CONSTRUCTIONAL APRAXIA IN PATIENTS WITH DISCRETE

MISSILE WOUNDS OF THE BRAIN

F. William Black and Richard L. Strub

(Louisiana State University Medical Center)

Many studies have appeared in recent years comparing the incidence

and severity of constructional apraxia associated with lesions in the two
hemispheres. Several reports have also commented on the importance of
rostral or caudal placement of lesions associated with constructional apraxia.
Both of these aspects of localization can best be studied in patients whose
lesions are confined to a single quadrant of the brain (Benson and Barton,
1970). Accordingly, we have studied constructional praxis in a group of
missile wound patients from the Vietnam Conflict who had neurosurgically
verified lesions limited to a single brain quadrant.
Differential levels of performance on test of constructional praxis by
patients with right and left hemisphere brain lesions has been noted for
some time (Critchley, 1953; Kleist, 1912). Studies of both the incidence
of constructional apraxia (Arrigoni and De Renzi, 1964; Piercy, Hecaen
and Ajuriaguerra, 1960) and the degree of severity of such defects (Arrigoni
and De Renzi, 1964; Costa and Vaughan, 1962; Piercy and Smyth, 1962)
suggest a higher incidence and degree of severity in patients with right
hemisphere lesions. An association between lesions of the nondominant
hemisphere and impaired constructional ability is well documented; a
similar impairment with lesions of the dominant hemisphere has also been
reported (Benton, 1962; De Renzi and Faglioni, 1967; McFie and Zangwill,
1960 ). In general, the literature suggests that constructional apraxia may
result from lesions in any quadrant of the brain (Benson and Barton, 1970;
Smith, 1966 ).
The validity of comparing right and left hemisphere lesion patients has
been strongly questioned because of the potential for sample bias; right
hemisphere lesion patients tend to have more extensive lesions at the
time of investigation (Smith, 1966; Warrington, James and Kinsbourne,
1966; Wolff, 1962). Patients with left hemisphere lesions tend to come to
the attention of professionals earlier because of the readily apparent language
deficits which follow such lesions. In a study of subjects with well matched
 Constructional apraxia 213

right and left hemisphere lesions, Arrigoni ana De Renzi (1964) failed to
demonstrate consistent significant differences in the severity of drawing
disability in the two samples. Conversely, Piercy, Hecaen and Ajuriaguerra
( 1960) reported qualitative differences in the constructional performance of
right and left hemisphere lesion patients. The differences in the severity
and frequency of constructional impairment, however, could not be accounted
for by the masking effects of paresis and dysphasia nor by unilateral impercep-
tion. By controlling the verbal demands of the tests used, De Renzi and
F aglioni ( 196 7) failed to demonstrate significant differences in either the
incidence or severity of constructional apraxia in patients with dominant
and nondominant hemisphere lesions. These and other similar reports suggest
that matching right and left hemisphere lesion samples on the criteria of
duration, nature (vascular, neoplastic, traumatic, etc.) and size of lesions
should be a necessary criterion prior to reporting on any differential
performance in lateralized samples. Unfortunately, with most clinical studies
utilizing samples of conveniences, such matching procedures have not
generally been carried out.
It is generally conceded that constructional apraxia is most often associated
with retro-Rolandic damage, specifically the parietal-occipital areas (Benson
and Barton, 1970; Critchley, 1953 ). Reitan (1964 ), however, found no
consistent significant differences in performance on a wide range of tasks
(Halstead Battery) between frontal and posterior lesion patients. Benson
and Barton ( 1970) studied disturbances in constructional ability in patients
with lesions localized to one of the four quadrants of the cortex (determined
by brain scan). This localizing design allows the investigation of both
laterality and caudality effects. The authors reported that lesions in the
right hemisphere produced more consistent disturbances in visual-spatial
performance than did similar left hemisphere lesions. There also was a greater
consistency of impaired performance with posterior lesions (in either hemi-
sphere) than with lesions localized in the anterior quadrants. Because of the
inherent problems in obtaining testable subjects meeting these stringent
localizing criteria from routine hospital populations, similar localization studies
have not thusfar been carried out. Such a research design would seem to
be the most efficient method of studying the differential effects of focal brain
lesions and deserves further use.
Although there have been some objections, it has been suggested that
penetrating missile wounds of the brain, because of their relatively discrete
nature in the absence of cerebral concussion, provide a unique opportunity
to study the effects of focal lesions (Simpson and Vega, 1971; Weinstein
and Teuber, 19 57). This is especially true in young previously healthy adults
without vascular or infectious complications of the initial brain injury. The
study of such subjects lessens the contamination of effects among the variables
of age, nature of neurological involvement, and duration of disease.
214 F. W. Black and R. L. Strub

Accordingly, the present study investigated the effects of side (right vs.
left hemisphere) and locus (anterior vs. posterior) of lesion upon constructional
praxis in patients with penetrating missile wounds of the brain.

