exaggeration of curvature (increased amplitude), from

- 300 -
which increased affactivity is to be inferred, also yields negligible
differences, flattening of figure 6 (the sign is 22-6) which, it is
claimed, reveals reduced affect does, however, occur more often in our
neurotic cases, although this fact in itself neither confirms nor
negates the psychodynamic imputation.
On the other hand, certain points of agreement between our
findings and those of Hutt are equally noteworthy. (1) Hutt states
that rotation "almost never occurs except in the case of psychosis"
(p. 13). The corollary, that rotation is very infrequent in normals
and neurotics, is borne out by our findings. Hutt's limitation of
rotation to psychotic cases might be questioned in the light of the
present writer's experience, however, since a number of children and
mental defectives have been observed to exhibit this phenomenon. It
will also be recalled that Fabian (38) reported a high incidence of
vertical rotation among his reading disability cases. (2) Hutt con
tends that reduction in size "connotes constriction, rigidity, or
stereotypy" (p. 12). The tendency for our neurotic subjects to make
their reproductions shorter in length than those of the controls is
in line with this interpretation, although other interpretations are,
of course, also possible. (3) Closely related to this point is Hutt's
interpretation of cohesion: "The compressed style is correlated with
depression, rigidity, and constriction; it indicates a sense of per
sonal inadequacy." (p. 12). Compression (the sign is cohesion:^) is
more common among our neurotic subjects, but again, this in itself
does not exclude other interpretations. (4) Our findings support
Hutt's statement that destruction of the gestalt is not typical of

- 301 -
the neurotic record.
To recapitulate, our findings are in agreement with Hutt's
"neurotic syndrome" with respect to preservation of the gestalt, ab
sence of rotation, reduction in size (for the variable of length
only), and change in curvature values (for flattening in figure 6
but not for modifications in other curved figures). The first two
of these signs do not, however, differentiate between normals and
neurotics, while the remaining two have seme discriminating power.
Disagreement is noted with regard to phallic symbols and indices
of sexual disturbance, regression, sketching, sequence, closure,
motor incoordination, occasional manifestions of perseveration and
fragmentation, "depressed" and "elated" slopes, condensation and
simplification, reversals, crossing difficulties, and angulation
difficulties. Many of his psychodynamic interpretations are also
voided in view of the fact that the deviations said to be indicative
of emotional and motivational processes occur equally often in all
groups of subjects. Although noted casually rather than systematically,
emotional exclamations and marked resistance to the test or parts of
it— said by Hutt to occur more frequently among neurotics— also appear
to be of limited value for differential diagnosis. The conclusion
seems warranted, therefore, that Hutt's method of evaluating the Bender-
Gestalt leaves much to be desired with regard to the interpretation of
the test performance of neurotics and that his personality evaluation
of normal and neurotic records is more deserving of healthy skepticism
-than uncritical acceptance.
Billingslea (26) investigated the incidence of deviations in

- 302 -
the reproductions of normals and neurotics without concerning himself
with the psychological meanings of these deviations) although he does
treat Hutt's indices of sexual disturbance on an operational basis.
The significance of his results is limited to some degree by the fact
that his control group consisted of only 50 subjects) but the areas of
correspondence and disagreement with the present investigation are wor
thy of note. (1) Speaking of cohesion) Billingslea says: "This factor)
as a whole is not discerning) but the trend exists for its low scores
(i.e., all test figures crowded into a small part of one sheet of paper)
to occur more often in the records of the Psychoneurotic group." (p. 14).
This statement is in virtually complete agreement with our findings
on cohesion. (2) He has this to say about sequence: "Logical or Ir
regular Orders are not discerning between the two groups even as trends;
Confused Order may be found in the records of either group." (p. 14).
Sequence was also non-discriminating in the present research. (3) Nei
ther total nor part rotation has significant validity) agreeing with
our findings. (4) On the factor of size, he asserts: "The trend is
for the Psychoneurotic subjects to produce smaller figures than the
Normal group) especially in the case of Figures 1) 2, 5, and 6." (p. 15).
Our findings are in substantial agreement) except for figure 5. (5) The
presence of erasures has low validity in both studies. (6) Shearing of
figure 5 (our sign is 67-5) shows negligible differences in both investi
gations. (7) Billingslea finds that the number of elements in figures
1) 2, 3) 5) and 6 (our scoring category is numeration) is not discerning.

Our findings are in agreement) except for figure 6 (the sign is 20-6)
in which neurotics tend to represent the number of sinusoidal waves
t
- 303 -
inaccurately. (8) Both studies indicate that downward slope in fig
ure 2 lacks discriminating power. (9) A number of signs occur rarely
or not at all in both studies, namely, substitution of numbers or let
ters for dots or circles, distortion beyond recognition, fragmentation,
ornamentation, perseveration, colliding, failure to cross the lines in
figure 6, closure of the square in figure 4 and the curve in figure 5 ,
pointed waves in figure 6 , and splitting in figures A and 4.
Some points of disagreement should be mentioned. (1) Dis
placement in figures 5, 6 and 8 are non-discriminating for Billingslea.
In the present study, this is true for figure 8 but not for figures 5
and 6 (the signs are 29-5 and 29-6), although 29-6 does turn out to be
non-discriminating for the cross-validation group. (2) Billingslea re
ports that horizontal irregularity is rare in figures 1 and 2. Our
findings indicate that this deviation is not uncommon (the signs are
36-1 and 36-2) but agree with Billingslea that it has low validity.
(3) Billingslea finds that regression (our scoring category is sub
stitution) occurs frequently among normals and neurotics, which is
consistent with our results but inconsistent with those of Hutt), but
he also reports that "the number of times it occurs in a single record
is significantly greater in those of the Psychoneurotic subjects" (p. 14)
a finding which we do not verify. (4) Billingslea finds that counting
is more common in the controls, whereas we find that counting is found
much more often among neurotics and that it is one of our most discrim
inating signs. Our findings in this regard hold up for both the initial
and the cross-validation groups and for both the Initial Test and the
Betest. Anastasi and Foley (8) in their experimental study of the draw-
- 304 -
ings of normals and abnormals also report a higher Incidence of counting
and exaggerated meticulousness among their abnormal subjects. (5) Sketch
ing distinguishes between normals and neurotics, according to Billingslea,
whereas in the present study it lacks validity. In this respect, too,
our findings are consonant with those of Anastasi and Foley (8), who point
out: "Going over the drawing repeatedly in pencil or crayon, thereby pro
ducing dark masses and early wearing out the paper, likewise failed to
differentiate between the two groups." (p. 188). (6) Multiple attempt
of a figure— a sign lacking in discriminating power in our study— occurs
significantly more often among neurotics in Billingslea*s research. (7)
Another finding not substantiated by our results is Billingslea*s state
ment that "indices of sexual disturbance" are more frequent among neuro
tics. Considering the fact that more than 100 deviations were analyzed
in both studies, our data and those of Billingslea show remarkably close
agreement in many particulars, with the very important exception that
Billingslea was unable to isolate signs which, in combination with other
signs, would separate normals from neurotics with a success well beyond
chance, while the present investigation demonstrates the feasibility of
this procedure. Our results agree with those of Billingslea and Hutt
with regard to the non-occurrence of distortion beyond recognition, the
rare occurrence of perseveration and fragmentation, and the tendency
toward reduction in size. Our findings agree with those of Billingslea
but disagree with those of Hutt with reference to total and part rotations,
difficulty with crossing in figure 6, reversals, downward slope, and fre
quency of regression. Billingslea agrees with Hutt and disagrees with us
regarding the higher incidence in neurotics of regression and indices of

- 305 -
sexual disturbance.
Although both Billingslea and Pascal and Suttell attempted
a quantification of the Bender-Gestalt, Pascal and Suttell (76) are the
only investigators other than the writer who have combined their signs
into a total score. Their research differs fro* that of the writer in
that (1) they did not systematically administer a recall test or a re
test, (2) they established the validity of their signs in terms of the
ability of each sign to distinguish between a control group and a com
bined neurotic and psychotic group, (3) sketching was prohibited, and
(4) the time required to reproduce the figures was not recorded. As
previously noted, the most remarkable aspect of their investigation is
the extremely high proportion of signs which they found to be discrim
inating. Whereas Billingslea and the writer found that only 15% to 20%
of their original signs met a satisfactory criterion of validity, Pascal
and Suttell report that over 50% of their signs are sufficiently sig
nificant to warrant consideration in the scoring of a given test re
cord. As indicated in Chapter II, reworking of their data by the wri
ter shows that several of their signs are not significant at the 5%
level, but the fact remains that the strongly positive nature of their
findings is in marked contrast to the more modest results reported by
Billingslea and the present investigator. That startling results are
unlikely is suggested by the inability of Bender and Woltmann to use
the test for purposes of differential diagnosis and by the high propor
tion of non-discriminating signs found by Anastasi and Foley (8) in
their attempt to quantify the characteristics of drawings of normals
and abnormals. In the latter study, of those signs which gave moderate
- 306 -
but consistent excess in abnormal frequency, only 5 out of 109 nor
mal-abnormal comparisons yielded critical ratios of 3.00 or more.
Of those signs showing consistently large differences, "only about
one-third of all the abnormal-normal comparisons yielded critical
ratios of 3.00 or more none of the frequencies was in itself large
enough to represent more than a small minority of abnormal cases."
(p. 191).*
Of those signs scored by Pascal and Suttell, the following
are confirmed by the present study, if by confirmation it is meant
that the sign meets our criteria of significance and consistency,
namely, that the sign be significant at the 5% level on either the
Initial Test or the Retest and significant at the 10% level on the
other, or that the sign be significant at the 1% level on either the
Initial Test or the Retest and consistent in the direction of the dif
ference on the other:
* Consistent with these observations are the findings of a recent

doctoral dissertation in which the Bender-Gestalt was employed
(Wilson, Robert G., A Study of Expressive Movement in Three Groups
of Adolescent Bovs: Stutterers. Non-Stutterers Maladjusted, and
Normals, by Means of Three Measures of Personality; Mirals Mvo-
kinetic Psvchodiagnosis. the Bender-Gestalt. and Figure Drawing.
September, 1950, unpublished doctoral dissertation on file at
Western Reserve University Library). Wilson scored the Bender-
Gestalt for nine variables which Billingslea found to be most dis
criminative between his normal and psychoneurotic groups (closure,
distortion, horizontal loops, vertical loops, distortion of angles,
■page cohesion, regression, sketching, counting and checking) and
then compared the scores of his groups of normals, stutterers, and
maladjusted non-stutterers. He found that only two of the 102
critical ratios between the means and between the percentages of
occurrence were significant at the 1% level and that none was sig
nificant at the 5% level, although the direction of the differ
ences was in favor of the normal group.
- 307 -
Pascal and Suttell1s Term Term Used in Present Study
Curve extra (figure 6) Numeration (20-6)

Asymmetry (figure 5 Asymmetry (27-5)
Compression Cohesion:^
Point of crossing (figure 6 Displacement (29-6)
Extension rotation (figure 5 Displacement (29-5)
Relative size Short figures
Pascal and Suttell signs significant at the % level on our Initial Test
but statistically unreliable on the Retest are:
Pascal and Suttell's Term Term Used in Present Study
Wavy line (figure 1) Horizontal irregularity (36-1)

Dot, dash, circle (figure 5) Substitution: dot-dash (14-5)
Curve not centered (figure 4) Displacement (29-4)
Signs statistically unreliable on our Initial Test but significant at
the 5% level on the Retest are: «
Pascal and Suttell1s Term Term Used in Present Study
Circles (figure 3) Substitution: dot-circle (15-3)

Number of dots (figure 1) Numeration (20-1)
Asymmetry (figure 6) Wave irregularity (29-6)
Distortion (figure 6) Distortion (47-6)
Not joined (figure 4) Splitting (44-4)
Tremor (figure 7) Tremulous line (5-7)
Ends not joined (figure 8) Closed-open (40-7)
From the above it is apparent that our results lend some support
to Pascal and Suttell's findings with respect to 16 signs, although only
6 of these meet the item validity criteria employed in the present study.
Unfortunately, however, the area of disagreement is great, as the fol
lowing list of signs, none of which is significant at the 5% level on
either our Initial Test or Retest, demonstrates:
Pascal and Suttell*s Term Term Used in Present Study
Wavy line (figure 2) Horizontal irregularity (36-2)

Dot, dash, circle (figures 1,3) Substitution: dot-dash (14-1, 14-3)
Substitution: dot-circle (15-1, 15-3)
Dash or dot (figure 2) ‘ Substitution: circle-dot (18-2)
Dashes (figures 1, 3) Substitution: dot-dash (14-1, 14-3)
- 308 -
Pascal and Suttell1s Term (Cont.) Term Used in Present Study (Cont.)
Circles (figures 1, 5) Substitutions dot-circle (15-1> 15-3)

Number of dots (figures 3, 5) Numeration (20-3, 20-5)
Number of columns (figure 2) Numeration (20-2)
Extra row (figure 3) Numeration (20-3)
Double row (figure 1) Continuation (78-1)
Workover (figures 1, 2, 3, 5, Retracing (11-1, 11-2, 11-3, H-5,
6, 8) 11-6, 11*£)
Second attempt(figures 1, 2, 3, Multiple attempt (7-1, 7-2,
4, 5, 6 ,7, 8) 7-3, 7-4, 7-5, 7-6 , 7-8)
Rotation (figures 1, 2, 3,4, 5, Total rotation (48-1, 48-2, 48-3,
b, 7, 8) 48-4, 48-5,.48-6, 48-7, 48-8)
Curve rotation (figure 4) Part rotation (49-4)
Design missing (figures 1, 2) Serial incompleteness (39-1, 39-2)
Design missing (figures 3, 4, Omission (54-3, 54-4, 54-5, 54,6,
5, 6, 7,8) 54-7, 54-8)
Deviation slant (figure 2) Angulation (31-2)
Guide lines (figures 2, 3, 4, Guides (9-2, 9-3, 9-4, 9-5, 9-6,
5, 6, 7,8) 9-7, 9-8)
Asymmetry (figures 3, 4) Asymmetry (27-3, 27-4)
Blunting (figure 3) Angulation (31-3) and distor
tion (47-3)
Distortion (figures 3, 4, 5, Distortion (47-3, 47-4, 47-5,
7, 8) 47-7, 47-8)
Tremor (figures 4, 6, 8) Tremulous line (5-4, 5-6, 5-8)
Ends not joined (figure 7) Closed-open (40-7)
Angles missing (figures 7, 8) Side reduction (68-7, 68-8)
Placement of Design A Figure A placement (8-A)
Overlapping of the designs Collision
Lines drawn to separate the de- Compartments
signs
Order Sequence
No order Sequence
Pascal and Suttell found no significant sex differences in test per
formance, nor have any other investigators reported such differences—
a fact which lends weight to our finding in the cross-validation study
that our scoring scheme has differential validity for both males and
females. They also report that test scores of adults from 15 to 50
years of age are unrelated to age and intelligence. Their correlation
of -.16 between Bender-Gestalt scores end Otis I.Q.'s is virtually
identical with our correlation of -.15 on the Initial Test and -.17 on
the Retest for our normal subjects. Their reported correlation of .03
- 309 -
between age and Bender-Gestalt score for their normal cases closely
approximates our correlations of .06 and .04 for the Initial Test and
the Retest. Similar correlations between intelligence and Bender scores
and between age and Bender scores are present in our neurotic subjects.
For children up to the age of 10 or 11, the correlation between age
and Bender-Gestalt score would undoubtedly be significant in size and
negative in direction, since young children reproduce the figures with
many inaccuracies. Pascal and Suttell, for example, report a correla
tion of -.58 for a small sample of children ranging in age from 6 years
and 3 months to 9 years and 3 months. we would agree with their conclu
sion to the effect that "Although the copying of simple designs may be
an important index of maturation warranting its use as an item in an in
telligence test for young children, it is not, and we would not expect
it to be, a discriminating test of intelligence for normal adults.11 (p. 22)
A finding of Pascal and Suttell which is very difficult to ex
plain, however, is the significantly lower scores made by their subjects
with a college education. They obtained a bimodal curve for their high
school and college subjects, leading them to devise separate norm tables
for the two groups. They could not compute a product-moment correlation
coefficient because they merely classified their cases as to whether or
not they had been to college or whether or not they had one or more years
of high school training, but their reported bi-serial r of -.38 between
test scores and this dichotomy cannot be attributed to sampling error.
In our study, the correlations between Bender-Gestalt scores and educa
tion are negative bublow (-.11 and -.10 for the Initial and Retest records
of the controls, respectively, and -.12 and -.06 for the two test records
- 310 -
of the neurotics), as one would expect in view of the lack of relation
ship between intelligence test scores and Bender scores. They comment
as follows:
We are at a loss to explain why this difference should

be there, especially in the light of the relation between
B-G score and I.Q., and can only speculate upon its sig
nificance. It may be that the college group was general
ly more test-wise and thus less disturbed by the testing
situation. It may be, as suggested by a psychiatrist
friend of ours, that going on to college in our competi
tive society is, in general, an adaptive act related in
some way to what we are measuring. Anyhow, the fact re
mains that we had two different populations based on
education. (76, p. 24).
More significant than the analysis of the areas of disagree
ment between Pascal and Suttell's investigation and the present research
with respect to specific signs is the complete failure of their scoring
scheme to differentiate between our normals and neurotics. The differ
ences between the means of the matched criterion groups do not even come
close to approaching statistical significance, while a breakdown into
educational groups discloses no relationship between education and test
score. This lack of correlation between educational status and Bender-
Gestalt* test performance is consistent with our findings for our scoring
scheme and with expectation based on the lack of relationship between
intelligence and Bender score in both studies. Thus we see that the
application of Pascal and Suttell's scoring method to our cases dis
agrees with the conclusions of these investigators in two important
respects: (1) their scoring scheme does not distinguish between nor
mals and neurotics, and (2) there is no need for separate norm tables
for subjects at the high school and college levels.
With regard to the scoring method developed in the present
study, the prediction that it would be successful in differentiating

