A COMPARESON OF INTENYIONAL
VERSUS ENCIDENTAL LEARNING IN
MEN?AL RETARDATES AS COMPARED
TO NORMALS WHEN MATCHED
BOTH 0N M. Ad AND C. As

“19563 {or ”‘0 Degree of M. A.
MECHIGAN STATE UNEVERSETY

David D. Robinson
1963

 

 
 
 

LIBRARY

Michigan State
University

 

 
 

 

 

A COMPARISON OF INTENTIONAL VERSUS INCIDENTAL
LEARNING IN MENTAL RETARDATES AS COMPARED
TO NORMALS WHEN MATCHED BOTH ON M.A. AND C.A.

By

David D. Robinson

AN ABSTRACT OF A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF ARTS
Department of Psychology

1963

ABSTRACT

A COMPARISON OF INTENTIONAL VERSUS INCIDENTAL
LEARNING IN MENTAL RETARDATES AS COMPARED
TO NORMALS WHEN MATCHED BOTH ON M.A. AND C.A.

By David D. Robinson

This study attempted to examine intentional and
incidental learning in mental retardates. Four questions
were posed. The first investigated the question of whether
the retardates were as well able to learn as the normals.

The second involved a comparison of performance on a per-
ceptual task between retardates, normals when matched on
M.A. and normals when matched on C.A. The third involved a
comparison of performance of normals and retardates on a
verbal task and the fourth question asked whether or not the
retarded subjects would forget more over time than the
normals.

A group of twenty male and twenty female retardates, a
group of twenty male and twenty female normals of equivalent
M.A. and a group of the same number of male and female normals
of the same C.A. were randomly selected from training schools
and children's homes. The subjects were assigned to either
the incidental or the intentional learning condition with
one restriction-~that there were twenty males and twenty

females under each condition.

David D. Robinson

A simple perceptual recognition test and a verbal recall
test were individually administered to each subject twice
with a fifteen minute time interval separating the first and
second presentation. A four-way analysis of variance was
used to analyze the three response measures.

No differences between normals and retardates with
respect to learning conditions were found. The retardates
and normals when matched on M.A. were found to perform less
well than the normals when matched on C.A. on one perceptual
task. Both groups of normal subJects performed better than
the retardates on the verbal task. The retardates forgot
no more than the normals over time. An interaction between
groups, learning condition and time was significant on two
of the three response measures and neared significance on
the third. A four-way interaction between groups, learning
condition, sex and time was nearly significant on one measure

of response.

APPROVED:

 

Major Professor

Datexw_

A COMPARISON OF INTENTIONAL VERSUS INCIDENTAL
LEARNING IN MENTAL RETARDATES AS COMPARED
TO NORMALS WHEN MATCHED BOTH ON M.A. AND C.A.

BY

David D. Robinson

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF ARTS

Department of Psychology

1963

To Brenda

ACKNOWLEDGMENTS

The author would like to express his gratitude to Dr.
M. R. Denny for his guidance and help which was given when-
ever needed, to Dr. R. M. McMichael for his assistance with
the statistical analysis and to Dr. J. Reyher for his review

and criticism of the thesis.

ll

TABLE OF CONTENTS

Page
ACKNOWLEDGMENTS . . . . . . . . . . . . . . . . . . . . 11
INTRODUCTION . . . . . . . . . . . . . . . . . . . . . 1
METHOD . . . . . . . . . . . . . . . . . . . . . . . . A
RESULTS . . . . . . . . . . . . . . . . . . . . . . . . 10
DISCUSSUION . . . . . . . . . . . . . . . . . . . . . . 17
SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . 2O
APPENDICES . . . . . . . . . . . . . . . . . . . . . . 23
REFERENCES . . . . . . . . . . . . . . . . . . . . . . 32

iii

3.

LIST OF TABLES

The mean age of retardates and normals when
matched on C.A. . . . . . . . . . . . . . . .