MATERIAL AND METHOD

Subjects
Subjects were patient referrals from the Neurology and Neurosurgery Services,
Fitzsimons Army Medical Center. All subjects suffered 'war-related unilateral brain
lesions penetrating the skull and dura, secondary to high velocity small sized
shrapnel wounds. Locus and limitation of lesions were documented by neuro-
surgical exploration. For the purpose of this study, the brain was divided into
quadrants with the Fissure of Rolando designated as the anterior-posterior line
of demarcation. This schema is consistent with that previously used (Benson
and Barton, 1970). From the population of 100 missile-wound patients with
available neuropsychological test results and complete medical records, 60 subjects
had discrete quadrant lesions. Fifteen could be assigned to each of the following
groups: right anterior, right posterior, left anterior, and left posterior brain
quadrants. The possibility of lesion in other than the primary quadrant was ruled
out by skull x-ray, EEG, and other neurodiagnostic procedure when deemed
necessary. All subjects were able to validly perform on the neuropsychological'
test battery and no potential subject required exclusion because of aphasia or
dementia. To preclude an obvious contamination of results, no patient with
clinically significant vascular or infectious complication of the initial brain lesion
or wounding of either upper extremity were included.
The diameter of entrance wounds in the samples was approximately equal,
with a range from .6 to 4.5 em and a median of 2 em. From the medical history
and longitudinal treatment records accompanying the subject, all subjects were
healthy prior to injury without history of neurological or psychiatric problems.
All subjects were right-handed on brief clinical testing and from self reports; no
information as to family history of handedness was obtained.
The age of the subjects ranged from 18 to 29 years, with a total sample
mean of 21.7 years (S.D. = 1.1). Age variations in the four samples were
insignificant, ranging from 21.1 in the right anterior sample to 22.5 in the left
posterior sample. The mean educational level for the total sample was 11.9 years
(S.D. = 1.6), with a sample range from 11.5 years in the right anterior sample
to 12.1 years in the right posterior sample. The mean WAIS Full Scale for the
total sample was 96.8 (S.D. = 12.3 ).

Tests
Subjects were routinely evaluated from 3 to 16 weeks post-m]ury with a
total sample mean of 7.4 weeks (S.D. = 2.2). There were no appreciable differences
in the mean time since injury in the four samples. All subjects were ambulatory
and no bedside testing w:>.s necessary. The Wechsler Adult Intelligence Scale and
Bender Gestalt Test were administered in the context of a more complete
neuropsychological evaluation as a part of the routine rehabilitation program.
Constructional performance measures selected for analysis were the W AIS Block
Designs subtest, W AIS Object Assembly subtest, and Bender Gestalt error scores,
 Constructional apraxia 215

(Koppitz, 1964). These measures were selected to assess a variety of constructional

tasks and as representative of those tests commonly used in clinical practice. The
W AIS Block Designs and Object Assembly subtests were administered and scored
in the standard manner and require no further description. Bender Gestalt
protocols were scored for rotation, perseveration, distortion, and integration errors
according to the Koppitz ( 1964) system, with total errors used as the criterion.

Statistics
A computer generated 2 X 2 (anterior- posterior locus versus right-left
lateralization) analysis of variance design was used to provide comparisons of
mean scores in the four quadrant samples. The Scheffe test for all possible
comparisons was used for specific intersample comparisons.

RESULTS

Means and standard deviations for all variables in the four samples are
presented in Table I.

TABLE I

Means and Standard Deviations for All Samples

WAIS WAIS
Block Designs Object Assembly Bender Gestalt
Mean S.D. Mean S.D. Mean S.D.

Left anterior 12.53 2.50 11.73 2.22 .53 1.30

Right anterior 11.20 2.27 10.73 3.45 1.33 1.54
Left posterior 9.73 2.99 8.47 3.04 1.47 1.77
Right posterior 7.40 2.06 7.27 3.31 3.40 2.20

W AIS Block Designs Subtest

The distribution of mean scores for this variable ranged from the
highest in the left anterior sample to the lowest in the right posterior sample.
The effect of lesion locus was significant favoring the anterior lesion samples
(F = 8.19, df 1, 56, p < .01) as was the side of lesion effect (F = 8.19
df 1, 56, p < .001 ). The interaction effect of locus and side of lesion failec
to reach significance (F = .61 ). The Scheffe Test indicated that the Block
Design performance of the right posterior sample was significantly les~
adequate than that of all other samples; while that of the left posterior
216 F. W. Black and R. L. Strub

sample was significantly less adequate than that of the left anterior sample.
All other comparisons were nonsignificant.

