Psychological Medicine, 1971, I, 292 29S

Developmental dyscalculia: a brief

report on four cases
P. D. SLADE AND G. F. M. RUSSELL 1
From the Institute of Psychiatry, University of London

SYNOPSIS Four patients with 'developmental dyscalculia' are described who were initially identified
on the basis of clinical tests, reports of longstanding and severe difficulties in arithmetic, and psycho-
logical test results. Descriptive psychological testing indicated (1) that the arithmetical deficits could
not be accounted for on the basis of a generally low level of intellectual functioning; (2) that of the
four basic arithmetical processes that of multiplication was relatively more deficient; (3) that this
relative deficiency in multiplication stemmed from a faulty grasp of basic multiplication tables; that
this was a real difficulty, experienced subjectively by the patient, and one that could not be accounted
for simply by carelessness; and finally, that the degree of difficulty experienced varied with the multi-
plication table involved. Two kinds of attempt were made to improve arithmetical functioning.
The first involved brief remedial coaching programmes directed at the basic multiplication tables:
relatively little success was obtained with this method. The second kind of attempt involved methods
of bypassing the necessity for learning multiplication tables: some success was obtained with these
methods. The discussion concentrates on some of the features which may provide possibly useful cues
for identifying 'developmental dyscalculia': these are classified in terms of: (1) pattern of deficit dis-
played; (2) qualitative features of arithmetical performance; and (3) response to remedial teaching.
Some suggestions are made concerning further lines of investigation.

The term 'acalculia' was first proposed by
Henschen (1919) to describe a condition in
which there was 'a disturbance in calculating,
produced by a focal lesion of the brain'. Since
that time the term has come to be used more
loosely, as a simple descriptive label, and in
many cases without implication that the con-
dition is necessarily acquired or that the dis-
turbance is due to a particular type of cerebral
pathology. In addition, the term itself has been
superseded by that of 'dyscalculia', with the
implication of a disorder rather than of a
complete absence of calculating ability.
Following Henschen, two major distinctions
have been recognized in the use of the term
'dyscalculia'—namely, (1) as a primary or a
secondary disorder, and (2) as a developmental
or an acquired disorder.
With respect to the first distinction, there has
Present address: Academic Department of Psychiatry,
at Royal Free Hospital School of Medicine, Friern Hospital,
NewSouthgate.LondonN.il.
been an unfortunate tendency in the literature to
apply the term 'dyscalculia' to any observed
disorder of arithmetical functioning, whether or
not the disorder itself can be understood as
secondary to other cognitive defects. It is clear
that a wide range of cognitive impairments—for
example, dementia, memory and retention dis-
orders, aphasic disorders, agnosic and apraxic
disorders, etc.—may affect the ability of a
patient to solve arithmetical problems, while the
patient's ability per se may not necessarily be
affected. Grewel (1952) proposed that this type
of condition be termed 'secondary' dyscalculia
as opposed to 'primary' cases in which 'cal-
culation itself is affected'. However, the dis-
tinction is a difficult one to draw, especially in
view of the controversy concerning the associa-
ted correlates of the disorder. Henschen (1919),
in his analyses, concluded that disturbances
of calculation produced by focal lesions were
i ar(T pi v ; n H P npnHpnt nf nnhacin noranhia
^rg?1V ^dependent Ot aphasia agraphia,
alexia, and other kindred disorders; while
 Developmental dyscalculia: a brief report on four cases
Gcrstmunn (1924) described a new syndrome
consisting of linger agnosia, agraphia, right-left
confusion, and acalculia, demonstrating that
these four disorders were associated with one
another with a significantly greater frequency
than expected by chance.
Although Henschen and Gerstmann differed
in terms of their views on the independence of
the disorder, both were concerned with dis-
turbances of calculation which had been acquired
as the result of a cerebral lesion. Many years
passed before Guttmann (1937) drew attention
to dyscalculia as a developmental disorder in
which the ability of the child to acquire calcula-
tion is itself disturbed. Since then a few papers
have appeared in the literature which have been
concerned with developmental cases of dys-
calculia and its relationship to other types of
disorders ( Strauss and Lehtinen, 1947; Kephart,
1960; Kinsbourne and Warrington, 1963b;
Johnson and Myklebust, 1967). So far, the
evidence is insufficient to arrive at any firm
conclusions concerning the correlates of develop-
mental dyscalculia or of what areas of the cortex
may be involved, if any. In addition, there seems
to be a relative dearth of information concerning
the phenomenology of the condition.
The term 'developmental dyscalculia' is
specifically reserved for cases in which dis-
abilities in arithmetic are known to have been
present from an early age; moreover there must
be no suggestion that the subject's arithmetical
abilities were formerly at a higher level and had
shown deterioration; and, finally, there must be
no evidence of gross cerebral disorder such as
TABLE 1
AGE, SEX, PSYCHIATRIC AND NEUROLOGICAL STATUS
C.S.
Age 16 yr. 6 mth.
Sex Male
Psychiatric status Episodic, acute
schizophrenic reaction
Neurological status
1. Left-right discrimination Slight difficulties
2. Finger agnosia Slight
3. Dyslexia Absent
4. Others Myoclonic tics
293
would account for the disability. The aim of this
paper is to describe some preliminary observa-
tions made on four adolescents with severe and
longstanding difficulties in arithmetical calcula-
tion, who may be considered examples of
'developmental dyscalculia'.
THE CASES
Since there is as yet no generally acceptable,
operational definition of dyscalculia, patients
were selected for investigation if they fulfilled
three criteria—namely:
1. If they performed badly on a series of clinical
tests involving simple calculating and money
problems;
2. If they complained of longstanding difficul-
ties in doing arithmetic or their school reports
indicated this to be the case;
3. If they obtained scores on various psycho-
logical tests which indicated severe backwardness
in arithmetical calculation, which was specific
and could not be accounted for by a generally
low level of intellectual functioning.
So far, four patients meeting these criteria
have been studied. The age, sex, psychiatric and
'neurological' status of these patients is shown in
Table 1. (The term 'neurological' is used for the
sake of brevity to describe disorders of higher
cerebral function such as impaired left-right
discrimination, finger agnosia, and dyslexia.)
It can be seen from Table I that the psychiatric
status of each of the cases is different, while
neurological correlates are present to some extent
in three cases and absent in the remaining one.
S.B. D.L. N.R.
15 yr. 16 yr. 6 mth. 14 yr.
Female Female Male
Anorexia nervosa Behaviour disorder Normal
Slight difficulties Correct Slight difficulties
Slight Absent Slight
Absent Absent Present
Nil Nil Nil
294 P. D. Slatle ami G. F. M. Russell