The means and pooled estimates of the variances
of the I.Q.'s for each group by learning
condition 0 O O O O O O O O O O O O O O O O O

A comparison of the correct responses with
the incorrect responses and the verbal
responses showing sources of variation and
Significant F-I‘atlos e e e e e e e e e e e e

The mean number of correct, incorrect and
verbal responses per subject by learning
condition across groups, tests, and sex . . .

The number of correct and incorrect responses
foreaChgrOUPeeeeeeeeeeeeeee

The number of verbal responses for each group .

The number of verbal responses in each learning
condition by time and the differences between

them....................

The scores of the male and female subjects by
learning condition by time on the three
response measures. e e e e e e e e e e e e e

The number of correct responses by group, by
learning condition by sex by time. . . . . .

iv

Page

11

12

12
13

14

15

16

LIST OF APPENDICES

Appendix

A.

The chronological ages of normals (C.A. Match)
and retardates by sex and by learning
condition . . . . . . . . . . . . . . . . . .

Subjects' I.Q.'s by group and by learning
condition 0 O O O O O O O C O O O O C O O O 0

Analysis of variance table for the correct
responses showing degrees of freedom, sources
of variation, mean squares and F-ratios which
were significant or approached significance .

Analysis of variance table for the incorrect
responses showing degrees of freedom,
sources of variation, mean squares and
F-ratios which were significant or
approached significance . . . . . . . . . . .

Analysis of variance table for the verbal
response measure showing degrees of freedom,
sources of variation, mean squares and
significant F-ratios . . . . . . . . . . . .

Raw score sums and sums of squares for the
correct responses by group, learning
condition, sex, and replication . . . . . . .

Raw score sums and sums of squares for the
incorrect responses by group, learning
condition, sex, and replication . . . . . . .

Raw score sums and sums of squares for the

verbal responses by group, learning condition,

sex and replication . . . . . . . . . . . . .

Page

24

25

26

27

28

29

30

31

INTRODUCTION

A survey of the literature indicates that very few
extensive studies of the learning process in mentally re-
tarded children exist. In fact, Sloan and Berg (30) found
that prior to 1957 only 23 papers had been published on this
subject.

Recent research has shown that the learning capabil-
ities of aments exceed prevailing notions. Clarke and
Hermelin (9) found that institutionalized subjects were
capable of rather complex motor skills such as the wiring
of television tubes and that they could begin a day's work,
maintain steady production and clean the shop at the end of
the day without supervision. Annett (2) concluded that,
even though the mentally retarded are slow to learn motor
skills and begin at a lower initial level of skill than
normals, they progress to a high level of efficiency with
training and many perform on a level equal to normals.

The implications of such studies are clear. If the
mentally retarded are capable of a much higher level of per-
formance than was previously thought possible, training pro-
grams might be geared to their actual capabilities resulting
in real improvement in performance, an increase in the retard-
ate's self-esteem and a hitherto untapped source of revenue

for the institution.

2

Several investigators have failed to find significant
differences in learning ability when defectives and normals
are matched on M.A. (8, 13, 27, 33). while others have found
marked differences (4, 16, 24, 32). Goldstein and Kass (18)
investigated incidental learning, comparing mental retardates
with gifted children, and concluded that the retardates per-
formed quantitatively and qualitatively as well on the gross
features of the task as the gifted children of the same M.A.
As the tasks became increasingly more complex, the retarded
children responded more frequently than the gifted children,
but with much less accuracy. 'Denny (11) has found that the
mentally retarded are well able to learn under intentional
conditions, but that they have difficulty learning incidentally.
According to Denny, this inability may be partly responsible
for their educational and social retardation.