- 311 -
well- adjusted from poorly-adjusted individuals is verified by the re
sults of the cross-validation study. The differences between the means
of these contrasting groups are smaller than those obtained in the initial
validation, but the differences emerge as statistically significant
despite the anticipated shrinkage in validity. The fact that the scor
ing scheme can be applied to non-veterans as well as veterans, to fe
males as well as males, and to individuals who are maladjusted but not
necessarily neurotic and the fact that differences in age (within the
range 18 to 50), education (within the high school and college levels),
and intelligence (for I.Q.'s of 85 and above) do not appreciably in
fluence the score indicates that the Bender-Gestalt can serve a useful
screening function not only in mental hygiene, clinics but also in voca
tional guidance and psychological counseling agencies dealing with nor
mal individuals and persons representing varying degrees of emotional
disturbance short of psychosis. It is recommended as a screening de
vice rather than as an instrument for elaborate personality interpre
tation because the differential incidence of most of the individual
figure deviations is not great and because many of the signs said to
have particularized meaning for neurotics prove to be non-discriminating.
Interpretations of "flexibility and spontaneity" based on logical se
quence, deterioration or emotional immaturity, based on substitution of
dashes or circles for dots, depression based on downward slope, ela
tion based on upward slope, instability based on curvature irregular-
ities, neurotic sex conflict based on closure or elongation and repeti
tive sketching of projections, inferiority or insecurity based on reduc
tion in size, and expansiveness or oppositional traits based on an in

- 312 -
crease In size appear to be premature In the light of the available
evidence. Clinicians working almost exclusively with abnormal patients
may establish correlations between some of these signs and information
gathered about the patient from behavioral, clinical, or psychometric
data, but it is submitted that they might well experience a keen feel
ing of disillusionment at the absence of such correlations in work
with comparatively well-adjusted people. In work with "normals" at
guidance agencies, for example, it is not uncommon for the clinical
psychologist to come up with a "neurotic Bender" on a client whose
performance on other tests is well within normal limits and whose per
sonal history gives no real evidence of serious emotional or social
difficulties. It is a great temptation under such circumstances to ig
nore vitiating data and to overinterpret the test on the grounds that
the Bender-Gestalt may be measuring something more "fundamental", such
as "latent" tendencies toward neurosis or even psychosis. Wechsler-
Bellevue scatters and Man and Woman drawings are occasionally subjected
to this error of the dramatic instance, sometimes to the point where it
is hard to find a "normal" record. It is felt that the Bender-Gestalt
may well contribute to the isolation of maladjusted individuals for
more intensive study but that it should be used as a supplement to,
rather than a substitute for, existing projective techniques, and that
a reasonable degree of consistency with other data should be demanded
before extravagant interpretations of personality dynamics are made.
The cut-off score to be used for screening purposes will de

pend upon the particular purpose to be achieved. If it is desired to
screen out a great proportion of the maladjusted individuals in a given
- 313 -
sample, the critical score must be lowered despite the relatively
high incidence of false positives. Reference to the distributions
of Initial, Retest, and Combined scores in Tables C and Cl should
prove helpful to the clinician in interpreting a given score in
terms of the chances that such a score is likely to be made ty an
emotional disturbed individual, and these tables may readily be con
verted into expectancy tables in personnel selection situations where
the interest is greater in group, than in individual predictions.
For general purposes, it is suggested that cut-off scores be set at
one standard deviation above the mean of the control groups in Ta
bles C and Cl. Disregarding fractions, these critical scores are
7, 6, and 13 for the Initial, Retest, and Combined scores, respec
tively. On the Initial Test, a score above 7 screens out 53.1# of
the maladjusted cases at the cost of 14.8# false positives; on the
Retest, a score above 6 screens out 48.4# at the cost of 16.7# false
positives; a Combined score above 13 picks up 50# of the maladjusted
cases at the expense of 11.1# .of the normals. A lowering of the cut
off score to 6 for the Initial Test, 4 for the Retest, and 10 for the
Combined score results in pickups of 70.3#, 71.9# and 70.3# of the
poorly adjusted cases, respectively, with false positive rates of
40.7#, 51.9#, and 40.7#, respectively.
If it is true that the Bender-Gestalt, as scored in the
present study, has a validity sufficient to warrant its conclusion
in a battery of psychodiagnostic tests even though its validity, when
used alone, is of modest proportions, is it also true that it has a
commensurate reliability? The whole question of the reliability of a

- 314 -
projective technique is itself fraught with problems since, as Sargent (81)
has observed, "in several studies of validity, indices based on predic
tion have proved to be higher than the reliability of the same ratings"
(p. 277), and Cronbach (34), referring to a widely used and highly re
garded projective teBt, remarks: "Objective treatments of TAT perform
ance are disappointingly unreliable.... Need scores assigned by four
judges for the same test records had an average correlation of only .57.
Correlations between scores obtained at separate times were low. Even
split-half coefficients were, at best, around .48 for need scores."
(p. 448).
How shall the reliability of the Bender-Gestalt be ascer
tained? Billingslea (26) attempted to apply the split-half technique
by correlating scores for indices from the first five figures with
those from similar indices on the last four figures but obtained co
efficients that were so low as to compel him to conclude that "the
instrument is either unreliable as far as the factors measured are con
cerned, or that the factors measured are not consistently expressed by
the individual with behavior patterns as judged" (p. 11). He goes on
to point out: "Bender was well aware of this characteristic of her
technique since she says, 'There is a tendency for the continuous ex
perimentation with the external stimulating pattern and the action ten
dencies of the sensory—motor—conceptions! person. The final pattern may
represent a momentary point of equilibrium between the balancing function
and is liable to change.' Hutt suggests oaution, but makes definite in
terpretative statements that, of necessity, assume consistency in this
perceptual behavior of the individual." (p. 11). Pascal and Suttell (76)
- 315 -
obtained scores for all deviations on each figure and then tried several
odd-even splits whieh resulted in coefficients of about .50. The low-
split-half coefficients are not surprising in view of the fact that this
technique gives misleading results when the two halves of the test are
not as equivalent as two parallel forms of the same test would be. How
an individual will respond to each figure depends on the properties of
that figure, its difficulty, its position in the series, space require
ments, familiarity, and similar factors plus attitudinal and motivation
al factors. Having difficulty with the angle, of the slanting columns
in figure 2 does not necessary imply angulation difficulties with fig
ures 3, 6, or 7, nor does it necessarily follow that a rotation of one
figure must be accompanied by rotations of all figures. One reason why
Hutt's diagnostic and interpretative statements are difficult to apply
is the fact that he tacitly assumed the existence of general factors
which were consistently manifested throughout the test. He speaks, for
example, of a general reduction in size of the reproductions of neurotics
as compared to normals, but as the results of our study clearly show,
this reduction applies much more significantly to certain figures than
to others, and there are actually figures in which there is either no
difference or<a reversal in the direction of the difference. Bender also
assumes general factors, as do many clinicians who speak loosely of such
deviations as regression, tremor, angulation, size, erasures, guide lines,
asymmetry, and modification of curvature values.
The split-half technique was therefore rejected in the present
investigation aB a method of determining test reliability. The concept
of reliability was, however, inherent in our method of determining item

- 316 -
validity, since only those signs which differentiated in a consistent
direction on both the Initial Test and the Retest were retained for
inclusion in the final scoring system. Correlation of the Initial and
Retest scores for the entire initial validation group of 393 cases re
sulted in a reliability coefficient of .67, while a similar correla
tion for the 82 male subjects in the cross-validation study gave rise
to a coefficient of .68. These figures are remarkably close to the
test-retest coefficient of .71 reported by Pascal and Suttell for 44
subjects after an interval of 24 hours. Visibly disappointed by these
low coefficients, Pascal (75) argues that there is no really satisfac
tory method of estimating the reliability of the test and that "it
would therefore seem that the best estimate of the test reliability
at the present time is the validity coefficient," and in their book
Pascal and Suttell (76) contend that "as users of the test we have not
been primarily concerned with reliability as long as validity has been
maintained." The present writer takes the position that Bender-Gestalt
test-retest correlation coefficients are not true reliability coeffi
cients but rather are underestimations of the true reliability. As is
true on most performance tests— and the Bender-Gestalt is no exception—
a test administered a second time does not measure the same factors
tapped by the initial administration. In the case of the Bender-Gestalt,.
the Initial Test constitutes an unfamiliar task, while the Retest repre
sents a relatively familiar situation. Each succeeding design can be an
ticipated before it is actually presented, and problems of size and spa
tial arrangement can be readily solved on the basis of prior experience.
Memory factors, such as remembering the number of dots in figure 1 or

- 317 -
the number of columns in figure 2, may obviate counting of the ele
ments in the stimulus figures or reduce the time required to repro
duce the figures. If the interval between test and retest is exces
sively long, these memory factors will be reduced to a minimum, but
changes in the subject's adjustment or motivation may preclude con
sistency of measurement. Analysis of our findings shows that there
is better organization and apportionment of space, resulting in more
logical sequence, that less time is required to reproduce the figures,
that the reproductions tend to be smaller, that counting is less fre
quent, and that scores are systematically lower on the Retest. It
may be inferred from the above, though it cannot be demonstrated, that
the subjects approach the Retest with greater confidence and that the
higher degree of structuration of the Retest makes it a more pleasur
able or less trying experience. The extent to which the intervening
Recall Test influences the correlation between the Initial Test and
the Retest cannot be determined, but it probably is not great in view
of the similarity of our test-retest correlation with that reported
by Pascal and Suttell, who did not employ the recall situation. It
is true, however, that our Retest scores are systematically lower
than the Initial Test scores, whereas Pascal and Suttell find a neg
ligible practice effect. They did not score for such factors as count
ing and time, however, both of which contribute markedly to a decreased
Retest score.
If it were possible to readminister the Bender-Gestalt under
the same conditions with which the subject approaches the test for the
first time, it is submitted that the reliably coefficient would be much

- 318 -
higher. If we think of reliability as accuracy of measurement (i.e.,
the extent to which the obtained score approximates the "true" score)
rather than as consistency of measurement, and if we further regard
test-to-test fluctuations as "error" rather than as "real" variables,
we may assume that the reliability of the Bender-Gestalt is satisfac
tory for practical clinical work. The reliability of the Combined
Initial-Retest score which is recommended where time permits is prob
ably higher than that of either the Initial Test or the Retest alone,
since the two administrations of the test constitute a better sample
of the subject's perceptual-motor behavior, and it is well-known that
increasing the length of a test tends to increase its reliability.
Although the aim of the present investigation has been to
ascertain and define the uses and limitations of the Bender-Gestalt
Test with regard to its ability to differentiate between relatively
well-adjusted and relatively poorly adjusted non-psychotic adults,
and although the focus of this study has been upon empirical or opera
tional validity rather than upon the underlying psychological pro
cesses that give rise to and are therefore mirrored in the test per
formance, our discussion would be incomplete if due attention were
not accorded to some fundamental theoretical considerations. We
shall therefore devote the remainder of this section to a considera
tion of the question: what does the Bender-Gestalt measure?
We would agree with previous investigators that neurotic
individuals give little or no indication of "loss of the visual-
motor gestalt function" and that any attempt to distinguish neuro
tics from normals on the basis of so gross a criterion as "marked"

- 319 -
deviation of the reproduction from the original is doomed to failure.
Pascal and Suttell (76) extend this statement to include the func
tional psychoses, noting that "measurement at the level of complex
ity indicated by estimation of whether or not the essential Gestalten
are reproduced does not seem to be of sufficient discriminating power
to distinguish between normal adults and those with psychogenic dis
orders. Adults of normal intelligence without known cortical damage
do not, in our experience, fail to reproduce the essential Gestalten."
(p. 8). We would agree with Hutt (56) that the Bender-Gestalt is a
"Partially structured" test in which the "test stimulus remains rela
tively structured or conventionalized, but the subject may respond in
a relatively free or individual manner to this stimulus". The Bender-
Gestalt materials are structured, but the situation is largely defined
by the subject since all questions regarding procedure, other than those
included in the instructions are referred back to him. The structura
tion of materials is, however, greater than in the case of such repre
sentative projective tests as the Rorschach, TAT, and Man and Woman Draw
ings, and the range of responses is more limited. Comparing the Bender-
Gestalt with the Man and Woman Drawings, for example, it may be observed
that (l) the Bender-Gestalt presents an objective stimulus to copy, while
the figure drawings require a construction, rather than a reproduction, in
the absence of an external stimulus, (2) the figure drawings permit more in
dividualized performance because of the fact that the subject is not stimulus-
bound, (3) the figure drawings are meaningful and social, while the Wertheimer
figures are less meaningful and less social in their implications, (4) the
figure drawings permit more projection and more meaningful thematic as
sociations, since the drawing is usually a portrayal of the self, and

- 320 - t
(5) the Bender-Gestalt arouses less resistance and produces less strain
because the task is simpler and failure is less apparent. Neverthe
less , we would expect that even in the copying of geometric designs the
final product can never be satisfactorily explained solely in terms of
the external stimulating pattern. If the copying of designs is depend
ent also upon a need pattern, we may regard the Bender-Gestalt as a form
of expressive movement and make the further assumption that variations
in performance are correlated with personality organization. The results
of the present study suggest, however, that this correlation is far from
perfect and that the deviations of the reproduction from the original
stimuli are probably the result of the interplay of a multiplicity of
factors, only some of which are directly related to personality expres
sion. As Gardner Hurphy (71) has pointed outs "organization is not
molded exclusively by the need pattern or by the stimulus pattern. This
definition, stemming chiefly from Koffka and Sherif, makes perception
an individual process. The outer world can never be so completely un
structured as to make perception depend solely upon the perceiver, but
it can never be so sharply and clearly organized as to obliterate indi
vidual differences among perceivers." (p. 353)•
Among those factors which influence the characteristics of
the reproductions on the Bender-Gestalt Test, the following appear to
be particularly significant: the nature of the stimulus figures, the
accuracy of the perception of the stimulus figures, drawing ability,
and the personality of the subject, including the variables of motiva
tion, attitude, and set, which in turn reflects ego needs, current con
flicts and preoccupations, and anxieties. (1) The nature of the stimulus
- 321 -
figures includes such attributes as the difficulty of the figures, their
familiarity, special peculiarities, familiar or unfamiliar line quality
(e.g., dots vs. continuous lines), and the unitary or continuous char
acter of the form itself (e.g., figures A and A vs. figures 1 and 2,
which consist of perseverated elements). (2) By accuracy of perception
is meant the extent to which the perception of the stimulus figure is
analytical or superficial and the extent to which the subject pays atten
tion to the model while he is drawing. If the perception of the stimu
lus figure is faulty, the reproduction will also necessarily be lacking
in fidelity. (3) The drawing ability of the subject refers to his abil
ity to represent his perception faithfully, including the variable of
motor coordination. It is dangerous to assume that a faulty reproduc
tion necessarily reflects a faulty perception, since perceptual accuracy
and facility of motor execution often fall short of perfect correlation.
The object which we draw in response to the request to draw a man, a
house, or a flower may differ markedly from our perception of those ob
jects and we are usually only too aware of the deviation. With psychotic
subjects out of contact with reality it may be argued that this aware
ness is at a minimum because of diminished capacity for self-criticism,
but in neurotic subjects one might assume that the awareness of the in
adequacy would be magnified rather than absent. On the Bender-Gestalt,
both normal and neurotic subjects frequently express some dissatisfac
tion with their drawings, indicating that we are not dealing solely with
a perceptual response but rather with perceptual-motor behavior. When
questioned subsequently, a number of subjects who drew dashes or circles
instead of dots readily admitted that they perceived the dots in the
- 322 -
stimulus figures but represented them otherwise because it saved time
or required less expenditure of effort. Subjects with faulty motor
control may likewise draw tremulous or wavy lines to represent straight
lines in the stimuli of which they are well aware.
If the Bender-Gestalt is to be considered as a projective
test, however, such factors as the nature of the stimulus figures, the
accuracy of the perception, and the drawing ability of the subject should
not be sufficient to account for all of the observed deviations nor for
the differences in test performance between well-adjusted and emotional
disturbed subjects, since these three factors may be assumed to be rela
tively constant for all adult subjects of essentially normal intelli
gence who have no known organic brain damage which would conceivably af
fect their graphic-motor behavior. The factors so far enumerated would
explain deviations common to normal and abnormal subjects, while per
sonality factors may be invoked to account for modifications which dif
ferentiate the two groups or which occur unexpectedly in normal records.
Among the more significant personality factors are variations
in motivation, attitude, and set. If motivation or cooperation is lacking
or if the subject regards the test as nonsensical or as a waste of time,
he may assume the attitude, "This is silly so why try hard" and as a re
sult produce a test record which is well below his ability level. Wheth
er the subject takes the test lightly or seriously, whether his attitude
is facetious or determined, will affect the quality of his reproductions.
Personality factors such as carelessness or impulsiveness may be reflected
in test performance in the form of inaccuracies, variations in line qual
ity, expansively scrawled figures, disregard of order or space allotments,

- 323 -
running together of two or more figures, and an exceedingly low time
score. Perfectionism or compulsiveness may be expressed in the form
of counting and recounting of the elements in the stimulus figures and
the reproductions, meticulous attention to detail, rigidity of sequence,
sketching or retracing of the reproductions, and such painstaking care
that the reproductions exhibit a draftsmanlike precision. A depressed
individual may require an extremely long time to complete the test and
may manifest much of the meticulousness of approach of the compulsive
individual. Bender reports that less depressed psychotics tend to draw
meticulously, and it might be noted in this connection that a scoring
system which concerns itself only with graphic deviations and ignores
a "methods sign" such as counting may result in a number of low-scoring
records for compulsive and depressed patients. Nervous tension may
result in a less efficient performance than the subject is capable of
giving under more optimal conditions, just as anxiety on the Wechsler-
Bellevue may produce temporary inefficiencies on subtests like Digit
Span, Block Designs, or Object Assembly. Subjects who lack confidence

•
in the non-verbal area or who have a negative reaction to drawing may
take the attitude, "I'm not good at this, so I'll get it over with and
go on to something else". The degree of ego-involvement is also a fac
tor. To the extent that the subject feels that poor performance on the
test will be a reflection on his ability, he is likely to try hard to
reproduce the stimulus figures faithfully in order to preclude a feeling
of humiliation or loss of status in the eyes of the examiner. If, on
the other hand, the subject is not ego-involved and does not regard the
test as a measure of anything which he deems important, the incentive

- 324 -
to a good performance may be quite lacking. If, in addition, the sub
ject is highly self-confident, he may dash through the test without un
due concern. Present set, which may or may not reflect habitual modes
of perception, constitute still another set of attitudinal factors.
If the subject structures the task in such a way that he assumes that
speed is more important than accuracy, he will disregard detail and
give his major attention to general features. Should he feel that he
will be heavily penalized for inaccuracies, he will subordinate speed
to accuracy and copy the designs with precise attention to detail.
Questions such as: How minutely are the figures to be copied?, Shall
I make the reproductions the exact size of the originals?, Shall I get
all the figures on one page?, Shall I make dashes instead of dots to
save time?, may be answered either in terms of present set or underlying
personality needs, such as tendencies toward perfectionism or impulsive
ness.
Pascal and Suttell (76) believe that the Bender-Gestalt measures
attitude toward reality— the ability to respond adequate to stimuli in the
environment— and this attitude, in turn, is correlated with "ego strength".
The test situation for the individual, once he is sub

jected to it, becomes a bit of reality with which he has
to cope, we would expect, therefore, that in those per
sons in whom the attitude toward reality is most disturbed,
we will find greater deviations from the stimuli.... This
continuum is correlated with a progression from less to
greater psychological disturbance.... this progression is,
in turn, correlated with decreasing ability to respond ade
quately to stimuli in the environment.... This enabling ca
pacity of the organism, i.e., to respond adequately to stim
uli in the environment, is sometimes referred to as a func
tion of the ego. If, for the purposes of this discussion,
we limit our definition of ego functioning to the ability
to reproduce faithfully the B-G drawings as presented, then
we may say that ego strength lies on a continuum from very
low to very high B-G scores, (p. 9).
- 325 -
It is the contention of the present writer that the Bender-Gestalt
may measure attitude toward reality in the case of psychotics showing
gross deviations of the reproductions from the originals but that this
concept does not apply to the ability of the test to distinguish be
tween normals and neurotics. Pascal and Suttell assume that all devi
ations are "abnormal", whereas the present study and that of Billingslea
have shown that some deviations are more characteristic of normals. Under
these circumstances, their concept of a continuum in "ego strength" loses
much of its cogency. The writer would agree that attitude and motiva
tional factors transcend such factors as drawing ability and the objec
tive character of the stimuli in accounting for the vast majority of
Bender-Gestalt modifications but submits that these attitudes may take
many different forms and may reflect a variety of ego needs. It it pre
cisely because many different needs and attitudes may give rise to the
same deviation and because many deviations may reflect a single need
that an attitude of healthy skepticism and critical empiricism is urged
before psychodynamic interpretations of specific gestalt modifications
are advanced by the clinical psychologist.