TABLK 2
WKCHSLKR INTKLLKiKNC'l-: SCAI.1-: IOR CIIII.DKI-.N (WISC) RI.SUI.TS

C.S. S.B. D.L. N.R. A vcraK?

Full-scale 1Q 106 96 88 117 101 -75

Verbal IQ 113 90 90 109 1005
Performance 10 97 103 89 124 103-25
Comprehension 16 10 II 12 12-25
Arithmetic 5 4 5 12 6-5
Similarities 17 II II 9 12
Vocabulary 14 8 6 16 II
Digit span 8 9 9 8 8-5
Picture completion 13 II 8 9 10-25
Picture arrangement 7 13 4 18 10 5
Block design 13 10 10 12 11-25
Object assembly 10 7 10 13 10
Coding 5 11 10 15 10-25

Results of intellectual testing were available for
all four cases in the form of the WISC: these are
shown in Table 2.
Table 2 shows that two of the cases were
slightly below average generally, while two were
somewhat above average. Their subtest scores
were generally characterized by inter-individual
variability, although the patients did tend to
obtain lower scores on the arithmetic, and, to
some extent, the digit span subtests.
PSYCHOLOGICAL FINDINGS
1. ATTAINMENT IN THE BASIC ARITHMETICAL
PROCESSES
As a result of clinical observations it was hypo-
thesized that the area of maximum difficulty for
these patients is in the processes of multiplica-
tion and division, in contrast with the processes
of addition and subtraction. In order to test this
hypothesis, the more difficult tests of the Schonell
Diagnostic Arithmetic Tests were administered,
the results of which are presented in Table 3.
These tests were used, despite the fact that they
were standardized more than 20 years ago on
a possibly non-representative sample of Bir-
mingham school-children, as the primary focus
of interest was their relative abilities in the
four basic processes, rather than their absolute
levels.
It can be seen from Table 3 that all four
patients obtained lower arithmetic ages on the
multiplication test compared with the tests of
addition and subtraction. Surprisingly, how-
ever, only one of the four patients obtained an
arithmetic age as low or lower on the test of
division. On the basis of these results it was
tentatively concluded that the area of maximal
defect in these four cases is in the basic process
of multiplication.