W AIS Object Assembly Subtest

The highest mean score was obtained by the left anterior sample and
the lowest by the right posterior sample. The analysis of variance indicated
a significant locus of lesion effect favoring the anterior lesion samples (F =
18.36, df 1, 56, p < .001 ). The effect for side of lesion was not significant
(F = 1.96) nor was the interaction effect (F = .02). The Scheffe Test
indicated significantly inferior performance by the right posterior sample
when compared with the left and right anterior sample; while the left
posterior sample performed significantly less adequately than the left anterior
sample. All other intersample comparisons were nonsignificant.

Bender Gestalt error scores

The lowest mean error score was in the left anterior sample and the highest
in the right posterior sample. There was a significant effect for locus of
lesion favoring the anterior lesion samples (F = 11.22, df 1, 56, p < .001)
and for side of lesion favoring the left hemisphere lesion samples (F = 9 .32,
df 1, 56, p < .01). The effect between locus and side of lesion was not
significant (F = 1.60). The Scheffe Test indicated that the performance of
the right posterior sample was significantly less adequate than that of both
the left and right anterior samples; while all other intersample comparisons
were not significant.

Incidence of constructional apraxia

W AlS Black Designs and Object Assembly subtest scale scores below
8, and more than 2 Bender Gestalt errors were operationally defined as
suggesting an impairment of constructional praxis. W AIS scale scores of 7
and under were chosen as criteria measures of constructional apraxia, as
scale scores of 8 and above are generally considered to represent norma·
performance (Wechsler, 1958). Although errorless performance on the Bender
Gestalt is expected by age 12 (Koppitz, 1964 ), an error score of 2 wa~
selected to allow for minor random errors due to haste or carelessness. Thf
percentages of subjects in the four samples demonstrating constructiona
apraxia for these measures appear in Table II.
 Constructional apraxia 217

TABLE II

Percentages of Subjects Demonstrating Constructional Apraxia On Selected Variables

Block Object Bender

Sample Designs Assembly Gestalt

Left anterior 0 7 7
Right anterior 7 27 13
Left posterior 20 38 27
Right posterior 67 53 53

DISCUSSION

This study was designed to investigate the effects of well-documented

brain lesions on a variety of commonly used tests of constructional ability.
The methodology employed was similar to that of Benson and Barton (1970)
in that subjects studied had relatively discrete lesions localized in and limited
to one of the four quadrants of the brain. Frequently used clinically oriented
constructional tasks were employed to maximize clinical utility of the results.
The results of this investigation clearly document the presence of differen-
tial impairment of constructional ability for all measures used, based upon the
locus of lesion (anterior versus posterior locus). Highly significant locus of
lesion effects were obtained for each criterion variable used, with the mean
performance of subjects with posterior lesions being uniformly more impaired
than that of similar subjects with anterior lesions. These results are in
general agreement with similar findings in previous studies (Battersby, Krieger
and Bender, 1955; Benson and Barton, 1970; Teuber, 1964) and suggest
that lesions of the frontal lobes generally result in a less significant impairment
of constructional praxis than do more posterior lesions. Conversely, Newcombe
(1969), in a study of subjects with longstanding penetrating missile wounds,
was unable to demonstrate consistent differences in performance on tests of
visual pattern identification and spatial orientation by anterior and posterior
lesion samples. She did, however, report significant interhemispheric
differences. In explaining our variance with Newcombe's study, it is important
to note that in the current study virtually all frontal lesions were well
anterior, with relatively few lesions in the peri-Rolandic area.
Statistically significant differences in performance on the W AIS Block
Designs subtest and Bender Gestalt error scores do suggest a significant
interhemispheric asymmetry of function with more impaired performance by
right hemisphere lesions subjects. No appreciable hemispheric difference in
performance was demonstrated for the W AIS Object Assembly variable.
Although performance on two of three criterion measures did demonstrate
218 F. W. Hlack and K. L. Strub