CHAPTER VII
SUMMARY AND CONCLUSIONS
The Bender-Gestalt Test was administered to 393 white male
adult veterans (108 neurotics and 285 controls) in an initial valida
tion study and to 108 white non-veteran adults of both sexes (64 poor
ly adjusted individuals and 54 controls) in a cross-validation study.
The test was administered twice to each subject in conjunction with an
interpolated test of immediate recall of the stimulus figures. A total
of 1533 test records (511 Initial Test, 511 Retest, and 511 Recall Test
records) was scored and analyzed. An objective scoring system consist
ing of 82 general categories and 312 specific signs was developed, and
the incidence of each these signs was determined for each test record
of the cases in the initial validation study. 30 signs meeting speci
fied criteria of statistical significance and consistency were isolated
and constitute the final scoring system for the Initial Test and the Re
test. An additional 6 signs were evolved for the scoring of the Recall
Test records. The original test records were then rescored with the se
lected signs of known discriminating power, and satisfactory separations
of the normal and neurotic cases were achieved. The test records of the
subjects in the cross-validation study were next scored with the same
signs, and statistically significant separations of the poorly adjusted

and well-adjusted cases obtained, although the anticipated shrinkage in
the discriminating power of the test scores eventuated. The results war-
- 327 -
rant the following conclusions:
1. Normals are much more alike than different with regard to
perceptual-motor response to simple geometric designs, to
the extent that such responses are measured hy reproduc
tion of the Bender-Gestalt figures from a model or from mem
ory. In this respect, the findings agree to a greater ex
tent with those of Bender and Billingslea than with those of
Hutt and Pascal and Suttell.
2. Gross destructions of the gestalt are quite rare in both nor
mals and neurotics, except when the reproductions are drawn
from memory in the absence of an objective stimulus, and
even under such circumstances the incidence of distortions
is approximately the same for the two contrasting groups.
If any separation of normals from neurotics is to be ef
fected, the evidence suggests that it must be achieved through
consideration of more subtle modifications of the test figures.
3. The hypothesis that deviations from the stimuli occur more
frequently in all instances in abnormal subjects is rejected,
although it is true that the vast majority (about 80%) of
discriminating signs do occur more often in neurotics, justi
fying the following formulation: Deviations of the reproduc
tions from the stimuli occur more often in the records of the
neurotic and maladjusted cases sampled in the present investi
gation, but certain deviations, though considerably less in
number, are more characteristic of relatively well-adjusted
individuals, suggesting that both types of deviations should

- 328 -
be considered in an evaluation of Bender-Gestalt test per
formance.
4* None of the "abnormal" signs occurs exclusively in the ab
normal subjects, and none of the "normal" signs is found
solely in the controls.
5. whether the degree of psychological disturbance progres
sively parallels an increase in Bender-Gestalt score with
in the neurotic and maladjusted groups cannot be ascer
tained directly from the present study, but it seems unlike
ly in view of the not uncommon instances of high-scoring
normals and low-scoring abnormals. It may be that the dif
ferences in test performance of non-psychotics and psycho-
tics are as much differences in kind as in degree, i.e.,
the discriminating signs may be qualitatively different
from those found valid in differentiating the groups in
cluded in the present investigation. The references in
the literature to such signs as fragmentation, gross dis
tortion, perseveration, and ornamentation in psychotic
cases— signs which hardly ever occur in the subjects here
in sampled— lend some support to this conclusion.
6. General factors said to apply to all or almost ail of the

figures, such as reduction in size, rotation, sketching,
or angulation, are largely non-existent, and any interpre
tation of test performance which assumes such factors is
likely to be extravagant, misguided, or inapplicable. The
evidence supports the view that deviations are, for the

- 329 -
most part, specific to specific figures.
7. The hypothesis that differences in test performance be
tween normals and neurotics will be accentuated in the
absence of an objective stimulus during recall is not
given strong support. Only 6 signs are discriminating in
the initial validation study, and the validity of the Re
call score approaches zero in the cross-validation study.
Even in the initial validation, the Recall score adds lit
tle to the differentiating power of the other scores, war
ranting the conclusion that the scoring of the Recall test
is essentially superfluous for the populations sampled.
8. The recall processes of neurotics, in so far as the Bender-
Gestalt figures are concerned, differ but little from those
of normal subjects with respect to such variables as extent
of recall, order of recall, and deviations of the recalled
material from the original stimuli.
9* The test figures most often recalled are not necessarily
the ones that are first recalled, the correlation between
extent of recall and priority of recall being only .55 for
the Total Control group and .25 for the neurotic cases.
10. Distortions are about four times as frequent on the Recall
Test as on either the Initial Test or the Retest, approxi
mately three out of every five subjects (neurotics and nor
mals alike) manifesting one or more instances. Most fre
quently distorted is figure 7, which is also one of the most
difficult figures to copy, followed by figures 4 and 6.

- 330 -
11. There is a tendency on the part of both normals and neuro
tics for the reproductions to be shorter in length and taller
in height that the stimulus figures. There is also a consist
ent tendency for the neurotics to make their reproductions
shorter and flatter than those of the controls, the tendency
being greater for length than for height.
12. Although the neurotics and the controls cannot be distin
guished on the basis of time per figure, there is a tendency
for the neurotics to take more total time on the Initial Test.
This tendency persists on the Retest, but to an insignificant
degree.
13. Scoring categories occurring very infrequently (less than
10%) in all subjects on the Initial Test and the Retest are:
tremulous line, erasure, multiple attempt, guides, retracing,
curvature flattening in figures 4 and 5 , perseveration, clo
sure of open figures, dissociation, simplification, distor
tion (except figures 7 and 8), total rotation, part rotation,
reversal, paper rotation, card rotation, fragmentation, omis
sion, recounting of test figures and reproductions, and con
tinuation.
14. Scoring categories occurring relatively frequently (20% or
more) in all subjects on the Initial Test and the Retest are:
over-shooting lines, substitution, wave irregularity in fig
ure 6, asymmetry, displacement, angulation, horizontal irreg
ularity in figure 2, gaps in closed figures, counting of test
figures and reproductions, leg shortening in figure 5 , projee-

- 331 -
tion angulation, and line sag.
15. Constituting the final scoring system evolved in this study
are 30 signs which discriminate consistently between nor
mals and neurotics on the Initial Test and the Retest and
which survive two weeding-out procedures, one based on all
393 subjects in the initial validation study and on on 138
matched normals and neurotics. These signs are as follows:
numeration and wave flattening in figure 6 , asymmetry and
displacement in figure 5, contiguity in figure A, parallel
lines in figure 8 , upward slope in figures 1 and 2 , sinistrad
direction in figures 4 and 5 , upward direction in figure 6 ,
counting of test figure in figures 1 , 2 , 3 , 5 , and 6 , count
ing of reproduction in figures 1 , 2 , 3 , and 6 , initial part
in figure 7, angle differentiation in figure 7, pairing of
dots in figure 1 , correction in figure 4 , sagging line in
figures A and 4 , cohesion:^, short figures, and time:139.
16. In the cross-validation study, none of the signs employed
in the final scoring system shows any large reversals of
direction on both the Initial Test and the Retest, while
17 or the 30 signs discriminate consistently on both tests
in the expected direction.
17. Neurotics and normals can be distinguished on the basis of
Initial, Retest, Recall, or Combined scores, but the Recall
score is much less discriminating than the others and fails
completely on cross-validation. With regard to discriminating
power for all the populations sampled, the Combined score is

- 332 -
somewhat superior to the Initial score, which in turn is
slightly ,r>re effective than the Retest score.
18. The expected shrinkage in validity is found in the cross-
validation study, resulting in a greater degree of overlap
ping of the distributions of the maladjusted and well-
adjusted cases, but differences between the means of the
two contrasting groups, significant at the 0 .1% level, are
nevertheless obtained for the Initial, Retest, and Combined
scores.
19. Although the differences in Initial, Retest, and Combined
scores are slight for the Below 70, 1 Above 70, and 2 Above
70 groups, they are in the expected direction, and the dif
ferences between the Below 70 and the 2 Above 70 groups are
significant at the 5% level for the Retest and Combined scores.
20. Among the more significant differences between performance on
the Initial Test and the Retest are better organization and
apportionment of space, shorter and flatter reproductions,
less time per figure (and hence less total time), and sys
tematically lower scores on the Retest.
21. Retest reliability (Initial vs. Retest scores) is represented
by a correlation of .67 for the initial validation study and
a correlation of .68 for the cross-validation study, but these
coefficients are regarded as an underestimation of the true
reliability of the test on the grounds that a second adminis
tration of the test does not duplicate the first administra
tion. The chief reason for this lack of replication is that

- 333 -
the Retest is less amorphously structured for the subject
because of increased familiarity with the requirements of
the test situation. Though not demonstrated, it is highly
probable that the reliability of the Combined score is con
siderably higher than that of the Initial Test or Retest
alone.
22. The scoring method evolved in this investigation is not
significantly related to sex, age (within the range of 18
to 50 ), education (for persons with eighth grade schooling
or better), and intelligence (for I.Q.'s of 85 and above).
23. Hutt's "neurotic syndrome" is in general not substantiated
by the present findings. Several of Hutt's psychodynamic
interpretations are also open to question in view of the
fact that many of the deviations said to be indicative of
emotional or motivational processes occur equally often
among normals and neurotics.
24-. The results agree with those of Billingslea with regard to
the fact that most of the modifications of the reproduc
tions found on the Bender-Gestalt do not differentiate
normals from neurotics but disagree with his conclusion
that a practicable scoring system which will distinguish
between these two groups cannot be established.
25. The scoring method developed by Pascal and Suttell fails
almost completely to differentiate the 138 normal and
neurotic subjects of the matched criterion groups. This
is true whether the groups are considered as a whole or

- 334 -
analyzed into educational sub-groups. Contrary to their
findings, there is no significant relationship between edu
cation and test performance for adults, and this statement
holds true whether Pascal and Suttell's scoring system or
the one evolved in the present investigation is employed.
26. The Bender-Gestalt, as scored in the present study, is recom
mended as a screening device to be used as a supplement to
other psycho-diagnostic tests rather than as an instrument
for the elaborate interpretation of individual personality
dynamics.
27. Among the factors believed to influence Bender-Gestalt test
performance, it is hypothesized that the nature of the stimu
lus figures, the accuracy of the perception, and the drawing
ability of the subject account for deviations common to well-
adjusted and poorly adjusted individuals and that personality
factors such as motivation, attitude, set, and ego-involvement
account for the differences in test performance between the
two groups.
- 335 -
BIBLIOGRAPHY
1. Allport, G. Vi., Change and decay In the visual memory image.

Brit. J. Psychol.. 1930, 21, 133-143.
-2. Allport, G. W., and Vernon, P. E., Studies in expressive movement.
New York: Macmillan, 1933.
3. Anastasi, A., and Foley, J. P., Jr., A survey of the literature on
artistic behavior in the abnormal: I. Historical and theoret
ical background. J. Gen. Psychol.. 1941, 25, 111-142.
artistic behavior in the abnormal: II. Approaches and inter
relationships. Ann. N. Y. Acad. Scl.. 1942, 42, 106 pp.
artistic behavior in the abnormal: III. Spontaneous produc
tions. Psvchol. Monogr.. 1940, 52, No. 6 , 71 pp.
6 . Anastasi, A., and Foley, J. P., Jr., A survey of the literature on
artistic behavior in the abnormal: IV. Experimental investi
gations.J. Gen. Psvchol.. 1941, 25, 187-237.
7. Anastasi, A., and Foley, J. P., Jr., An analysis of spontaneous
artistic productions by the abnormal. J. Gen. Psvchol.. 1943»
28, 297-313.
8 . Anastasi, A., and Foley, J. P., Jr., An experimental study of the
drawing behavior of adult psychotics in comparison with that
of a normal control group. J. E x p . Psvchol.. 1944* 34, 169-194.
9. Angyal, A., The perceptual basis of somatic delusions in a case
of schizophrenia. Arch. Neur. and Psvchiat.. 1935» 34, 270-279.
10. Anonymous., A guide to the use of the Bender-Gestalt drawings.
Neuropsychiatric Service, Psychology and Social Work Section,
Mason General Hospital, L.I., N.Y., 1945*
11. Army Air Force Aviation Psychology Program Research Reports.
Report No. 15: The psychological program in Armv Air Force
convalescent hospitals. (Restricted).
12. Ash, P., The reliability of psychiatric diagnosis. J. Abnorm.
Soc. Psvchol.. 1949, 44, 272-276.
13. Barkley, Bill J., A note on the development of the Western Re
serve hapto-kinesthetic gestalt test. J. Clin. Psychol..
1949, 5, 179-180.
14. Beck, S. J., Rorschach's test. Vol. I, Basic processes. New
York: Grune and Stratton, 1944-
15. Becker, P. F., Das zeichnen schizophrener. Z. Ges. Neurol.
Psvchiat.. 1934, 149, 433-489.
16. Bell, J. E., Prp-1 ective techniques: a dynamic approach to the study
of the personality. New York: Longman, Green, 1948.
17. Bender. L., Principles of gestalt in copied form in mentally defec
tive and schizophrenic persona. Arch. Neurol. Psychiat.. 1932,
28, 661-673.
18. Bender, L., Disturbances in visuomotor gestalt function in organic
brain diseases associated with sensory aphasia. Arch. Neurol.
Psychiat.. 1933 , 30, 5H-537.
- 336 -
19. Bender, L., Gestalt functions in visual motor patterns in organic

disease of the brain, including dementia paralytica, alcoholic
psychoses, traumatic psychoses, and acute confusional states.
Arch. Neurol. Psvchiat.. 1935, 33, 1-24.
20. Bender, L., A visual motor gestalt test and its clinical use. Res.
Monographs, No. 3* New York: American Orthopsychiatric Assn.,
1938.
21. Bender, L., The Goodenough test in chronic encephalitis in children.
J. Nerv. Ment. Pis.. 1940, 91, 277-286.
22. Bender, L., Instructions for the use of visual motor Gestalt test.
New York: American Orthopsychiatric Assn., 1946.
23. Bender, L., Childhood schizophrenia. Amfir. J. Orthopsvchiat..
1947, 17, 40-56.
24. Bender, L., Psychological principles of the visual motor gestalt
test. Trans. N. Y. Acad. Sci.. 1949, 11, 164-170
25. Bender, L., Curran, F. J., and Schilder, P., Organization of memory
traces in the Korsakoff syndrome. Arch. Neurol. Psvchiat..
1938, 39, 482-487.
26. Billingslea, F. Y . , The Bender-Gestalt: an objective scoring
method and validating data. J. Clin. Psvchol.. 1948, 4, 1-27.
27. Boring, E. G., Sensation and perception in the history of experimental
psychology. New York: D. Appleton Century, 1942.
28. Brill, M., Study of instability using the Goodenough Drawing Scale.
J. Abnorm. Soc. Psvchol.. 1937, 32, 288-302.
29. Brown, J. F., and Rapaport, D., The role of the psychologist in the
psychiatric clinic. Bull. Menninger Clinic. 1941» 5> 75-84*
30. Brown, Growth of "memory images". Amer. J. Psvchol.. 1935>
47, 90-102.
31. Bruner, J. S., and Goodman, C. C., Value and need as organizing
factors in perception. J. Abnorm. Soc. Psychol.. 1947, 42, 33-44*
32. Cameron, D. E., Studies in depression. J. Ment. Sci.. 1936, 82,
148-161.
33* Coffin, T. E., Some conditions of suggestion and suggestibility:
a study of certain attitudinal and situational factors influencing
the process of suggestion. Psychol. Monogr.. 1941, 53, No. 24 .
34* Cronbach, L. J., Essentials of psychological testing. New York:
Harper, 1949.
35* Deri, S. K . , Introduction to the Szondi Test: theory and practice.
New York: Grune and Stratton, 1949.
36. Des Lauriers, A., and Halpern, F., Psychological tests in childhood
schizophrenia. Amer. J. Ortnopsvchiat.. 1947, 17, 57-67.
37. Ellis, A., and Conrad, H. S., The validity of personality inventories
in military practice. Psychol. Bull.. 1948. 45, 385-426.
38. Fabian, A. S., Vertical rotation in visual-motor performance— its
relationship to reading reversals. J. Educ. Psvchol.. 1945»
36, 129-154*
39* Fehrer, E. V., An investigation of the learning of visually perceived
forms. Amer. J. Psychol.. 1935* 47, 187-221.
40. Foster, J. C., Significant responses in certain memory tests. J.
AppI . Psvchol.. 1920, 4, 142-153*
- 337 -
4-lj. Frank, L. K . , Projective methods for the study of personality.