TABLE 3
RESULTS ON THE SCHONELL DIAGNOSTIC ARITHMETIC TESTS! (ARITHMETIC AGES)

C.S S.B. D.L. N.R. A verage

Arithmetic process (yr.) (mth.) (yr.) (mlh.) (yr.) (mth.) (yr.) (mth.) (yr.) (mth.)

Graded
Addition 10 5 10 3 10 6 11 6 10 8
Subtraction 9 11 9 7 8 9 10 9 9 9
Multiplication 8 11 8 8 7 II 9 8 8 9-5
Division 10 1 9 8 8 8 9 0 9 4-25
Average 9 10 9 6-5 8 11-5 103
 Developmental dyscalciilia: a brief report on jour cases
2. KNOWI.I'IKii; AND GRASI' <)!•' HASIC MUI.TI-
1'I.ICAIION TARI.I-S
On the basis of further clinical observation it was
hypothesized that the defect in the process of
multiplication may be due to a faulty or non-
existent grasp of the basic multiplication tables.
Two subsidiary hypotheses were derived—
namely:
1. That errors on the multiplication tables are
due to a real, experienced difficulty on the part
of the patient and not simply the result of care-
lessness, and
2. That the degree of difficulty experienced will
vary with the particular multiplication table
involved.
In order to test these three hypotheses a
Multiplication Tables Test was developed. This
consisted of a set of 36 multiplication problems
involving every possible combination of the
numbers 4 to 9—that is, 4 x 4 , 4 x 5 - > - 4 x 9 - »
9 x 4 , 9 x 5 -> 9 x 9 . Each problem was typed
out on a separate index card. Problems were
presented one at a time in random order. In
addition, each problem was timed individually
and the patient asked to rate his/her subjective
confidence in the solution, after each problem,
on a simple five-point scale—that is, very
confident, fairly confident, doubtful, fairly
unconfident, very unconfident. Four measures
were derived from this test:
i. Number of problems correct and number of
problems incorrect.
ii. Total mean time, mean time for problems
correct, and mean time for problems incorrect.
iii. The relationship between objective score
and subjective confidence.
iv. Number correct on each of the six mul-
tiplication tables—that is, 4 x , 5 x , 6 x , 7 x ,
8 x and 9 x.
This test was administered to three of the four
patients (S.B., D.L., and N.R.); the results
showed that:
1. All three patients produced incorrect
answers to approximately one third to one half
of the simple multiplication problems, thereby
providing some confirmation for the major
hypothesis of a faulty grasp of basic multiplica-
tion tables.
2. With respect to the first subsidiary hypo-
thesis all three patients spent approximately
twice as long on problems which they failed
295
compared with problems to which they obtained
correct answers and they all showed a strong
relationship between their objective performance
and their subjective confidence ratings. If their
errors were simply due to carelessness, one
would not expect to find any difference in terms
of time for the two types of problems nor any
systematic relationship with subjective confidence
level.
3. With respect to the second subsidiary
hypothesis, all three patients obtained their
highest score on the 5 x table and their lowest
scores on the 8 x and 9 x tables.
On the basis of these results, it was therefore
tentatively concluded that the basic difficulty in
the process of multiplication stems from a faulty
grasp of basic multiplication tables, that this is a
real difficulty, experienced subjectively by the
patient, and one that cannot be accounted for
simply by carelessness, and that the degree of
difficulty experienced varies with the multiplica-
tion table involved.
3 . ATTEMPTS AT IMPROVING ARITHMETICAL
FUNCTIONING
a. COACHING IN THE BASIC MULTIPLICATION
TABLES
Two methods of achieving this have so far been
attempted with individual patients. The first
method involved a relatively uncontrolled
learning situation with feedback, while the
second was based on controlled learning trials.
1. UNCONTROLLED LEARNING SITUATION WITH
FEEDBACK D.L., whose main difficulties lay
with the 6 x, 7 x, 8 x, and 9 x tables, was
presented with a set of these tables and encour-
aged to practise them whenever she had the time
(plenty of time was available to her to do this).
She was then retested on the Multiplication
Tables Test four days later, and her results dis-
cussed with her. She was then encouraged to
practise them for another two days and then
retested again. Her results showed very little