a clear laterality effect, these interhemispheric effect did not approach the
magnitude of the locus of lesion effect; anterior lesion subjects consistently
performed better than similar posterior lesion subjects.
These results suggest a stronger effect for the locus of lesion (anterior
versus posterior) than for side of lesion (right versus left) and differ from
some previous reports. A number of these investigations have indicated a
strong and consistent laterality effect on constructional tests (Arrigoni and
De Renzi, 1964; Newcombe, 1969; Reitan, 1955; Rubino, 1970; Simpson
and Vega, 1971)~ Other research, however, has reported that mean group
differences on non-verbal measures by subjects with unilateral lesions tend
to be negligible (Benton, 1962; De Renzi and Faglioni, 1967; Heilbrun, 1956;
Smith, 1966 ). The previously mentioned difficulties in adequately matching
samples with unilateral lesions is undoubtedly of some major significance in
explaining these apparent differences in obtained results. The more closely
samples are matched for lesion size, duration and type, the less apparent
are interhemispheric differences on constructional measures. Utilizing a similar
lesion localization schema, Benson and Barton ( 1970) reported significant
differences for right and left hemisphere lesion subjects on measures of
reaction time and the token pattern test, but obtained no significant
hemispheric performance differences on tests of drawing from memory and
copy, puzzle construction, stick pattern reversal, and template matching.
Both the current and most previous results suggest a moderate asymmetry
of function on tests involving constructional ability according to laterality
of lesion, but a substantially more significant and consistent effect of the
caudality dimension.
For all criterion n;.easures, both the patterns of absolute mean scores
and the incidence of constructional apraxia as operationally defined varied
as expected. The universal range from least to most impaired samples was
left anterior, right anterior, left posterior, and right posterior. Using the
Scheffe test for all possible comparisons, the mean performance of the
right posterior sample was consistently more impaired than that of both
anterior samples for each measure. For the W AIS Block Designs subtest,
the performance of the right posterior sample was also significantly more
impaired than that of the left posterior sample. The mean performance of
the left posterior sample was significantly more impaired than that of the
left anterior sample for three of four variables studied. No other intersample
comparisons showed significant differences. Despite the generally greater
impairment of performance in the right hemisphere lesion sample as a whole,
it should be noted that not all subjects with such lesions showed evidence
of constructional apraxia. Depending upon the criterion measure, from 47
to 3 3% of right posterior lesion subjects did not demonstrate operationally
defined constructional apraxia. Similarly, from 100 to 62% of left hemisphere
lesion subjects failed to show evidence of constructional apraxia. Only one
 Constructional apraxia 219

subject with a left anterior lesion showed constructional apraxia, and then
::m only two of three measures.
The following factors are important in interpreting the high incidence
Jf intact constructional praxis in many subjects: relatively discrete lesions
Jf short duration, the subjects' general good health prior to injury, immediate
neurosurgical care after injury, and the lack of significant cognitive impairment
in the majority of subjects. In contrast to other studies reporting significant
impairments in subjects of middle or old age with longstanding disease and
·nore generalized cognitive deficits, these subjects, despite their recent brain
trauma, were relatively intact both intellectually and neurologically.
These results are consistent with previous suggestions that the localization
:)f brain lesions by the use of tests of constructional praxis is highly dependent
upon the nature of the constructional test employed. The W AIS Block
Designs and Object Assembly subtests and the Bender Gestalt error score
were relatively sensitive to specifically localized lesions, primarily those in
the right posterior quadrant, for the majority of subjects.

SUMMARY

Sixty patients with missile wounds confined to one of the four quadrants of
the brain were investigated. All patients had neurosurgical verification of the
limits of their lesions. The incidence and severity of constructional apraxia was
studied using the W AIS Block Designs and Object Assembly subtests, and the
Bender Gestalt Test. A uniformly significant caudality effect was obtained, with
more posteriorly localized lesions resulting in more severe constructional apraxia.
A significant laterality effect was obtained on two of three criterion measures
with uniformly inferior performance by patients with right hemisphere lesions.
The magnitude of the laterality effect, however, was less than that of the
:audality effect for all criterion variables. The degree of severity of constructional
apraxia in patients with right posterior lesions was uniformly greater than that of
patients with other quadrant loci. The incidence of constructional apraxia in the
four quadrants varied as expected with the left anterior lesion sample showing
very little evidence of constructional apraxia, while the right posterior sample
showed a high incidence of such deficits. The absolute incidence of significant
:onstructional apraxia in all samples was surprisingly low. This finding might be
partially accounted for by the age and general good health of the subjects studied,
the relative absence of general cognitive impairment in the majority of subjects,
and the discrete nature of the lesions.