J. Psvchol.. 1938, 8, 389-413.
42. Gibson, J. J., The reproduction of visually perceived form. J. E x p .
Psvchol.. 1929, 12, 1-39.
43. Guilford, J. P., Fundamental statistics in psychology and educa
tion. New Yorks McGraw-Hill, 1942.
44. Hanawalt, N. G., Memory traces for figures in recall and recogni
tion. Arch. Psvchol.. 1937, No. 216.
45. Harrower, M. ft., Changes in figure ground perceptions in patients
with cortical lesions. Brit. J. Psvchol.. 1938, 30, 47-51.
46. Harrower, M. R., Neurotic depression in a child. In Burton, A.,
and Harris, R. H. (Ed.), Case histories in clinical and abnormal
psychology. New York: Harper, 1947, 519-529.
47. Harrower, M. R., Group techniques for the Rorschach Test. In Abt,
L. B., and Beliak, L., (Ed.), Projective psychology. New York:
Knopf, 1950, 146-184.
48. Harrower-Erickson, M. R., The value and limitations of the so-called
"neurotic signs". Ror. Res. Exch.. 1942, 6, 109-114.
49. Hathaway, S. R., and McKinley, J. C., Manual for the Minnesota
Multiphasic Personality Inventory. New York: Psychological
Corporation, 1945.
50. Hertz, M. R., Scoring the Rorschach Test with specific reference
to the normal detail category. Amer. J. Orthopsvchiat.. 1938,
8 , 100-121.
51. Hertz, M. R., The shading response in the Rorschach inkblot test:
a review of its scoring and interpretation. J. Gen. Psvchol..
1940, 23, 123-167.
52. Hertzman, M., Recent research on the group Rorschach Test. Ror.
Res. Exch.. 1943, 7, 1-6.
53. Hovland, C. I., Experimental studies in rote-learning theory.
I. Reminiscence following learning by massed and distributed
practice. J. Exp. Psychol.. 1938, 22, 201-224.
54* Hunt, J. McV., and Cofer, C. N., Psychological deficit. In Hunt,
J. McV. (Ed.), Personality and the behavior disorders. Vol. 2.
New York: Ronald, 1944, 971-1032.
55. Hutt, M. L., The use of projective methods of personality measure
ment in army medical installations. J. Clin. Psvchol.. 1945,
1, 134-140.
56. Hutt, M. L., A tentative guide for the administration and interpre
tation of the Bender-Gestalt Test. U. S. Army: Adjutant General's
School, June, 1945 (Restricted).
57. Kadis, A. L., Finger-painting as a projective technique. In Abt, L. E.,
and Beliak, L., (Ed.), Projective psychology. New York: Knopf,
1950, 403-431.
58. Kisker, G. W., A projective approach -bo personality patterns during
insulin shock and metrazol convulsive therapy. J. Abnorm. Soc.
Psvchol.. 1942, 37, 120-124.
59. Kuhlmann, F., On the analysis of memory consciousness: a study in
mental imagery and memory of meaningless visual forms. Psychol.
Rev.. 1906, 13, 316-348.
- 338 -
60. Lawshe, C. H., and Baker, P. C., Three aids in the evaluation of
the significance of the difference between percentages. Educ.
Psvchol. Measmt.. 1950, 10, 263-270.
61. Loeblowitz-Lennard, H., and Eiessman, F., Jr., Recall in the Thema
tic Apperception Test: an experimental investigation into the
meaning of recall of phantasy with reference to personality
diagnosis. J. Pers.. 1945, 14, 41-46.
62. Lord, E., and Wood, L., Diagnostic values in a visuo-motor test.
Amer. J. Orthoosvchiat.. 1942, 12, 414-428.
63. Machover, K., Personality pro.iection in the drawing of the human
figure. Springfield, 111.: Charles C. Thomas, 1949.
64 . Mailer, J. B., Personality tests. In Hunt, J. McV. (Ed.), Persona
lity and the behavior disorders. Vol. 1. New York: Ronald,
1944, 170-213.
65 . Marks, E. S., Skin color judgments of Negro college students.
J. Abnorm. Soc. Psvchol.. 1943, 38-370-376.
66 . Mira, E., Myokinetic psychodiagnosis: a new technique for exploring
the conative trends of personality. Proc. Roy. Soc. Med.. 1940,
33, 173-194.
67. Mira, E., Psychiatry and war. N. Y. Academy of Medicine, Salmon
Committee for Psychiatry and Mental Hygiene: Thomas william
Salmon Memorial Lectures. New York: Norton, 1943*
68. Monroe, R. L., The inspection technique. Ror. Res. Exch.. 1944,
8, 46-70.
69 . Monroe, R. L., Prediction of adjustment and academic performance of
college students by a modification of the Rorschach method.
Ap p I . Psychol. Monogr.. 1945, No. 7.
70. Monroe, R. L., The inspection technique for the Rorschach protocol.
In Abt, L. E., and Beliak, L.,(Ed.), Protective psychology.
New York: Knopf, 1950, 91-145.
71. Murphy, 0., Personality; a biosocial interpretation. New York:
Harper, 1947.
72. Napoli, P. J., Finger-painting and personality diagnosis. Genet.
Psychol. Monogr.. 1946, 34, 192-230
73. Orenstein, L. L., and Schilder, P., Psychological considerations of
the insulin treatment in schizophrenia. J. Nerv. Ment. Pis..
1938, 88, 397-413.
74* Otis, A. S., Manual for the Otis Quick Scoring Mental Ability Tests-
Gajnma Test. Form AM. Yonkers: World Book Company, 1940.
75* Pascal, G. R. Quantification of the Bender-Gestalt: a preliminary
report. Amer. J. Orthoosvchiat.. 1950, 20, 418-423.
76. Pascal, G. R., and Suttell, B. J., The Bender-Gestalt Test. New
York: Grune and Stratton, 1951.
77. Perkins, F. T., Symmetry in visual recall. Amer. J. Psychol..
1932, 44, 473-490.
78. Piotrowski, 4. , On the Rorschach method and its application in
organic disturbances of the central nervous system. Ror.
Res. Exch.. 1936, 1, 23-40.
79. Piotrowski, Z .., Positive and negative Rorschach organic reactions.
Ror. Res. Exch., 1940, 4, 147-151.
- 339 -
80. Sapas, E . , Zeichnerische reproduktionen einfacher figuren durch

geisteskranke. Schweiz. Arch. Neurol. Psvchiat.. 1918, 4,
140-152.
81. Sargent, H., Projective methods: their origins, theory, and
application in personality research. Psvchol. Bull.. 1945»
42, 257-293.
82. Saudek, R.. Experiments with handwriting. New York: Morrow, 1928.
83. Schilder, P., Psychic disturbances after head injuries. Amer. J .
Psvchiat.. 1934, 91, 155-188.
84. Schilder, P., The image and appearance of the human body. London
Routledge, 1935.
85. Schilder, P., Notes on the psychology of metrazol treatment of
schizophrenia. J. Nerv. Ment. Pis.. 1939, 89, 133-144.
.
86 Spiegel, H., Shor, J., and Fishman, S., An hypnotic abliation
technique for the study of personality development.
Psvchosom. Med.. 1945, 7, 273-278.
87. Stainbrook, E . , and Lowenbach, H., Writing and drawing of psycho
tic individuals after electrically induced convulsions.
J._ Nerv. Ment. Pis.. 1944, 99, 382-388.
.
88 Street, R., The gestalt completion test and mental disorder.
J. Abnorm. Soc. Psvchol.. 1934, 29, 141-142.
89. Wayne, D. M., Adams, M., and Rowe, L. A., A study of military
prisoners at a disciplinary barracks suspected of homosexual
activities. Mil. Sure.. 1947, 101.
90. Wechsler, D., The measurement of adult intelligence. Baltimore:
Williams and Wilkins, 1944.
91. Wechsler, D., and Hartogs, R . , The clinical measurement of
anxiety. Psvchiat. Quart.. 1945, 19, 618-635.
92. Wechsler, I., A textbook of clinical neurology (6th ed.).
Philadelphia: Saunders, 1947.
93. Wertheimer, M., Studies in the theory of gestalt psychology.
Psychol. Forsch.. 1923, 4, 301-350.
94. White, R. W., Interpretation of imaginative productions. In
Hunt, J. McV. (Ed.), Personality and the behavior disorders.
Vol. 1. New York: Ronald, 1944, 214-251.
95. Wolff, W., The experimental study of forms of expression. Char
acter and Pers., 1933, 2, 168-176.
96. Wolff, W., Involuntary self-expression in gait and other move
ments: and experimental study. Character and Pers.. 1935,
3, 327-344.
97. Wolff, W., Projective methods for personality analysis of expres
sive behavior in preschool children. Character and Pers..
1942, 10, 309-330.
98. Wolff, W., Graphometry: a new diagnostic method. Psvchol. Bull..
1942, 39, 456 (abstr.).
99. Wolff, W., The expression of personality. New York: Harper, 1943.
.
100 Wolff, W., The personality of the preschool child. New York:
Grune and Stratton, 1946.
101. Wolff, W., Diagrams of the unconscious. New York: Grune and
Stratton, 1948.
- 340 -
102. Woltmann, A. G., The Bender Visual-Motor Gestalt Test. In

Abt, L. E., and Beliak, L. (Ed), Protective psychology.
New York: Knopf, 1950, 322-356.
103. Wood, L., and Shulman, E., The Ellis visual designs test.
J. Educ. Psvchol.. 1940, 31, 591-602.
104. Woodworth, R. S., Experimental psychology. New York: Henry
Holt, 1938.
105. Zubin, J. A., Quantitative approach to measuring regularity of
succession in the Rorschach experiment. Character and Pers.,
1941, 10, 67-78.
106. Zubin, J. A., Psychometric approach to the evaluation of the
Rorschach test. Psvchiat.. 1941, 4, 547-566.
107. Zubin, J. A., Chute, E., and Verniar, S., Psychometric scales
for scoring Rorschach test responses. Character and Pers..
1943, 11, 277-301.
- 341 -
APPENDICES
Table LVI. Scoring of the Initial Test Record of Each Case in the
Below 70 Matched Criterion Group on the Basis of the
Selected Signs Shown to Have Consistent Validity on
the Initial Test and the Retest
Table LVII. Scoring of the Initial Test Record of Each Case in the
Neurotic Matched Criterion Group on the Basis of the
Selected Signs Shown to Have Consistent Validity on
the Initial Test and the Retest
Table LVIII. Scoring of the Recall Test Record of Each Case in the
Below 70 Matched Criterion Group on the Basis of
Selected Signs Shown to be Significant at the % Level
Table LVIX. Scoring of the Recall Test Record of Each Case in the
Neurotic Matched Criterion Group on the Basis of Se
lected Signs Shown to be Significant at the 5% Level
Table LX. Scoring of the Retest Record of Each Case in the Below
70 Matched Criterion Group on the Basis of the Se
lected Signs Shown to Have Consistent Validity on the
Initial Test and the Retest.
Table LXI. Scoring of the Retest Record of Each Case in the Neuro
tic Matched Criterion Group on the Basis of the Selected
.. Signs Shown to Have Consistent Validity on the Initial
Test and the Retest
Table LXIII. Initial, Retest, and Combined Scores of Each Case in

the Below 70 Matched Criterion Group
Table LXIV. Initial, Retest, and Combined Scores of Each Case in

the Neurotic Matched Criterion Group
Table LXV1I. Initial, Retest, Recall, and Total Scores of Each Case
in the Below 70 Matched Criterion Group
Table LXVIII. Initial, Retest, Recall, and Total Scores of Each Case
in the Neurotic Matched Criterion Group
Table LXXIII. Initial, Retest, and Combined Scores of Each Case in

the Below 70 Unmatched Criterion Group
Table LXXIV. Initial, Retest, and Combined Scores of Each Case in

the Neurotic Unmatched Criterion Group
- 342 -
Table LXXVIII. Initial, Retest, and Combined Scores of Each Case

in the 1 Above 70 Criterion Group
Table LXXIX. Initial, Retest, and Combined Scores of Each Case

in the 2 Above 70 Criterion Group
Table LXXXII. Scoring of the Initial Test Record of Each Case

in the Below 70 Matched Criterion Group, Using
Pascal and Suttell’s Signs
Table LXXXIII. Scoring of the Initial Test Record of Each Case

in the Neurotic Matched Criterion Group, Using
Pascal and Suttell* s Signs
Table LXXXVII. Scoring of the Initial Test Record of Each Case

in the Total Male TAC Satisfactory Adjustment
.Group on the Basis of the Selected Signs Shown
to Have Consistent Validity on the Initial Test
and the Retest
Table LXXXVIII. Scoring of the Initial Test Record of Each Case in

the Total Male TAC Unsatisfactory Adjustment Group
on the Basis of the Selected Signs Shown to Have
Consistent Validity on the Initial Test and the
Retest
Table LXXIX. Scoring of the Retest Record of Each Case in the

Total Male TAC Satisfactory Adjustment Group on
the Basis of the Selected Signs Shown to Have
Consistent Validity on the Initial Test and the
Retest
Table XC. Scoring of the Retest Record of Each Case in the

Total Male TAC Unsatisfactory Adjustment Group on
the Basis of the Selected Signs Shown to Have Con
sistent Validity on the Initial Test and the Retest
Table XCII. Initial, Retest, and Combined Initial and Retest

Scores for Each Case in the Total Male TAC Satis
factory Adjustment Group (Matched and Unmatched)
Table XCIII. Initial, Retest, and Combined Initial and Retest

Scores for Each Case in the Total Male TAC Unsatis
factory Adjustment Group (Matched and Unmatched)
- 343 -
TABLE LVI
Scoring of the Initial Test Record of Each Case in the Belov 70

Matched Criterion Group on the Basis of the Selected Signs Shown
to Have Consistent Validity on the Initial Test and the Retest
____________________________ Case No._____________________________