improvement in terms of number of problems
correct, a large improvement in terms of time
taken and no change in terms of subjective
confidence.
2% P. D. Slade and G. F. M. Russell
2 . CONTROIXHI) I.HARNING TRIALS S . B . , w h o WilS
found to have difficulties with the 4 x, 6 x, 7 x,
8 x , and 9 x tables, was given five learning
sessions, on five consecutive days, each session
being devoted to one particular table. The
procedure employed was based on the normal
paired-associate learning technique. S.B. was
first shown a set of 12 cards on which both
problems (stimuli) and answers (responses) were
presented—for example, 4 x 8 =32. She was then
shown 12 cards on which the problems (stimuli)
alone were presented and was asked to supply
the answer. Learning and test trials were repeated
alternately in this manner until S.B. had pro-
duced five consecutive correct answers to each
problem.
It was decided to standardize the format,
procedure and criteria in this way as, with the
usual form of remedial teaching, it is often
difficult to know afterwards what and how much
has been learned and to what degree. With the
technique adopted here it is possible to state
that each problem has been learnt to at least a
criterion of five consecutive correct responses.
Four days, and again one month, after this
brief remedial coaching programme had been
completed, the Multiplication Tables Test was
repeated. The results indicated very little
improvement in terms of 'no correct' score,
although, again, there was a large improvement
in terms of 'time taken'.
b. ATTEMPTS AT AVOIDING NECESSITY OF LEARN-
ING BASIC MULTIPLICATION TABLES
Two methods of achieving this have so far been
attempted with individual patients. The first
method was simply to provide patients with a set
of multiplication tables and see whether their
ability at multiplication and division problems
was consequently improved. The second method
involved an attempt to teach a patient the
Trachtenberg System of calculating, in which all
the basic arithmetical processes are mediated by
the processes of addition and subtraction.
1. USE OF MULTIPLICATION TABLES IN MULTIPLICA-
TION AND DIVISION The graded multiplication
and graded division tests of the Schonell Diag-
nostic Arithmetic Tests were readministered to
D.L. and N.R. On this occasion, however, they
were provided with a complete set of multiplica-
TABLE4
KliSUl.TS ON THIv SCTIONKLI. MULTIPLICATION AND DIVISION
TESTS, WITH AND WITHOUT MULTIPLICATION TABLES
D.L. N.R.
(yr.) (mill.) (yr.) (mlh.)
Multiplication
Without 1 1! 9 8
With 9 2 9 II
Division
Without 8 8 9 0
With 9 8 II 0
tion tables which they could use in answering
the problems. The results of this procedure,
which are presented in Table 4, indicate that
both patients improved their performance
when they were allowed the use of a set of
multiplication tables.
2. COACHING IN TRACHTENBERG SYSTEM C.S.
was given weekly coaching in the Trachtenberg
System (Cutler and McShane, 1962) by a
qualified teacher accustomed to the coaching of
psychiatric patients. In this system, multiplica-
tion is carried out by means of a complex
sequence of additions, while division is similarly
mediated by a complex sequence of subtractions.
During the first 11 months of coaching most of
the time has been spent on mastering the pro-
cedure involved in multiplication. After four
months coaching C.S. exhibited an increase of 12
months in arithmetic age on the Schonell graded
multiplication test over the previous baseline,
while after 11 months the increase was 15
months. Thus, there was some evidence that
coaching in the Trachtenberg System led to an
improvement in multiplication ability in the
case of C.S., although the increases observed
were not as large as had been hoped for.
CLINICAL OBSERVATIONS
Clinically, a number of typical features in the
procedures adopted by these patients for tackling
arithmetical problems were identified and these
will now be outlined briefly. Firstly, the use of a
'dotting' system was noted quite commonly. This
system was used by patients for all four basic
processes but most particularly for multiplica-
 Developmental dyscalcnlia: a brief report on four cases
tion. An extreme example of this was shown by
one patient, D.L., who solved the problem 9 x 7