In past research defectives have been compared with nor-
mals of the same M.A., but rarely with normals of the same
C.A. The present study has utilized two control groups in
an attempt to better identify the retardates' deficit. The
purpose of this study was to compare incidental and intentional
learning in mentally retarded subjects when matched with nor-
mals on C.A. Four questions are asked:

In all groups, there will be a performance difference
between the intentional and incidental conditions, but will
the difference be greatest in the mentally retarded group?
Hebb theorizes that learning efficiency increases with chron-

ological age and several investigators (1, 13, 14, 27. 33)

3
have found results supporting this notion. Other investi~
gators (6, 10, 16, 24, 26, 30, 32) have noted that retardates
experience Special difficulty on verbal tasks. Two questions
are raised: Will the normals perform better than the
retardates on a perceptual task? If not, will there be a
difference between these groups on a task of verbal learning?
A fourth question to be investigated is, Will the retardates
forget more than the normals over time because of a lack of

ability for symbolization?

METHOD

Subjects
Each of the three groups consisted of forty subjects,

each of whom were randomly assigned to the incidental or the
intentional learning condition with the restriction that there
were ten females and ten males under each condition in each
group. Badit (3), Goldfarb (17), Sarason (28) and Spitz
(31) have found institutionalization to be detrimental to
the capacity for abstract thinking, and in order to eliminate
institutionalization as a confounding variable, all subjects,
both retarded and normal, were randomly selected from rosters
of those who had been institutionalized for a period of at
least six months. No subject had observable motor defects
nor a history of seizures and none had been diagnosed as
psychotic. However, in the group of older normals (C.A.
match) there was a preponderance of character disorders.

Retarded group.--This group consisted of educable
retardates from the State Home and Training School, Lapeer,
Michigan. They ranged in age from 13 to 18 years with a mean
age of 16.1 and their Stanford-Binet I.Q.'s ranged from 60
to 80 with a mean of 68.3. The subjects selected had been
diagnosed by the medical staff as having mental deficiency
due to a familial or idiopathic etiology.

Normals (C.A. match).--This group consisted of twenty
inmates of the Boy's Vbcational School, Lansing, Michigan,

4

5
ten inmates of the Girl's Training School, Adrian, Michigan,
and ten female inmates of the Oakland County Children's Home,
Pontiac, Michigan. These subjects ranged between thirteen
and eighteen years of age with a mean age of 15.9 years.
Their mean WISC or WAIS I.Q. score was 101.8 with a range
from 90 to 126. All subjects in this group were confined
by court order for such offenses as car theft, assault,
breaking and entering or for sexual delinquency.

Normals (M.A. match).--This group consisted of 8 males
and 2 females from St. Vincent's Home, Lansing, Michigan,

12 males from St. Francis' Home, Detroit, Michigan, and 18
females from Guardian Angel Home in Detroit. Their mean age
was 10.7 and ranged from 8.0 to 14.0 years. Their mean
WISC, Stanford-Binet or California Test of Mental Maturity
I.Q. was 100.8 ranging from 90 to 119. In this group there
were no subjects who presented problems of a behavioral

or psychological nature.

A Kruskal-Wallis test (29) indicates that there are no
significant age differences between the sexes, learning con-
ditions, or groups when comparing retardates and normals when
matched on C.A.; Table 1 shows the mean ages for the retarded
subjects and for the normals when matched on C.A. Student-
ized tests (34) indicate that there are no significant differ-
ences in 1.0. between the incidental and intentional learn-
ing subjects; no significant differences in 1.0. between the
two groups or normals, regardless of sex; and no significant

differences.in I.Q. among retardates across learning conditions.

6

Table 1.--The mean age of retardates and normals when matched

 

 

 

 

on C.A.
Retardates Normals (C.A. Match)
Intentional Incidental Intentional Incidental
Males 15.4 16.7 16.1 15.9
Females 16.0 16.3 15.4 16.0

 

Table 2 shows the means and pooled estimates of the variances

of the I.Q.'s for each group by learning c0ndition.*

Table 2.--The means and pooled estimates of the variances of
the I.Q.'s for each group by learning condition.