-Sign_________ 1 2 3 4 5 6 7 8 9 10 11 1 2 .1 3 14 15 16 17 18
20-6 - - - - -
22-6
27-5 - - - - -
29-5
30-A - -
33-8
55-1
55-2
57-4+ ♦ ♦ ♦ «.
57-5
58-6* ♦ ♦ ♦
6o-l - — - - - - - - -
60-2 - - - - - - - - -
60-3 - -
60-5 - - - -
6 0 -6 —
61-1 - - -
61-2 - _ _ _ _ _
61-3
61-6
65-7* ♦ ♦ ♦ + ♦ + + + + + + + t + +
66-7 - - -
70-1+ ♦
72-4 - -
75-A - _ _ _ _ _ _
75-4 _ _ _ _ _
Cohesion:^
Times 139* * +
Short Figures - - - -
Initial Score 8 4 0 4 6 4 5 1 2 6 3 8 1 5 1 7 6 4
(continued)
- 344 -
TABLE LVI (continued)
Case No.
-Sign 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34
2 0 -6
22-6 — — —
27-5 _
29-5 — —
29-6 - — — _ — —
30-A — — - —
33-8 -
55-1 -
55-2 -
57-4* ♦ * *
57-5
58 -6+ + * * + * ♦ ♦
60-1 — — - — — —
60-2 _ — _ _ _
60-3 _
60-5 _
60-6 _
61-1 _ _ — _
61-2 — _ _ — _
61-3 — — —
61-6
65-7+ ♦ + ♦ ♦ ♦ ♦ ♦ ♦ * + * ♦
66-7
70- 1* ♦ ♦ *
72-4 - — -
75-A - - - - - - - -
75-4 - - - - -
Cohesion:^
Time: 139* ♦ ♦ ♦ ♦ ♦ ♦
Short Figures — — — — — — — — — —
Initial Score 0 0 2 4 *2 2 6 11 2 8 4 4 0 2 4 3
(continued)
- 345 -
Case No.
Sign_________35 36 37 38 39 40 41 42 43 44 45 46 .47 48 49 50
20-6 - -
22-6 - -
27-5 - -
29-5
29-6 - - - - - - - - _ _ _
30-A - - -
33-8
55-1
55-2
57-4* * * * + * +
57-5
58-6* ♦
60-1 - -- - _ _ _ _ _
60-2 - - - -
60-3
60-5 - -
60-6
61-1 - _ _
61-2 - - - - -
61-3
61-6
65-7+ ♦ + ♦ ♦ ♦ ♦ * ♦ ♦ ♦ ♦ +
66-7 - - - -
70-1* * + + + * +
72-4 - - _
75-A - - - - - - -
75-4 - - - -
Cohesion:^
Time:139+ ♦ ■ + + + ♦ ♦ *
Short Figures - -
Initial Score +2 3 6 3 4 7 1 5 4 5 5 4 0 0 4 4
(continued)
- 346 -
Case No.
Sign 51 52 52 54 55 56 57 58 59 60 61 62 63 64
20-6 _
22-6 - - - -
27-5 - - -
29-5 - - - -
29-6 - - - - - -
30-A - -
33-8
55-1 - -
55-2 - - -
57-4+ * + + + + + +
57-5
58-6+ + + + + *
60-1 — - - -
60-2 - - -
60-3 -
60-5 -
60-6
61-1 - -
61-2 - -
61-3
61-6
65-7* + + * + + + + ♦ +
66-7 — - - - -
70-1+ + ♦ + +
72-4 -
75-A - - - - - - -
75-4 - - -
Cohesion: •§
Time:139* + + + ♦ + + ♦ ♦
Short Figures —
Initial Score * 4 + 1 3 4 + 1 9 0 1 2 4 3 3 1 0
(continued)
TABLE LVI (concluded)
Case No.
Sign 65 66 67 68 69 Total
20-6 15
22-6 —
12
27-5 12
29-5 8
29-6 - - - 38
30-A - -
15
33-8 4
55-1 -
5
55-2 7
57-4+ + + ♦ 23
57-5 0
58-6+ + * 18
60-1 - - 30
60-2 - - 30
60-3 6
60-5 10
60-6 3
61-1 — 18
61-2 - - -
24
61-3 6
61-6 2
65-7* + + + + * 53
66-7 14
70-1+ ♦ + 16
72-4 - - 12
75-A - - 32
75-4 - - 21
Cohesion:f 0
Time:139+ + + 25
Short Figures — 23
Initial Scores 0 6 *2 4 2
- 348 -
I I I I I
CM
v£> I I I I + I
CM
T3
©
IT I I * I ♦ I CM
CM 2
+»
II I I * I 1 * 1 CfN oa
Group
o
CO I I I ♦ I I I I I lll*i O
CM rH
Test Record of Each Case in the Neurotic Matched Criterion
CM I I II I I I * I * ir\
CM
of the Selected Signs Shown to Have Consistent Validity
rH I I I I I I I * l
CM
O I I I I I I * * I I NO
CM
I I II I I I * I 1 * 1 CO
I I I I I r-
II I * I I c-
vO I ill I I I * ♦ I + to
H
O ur> II il I+ * I III i to
21
©
02 I I + I I I I I I * I o
rH
o
I II I I I I I I I I I
3
II I I * I I I * CO
a
I * I
I * I cr\
cr I I I I I I
co I I I I I ' I a
I I l I O'
on the Basis
of the Initial
'O I I I I I I
ir I I I I I ♦ I I I o
rH
I I II * I I I t"-
on I I I I I I I I
CM I I I I I II
Scoring
3
I I I m
01 ©
® fn
O
Htol 3 O
•• * Go CO
0 O' iH
o rH
rH rH a)
* 4i * * Ifl «•-P
v fl'O v C M A 'O < l « H C M C M O M O vO H CM t 'r l sf © © h
i ij i i i i i ij i WiO O
OCMC'O'O'Ol'MAiftf'^’ i OiO Qi HlHirtH H i
i ni.v o
L p1OiI jJ.f v l1A O
*'-Sn§J S2 •s
CO CMCMCMCMCMtrv(»1«Mrv>Airv»r»M5'0'0'0'0'0'0'0'0'0'OC^C^C'CÔHCO
- 349 -
i i i i 00
i + i i i
*8
i- i i i ii ♦ ir\ §
1
i i i i i i it ♦ i
i i i i ii + C"
5
i i i i i i C"
i i i i i i i i i ♦ i o
i i ♦ i CM
i i i i i i ii i i i i i i i nO
rH
i i i i ii i i i i i i i C*>
rH
i i i i i i i i i i i i i a
i i i i i i vO
i i i ♦ i i •sf
i i ♦ i i ir\
i i i m
i i i i 111*1 oo
i i i i i ii * i c~
i i i i i i i i i i i i rH
rH
ii iii i 00
i i ii ii I>
i i i i I"
i 'i t i i i i •J5
i i i i i i i 00
TABLE LVII (continued)
i i i i i i ♦ i i i * i O
rH
i i i i i i i i i i i
a ©
a f*
O
O
CO
flO'tI
O rH
Qj
•rH
vO vp tfWO «3j to H C$ ir\"S rH CM r*'* VNsO rH CM rr\vO I?- t" H -a) --t 2 ® t* -P
OÊ>^0'oAjNukt^t>OOQcioQOrHrHHHiA\l!>OCMiAi£\'oiijS
CO CMCMCMCMCMCr%rr\»A»rviA>A«r>'5v0v0'®'^s0'0^v0s0\0C'CiC,-r'OeHC0
- 350 -
(^rlOlrt'TrlrlHH -4 ■>* cr\<n rH C*'*C*\ H cr\ rl r-t C*'*
oO
v> I I ♦ CM
l l I I I I I ♦ i
—I
i
—I
I' I I
ve I I C~
I I I II I + t-
vO I I I I I I ♦ I I ♦ o
H
vd I I I I I O'
cn I I I I+ I
vO »n
it cm I I I
0 vO II I * III I **>
a
a
vO t II ♦ I I I o
rH
s I I III I I I I CM
i
—I
O' I I I
u> I I I I + I CO
to I I I
*n I + I II III ♦ I I
e- II II II I♦
m O'
vO I ill
m III ♦
3
*
«v I I
in CM
vn
I I I I ♦ III I I I I
(concluded)
3
,*n I I I I I I
3
a ®
8
H|0l
♦i
TABLE LVII
o
a O' -h CO
o I*, I—I
a?
g vOvOMVUp\q<t1»HCM-$»nvSrHCMr^»n'qr4CM(nvpi>n-rH'iJ'^)Nf® " **
•r1 O £}t^CT'Ôc^<A»Ac^C^«<iQiQ{ ^ H 3 3 ,J|iAj>cic^V r\i£'S'Sia
03 CM 02 02 02 02 CO CO lA V \ 1A IA *A '•O vD \Q \Q \Q \Q \Q \Q \Q *1$ f * I> o E-t CO
TABLE LVIII
Scoring of the Initial Test Record of Each Case in the Below 70

Matched Criterion Group on the Basis of the Selected Signs Shown
to be Significant at the 5% Level
Case No.
Sien 1 2 3 4 5 7 8 9 10 11 12 13 14 _ 15_ 16 17 1?
Cohesion: ^ -
Closure-4*
Short Figures (4)
A (5-9)* ♦ ♦ * ♦
6 (1-4)+ * ♦ ♦ ♦ ♦ ♦ * + ♦ ♦ ♦
34 or less
Recall Score 0 0 + 1 +2 0 1 0+2 0_ +1 1 +2 0 0 +1 ♦1 *1 +2
Cas_e No_».
Sien 19 30 21 22 3? 24 3? 36 27 28 29 30 21 ?2 22 24
Cohesion: jz
Closure-4+ * * ♦
Short Figures (4) - - -
A (5-9)+ * ♦ ♦ * *
6 (1-4)+ + + * ♦ ♦ ♦ ♦ ♦ ♦ +
34 or less
Recall Score +1 0 ■'1 +2 +1 ♦1 *1 ♦1 ♦1 1 ♦3 0 0 0 ♦1 +3
Case No.
CO
*1
Sien 25 26 27 29 40 41 42 42 _44 45. 46 47 48 49 50
Cohesion: \
Closure-4+ ♦ * + +
Short Figures (4) - -
A (5-9)+ ♦ * * + ♦ + ♦ ♦ *
6 (1-4)+ ♦ * ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦
34 or less
Recall Score ♦2 ♦2 l *1 ♦2 0 +1 ♦1 *1 0 0 ♦2 ♦1 t l t2 *1
(continued)
- 352 -
TABLE LVIII (concluded)
Case No,■
Sien 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66
Cohesion: £
Closure-^* * + + ♦
Short Figures (4) -
A (5-9)4 *■ * * 4 ♦ *
6 (1-4)* ♦ * ♦ * * ♦ ♦ + *
3i or less
Recall Score 4? ♦1 0 +1 1 0 0+2 +3 +2 ♦1 *1 0 ♦2 ♦2
Case No
Sien 67 68 69
Cohesion:^-
Closure-4* ♦
Short Figures (4)
A (5-9)4 ♦
6 (1-4)4 *
3i or less
Recall Score .. *£._ .0 4I
- 353 -
TABLE LVIX
Scoring of the Recall Test Record of Each Case in the Neurotic Matched
Criterion Group on the Basis of Selected Signs Shown to be
Significant at the % Level
Case No.
Sien 1.2 3 4 5 6 7 $ 9 10 11 12 13 14 15
Cohesion: ^
Closure-4*
Short Figures (4) - - - -
A (5-9)* + *
6 (1-4)* * * + ♦ + ♦ ♦ *
3k or less — —
Recall Score 0 0 0 .+1 X - 1 ♦1 ♦1 0 ♦1 l 1 1
Case No.
Sign 16 17 18 19 20 21 22 23 24 25 _26 27 28 29 30
Cohesion: \
Closure-4* *
Short Figures (4) - — - -
A (5-9)* * ♦ *
6 (1-4)* * * * ♦ * *
Jk or less —
Recall Score o _*x_ .+1 I ___ Q _fci_ 1 2 * ♦2 9 0
Case No.
Sign 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45
Cohesion: ^
Closure-4*
Short Figures (4) - - - -
A (5-9)* ♦ * * +
6 (1-4)* + + * + * ♦ +
3k or less —
i*4
0 0 0
o
Recall Score *1 l *1 1 0 ti 0 *1 0
♦
2
(continued)
- 354 -
TABLE LVIX (concluded)
Case Ho.
Sien 46 47 48 49 50 51 52 53 54 55 56 57 58 59 6o
Cohesion: \
Closure-4* +
Short Figures (4) - - - - - -
A (5-9)* ♦ ♦ ♦ * ♦ +
6 (1-4)* ♦ + ♦ ♦ + *
or less
Recall Score 0 +1 tl t2 +2„ 0 1 2 ' +1 tl 0 2 *1 0 0
Case No.
Sien 61 62 63 64 65 66 67 68 69
Cohesion: £
Closure-4* * ♦ ♦
Short Figures (4)
A (5-9)* ♦
6 (1-4)* ♦
3^ or less —
Recall Score 0 0 *2 1 0 0 1 tl tl
to
I
h
o
d x
O P
■H *H
M TJ
© P
P P
©
>
0
P
a
©
1 p
CQ
co
©
o P
o ©
R ©
© p
* ©
o
—I•i ^
I
© ©
PQ
55 *0
©
5 63
a5 *
P CQ P
ra
V © ©
© d EH
© UO
o
CQ ©
•d
o
©
w
O «
«P
t3 CQ P
Jh
o ©
O
© 5
PS
CO rH
p o
©
©
p
in
P©S cd nO
0
SJ
«H
o d
o
bo
5
u
o
o
CQ vO
hJcm a
» 4> W
d on p
O PM*,
g
P
'OvOir\inNO«©coHCNl'4’kr\NOHoic>^crlvOHcy
, i i i t i » js a o p
sQ \Q vQ vQ v‘*"8
O CHÔ'OC>rr\UMnt~C-CO
OCHÔ'odAuMAf-C'-CO j \Q P ™ if> If* ° -3 -C
j~s>c-C'-C'-C'-ohcq
cvjcvojcvc\jf^rc\uMrMr»vA«rv
- 356 -
s|
IT
1 1 ♦ 1 ♦ 1 CM
<n ■ I I 1 1 I * 1 II +1 00
m
CM ♦ 1 +1 o
w>
CM
H
m 1 1 * + * * ♦ ■(
o 1 1 1 + 1 toy -
in
1 ♦ 1 1 CM
<9
s
* 1 O
l>
sf
1 + 1 * O
NO
S
1 1 1 1 1 * ♦ 1 + coy
m
sf
1 1 I <• + r—l
s 1 i—1
s
to I + O
sf
CM
s
1 III 1 ♦ * * CM
•
o r- 1 * 1 i-l
S3 SI
e
a O 1 1 I I 1 ♦ + + CM
tf Sj
o
O' •I 1 + 1 I ♦ * 1 CM
CO
00
on 1 1 + 1 + 1 CM
e' 1 I + 1 CM
en
S3 II II 1 + * 1 -t
co
m
CO 1 1 + 1 ♦ 1 CM
sl
CO 1 1 + + 1 + 1 i—i
CO I 1 1 + CM
CO
CM 1 1 + III -4-
CO
i-1
H * ♦ 1 -V
cn
O CM
CO ♦ *
(continued)
O'
CM
1 1 1 1* C0\
CQ
Retest Score
B O' -H
LX
O fo\ (*«
•H r-1
•k + CQ •• +S
TABLE
vO in ifl'O <! to rl Cl se ITSvO r-l CM C^ lf\\0 r-i CM CO\vO r~ C^H st«l t ® 9 b

I 1 1, J 1 1 1 1 1 1 J J 1 I I I I I 1. 1. 1 1
O CMC'-0'O'OC0\v\ir»£'-tJ- 0 0 O O O Q Q r - I H r l r l
CM cm cm CMCMro\co\mir\ir\ir\mvOvO\0'Jj'S'OÔvOsO
- 357 -
rH
S
o CM rH rH c0!f-H rH rH rH rH rH H ^trHH rH CM (A CM
61
O'
vO
I * > J I H* I + m
58 II I I + I + (A
n- + I
SO
I I I I I
$
1 1 * 1 *
NO
fl I I + O
o m3
z o
I I I * rr\
® vO
M
td
u CM • I I * ♦ + O
sO
rH
vO
I I I I * + I I * I
8 • I I I -*
O'
m I CM
to I I * H
in
C- II I + CM
in
vO I I I i m
in
in
TABLE LI (concluded)
m * I rH
♦
to
®.
f
®
-#i>
* bo
8
fl O'-rH m
O CACn
s •H rH +>
*r- Hr i . * + m
a •.• p»
. h t»
i I. j i i ii
. ..
i i i i i i i j i i i i i i i i j i i i ..a
o W« MM
U>vO r l N P M n ' O r l (M f^ v O C - C - r H ^ t - f l - t O ®
CO f l ia oo -p
SS^{^Ta^0^ 2 2 ^ * f' l^‘ ts' c^ a)C2 <2 Q c2 0 ^ - l r^ ,-,,-, Vf' v0 o cm in in 1o
CM CM CM CM CM fc\ m in in m m mi'O sO MO >0 sO sO nO nO iO nO \£> t" C~t - C 3O HHI O
o ■H xi
p XI
CO
®
(A
- 358 -
I t i i i i i i i i
3
i i i i i
i + i c^ ■p
d
o
i i i i♦ i + Cf\ u
i i i i i i + i NO
§
# ii i i i# i -J3
o
S i i + i i <*>
8 k
•H -P i i
fn *H CM i # i + CM
d) TJ
-P T-j
•H n i i ii i i + +
F-i 0] I nO
O >
T3 -p + i I i+ i
0) a V\
Si 3
o
■p 01
•nH i i CM
a a -p
O M i i * NO
-p
o « ®
NO i i i i ♦ i ♦ I in
® W ®
23
o
® •p
5 i i i + ♦ i li i
a 5 P*
I8
iiS "tQ
i + i i* i ■sf
a ®
a
(0
®
© Eh
o §>iH
•H CD
CO t «
CM i i
i i i i +
+
ii i*
i
CO
•8 +>
-H
£ sS I+ CM
o
o ® ®
to •3 5 cn
o ra a
« o i i I i i i
®
Pi 3
-P Cm
i i i I i ii i i i cn
a O rl
® i i i i t-
■P <n
® iH
ai a i i i i i I i i ♦ i i
« s 3
5 ® i i I i i I# i i i rl
rH
Cn 5 i ii i + i '•*
I
h
o
o
i i •
ii
i
iI i i
i i
i
On
ON
10 i ill
Ih
O
h**
•• *
+ _ O
d ON .|H CO
c O C*\ El<
6T •H i—I -p
•rl W •• -P ©
© ® u «
oa i i j t u t i i j t j § i o -p
cm cm cMoi&'f’^rr\!r\irNtr\!]r\irN si ©
CO P3
- 359 -
r»> II I I 1111*1 to
IT
(continued)
Oi I I II I II I I I I I CM
i
—i
*C II II ♦ III vO
O I I I I I + I + st
IT
I I II I II I II I 3
I I I ♦ I I I I I * I I to
S|
III I till I I + I
3
M3 I III I I I 1 * * 1 I CO
I I 4* I III vO
I III I I I I o
ro
si I I III I II + I O'
o III I I I I I I I I CM
rl
a
m I I I I v\
m
cd
o
I I I * I II + I m
O'
CO
I I I I * I I + I ir\
co I + CM
CO
c* + I II * I co
C*'
VO I I I I I II * I
CO
IT
C*'
I I III* I * I
si III I III O'
CO
<o
CO
I I St
CM
co
I I I I I I I I
3
H
co I I * I I ♦ I v\
O till I I I * I rl
TABLE LXI (continued)
CO H
O' I I I I I I I I * I II * I
CM 3
at
at e
u
H|W
•• * I
o
o
?
fl O' Tl
o CO &4
03
g pf +J
•• -P CO
• rl vO M3 Cf\ CTcvO CO H CM -t ITCsS H CM CO iftvO H CM Cf\M) > > rl st a| « ® h
03 J I J I I J I I I J J I I I I III I I I I II I I I J 3 B O
b « t ' 0 ' 0 ' 6 « 3 K M n > M o 6 6 g o 6 H H H H «^s6 6 cm jcmt* o 3 .a V
cm cm cm <McMcoootrvv\crcir\m'5'0'5\0'0\OvOsO'0'C'Si>£^r'0 o e-t ca Bj
CQt-30~J-a-<j->JQsOsO'OsO'0'OsO'0'0'OsUxUlVJxVnvn\jjVjjfO!'jjofOV)
TABLE LXI (concluded)

ff
£ 6T B - T T f ? f T I T Yr ? ? ? ? ? ? T3 73Y, Y' Y ? f f ? V ? CO
sr 4 a> ® M > ^ H ' J > J O ' W M H O " J i U ) M H O ' W ^ - N H 0»t» O'y'Vft O'O' P*
«4 •• w 4 4 4 +
frir
CD
S
P
**) \jJ o
C
fto p- 'P p
o 1 ^
3 ®
CD
ax
VJl 4 1 4 1 1 4 1 I 1 1 I
M 4 ax
4 1 1 1 1 Ox
ax
O' 1 1 14 1 1 I | O'
ax
-J 1 4 I I 1 1 1 1 I -J
sO ax
1 1 1 4 1 1 II 1 1 I ax
ax
-J 14 II 1 4 1 1 1 1 1 -£>
sO 1 1 1 4 1 1 1 1 I I I 8
H O'
M 1 1 1 14 1 1 1 1 1 I I I H
O'
Case No.
Ox 14 II 4 14 1 1 II M
JO 1 1 O'
uJ
ca 1 1 II 1 1 1 1 £
H 4 1 1 4 1 8
ax 4 1 1 1 1 1 1 1 1 1 8
O' I 4 1
O'
I I 1 I I *3
* - 4 1 4 1 1 1 1 I 8
O'
H 4 1 1 sO
t-3
O
^ ^ s O ^ V > J M->3<IHa»sOso!^vO-^t''Ul£>X C»f c o S o ' ®
#_1
- 361 -
TABLE LXIII
Initial, Retest, and Combined Scores of Each Case in the