Acknowledgments. The authors wish to thank D. Frank Benson, M.D.,

Director of the Neurobehavioral Unit, Boston Veterans Administration Hospital,
for his constructive comments on preliminary versions of this report.
220 F. W. Htack and K L. Strub

REFERENCES

ARRIGONI, G., and DE RENZI, E. (1964) Constructional apraxia and hemispheric locus of lesion,
"Cortex," 1, 170-197.
BATTERSBY, W., KRIEGER, H., and BENDER, M. (1955) Visual and tactile discriminative learning
in patients with cerebral tumors, "Am. J. Psycho!.," 68, 562-574.
BENSON, D., and BARTON M. (1970) Disturbances in constructional ability, "Cortex," 6, 19-46.
BENTON, A. ( 1962) Clinical symptomatology of right and left hemisphere lesions, in Inter-
hemispheric Relations and Cerebral Dominance, ed. by V.B. Mountcastle, John Hopkins
Press, Baltimore.
- , and FoGEL, M. (1962) Three dimensional construction praxis, "Arch. Neural.," 7, 347-354.
CosTA, L., and VAUGHAN, H. (1962) Performance of patients with lateralized cerebral lesion.
I: Verbal and perceptual tests, "J. Nerv. Ment. Dis.," 134, 162-168.
CRITCHLEY, M. (1953) The Parietal Lobes, Arnold, London.
DE RENZI, E., and FAGLIONI, P. (1967) The relationship between visuo-spatial impairment and
constructional apraxia, "Cortex," 3, 327-342.
HEILBRUN, A. (1956) Psychological test performance as a function of lateral localization of
cerebral lesions, "J. Comp. Physiol. Psycho!.," 49, 10-14.
KLEIST, K. (1912) Der Gang und der gegenwartige Stand der Apraxieforschung, "Ergbn. Neuro.
Psychiat.," 1, 342-452.
KoPPITZ, E. (1964) The Bender Gestalt Test for Young Children, Grune and Stratton, New
York.
McFm, J., and ZANGWILL, 0. (1960) Visual-constructive disabilities associated with lesions of
the left cerebral hemisphere, "Brain," 83, 243-260.
NEWCOMBE, F. (1969) Missile Wounds of the Brain, Oxford University Press, London.
PIERCY, M., HE.cAEN, H., and AJURIAGUERRA, J. (1960) Constructional apraxia associated with
unilateral cerebral lesions; left and right cases compared, "Brain," 83, 225-242.
- , and SMYTH, V. (1962) Right hemisphere dominance for certain nonverbal intellectual skills,
"Brain," 85, 775-790.
REITAN, R. (1955) Certain differential effects of left and right cerebral lesions in human adults,
48, 474-477. .
- (1964) Psychological deficits resulting from cerebral lesions in man, in The Frontal Granular
Cortex and Behavior, ed. by J. M. Warren and K. Akert, McGraw-Hill, New York.
RuBINO, C. (1970) Hemispheric lateralization of visual perception, "Cortex," 6, 102-120.
SIMPSON, C., and VEGA, A. (1971) Unilateral brain damage and patterns of age-corrected W AIS
subtest scores, "J. Clin. Psycho!.," 27, 204-208.
SMITH, A. (1966) Certain hypothesized hemispheric differences in language and visual functions
in human adults, "Cortex," 2, 109-126.
TEUBER, H.L. (1964) The riddle of frontal lobe functions in man, in The Frontal Granular
Cortex and Behavior, ed. by I. Warren, and K. Akert, McGraw-Hill, New York.
WARRINGTON, E., }AMES, F., and KINSBOURNE, M. (1966) Drawing disability in relation to
laterality of cerebral lesion, "Brain," 89, 53-82.
WECHSLER, D. (1958) The Measurement and Appraisal of Adult Intelligence (4th Edition), Wil-
liams and Wilkins, Baltimore.
WEINSTEIN, S., and TEUBER, H.L. (1957) Effects of penetrating brain injury on intelligence
scores, "Science," 125, 1036-1037.
WoLFF, H. (1962) Discussion of Teuber's paper, in Interhemispheric Relation and Cerebral
Dominance, ed. by V.B. Mountcastle, Johns Hopkins Press, Baltimore.

F. William Black, Ph. D., Assistant Professor of Neurology, and Richard L. Strub, M. D., Assistant Professor of
Neurology, Neurobehavior Section, Department of Neurology, LSU Medical Center, 1542 Tulane Avenue, New
Orleans, Louisiana 70112.