 

 

 

Retardates Normals Normals

(C.A. Match) (M.A. Match)

Intentional x = 67.95 i = 102.45 Y = 102.45
Incidental i = 68.65 "x = 101.20 B? = 99.5
Pooled s2 : Pooled $2 : Pooled s2 =

29.1 62.03 42.46

Experimental Task

The eXperimental task involved the recognition of a
series of stimuli immediately after its initial presentation
and again following a 15 minute rest period. The stimuli
were 4-inch by 5-inch full length photographs of college
women standing in front of a plaster wall. Four photographs
were mounted on each of six 9é-inch by 11é-inch white folders.

There were, then, 24 different photographs. On each

 

*Appendix A reflects the chronological ages of the
retardates and normals (C.A. match) and Appendix B lists I.Q.'s

of all subjects.

7

photograph was a strip of paper approximately 3-inches by
3/8-inch, on which a name, the verb "is" and a descriptive
adjective was typed in capital letters, such as "MARY IS
SWEET" or "JUDY IS PRETTY." A second set of the same photo-
graphs without the printed strips was mounted on six other
folders which were used as recognition stimuli. The photo-
graphs in the second series were randomly assigned to the
folders to eliminate the "one picture per folder" response

cue.

Procedure
Each subject was seen individually and the following

instructions were read from a card:

Incidental condition

Here are some cards with girls' pictures on
them. Each girl has a name and something about
her on her picture. I'd like you to pick out
the girl you like best on each card and read
what it says on her picture, or if you have
trouble with the words, I'll read it to you.

Intentional condition

Here are some cards with girls' pictures on
them. Each girl has a name and something about her
on her picture. I'd like you to pick out the girl
you like best on each card and read what it says
on her picture, or if you have trouble with the words,
I'll read it to you. Look at each card and remember
the girls you choose. Remember their names and some-
thing about them. Later on I'm going to ask you to
find them again and tell me their names and some-
thing about them.

After the subject made his choice, the experimenter handed

8
him the second set of cards, the recognition stimuli, and
said:

Here is another set of the same pictures but
these are arranged differently; they're all mixed
up. I'd like you to pick out the girls you just
chose--the ones you liked best--and tell me their
names and something about them. Sometimes you may

not find a girl you chose or you may find more
than one on one card. Okay? See if you can find

them.

A non-correction procedure was used throughout. Ellis
and Pryer (15), Zigler, Hodgen and Stephenson (35) and Gordon,
O'Connor and Tizard (19) report that encouragement is a
sufficient incentive in learning problems with mental
retardates, and it is a well-established fact that normals
are also influenced by social rewards. In the present study
verbal reinforcement such as "good," "okay" and "uh-huh" was
given regardless of the correctness of the response in order
to maintain a fairly high rate of response.

The selection of the preferred stimuli (test 1) and
the presentation of the recognition series took approximately
eight minutes, regardless of group or learning condition.
After the recognition series was presented, the subject
was requested to wait in another room for fifteen minutes
while the next subject was seen by the experimenter. After
the fifteen minute rest period, the recognition series was
presented again (test 2). The subjects' behavior during
the fifteen minute interval was limited to conversation with

other subjects who were waiting to see the experimenter a

9
second time. Rehearsal was not controlled for. Naive
subjects were prevented from communicating with those

familiar with the experimental task.

RESULTS

A four-way analysis of variance comparing groups,
learning conditions, sex and one replication (tests 1 and 2)
was used (12). Three response measures were selected for
analysis: number of correct responses, number of incorrect
responses and number of correct verbal responses (Appendices
C, D, and E). A correct response was defined as the recog-
nition of a previously selected stimulus and an incorrect
response was defined as either the rejection of a selected
stimulus or the acceptance of a stimulus not previously
chosen. The criterion for a verbal response was the recall
of the name or descriptive adjective associated with a
correctly recognized stimulus. Table 3 on page 11 is a
summary of the three measures analyzed. The raw score means
and sums of squares are shown in Appendices F, G, and H.

Subjects under the intentional learning condition
performed better than those under the incidental condition
on all three response measures (corrects p‘7.025, incorrects
p4.005, verbals p4 .005). These results, shown in Table 4,
page 12, are consistent with the conclusions of others who
have investigated the importance of appropriate and inappro-
priate learning sets in learning (5. 7, 20, 25).