Below 70 Hatched Criterion Group
Case Number
1 2 3 4 5 6 7 8 9 10 11 12 13 14
Initial Score 8 4 0 4 6 4 5 1 2 6 3 8 l 5
Retest Score 6 4 1 7 6 3 2 }. ♦1 0 4 7 l 3
Combined Score H 8 1 11 12 7 7 2 1 6 7 15 2 8
Case Number
1$ 16 17 18 19 20 21 22 23 24 25 26 27 28
Initial Score 1 7 6 4 0 0 2 4 ♦2 2 6 11 2 8
Retest Score ♦1 7 ') 2 3 ♦3 0 2 tS ? 1 2 2 6
Combined Score 0 14 11 6 3 *3 2 6 +4 5 7 13 4 14
Case Number
29 30 31 3? 33 34 35 36 37 38 39 40 41 42
Initial Score 4 4 0 2 4 3 +2 3 6 3 4 7 1 5
Retest Score 3 +2 +1 4 2 l 4 2 % 2 2 1 2
Combined Score 7 2 *1 6 6 4 0 7 8 5 6 9 2 7
(continued)
- 362 -
TABLE LXIII (concluded)
Case Number
43 44 45 46 47 48 47 50 51 5? 5? 54 55 56
Initial Score 4 5 5 4 0 0 4 4 ♦4 ♦1 3 4 ♦1 9
Retest Score 0 l 1 3 0 0 2 3 +2 0 8 2 tl 5
Combined Score 4 6 6 7 0 0 6 7 ♦6 ♦l 11 6 +2 14
Q_a.se Number
17 58 59 60 61 62 6? 94 65 66 67 9e 69
Initial Score 0 1 2 4 3 3 1 0 0 6 +2 4 2
Retest Score 2 1 4 4 Q 3 9 l 5 0 3 ,3.___
Combined Score 2 2 4 8 7 3 4 0 1 1 1 + 2 7 5
- 363 -
TABLE LXIV

Neurotic Matched Criterion Croup
Case Number
1 2 3 4 9 6 7 8 9 3-0 11 12 1? 14
Initial Score 5 10 12 7 10 7 9 12 11 5 A 8 16 10
Retest Score 4 9 9 A lj. 1? 7 13 7 3 ? A 8 A
Combined Score 9 19 21 11 21 19 16 25 18 8 6 12 24 14
_Caae_Number
15 16 17 18 19 20 21 22 23 24 25 26 27 28
Initial Score 8 8 7 7 8 6 7 5 10 5 2 4- 7 12
Retest Score 7 5 6 2 5 6 2 2 6 6 ? 3 7 11
Combined Score 15 13 13 9 13 12 9 8 16 n 5 7 U 23
Case Number
29 30 31 22 22 34 25 26 37 28 29 40 41 42
Initial Score 10 8 6 7 7 8 11 7 8 5 5 4 6 14
Retest Score 12 11 5 10 4 9 5 7 2 2 5 5 5 12
Combined Score 22 19 11 17 11 17 16 14 11 7 10 9 11 26
(continued)
- 36A -
TABLE LXIV (concluded)
_________________ Case Number

A3 AA__A5_A6 A? A8 A9 50 51 52 .53 5A 55 56
Initial Score 13 16 2 9 7 7 12 5 3 S 10 10 2 10
Betest Score 9 9 6 $ 10 S 11 A 6 12 8 5 2 6
Combined Score 22 25 S 17 17 15 23 9 9 20 18 15 A 16
Case Number
57 5? 59 60 61 6?? 63 64 65 66 67 68 69 _ .
Initial Score 9 11 8 12 10 13 5 9 10 7 7 11 2
Retest Score 7 9 7 9 11 f 2 9 1 8 6 A 1
Combined Score 16 20 15 21 21 18 7 17 11 15 13 15 3
»
- 365 -
TABLE LXVII
Initial, Retest, Recall, and Total Scores of Each Case in the

Below 70 Matched Criterion Group
Case Number __
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Initial Score 8 4 0 4 6 4 5 1 2 6 3 8 l 5 1
Retest Score 6 4 1 7 6 3 2 1 *1 0 4 7 1 3 ♦1
Recall Score 0 0 tl +2 0 1 0 *2 0 *1 1 +2 0 0 tl
Total Score 14 8 0 9 12 8 7 0 1 5 8 13 2 8 ♦1
Case Number
16 17 18 19 20 21 22 2? 24 2? 26 27 29 29 JO
Initial Score 7 4 0 0
6 2 A ♦2 2 6 11 2 8 4 4
Retest Score 7
5 2 3 ♦3 0 2 ♦2 3 1 2 2 6 3 ♦2
Recall Score *1 +1 *2 ♦1 0 1 ♦2 +1 tl ♦1 ♦1 ♦1 1 ♦3 0
Total Score 13 10 4 2 ♦3 3 A +5 4 6 12 3 15 4 2
Case Number
31 32 33 34 35 39 37 39 39 40 41 42 43 44 45
Initial Score 0 2 3 ♦2
4 3 6 3 4 7 1 5 4 5 5
Retest Score +1 4 1 2 4
2 2 2 2 2 1 2 0 1 1
Recall Score 0 0 fl *2 ♦2 t3 1 tl f2 0 ♦1 +1 +1 0 0
Total Score ♦1 6 5 2 ♦2 4 9 4 4 9 1 6 3 6 6
(continued)
- 366 -
TABLE LXVII (concluded)
Case Number
to
46 47 48 49 50 51 52 5? 54 55 56 57 59
Initial Score 4 0 0 4 4 +4 ♦1 3 4 ♦1 9 0 1 2
Retest Score 3 0 0 2 3 *2 0 8 2 *1 5 2 1 2
Recall Score *2 ♦1 ♦1 +? ♦l +2 ♦1 0 ♦1 1 0 0 *2 .♦3.
Total Score 5 *1 *1 3 6 +8 *2 11 5 ♦1 14 2 0 1
Case Number
60 62 63 64 65 66 67 68 69
Initial Score 4 3 3 1 0 0 6 +2 4 2
Retest Score 4 4 0 3 0 1 5 0 3 3
Recall Score *2 +1 +1 P ♦2 +2 +1 *2 0 ♦1
Total Score 6 6 2 4 ♦2 ♦1 10 ♦4 7 4
- 367 -
TABLE LXVIII
Initial, Retest, Recall, and Total Scores of Each Case in the

Neurotic Matched Criterion Group
Case Number
1 2 3 4 5 6 7 8 9 10 11 12 1? 14
Initial Score 5 10 12 7 10 7 9 12 11 5 4 8 16 10
Retest Score 4 9 9 4 11 12 7 13 7 3 2 4 8 4
Recall Score 0 0 0 *1 1 1 1 1 *1 ♦J- 0 +1 1 1
Total Score 9 19 21 10 22 20 17 26 17 7 6 11 25 15
Case Number
19 16 17 118 19 20 22 2? 24 25 26 27 28
Initial Score 8 8 7 7 8 6 7 5 10 5 2 4 7 12
Retest Score 7 5 6 2 5 6 2 3 6 6 3 3 7 11
Recall Score 1 0 ♦1 ♦1 1 ♦1 0 ♦1 1 2 ♦1 1 1 t2
Total. Score 16 13 12 8 14 11 9 7 17 13 4 8 15 21
Case Number
29 90 91 92 99 94 99 96 97 38 29 40 41 42
Initial Score 10 8 6 7 7 8 11 7 8 5 5 4 6 14
Retest Score 12 11 5 10 4 9 5 7 3 2 5 5 5 12
Recall Score 0 0 tl tl 0 1 tl 1 2 0 ♦1 0 +1 0
Total Score 22 19 10 16 11 18 15 15 13 7 9 9 10 26
(continued)
*
- 368 -
TABLE LXVIII (concluded)
Case Number
4? 44 45 46 47 IB £9 50 51 52 5? 54 55 56
Initial Score 13 16 2 9 7 7 12 5 3 8 10 10 2 10
Retest Score 9 9 6 8 10 8 11 £ 6 12 8 5 2 6
Recall Score 0 0 0 0 tl +1 +2 *2 0 3, 2 +1 tl 0
Total Score 22 25 8 17 16 U 21 7 9 12 20 14 3 16
Case Number
57 58 59 60 $1 62 63 64 65 66 67 68 69
Initial Score 9 11 8 12 10 13 5 9 10 7 7 11 2
Retest Score 7 9 7 9 11 5 2 8 l 8 6 4 1
Recall Score 2 ♦1 0 0 0 0 *2 1 0 0 1 *1 +1
Total Score 18 19 15 21 21 18 5 18 11 15 14 14 2
- 369 -
TABLE LXXIII

Below 70 Unmatched Criterion Group
Case Number
70 71 72 73 1L 75 76 77 78 79 80 81 82 83
Initial Score 4 6 1 3 2 3 4 6 3 0 5 6 2 6
Retest Score A 3 A 3 0 3 2 2 2 A 1 3 0 1
Combined Score 8 9 5 6 2 6 6 8 5 4 6 9 2 7
Case Number___________________
8A 85 86 87 88 89 90 91 92 93 9L 95 96 97
Initial Score2 2 5 6 5 7 6 5 3 3 7 7 3 1
Retest Score 2 0 6 *1 *1 5 0 2 1 2 4 2 3 4
Combined Score 4 2 1 1 5 412 6 7 4 5 1 1 9 6 5
Case Number
95 99 100 101 102 103 10A 105 106 107 108 109 110 111
Initial Score 8 1 *1 4 3 3 2 0 5 7 2 2 1 7
Retest Score 0 , ?_3_ 2 6 2 5_ 0 5 9 tl 4 6..J5
Combined Score 8 3 2 6 9 5 7 0 10 16 1 6 7 1 2
(continued)
- 370 -
TABLE LXXIII (concluded)
Case Humber
112 113 n L 115 116 117 118 119 120 121 122 123 124 125
Initial Score 4 1 2 3 4 3 4 5 8 0 5 +2 3 8
Retest Score 2 0 3 2. S.. 0 1 ___6 1 .5 1 2 0
Combined Score 6 1 5 5 9 3 5 11 16 1 10 *1 5 8
Case Humber
226 127 128 129 130 131 132 133 134 135 136 137 138 139
Initial Score 0 3 8 *1 4 5 1 1 2 6 4 0 2 6 1
Retest Score2 3 6 0 2 2 A 0 0 A 5 A 7 4
Combined Score 2 614 +1 6 7 15 2 6 8 5-6 13 5
Case Mnpihar___________________
140 141 142 143 144 145. 146 147 148 149 150 151 152 153 154 155
Initial Score 0 4 6 2 0 4 5 6 4 2 5 6 2 3 1 2
Retest Score 3 0 7 +2 0 2 3 7 3 1 4 2 +2 0 +2 .3
Combined Score 3 4 12 0 0 6 8 13 7 3 9 8 0 3 +1 5
- 371 -
TABLE LXXIV

Neurotic Unmatched Criterion Group
Case Number
70 71 7? 7? 74 75 76 ...77 78 79 80 81 82 82
Initial Score 4 16 16 4 3 10 11 12 8 8 13 11 11 3
Retest Score ? 11 12 5 1 5. 9. 10 1 7 12 9 3 3
Combined Score 7 27 29 9 4 15 20 22 9 15 26 20 16 8
Case Number
84 95 $6 97 89 89 90. 91 92 9? 94 93 96 97
Initial Score 9 10 6 14 14 4 13 3 13 9 5 7 9 l
Retest Score 9 5 <? 8 10 6 8 5 9 3 ? ? 9 6
Combined Score 18 15 12 22 24 10 21 8 22 12 8 10 18 7
Case Number
98 99 100 101 102 103 10A 105 106 107 108
Initial Score 6 5 7 8 10 8 12 7 10 9 6
Retest Score 4 8 9 8 8 10 10 L 5 6 6
Combined Score 10 13 15 16 18 18 22 11 15 15 12
- 372 -
TABLE LXXVIII

1 Above 70 Criterion Group
Case Number___________________
1 2 3 4 _ 5 6 7 8 9 10 11 12 13 1A
Initial Score 5 2 2 3 0 3 5 0 4 3 4 5 10 2
Retest Score 4 3 0 4 2 5 Z . 2 2 2 6 3 6 2
Combined Score 9 5 2 7 2 8 9 2 6 5 10 8 16 4
■
______________ Case Number____________________
15 16_ 17 18, -JL9- 20 21 22 23 24 25 26 27 28
Initial Score 6 8 3 7 2 2 4 1 3 5 13 8 0 7
Retest Score 6 4 2 2 0 2 1 2 3 8 Z . 1 1 2 3
Combined Score 12 12 5 9 2 4 5 3 6 13 17 19 2 10
29 30 31 32 33 34 31 3$ 37 38 39 40 41 42
Initial Score 0 2 0 6 7 5 5 6 2 3 2 5 4 3
Retest Score 4 +1 +2 4 3 3 3 4 0 2 3 4 4 3
Combined Score 4 1 ♦2 10 10 8 8 10 2 5 5 9 8 6
(continued)
- 373 -
TABLE LXXVIII (concluded)
CaseNumber
44 45 4$ 47 48 49 50 51 52 53 54 55 56
Initial Score 1 2 0 3 3 S 10 5 5 4 5 9 7 ♦1
Retest Score ♦1 5 0 1 ♦? 5 8 *2 1 0 3 7 3 1
Combined Score 0 7 0 4 l 13 18 3 6 4 8 .16 10 0
Case Number
57 5$ 59 60 61 62 62 64 65 66 67 68
Initial Score l 2 0 1 5 0 2 5 3 6 3 4
Retest Score ? 1 0 0 7 1 2 2 1 5 2 4
Combined Score 4 3 0 1 12 1 4 7 4 11 6 8
- 374 -
TABLE LXXIX

2 Above 70 Criterion Group
_______________________ Case Number___________________

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Initial Score 0 4 5 2 5 9 9 10 0 1 5 2 2 6 2
Retest Score 0 0 4 2 4 5 8 4 3 1 3 1 4 3 3
Combined Score 0 4 9 4 9 1 4 17 1 4 3 2 8 3 6 9 5
Case Number ________

16 17 18 19 20 21 22 23 24 25 26 27 28 29
Initial Score 4 3 0 5 2 10 5 4 5 4 *2 3 3 4
Retest Score 2 *1 0 4 3 7 3 5 5 6 2 4 4 4
Combined Score 6 2 0 9 5 17 8 9 10 10 0 7 7 8
_______________________ Case_Number______
-3Q_ 31 32 33 34 35 36 37 38 33 40 41 42 43.
Initial Score7 4 4 8 3 3 0 9 5 8 2 5 1 4
Retest Score 1 7 1 4 3 5 +2 1 1 5 7 1 3 3 5
Combined Score 8 11 5 12 6 8 +2 20 10 15 3 8 4 9
(continued)
- 375 -
TABLE LXXIX (concluded)
Case Number
J*L 45 46 47 48 49 50 51 52 _53 54 55 56 57
Initial Score 1 1 8 2 3 7 5 5 1 4 2 4 4 1
Retest Score 3 2 1Q 1 2 5 4 A 1 4 3 3 +1 3
Combined Score 4 3 18 3 5 12 9 9 2 8 5 7 3 4
Case Number
to
IP
59 6 0 6 1 62
Initial Score 5 3 6 +1 5
Retest Score 4 0 2 0 3
Combined Score 3 8 tl 8
- 376 -
TABLE LXXXII
Scoring of the Initial Test Record of Each Case in the