by making nine groups of seven dots and then
proceeding to count up all the dots.
A second feature which was noted typically
was the tendency to break down multiplication
problems into a series of apparently simpler
stages. For example, one patient tackled the
problem 8 x 7 by first solving 2 x 7 = 14 and
then adding 14 four times to obtain 56. When a
patient adopts this particular strategy it was
noted that he often becomes muddled and has
difficulty in aligning the figures appropriately in
a column so that failure is a common result.
A third feature which was observed was that
of gradual approximation. An example of this
was provided by one patient who attempted to
solve the problem 102-^-6 by first multiplying
8 x 6 (breaking this down into stages as des-
cribed above) to obtain 48 and then adding 48
to itself to arrive at an approximation to 102.
The three features described above can all be
seen as adaptive manoeuvres on the part of the
patients to deal with problems, which, for them
are too difficult. However, because they generally
involve an increase in the number of stages
required they often lead to more incorrect than
correct solutions. Certain other features were
noted which reflect a more basic difficulty on the
part of the patient. One such feature is the
tendency to reverse two-digit numbers when a
carrying operation is required. For example, one
patient, faced with the problem 17x5, multiplied
5 x 7 correctly to obtain 35, but then proceeded
to fill in 3 in the digits column and carry forward
5 to the 10s column. Another common feature
was the variability in performance demonstrated
by all the patients. This affected all four basic
processes, particularly multiplication, and both
easy and difficult problems. For example, even
the very simple problem, 8 + 1, was on occasions
solved incorrectly by the patients who normally
had no difficulty with this.
DISCUSSION
The point which was made at the beginning of
this paper was that, as yet, there is no readily
available, adequate operational definition of
developmental dyscalculia. The findings des-
cribed in this report, because of their somewhat
297
limited scope of inquiry and the small number of
cases involved, do not permit this gap to be
filled. However, they do suggest a tentative set of
criteria for identifying cases of developmental
dyscalculia. Three kinds of criteria seem relevant
in this respect—namely, (1) pattern of deficit
displayed; (2) qualitative features of arithmetical
performance; and (3) response to remedial
teaching.
The pattern of deficit displayed by the four
patients studied has been elaborated at some
length. The psychological findings suggest that
one of the areas of maximal deficit may reside
in the process of multiplication and that this may
stem from a faulty grasp of the basic multiplica-
tion tables, especially the 8 x and 9 x tables.
However, it is clear that the deficit in multiplica-
tion is only relatively more severe, and that the
patients exhibit varying degrees of difficulty with
all four basic processes.
The qualitative features of performance were
described under the heading of clinical observa-
tions. The major features noted were the use of
the dotting system, the tendency to break down
multiplication problems into simpler stages, the
use of a procedure of gradual approximation,
the tendency to reverse two digit numbers when
a carrying operation is required, and, finally, the
tendency to exhibit performance variability. All
these features were displayed by the four
patients to some extent and would seem to pro-
vide useful indices for purposes of identification.
The third criterion, and perhaps the most
important, is the apparent intractability of the
deficits and the poor response to various attempts
at treatment. In addition to the minor attempts
at remedial coaching described here, three of
the patients (C.S., S.B., and N.R.) had under-
gone prolonged and intensive remedial pro-
grammes in arithmetic with relatively limited
benefit.
In addition, it has been emphasized by earlier
writers that dyscalculia may be associated with
defects of higher cortical functioning, such as
failure of left-right discrimination, finger ag-
nosia, and dyslexia. Kinsbourne and Warring-
ton (1963b) reported a series of cases of'develop-