The lack of a significant groups x learning condition
‘interaction indicates that, comparatively speaking, the
retardates were as able to learn incidentally as the normals.

1O

Table 3.--A comparison of the correct responses with the
incorrect responses and the verbal responses showing sources
of variation and significant F-ratios.

 

 

 

Source Corrects Incorrects Verbal
F-Ratio F-ratio F-ratio
Between
Group (G) 2.73* 3.65** 4.7***
Learnin Condition
(L 6.2 **w 11.05**** 10.5****
Sex (S) n.s. n.s. n.s.
Within
Times (T) n.s. n.s. 16.1****
Int: T x L n.s. 3.46* n.s.
Int: T x L x S 3.72* 4.46** 6.2***
Int: G x L x S x T 2.80* n.s. n.s.

 

*(.104p<.05)

**(.05< p< .025)
***(.025<p<.01)
****(.005< p<.001)

12

Table 4.--The mean number of correct, incorrect and verbal
responses per subject by learning condition across groups,
tests, and sex.

 

 

 

Intentional Incidental P
Corrects 5.4 2.9 .0254 pit—.01
Incorrects 3.9 5.1 .0054 D4 .001
Verbal 2.9 1.8 .0054 p4 .001

 

There was a significant main effect (.05<jp4 .025)
between the three groups on the response measure of incorrects
and a nearly significant main effect (.101 p4 .05) on the
corrects. Analysis of both measures indicated that the
older normals (C.A. match) performed significantly better
than those matched on M.A. with the retardates and better
than—the retardates. The younger normals (M.A. match) differed
non-significantly from the retarded subjects, as shown in
Table 5.

Table 5.—-The number of correct and incorrect responses for
‘ each group.

W

 

Retardates Normals Normals
(M.A. Match) (C.A. Match)
Corrects 240 248 280
Incorrects 398 388 312

 

These findings are in agreement with Hebb (21) who theorizes
that learning efficiency increases with chronological age.
Benoit (4) discussed Hebb's theory and concluded that learn-

ing efficiency is not a simple function of measured intelligence,

13
thus supporting Hebb. Others (1, 14, 13. 27. 33) have re-
ported similar results.

An analysis of the verbal responses indicated a signifi-
cant main effect between groups (.0254 p<.01). The per-
formance difference of the normal groups was non-significant
and both normal groups performed significantly better than
the retardates on this response measure. These findings are
consistent with those of Berkson (6), Cruickshank (10) and
others (26, 16, 24, 30, 32). Table 6 indicates the number of

verbal reaponses for each group.

Table 6.--The number of verbal responses for each group.

 

Retardates Normals Normals
(M.A. Match) (C.A. Match)
137 182 230

 

A non-significant time x group interaction showed that
the retardates forgot no more over time than did the normals.
Examination of the corrects and incorrects reveals no sig-
nificant performance difference between test 1 and test 2,
but significantly more verbal responses were produced during
test 1 (.005€p< .001).

There was a nearly significant (.10<lp4'.05) time x
learning condition interaction on the measure of incorrect
responses (see Table 7, page 14) which became more pronounced
when the sexes were examined separately. The time x learn-
ing condition x sex interaction tended to be significant
(.106p4 .05) on the corrects, reached the .05 level on the

14

Table 7.--The number of verbal responses in each learning
condition by time and the differences between them.

 

 

 

Intentional Incidental Difference
Test 1 226 312 86
Test 2 252 309 57
Difference 26 3

 

incorrects and the .025 level on the verbal measure. Con-
sidering the corrects and incorrects, the females were
better intentional learners on test 1 than the males and the
males were better incidental learners on test 1 than the
females (see Table 8, page 15). The sex differences between
learning conditions on test 2 were non-significant on the
corrects and incorrects. Analysis of the verbal measure
indicated that the females far surpassed the males on the
intentional test 1 and on the incidental condition test 2
but that the sex differences between incidental test 1 and
intentional test 2 were non-significant.