Below 70 Hatched Criterion Group,
Using Pascal and Suttell's Signs
Case Number
1 2 3 4 5 6
Fieure Sign wt. Sign Wt. Sign wt. Sign wt. Sign wtt Sign wt.
1 7 2 1 3 5 10 7 2
8 3
1 2
2 1 2 7 4 3 3 1 2 1 2
6 3
3 1 3 5 2 5 2 2 3 1 3
5 2 2 3 1 3 2 3
10 2 2 3
4 3 1 1 3 8 4 1 3
11 3 3 1
8 4
5 2 3 5 2 1 3 5 2 2 3 1 3
2 3 5 2 2 3
5 2
6 7 A 3 2 7 4 5 1
5 1
7 5 1 5 2 1 8 5 3 9 3
6 A 5 4 3 3
7 8 7 8
2 3
8 5 2 5 1 2 3 1 8 5 2
6 4 5 1
Conf. 5 2 1 2
2 6 2 4
5 2
Raw Score 45 10 34 53 23 43
Z score 79 41 67 87 55 77
(continued)
- 377 -
TABLE LXXXII (continued)
Case Number
* $ 9 10 J1 12
Fieure Sian Wt. Sian WtT Sian wt. Sian Wt. Sian Wt. Sian Wt.
1 1 2 2 3 2 3 2 3
1 2 7 2
2 1 2 1 2 1 2 1 2
6 3 6 3
3 2 3 2 3 2 3 2 3
4 11 3 1 3 1 3
3 1
5 11 3 1 3 2 3 5 2 2 3
1 3
6 5 1 1 3 7 4 3 2
7 5 2 5' 1 5 1
6 4
8 6 4 6 4
Conf. 5 2 5 2
Raw Score 9 16 20 *2 23 18
Z Score 40 AS 52 44 55 50
(continued)
- 378 -
Case Number
13 H _ 15 _ 16 17 18
Figure Sign Wt. Sien wt. Sign Wt. Sign wt. Sign Wt. Sign Wt.
1 1 2 1 2 1 2 2 3 1 2
3 2 2 3
2 1 2 1 2 1 2 1 2 1 2
6 3 6 3
3 2 3 2 3 2 3 1 3
10 2 10 2 5 2
2 3
A 1 3 3 1 1 3
5 5 2 5 2 1 3 2 3 1 3
2 3 2 3 10 2 5 2 5 2
10 2 11 3
1 3
6 1 3 1 3
6 2 7 A
5 1
7 5 6 9 3 5 1 3 3
6 A 5 1
5 3 2 3
8 5 5 10 3 5 1 11 8 5 1 5 1
5 1
Conf. 2 2 A 8 5 2
Raw Score 25 aa 22 21 32 18
Z Score 57 78 5A 53 65 50
(continued)
- 379 -
Case Number
19 20 21 . 22 23 24
Figure Sign Wt. Sign Wt. Sign Wt. Sien wt. Sign Wt. Sign Wt.
1 2 3 1 2
5 2 3 2
2 7 2 1 2 1 2 1 2 6 3 1 2
1 2 8 8
3 1 3 2 3 1 3 3 2
5 2 5 2
4 1 3 1 3 3 1 1 3 1 3
3 1 8 4
4 4
5 1 3 2 3 5 2 3 2 2 3
5 2 5 2
6 1 3 7 4
7 10 8 7 8 1 8 6 4
5 2
6 4
8 2 3 6 4 6 4 7 8
5 1 2 3
10 3
Conf. 1 2
Raw Score 10 26 21 35 28 35
Z Score 41 59 53 68 61 68
(continued)
- 380 -
Case Number
25 26 27 28 _ 29 30
Fieure Sien Wt. Sign Wt. . Sign Wt. Sien Wt. Sien Wt. Sien Wt.
1 2 3 2 3 1 3
7 2 7 2
2 1 2 1 2 6 3
11 3
3 1 3 1 3 1 3
2 3 5 2 5 2
5 2 2 3
4 8 4 3 1 1 3
1 3 4 4
* /
5 2 3 1 3 1 3 9 2
5 2 5 2 5 2
6 5 2 4 8 1 3 1 3
11 3 2 2 3 4
7 5 3 6 4 2 3
10 3 5 1
7 4
8 2 3 6 4 6 4
10 3
Conf. 1 2 2 2
2 2
6 8
Raw Score 49 16 20 22 27 13
Z Score 83 48 52 54 60 45
(continued)
- 381 -
Case Number
31_ 32 33 34 35 36
Figure Sien Wt. Sien Wt. Sien Wt. Sien Wt. Sien Wt, Sien Wt.
1 2 3 1 2 6 8 1 2 3 2 3 2
2 3 2 3
2 1 2 1 2 1 2 1 2
8 8
3 2 3 1 3 1 3 1 3
2 3 2 3 5 2
5 2 5 2
4 3 1 1 3 1 3 3 1 1 3
3 1 3 1 4 4
11 3
5 1 3 2 3 2 3 2 3 2 3
5 2 5 2
11 3
6 5 1 11 3 7 4 1 3
7 5 1 2 3 5 4 5 1 3 3
5 1 6 4
7 8
8 5 1 6 4 5 1
Conf. 7 8
Raw Score 12 36 12 37 26 34
Z Score 44 69 44 70 59 67
(continued)
- 382 -
Case Number
37 .. 38 39 - AO A1 A2
Fieure Sien Wt. Sign Wt. Sign Wt. Sign wt. Sign. Wt. Sign Wt,
1 2 3 2 3 1 2 2 3 1 2
8 3 2 3 3 2
2 1 2 1 2 1 2
6 3
3 2 3 1 3 2 3 2 3
2 3
A 10 2 3 1 3 1 1 3 3 1
1 3 9 8 1 3
5 5 2 5 2 1 3 2 3 2 3 2 3
5 2
6 1 3 1 3
6 A
7 5 1 7 8
8 10 3 2 3 10 3 5 1
10 6
Conf. 5 2
Raw Score 7 18 20 A3 17 17
Z Score 38 50 52 77 A9 A9
(continued)
- 383 -
Case Number
43 . AA A5 46 A7 48
Figure Sien Wt. Sien Wt. Sien Wt. Sien Wt. Sign Wt. Sign. Wt.
1 2 3 1 2 1 2
2 11 3 6 3 1 2
6 3
3 ' 1 3 1 3 2 3 2 * 3
2 3 2 3
5 2 5 2
A 1 3 3 1 3 1 3 1 8 4
5 1 3 1 3 5 2 5 2
2 3 5 2 2 3
6 5 3 7 A 1 3 5 1 1 3
7 9 3 5 2 9 3 9 3
5 1
8 5 2
Conf. 1 2 1 2 4 8
5 2 2 2
Raw Score H 31 17 17 16 17
Z Score 46 64 49 49 53 54
(continued)
- 384 -
Case Number
.42 _ 59 51 52 53 54
Fieure Sian Wtr ■Sign Wt. Sien wt. Sien W^t Sien Wt. Sien Wt.
1 1 2 1 2 8 3 2 3 2 3
3 2
S 3
2 1 2 1 2 1 2 1 2 1 2 7 4
2 3
3 1 3 1 3 1 3 2 3 1 3
3 2 5 2 2 3 5 2
5 2
4 3 1 8 4 1 3
1 3 3 1
5 1 3 5 2 1 3 1 3 1 3 2 3
5 2 5 2 2 3
2 3 2 3 5 2
6 3
6 2 2 7 4 2 2 1 3
1 3
7 5 1 2 3 2 3 10 8
7 8 5 1
2 3
8 5 2 3 3 2 3 2 3 2 3 6 4
8 2 5 1
6 4
Conf. 7 8 2 4
Raw Score 31 21 31 39 38 26
Z Score 70 59 70 70 77 64
(continued)
- 385 -
Case Number
55 56 __57 . 58 . 59 60
Fieure Siffa wt. Sien Wt. Sien Wt. Sien Wt, Sien Wt. Sien Wt.
1 2 3 8 3 1 2 2 3 2 3 8 6
7 2 3 2
2 1 2 1 2 1 2 6 3 1 2
8 8 1 2
3 3
3 2 3 4 8 2 3 1 3 2 3 1 3
5 2 5 2
2 3
4 1 3 8 4 3 1
1 3
5 5 2 5 2 3 2 2 3 5 2 5 2
2 3 A 8 5 2 6 3
6 1 3 7 4 7 4 5 1
7 5 1 2 3 5 1 5 2 9 3
5 2 6 4
8 5 1 5 2 5 1 2 3 2 3
2 3 6 4 6 4
5 1
Conf. 2 2
Raw Score 20 34 32 39 27 18
Z Score 58 73 71 79 65 55
(continued)
- 386 -
Case Number
61 62 63 hL 65 66
Fieure Sien Wt. Sien Wt. Sien Wt. Sien_ Wt. Sien Wt. Sien Wt.
1 2 3 1 2 2 3 5 6
*
2 3
2 1 2 1 2 1 2 6 3 6 3 1 2
1 2 1 2
3 2 3 2 3 2 3 1 3 2 3
7 8 5 2
A 1 3 1 3 1 3 3 1 1 3
8 A
5 2 3 1 3 5 2 5 2
1 3 2 3
6 1 3 3 2 7 A 11 3 A 8
A 8
7 6 A 5 1 6 A 2 3
5 1 6 A
8 6 A 7 A 2 3
5 1
Conf. 1 2 5 2 5 2
5 2
Raw Score 32 23 38 11 22 32
Z Score 71 61 77 A8 60 71
(continued)
- 387 -
TABLE LXXXII (concluded)
Case Number
67 68 69
Fieure Sien Wt. Sien Wt. Sien Wt.
1 2 3
2 1 2 1 2
3 3
3 2 3 2 3
5 2
10 2
4 8 4 1 3
5 1 3 1 3
2 3 5 2
6 7 4 12 8
7 4
7 6 4 6 4
5 1
8 2 3 6 4
6 4 1 8
Conf. 1 2 2 2 5 2
5 2
Raw Score 23 49 18
Z Score 61 90 55
- 388 -
TABLE LXXXIII
Scorlpg of the Initial Test Record of Each Case In the Neurotic Hatched
Criterion Group, Using Pascal and Suttell's Signs
Case Number
1 2 3 4 5 6
Fieure Sien Wt. Sien wt. Sien Wt. Sien Wtr Sien Wt. Sien Wt.
1 1 2 1 2 1 2 2 3 2 3
2 3 2 3 9 8
7 2
2 1 2 1 2 1 2 1 2 6 3 1 2
2 8 6 3
3 2 3 2 3 1 3 2 3
2 3 12 8
10 2
4 3 1 1 3 3 1 1 .3 3 1
11 3 1 3 11 3
8 4
5 2 3 1 3 4 8 5 2
2 3 5 2
5 2
6 1 3 1 3 1 3
2 2 2 2
4 8
7 5 2 5 2 7 8 2 3 7 8
9 3 6 4 6 4
8 5 1 10 3 1 8 6 4 11 8
5 1 2 3 2 3
7 8
Conf. 2 2 5 2
Raw Score 19 4 54 48 34 61
Z Score 51 35 88 82 67 96
(continued)
- 389 -
TABLE LXXXIII (continued)
Case Number
7 8 9 10 11 12
Figure Sign Wt. Sign Sign Wt. .Sign Wt. Sign WtT Sign Wt.
1 2 3 1 2 7 2 1 2 1 2
7 2 3 2
9 8
2 1 2 1 2 1 2 1 2
2 3 12 8
3 1 3 2 3 10 2 10 2 12 8
5 2 2 3
2 3
4 1 3 3 1 3 1 3 1
11 3 1 3
5 2 3 2 3 5 2 1 3 5 2 5 2
1 3 10 2 2 3 1 3
6 3 2 3 2 1 3 3 2
5 1
6 8
7 5 1 2 3 5 1 5 2 2 3
2 3 7 4 9 3
1 8
8 5 1 5 4 5 1 5 3
7 8 9 2
Conf. 5 2 6 8 7 8
Raw Score 34 48 15 16 53 23
Z Score 67 82 47 48 87 55
(continued)
- 390 -
Case Number
1? 14 15 16 17 18
Fieure Signt wt. Sign Wt*. Sign Wt. Sign. Wt, Sign Wt. Sien Wt.
1 7 2 1 2 2. 3
3 2 3 A 8 2 3 3 2 1 3 2 3
10 2 5 2
A 1 3 3 1 1 3
5 5 2 2 3 1 3 2 3
2 3 2 3
10 2
6 1 3 3 2 3 2
5 2
11 3
7 3 3 5 1 5 1 9 3 5 1 9 3
6 A 5 3
8 5 1 2 3 5 3 9 3
7 8 5 2
Conf. 3 3 5 2 5 2
Raw Score 23 10 27 15 13 29
Z Score 55 41 60 47 45 62
(continued)
- 391 -
________________________ Case Number________________________

19________ 20________ 21________22________ 22________ 24
Figure Sign Wt. Sign Wt. Sign Wt. Sign Wt. Sign Wt. Sign Wt.
77 -r’J
1 2 3 2 3 3 2 1 2 2 3
7 2
2 1 2 9 2 1 2 7 4
3 1 3 9 2 2 3 2 3 1 3
5 2
4 3 1 3 1 3 1 1 3 1 3
1 3
5 5 2 2 3 2 3 3 2 1 3 2 3
1 3 5 2 5 2 12 8
9 2
6 1 3 3 2 13 8
7 5 2 8 2 5 2 2 3
8 5 1 2 3 6 4 2 3
Conf. 3 3
Raw Score 27 19 11 14 19 34
Z Score 60 51 43 46 51 67
(continued)
- 392 -
Case Number
25 26 27 28 29 30 _
Fieure Sien Wt. _ Sign Wt. Sien Wt,. Sien Wt. Sien Wt. Sien Wt.
>
1 1 3 2 3 2 3 7 2
2 3
2 2 2 1 2 1 3 1 3 6 3
9 2 6 3 1 2
3 4 8 2 3 2 3 2 3 1 3
9 2 7 8 4 8
5 2
4 1 3 8 4 11 3
10 2 3 1
1 3
5 4 8 5 2 5 2 2 3 1 3 2 3
5 2 2 3 5 2
9 2 10 2
6 9 2 1 3 1 3 5 1 7 4
7 8 2 5 2 5 1 1 8
2 3
7 4
11 8
8 8 2 3 3 2 3
6 4
Conf. 1 2 7 8 1 2
5 2
Eav Score 45 24 7 17 62 34
Z Score 79 56 38 49 97 67
(continued)
- 393 -
Case Number
31 32. 33 34 35 36
_Jigure Sien Wt. Sien Wt. Sien Wt,. Sien Wt. Sign wt. Sign Wt.
1 2 3 7 2 1 3 1 3
9 8 3 2
2 1 3 1 2 1 2 1 2
6 3 3 3 6 3
3 3 2 2 3 10 2 1 3
2 3
5 2
7 8
4 1 3 10 2 1 3 3 1
3 1 3 1
11 3
5 3 2 5 3 5 2 2 3 1 3 1 3
5 2 * 10 2 5 2 5 2 3 2
5 2
6 3
6 1 3 9 2 1 3 3 2
7 8 2 5 1 5 1
10 8 9 3
8 8 2 2 3
Conf. 5 2 1 2 5 2
6 8
Raw Score 17 15 47 19 11 42
Z Score -47 81 51 43 75
(continued)
- 394 -
TABLE L x m i l (continued)
Case Nuuber
37 -3.8 _ _39 40 41 42
Fieure _ Sien Wt. Sien Wt. Sien Wt. Sien wt. Sien Wt. Sien wtt
1 3 2 1 2 2 3 2 3 2 3 1 2
5 2 3 2 3 2
2 13 8 1 2 1 2 1 2
6 3
3 3 2 2 3 2 3 2 3
5 2
1 3
4 3 1 1 3 3 1 3 1
4 4 11 3
3 1
5 3 2 5 2 2 3 2 3 1 3
7 2 5 2 5 2 5 2
2 3
6 7 4 4 8 3 2 1 3 3 2
2 2
7 5 1 3 3 3 3 3 3 3 3 5 1
8 2 3 2 3 2 3 2 3 6 4
5 1
t
Conf. 1 2 1 2
2 4
Raw Scere 27 30 14 36 20 25
Z Score 60 63 46 69 52 57
(continued)
- 395 -
Case Number
A3 AA A5 A6 A7 4$
Fieure Sien Wt. Sien Wt, Sien Wt,. Sien W*t Sien Wt. Sien wtT
1 2 3 2 3 2 3 7 2 2 3
7 2 5 A
7 2
2 1 2 1 3 1 2
6 3
3 10 2 1 3 2 3 10 2 1 3
2 3 5 2 5 2
1 3 2 3
A 1 3 3 1 1 3 3 1 3 1 3 1
3 1 1 3 1 3 1 3
5 2 3 2 3 9 2 10 2 2 3
10 2 5 2 2 3 5 2
1 3 5 2 2 3
6 1 3 1 3 11 3 3 2
7 A A 8
7 9 3 1 8 2 3
5 1
6 A
8 2 3 10 3 10 3 6 A 3 3
9 2 5 1
6 A
Conf. 5 2 A 8
2 A
Raw Score A3 A2 A5 27 10 20
Z Score 77 75 79 60 A7 58
(continued)
- 396 -
Case Number
49 ... . 50 51 52 53 54
Fieure Sien Wt. Sien Wt. Sien Wt.. Sien Wt. Sien Wt. Sien Wt.
1 1 2 5 18 2 3
2 3 2 3
2 1 2 7 12 1 2
3 2 3 1 3 1 3 2 3 1 3 2 3
5 2 2 3 5 2
2 3 5 2 10 2
11 3
4 3 1 1 3 11 3 11 3 1 3
3 1
4 4
8 4
5 5 2 1 3 2 3 5 2 5 2 2 3
2 3 5 2
6 1 3 3 7 4 11 3 1 3
3 2 5 1
7 2 3 9 6 2 8
5 1
8 2 3 2 3 10 6
Conf. 1 2 1 2 2 2
5 2 2 2 3 3
2 2 7 8
Raw Score 17 34 62 22 35 27
Z Score 54 73 104 60 74 65
(continued)
- 397 -
Case Number
_ j55 56 57 58 59 60
Fieure Sien Wt. Sien Wt. Sien Wt. Sien Wt. Sien Wt. Sien Wtf
1 3 2 2 3 1 2 3 2 4 8 1 2
2 5 5 1 2 6 3 i 2
3 3 6 3
3 2 3 1 3 12 8 1 3 4 8
4 8 5 2
7 8 10 2
4 3 1 12 8 1 3 1 3 6 3
3 1
5 2 3 5 2 1 3 2 3 9 2 1 3
1 3 4 8 5 2 5 2
2 3
6 3 4 3 2 5 1 1 3
5 1
7 5 2 10 8 5 1 10 8
5 1
8 7 8 5 3 6 4
5 1
Conf. 2 4 5 2
Raw Score 30 42 37 22 31 26
Z Score 69 82 76 60 70 64
(continued)
- 398 -
Case Number
61 62 63 64 65 66
Figure Sign Wt. Sign Wt. Sign Wt. Sign Wt. Sign wt. Sign Wt.
1 1 2 1 2 8 3 1 2 1 2
3 2 1 2 3 2
2 6 3 1 2 1 2 1 2 7 2
6 3
3 A 8 2 3 2 3 1 3
5 2
4 8 4 8 A 1 3 3 1 1 3
1 3
5 1 3 5 2 1 3
6 3 2 3
5 2
2 3
6 7 4 3 2 1 3 7 4 3 2 3 2
7 4 1 3
7 6 4 5 1 9 3 1 8 5 1 1 8
5 2
8 6 4 5 2 8 2 10 3 5 1
Conf. 7 8 3 3 2 2
Raw Score 31 42 23 25 11 31
Z Score 70 82 61 63 48 70
(continued)
- 399 -
TABLE LXXXIII (concluded
Case Number
67 68 69
d
Fieure wt. Sien Wt. Sien Wt.

S
1 2 2 1 2
3 2
2 1 2
6 3
3 3 2 2 3 2 3
10 2
4 3 1 3 1 3 1
6 3
8 4
5 1 3 1 3
3 2 5 2
6 1 3 3 2
5 1
7 9 3
5 1
6 4
8 5 2 10 3 2 3
Conf. 5 2 1 2
2 2 7 8
4 8
Raw Score 47 31 7
Z Score 87 70 43
- 400 -
I I I I * I * O'
I * I * -*
3
I + * 0
1
fl I I * CO
© 81 1O
© I I * I CO
a
•5*
01
$$ I + + Nt
33
ch I I ♦ st
CO *H
21
<5 O
ra o I I I * IT\
o ©
^ > p I I I I I + * I ■'*
Eh © ra
tc ©
© p
I I + I vO
35S
I I I *
3o «I! I-g
I I I I I O'
EH
® E-
3 I I ♦ I I I co
w
CO © a I I sO
© P ©
n o<H
©H
a I II III ♦ C~
© © P
u33
.gCOI-l
0
H * III II * sO
o> I I I I I ♦ I * I I CO
d jj
° 'oa o
£ CO * I I I I I I I* O'
Th « fei
O -rl -P * o
O © iH
© © Tl
vO I I I I * c-
-P © ©
2 3 s-
in I I I I I I I I *
H a a
*1 •st I I + CO
3 S’
M C5 co I I * CO
©
3 01 I I I I * in
Vi
o
t© I ♦ I I
3 to
©
©
a
2 H|CU o
o
(3 a
O<*jfx,
cr
co
I5*? TfTfl cr\vr\\OHCM_.