mental Gerstmann's syndrome' in children, in
whom this was so. But, their cases cannot be
considered 'developmental' in the sense of the
definition used here, as they were all found to
298 /'. D. Slaclc ami G. /•. M. Russell
have detectable cerebral pathology of one kind or
another. In their papers (1963 a, b) Kinsbourne
and Warrington hypothesize that the basic deficit
underlying the Gerstmann syndrome may be
one of a failure in sequential ordering: consistent
with such an explanation was the tendency
observed in at least one of our patients to reverse
two-digit numbers and to have difficulty aligning
figures correctly in columns. However, a
'sequential ordering' deficit cannot seemingly
account for all the difficulties we observed.
Hecaen, Angelergues, and Houiller (1961)
studied a series of 183 patients with acquired
arithmetical disorders. On the basis of their
clinical and statistical findings they divided their
cases into three separate categories—namely,
(1) dyscalculia of spatial-deficit origin; (2)
acalculia related to specific disabilities in reading
and writing numbers, and (3) pure dyscalculia
(anarithmetie) in which the basic arithmetical
operations are themselves impaired. None of our
patients showed the type of disabilities indicated
by the second category of Hecaen et al.; they were
all able to read and write numbers with ease. If
anything, they would seem to represent a mixture
of the disabilities described by their first and
third categories. But it is not clear as yet to what
extent cases of acquired and cases of develop-
mental dyscalculia can be compared in terms of
their phenomenological aspects.
Rutter, Yule, Tizard, and Graham (1966) have
also reported a statistical association between
severe reading and arithmetical retardation in
normal children. Further cues for the identifica-
tion of developmental dyscalculia and its under-
lying mechanisms may therefore be forthcoming
from a more intensive investigation- of its
correlates.
To return to the findings of a faulty grasp of
multiplication tables, two possible explanations
would seem to merit further attention. The first
one can be expressed in terms of an hypothesis of
a simple deficit in learning the association bet-
ween numbers or number combinations, while
the second involves the hypothesis that the
present deficit is due to faulty past learning
which competes with and interferes with any
attempt to establish correct learning. From the
point of view of trying to develop an effective
programme of remedial education, it would be
important to distinguish between these two
alternatives.
REFERENCES
Cutler, A., and McShane, R. (1962). The Traclitenberg Speed
System of Basic Mathematics. Pan Books: London.
Gerstmann, J. (1924). Fingeragnosie: einc umschriebenc
Stoning der Orienterung am eigener Kb'rper. Wiener
klinische Wochenschrift,31, 1010-1012.
Grewel, F. (1952). Acalculia. Brain, 75, 397-407.
Guttmann, E. (1937). Congenital arithmetic disability and
acalculia (Henschen). British Journal of Medical Psychology,
16, 16-35.
Hecaen, H., Angelergues, R., and Houiller, S. (1961). Les
varietes cliniques des acalculies au cours des lesions
retrorolandiques: approche statistique du probleme.
Revue Neurologique, 105, 85-103.
Henschen, S. E. (1919). Uber Sprach- Musik- und Rechen-
mechanismen und ihre Legalisation im Gehirn. Zeitschrifi
fiir die gesamte Neurologie und Psychiatrie, 52, 273-298.
Johnson, D. J., and Myklebust, H. R. (1967). Learning
Disabilities. Grune and Stratton : New York.
Kephart, N. (I960). The Slow Learner in the Classroom.
Merill: Columbus.
Kinsbourne, M.,and Warrington, E. K. (1963a). The develop-
ment of finger differentiation. Quarterly Journal of Experi-
mental Psychology, 15, 132-137.
Kinsbourne, M., and Warrington, E. K. (1963b). The
developmental Gerstmann's syndrome. Archives of
Neurology, 8,490-501.
Rutter, M., Yule, W., Tizard, J., and Graham, P. (1966).
Severe reading retardation: its relationship to maladjust-
ment, epilepsy and neurological disorders. In What is
Special Education? Proc. First International Conference,
Association for Special Education, London, 1966.
Strauss, A. A., and Lehtinen, L. E. (1947). Psychopathology
and Education of the Brain-Injured Child. Grune and
Stratton: New York.