A trend appearing on the correct response measure was
the group x learning condition x sex x time interaction
(.104 pt .05) in which there was a sizeable intentional-
incidental difference among female defectives on test 1.

In Table 9 on page 16, where the mean intentional-incidental
difference in number of correct responses on test 1 was 6.7.
that difference among female retardates was 17, nearly
double that of the next largest difference. Only on this

four-way interaction was there an indication of a deficit

15

Table 8.--The scores of the male and female subjects by

learning condition by time on the three response measures.

Numerical values are representative of the total number of
responses in each category.

 

 

 

 

 

 

 

j. Intentional Incidental
Test 1 Test 2 Test 1 Test
Corrects
Males 103 100 99 93
Females 110 101 78 87
Differences 7 1 21 6
Incorrects
Males 118 123 144 151
Females 108 129 168 158
Differences 10 6 24 7
Verbal
Males 88 75 58 33
Females 107 77 66 66

Differences 19 2 8 33

 

16

Table 9.--The number of correct responses by group, by learn-

ing condition by sex by time. The largest difference in the

table, the learning condition difference among female retardates
on Test 1, is underlined.

 

 

 

J
__’_v

 

 

 

 

Normals Normals Retardates
(M.A. Match) (C.A. Match)
Int. Inc. Int. Inc. Int. Inc.
Males
Test 1 32 30 36 40 35 29
Test 2 35 27 35 37 30 29
Females
Test 1 36 27 37 32 27 20
Test 2 34 28 32 33 35 26

 

in incidental learning among retarded subjects and the differ-

ence in this case is dependent upon sex and time.

DISCUSSION

Several conclusions can be drawn from the results of
this study. One, that under conditions of perceptual
recognition and verbal recall, learning is more efficient
under intentional conditions than under incidental. Two,
that older normals perform better than younger normals and
retardates on perceptual tasks. Thirdly, on verbal tasks
both younger and older normals perform better than retardates.
When tested for recognition immediately after presentation
of visual stimuli, females appear to be better intentional
learners than males, although males are superior on incidental
tasks. The female retardates performed poorly under the
incidental condition on test 1 but the hypothesis that the
retardates are not as able to learn incidentally as the
normals is not supported. The retardates forgot no more
over time than the normals.

Since the comparison of groups and the time x learning
condition x sex interaction on the analysis of the number of
incorrect responses are significant, it was expected that
the same comparisons would be significant in the analysis
of the correct responses. However, the comparison of groups
and the above mentioned interaction on that analysis fail
to reach significance. It is likely that the restricted
range of possible correct scores (zero to six) was responsible

17

18
for the failure of this measure to reach significance and
that differences do, in fact, exist.

Contrary to expectation, the retardates were as able
to learn incidentally as the normals. This finding does not
support the conclusions of Denny (11), of House (27) and of
House, Orlando and Zeaman (23). In the above studies no
special attention was paid to any particular stimuli under
the incidental condition but in the present study the sub-
jects under that condition were instructed to select the
girls they liked best. Special attention, therefore, was
drawn to the appropriate stimuli and the experimental task
may have been made simpler than those in the studies cited
above. Goldstein and Kass (18) reported that on incidental
learning tasks, retarded children perform as well quantita-
tively and qualitatively as gifted children of the same M.A.
and that they perform less well than the normals only when
the incidental learning tasks increase in complexity. In
the present study the failure to find greater differences
between learning conditions among retarded subjects than
among normals may have been due to the simplicity of the
experimental tasks.

The fact that the retarded subjects forgot no more than
the normals from test 1 to test 2 is interesting but may
need qualifying by the following factors: (1) the simplicity
of the tasks, (2) the short retention interval, (3) the
Opportunity for rehearsal though this should work in favor
of the normals, (4) the possibility that the retardates were

19
more highly motivated than the normals as a result of their
being experimental subjects.