1 I. I JLJt
.. . , . ©_ «
I I 1 4
_a
•rl rH
M •» P
8 0
3
CO
ooiC'-a'O'Ocîrvint'- ■ir 2
>
CMCMOIC\lC\icOfc\UMn»r»in>rt
—
\ in2'O2 v3
o '23> 3
'S > > > P o hffl 3
- 401 -
O' I I♦ I I
CO
00 I I I♦ I sO
CO
I> I I
co
NO I I I II NO
ON
«N
CO I I o\
CC I I o>
u
co I I I co
1
I CO
CM I I I I in
© CO
©
©
a rl I I + I I co
CO
o I ♦ I CM
CO
O' I II I II I
CM
to
CM I I II II
l> I I ♦ III III C-
(concluded)
CM
vO
CM
III III ♦ I «*
CO
<D
H|N i IO
TABLE LXXXVII
CO
2 <*5
•qH o\
f—ic*_ 3
♦ ♦ ♦ ♦ CO •• «P
nO nO in U>nO I AnO H CM <r\\0 C*- C*- rH fl> ® U 3
•r
w CMNc\i(^(Mr'M*Mr\v\u\UMr\'«'«\0'S'3vO'OvOiOw'C> O E-I to
- 402 -
3 I I I 00
cr CM
CM
* COV
c\
Adjustment
CM I I♦ I I I ♦ £>
O I I + *r»
CM
O' I I I I
I I I+
the Initial Test Record of Each Case in the TotalMale TAC Unsatisfactory
00
e- I I I+
Group on the Basis of the Selected Signs Shownto Have Consistent
H 00
vO I I I 114* +11 O'

H
I I I+ II O'
I I I I I I
Validity on the Initial Test and the Retest
a
h CO I I t*\
H
I Ol
I I I ♦ vO
SB
®
U I I ♦ I VN
id
u
o I II I I♦ +
rH
O' I I I ♦ I +
00 I I I O'
r- I* I II II II I I
3
vO I+ II I
a
IT ♦ I c*-
I I I 00
CO I I + till 11+ I
3
CM ♦ II II I I I O'
I I I 'O
«a
Scoring of
9
Him §
o
+ §0 CO
S3 O' -rl
O rl
•H H
'OvOlAinvO«t|OOHCM4rir\'$HCMrr\ ir\\Q H CM cr\sO C*-H «3| ® ® £
C0 CMCMCMCMCMOr\rr\if\ir\i/'vu-\ir\vO'0'5'5'JDvO'0'0'0'S'OCit'-C'-P~OE-'C/3 ■A
- 403 -
II I I♦ I I o
I I II I I I * + to
I I II 1 1 + I O'
I I + I «r>
O' I I I+ I II I I+ O'
co
to I I I I ♦ + I
co •>*
C" I I II I I I I I♦ O'
CO
sD I I I
CO II I to
it II II I III ♦ I
CO
3
I I I I I I I + I O'
a CO II II I * I * II
« CO vr>
o
CM I I I I + I
CO I I I O'
H
CO I I ♦ I + + I CO
O I ♦ + I I I >*
CO
CM I I I I I I I + I O'
$ I I ♦ ♦ I CM
CM
I ♦ I I II I I I♦
3
(concluded)
v£> I I ♦ ♦ I I II I I I I I
CM 3
»T I + I II + II 'O
CM
a ©
U
o
TABLE LXXXVIII
r-4N O
*i CO
d O ' «H
O CO ft, r -i
■H r—t
'© 'O v\ irt'O <t CO H CM -$ H CM CO UO'P H CM COM? C~-r?
ca ••4*
«3j '■<■ © 3
S I I I I I^ © ®
8 thn -p
•H O CM C" O' O' c _!> C O__»_ O_M O l >.C _^ t O c i c i Q Q Q i H r - I H H jA'i {i li «0 V\ O *9 5
CO C M C M C M C M C M r O C O MIT
O»UV\
O UIO
T»U O J^n sO vD \0 w W vO vO vD vO \o M
- 404
rH
+
Group
♦ I V\
of the Retest Record of Each Case in the Total Male TAC Satisfactory Adjustment
on the Basis of the Selected Signs Shown to Have Consistent Validity
CO
♦ I
♦ I «rv
C^
l vr>
* I cr\
+ CM
O'
«o + I C"
V0
3
CM
cn
rl
U\
m
Scoring
e
s
O
If?lO
*'-&
H o
CO
.33* -p
vO NO »A «A i-l CM <Vk\0 t>
«n •• 4» m
S ffl 9 fct
T i i j I i II i 9©
c^min>>„ (OQQQ
----- w m ir\\oo vO
cMc\jcMCMc\icf\c»\»rvtf\»r\»r\uv
- 405 -
O' I ♦ ♦ I CM
co
♦ I I I I x*
c- I ♦ I CM
co
NO
C"
I I I I US
IT
C»' I I I ♦ CM
I I ♦ I CM
CO I I I OS
CO
CM rl
CCi
® I I CM
02 CO
0)
O
o I H
CO
O'
Oi I I I I I I NO
«0
CM
I I I I X*
CM
I I I I I I III
NO I II I I I ♦ X*
CM
UN I III I I I I I tx
CM
(concluded)
I I ♦ II I ♦ CM
<0
b
hn
••+00& o
o
riO'-H CO
TABLE LXXXII
ocnpn
•HiH _ -p
CQ••-P 02
a> ® u ®
fi •p
£
n n t n t i h f m n i i snO'nO x
CMCMCMCMCMC'SC'SUSUSUSUSUS'S'S^'S h p - ? ' * O■SJ1.8
nO nO nO^ ' ^ C x H CO £
- 406 -
I I I I I
I I ♦ ♦ I
I ♦
I
c5
CM cn -p
a
o
-P
I I I I ♦ ♦ I u\
o
S
$a O' I I I I I II ♦ CO
3
I I I I I I I * ON
'O
^5
I I I I * NO
S3
0 > nQ I I I I I I I * * <0
C0
Ch|-P
CO 0
•rt ® IT I I I I I I I * * CO
•P -P
a) 01
QQ •H
I I I I I I I I II *
s S
O -P a
cj a
1 ® -5 r-(
® >0 (®4
*3
a
33 ®
S 3
la I * I I I •st
H I I I I I I I +
at
■P gl
a NO
a H
O O as
I I I II I I I I III*
CO -P 3
I a
®
I •H H
H II III I* I I I I III * I
3
•rf CO at
® T3
•H
-P
I I I III I I CO
a © -H
as-P d
O O M I III I 1 * 1 * 1 1 1 1 1 1 1 1 1 I I
®
J3 r-l ®
o © jg
II II II II I * I
a -© *d 3
I I I I I ♦ * cn
o I I I -a-
o a
© T<
Pi a
« I I I I I ♦ I I I I I II I I * I* I
-p m H
a
© o
S3 I ♦ I II I I II I ON
“© O
a
I I I I I I NO
5
tH
o a ©
bo Ih
HUM O
.1 O
h fl CT'rl to
o O coh
o
co
a
6 3
■H
CO tcMCMCMCMCM<nc<Mrv<r»uMr>vs'®'«'®'«'®'>0'^>'0'0'*>'i)P-rxCxtxoE-«
n m x m s f o m i l Z U U U U ^ £
- 4.07 -
i i i i i i ♦ i i * i * i
5| ' ' • + i i i i 1 i i i i ♦ + i i j-j
!^ • l I l l i i * i |>
l ♦ + l
O' li l* l I I l I I I l i
CP
$ I I l I* + I ♦ I rr\
& I I * 1 1 1 1 1 1 1 1 1 I r-j
vO
CP I I I I I lA
IT I I I I I * I I to
CP
u
I I I I I I * * ^
®|?3| II I I + I I * -«*
oa
C0
O CM I I♦ I I ♦ II ♦ I’ '4’
CP
CP + * * I H
*
o ♦ I II *1 cp
CP
O' II I I I I I * t>
CM
to • I I I I * I I * 1A
CM
♦ I I I I I I I •I I ♦ I I g
| | * * 1 1 1 1 1 1 1 1 1 «q
JQ I I I ♦ I I I I I * | |
^ 1 I I I I I * I I C*-
TABLE XC (concluded)
CO
© Q
O crjfe
* *H H
£ v9 ' ? ‘
C'l
O ' 9 <? C9 r^ rHÎ' H CM CA 1AM? H CM CPvO t> C" rH ~4 <4 -•* ® ® tn £
tr
CO CMcSi
cm cm cm cm S\rr\
MCMCMCMCACA 0>m5 m9
5 m5 vcj
i a »AIAiA I A ’
^ vt!
5'©'0'0 $4 Ci O -S
0 s0 m ?C'-C*C'*C'-OE“ 5
ICQ ■#
(2
- 408 -
.TABLE XCII
Initial, Retest, and Combined Initial and Retest Scores for Each Case
in the Total Male TAC Satisfactory Adjustment Group
(Matched and Unmatched)
Case Number
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Initial Score 7 5 3 3 11 7 0 9 8 6 7 6 3 9 6
Retest Score 5 1 3 2 11 6 l 7 3 2 ? 5 7 7 4
Combined Score 12 6 6 5 22 13 l 16 11 8 10 11 10 16 10
Case Number
16 17 1$ 19 20 21 22 23 24 25 26 27 28 29 30
Initial Score 4 5 4 4 2 8 3 0 4 9 4 7 7 7 2
Retest Score 5 3 8 4 5 4 ♦1 4 2 7 4 7 4 6 1
Combined Score 9 8 12 8 7 12 2 4 6 16 8 14 11 13 3
Case Number
31 32 33 34 35 36 37 38 39
Initial Score 3 5 3 3 3 6 3 6 8
Retest Score 2 1 3 2 2 5 2 4 2
Combined Score 5 6 6 5 5 1 1 5 10 10
- 409 - tIBHARJ «
XOltt OUltBBSn.
TOlVEHsnr HEIftfr^
TABLE XCIII
Initial, Retest, and Combined Initial and Retest Scores for Each
Case in the Total Male TAC Unsatisfactory Adjustment Group
(Matched and Unmatched)
Case Number
1 2 3 4 5 6 7 . 8 9 10 11 12 13 14 1?
Initial Score 6 9 11 8 7 12 16 9 7 10 5 6 3 11 9
Retest Score 6 9 1? 4 3 10 16 8 12 10 6 4 1 11 8
Combined Score 12 18 24 12 10 22 32 17 19 20 11 10 4 22 17
Case Number
16 17 18 19 20 21 22 23 24 25 2$ 27 28 29 30
Initial Score 9 8 8 7 5 7 3 2 3 6 11 10 2 9 4
Retest Score 8 6 9 8 5 3 1 7. 7. 8 8 10 5 7 3
Combined Score 17 14 17 15 10 10 4 9 15 14 19 20 7 16 7
Case Number
33- 32 33 34 35 36 37 38 39 40 41 42 43
Initial Score 3 9 5 9 10 8 9 4 9 5 9 8 9
Retest Score +3L 4 4 4 8 5 11 3 31- 0 7 H- 7.
Combined Score 2 13 9 13 18 13 20 7 20 5 16 19 16

Xerox University Microfilms: 300 North Zeeb Road Ann Arbor, Michigan 48106

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Xerox University Microfilms: 300 North Zeeb Road Ann Arbor, Michigan 48106

Uploaded by

Copyright:

Available Formats

INFORMATION TO USERS

T h e following explanation of techniques is provided to help you understand

5. PLEASE NOTE: Some pages may have indistinct print. Filmed as

Xerox University Microfilms

l.°Adjustment (Psychology) 2.Gestalt

Xerox University Microfilms, Ann Arbor, Michigan 48106

TH IS DISSERTATION HAS BEEN M IC R O FILM ED E XA C TLY AS RECEIVED.

A QUANTIFICATION, STANDARDIZATION, AND VALIDATION OF THE

BENDER-GESTALT TEST ON AN ADULT POPULATION IN TERMS OF ITS ABILITY TO

DIFFERENTIATE NORMAL AND PSYCHONEUROTIC LEVELS OF ADJUSTMENT

A dissertation in the department of Psychology submitted

The writer wishes to express his gratitude to Professor

Thomas N. Jenkins, whose incisive suggestions and perspicacious

observations contributed innumerable insights to the present in­

vestigation and whose kindly encouragement gave impetus to the ex­

ploration of promising leads. The deepest appreciation is expressed

veterans and non-veteran clients. Heartfelt thanks are likewise prof­

fered to Dr. Bernard Locke, chief clinical psychologist of the Mental

Hygiene Unit of the Brooklyn Hegional Office of the Veterans Adminis­

tration, and to Dr. Richard H. Paynter, chief clinical psychologist of

Administration, whose generous cooperation made the testing of neurotic

out-patients possible. Finally, the writer wishes to acknowledge his

standing served as a never-ending source of inspiration and encourage­

ment and to whom this study is affectionately dedicated.

II. STATEMENT OF THE PROBLEM............................... 38

A. The Test Materials ...... 65

B. Administration of the Test....... 67

C . The Subj acts .......... 75

IV. THE SCORING SYSTEM.......................... 83

A. Global Signs on the InitialTest and the Retest....... 121

B. Individual Figure Signs on the Initial Test and

C. The Recall Test...... 184

D. Scoring of the Test Records of the Matched Criterion

E. Scoring of the Test Records of the Unmatched Criterion

F. Effect of Education, Age, and Intelligence upon Test

G. Scoring of the Initial Test Records of the Matched

H. Results of the Cross-Validation Study............ 252

VI. . DISCUSSION....... 286

VII. SUMMARY AND CONCLUSIONS................................. 326

APPENDICES............................................... ..... 341

I. AGE COMPOSITION OF THE CRITERION GROUPS.................. 79

II. EDUCATIONAL STATUS OF THE CRITERION GROUPS............... 80

III. MARITAL STATUS OF THE CRITERION GROUPS................... 81

IV. MILITARY SERVICE AFFILIATION OF THE CRITERION GROUPS...... 81

VI. MEANS AND STANDARD DEVIATIONS OF THE LENGTHS OF EACH OF THE

VII. MEANS AND STANDARD DEVIATIONS OF THE LENGTHS OF EACH OF THE

VIII. COMPARISON OF THE ACTUAL LENGTHS OF THE STIMULUS FIGURES

IX. MEANS AND STANDARD DEVIATIONS OF THE HEIGHTS OF EACH OF

X. MEANS AND STANDARD DEVIATIONS OF THE HEIGHTS OF EACH OF

XI. COMPARISON OF THE ACTUAL HEIGHTS OF THE STIMULUS FIGURES

XII. PERCENTAGE DISTRIBUTION OF SHORT FIGURES (REPRODUCTIONS AT

XIII. PERCENTAGE DISTRIBUTION OF SHORT FIGURES (REPRODUCTIONS AT

XIV. PERCENTAGE DISTRIBUTION OF SHORT FIGURES (REPRODUCTIONS AT

XV. PERCENTAGE DISTRIBUTIONS OF LONG FIGURES (REPRODUCTIONS AT

XVI. PERCENTAGE DISTRIBUTION OF FLAT FIGURES (REPRODUCTIONS AT

XVII. PERCENTAGE DISTRIBUTION OF FLAT FIGURES (REPRODUCTIONS AT

XVIII. PERCENTAGE DISTRIBUTION OF TALL FIGURES (REPRODUCTIONS AT

XVIX. PERCENTAGE DISTRIBUTION OF TALL FIGURES (REPRODUCTIONS AT

XX. PERCENTAGE DISTRIBUTION OF THE NUMBER OF SEQUENTIAL CONNEC­

XXI. PERCENTAGE DISTRIBUTION OF THE NUMBER OF SEQUENTIAL CONNEC­

XXII. MEDIANS OF THE TIME REQUIRED TO REPRODUCE EACH FIGURE ON

XXIII. MEDIANS OF THE TIME REQUIRED TO REPRODUCE EACH FIGURE ON

XXIV. MEDIANS OF THE TOTAL TIME REQUIRED TO REPRODUCE ALL THE

observations contributed innumerable insights to the present in

vestigation and whose kindly encouragement gave impetus to the ex

veterans and non-veteran clients. Heartfelt thanks are likewise prof

Hygiene Unit of the Brooklyn Hegional Office of the Veterans Adminis

standing served as a never-ending source of inspiration and encourage

XX. PERCENTAGE DISTRIBUTION OF THE NUMBER OF SEQUENTIAL CONNEC

XXI. PERCENTAGE DISTRIBUTION OF THE NUMBER OF SEQUENTIAL CONNEC

XXX-A. ANALYSIS OF THE SCORING CATEGORIES INTO INDIVIDUAL FIGURE SCOR

XXZIII. PERCENTAGE INCIDENCE AND CRITICAL RATIOS OF THE DIFFER

XLI. FREQUENCY OF RECALL AND ORDER OF RECALL, EXPRESSED AS PER

XLII. FREQUENCY OF RECALL AND ORDER OF RECALL, EXPRESSED AS PER

LIII. PERCENTAGE INCIDENCE OF THE TENTATIVE SIGNS ON THE IMMEDI

LIV. PERCENTAGE INCIDENCE OF THE TENTATIVE SIGNS ON THE IMMEDI

LV. PERCENTAGE INCIDENCE OF THE TENTATIVE SIGNS ON THE IMMEDI

LXV. DISTRIBUTION OF COMBINED INITIAL AND RETEST SCORES, CUMULA

LXIX. DISTRIBUTION OF TOTAL SCORES (INITIAL PLUS RETEST PLUS RE

LXX. DISTRIBUTION OF COMBINED INITIAL AND RECALL SCORES, CUMULA

LXXI. DISTRIBUTION OF COMBINED RETEST AND RECALL SCORES, CUMULA

LXXV. DISTRIBUTION OF COMBINED INITIAL AND RETEST SCORES, CUMULA

LXXXIV. DISTRIBUTION OF THE HAW SCORES, MEANS, AND STANDARD DEVIA