The significant time x learning condition x sex inter-
action for the incorrect responses and nearly significant
interaction for the correct responses may reflect a fairly
common phenomenon in our culture: on perceptual tasks females
are better intentional learners than males, but males outdo
females on incidental learning. On the verbal reaponse
measure, however, the females demonstrated their superiority

by performing better than the males in every case.

SUMMARY

This study attempted to examine intentional and inci-
dental learning in mental retardates. Four questions were
asked: In all three groups subjects in the intentional
learning condition should perform at a significantly higher
level than those under the incidental condition, but would
the difference between intentional and incidental conditions
be greatest for the retarded subjects? The second question
asked whether on a perceptual task the retardates would per-
form as well as the normals when matched on M.A. but less
well than those when matched on C.A. The third asked whether
the retarded subjects would perform less well than either of
the normals on a verbal task. The fourth asked if the
amount of forgetting over time would be greater for retard-
ates than for normals.

A group of twenty males and twenty female retardates,

a group of twenty male and twenty female normals of equivalent
M.A. and a group of the same number of male and female normals
of the same C.A. were randomly selected from training schools
and children's homes. All subjects had been institutional-
ized for a minimum of six months and none had observable
physical defect or psychosis. The subjects were randomly
assigned to either the incidental or intentional learning
condition with the restriction that there were ten females and
ten males under each condition.

20

21

A simple perceptual recognition test and a verbal recall
test were individually administered to each subject twice
with a fifteen minute time interval separating the first
and second presentation.

A four-way analysis of variance was used to analyze
the three response measures. A non-significant groups x
learning condition interaction on all three response measures
showed that under the incidental condition the retardates
were as able to learn the experimental task as the normals.
The second question was answered by the significant F-ratio
on the incorrects, and a nearly significant F-ratio on the
corrects, indicating that the normals when matched on C.A.
performed better than the retardates or the normals when
matched on M.A. Both groups of normals performed better
than the retardates, thus answering the third question.

The retardates were found to forget no more over time than
the normals.

The time x learning conditions x sex interaCtion also
neared significance on the corrects and was highly significant
on the incorrects and the measure of verbal responses. The
group x learning condition x sex x time interaction was
nearly significant.

This study found no evidence that retardates are less
able to utilize incidental learning than normals, but the
task may not have been of sufficient complexity to detect
such a difference if it exists. These findings would tend to

indicate that training programs for the educable mentally

22
retarded should be set up in such a way that minimal emphasis
be placed on verbal tasks and that an emphasis on perceptual
tasks will lead to optimal learning. The results of this
study also indicate that females and males learn differently
under different learning conditions and perhaps the greatest
benefits could be accrued by separating the sexes in learning
situations. Further indicated is the proposition that
retardates, once having learned a stimulus-response pattern,

will forget no faster than normals over a short term.

APPENDICES

THE CHRONOLOGICAL AGES OF NORMALS (C.A. MATCH) AND RETARDATES

24

APPENDIX A

BY SEX AND BY LEARNING CONDITION

 

‘1

 

 

-

:—

 

 

Normals (C.A. Match) Retardates
INT FREQ INC FREQ INT FREQ INC FREQ
Males
13 13 13 13 1
14 2 14 2 14 4 14
15 15 1 15 2 15 1
16 3 16 4 16 2 16 3
17 5 17 5 17 17 4
18 18 18 2 18 1
Females
13 1 13 13 13 1
14 2 14 14 2 14 2
15 1 15 3 15 2 15
16 4 16 5 16 1 16 1
17 2 17 1 17 4 17 2
18 18 1 18 1 18 4

 

_ ‘flTI_-_‘-'_‘ "n:‘m___r

25

 

 

 

 

 

 

 

 

 

 

 

APPENDIX B
SUBJECTS' I.Q.'S BY GROUP AND BY LEARNING CONDITION
Retardates Normals Normals
(C.A. Match) (M.A. Match)
IQ ___E8EQ__ IQ FREQ IQ FREQ__
INT INC INT INC INT INC

60 5 90 1 1 90 1 1
61 2 91 1 1 91
62 1 92 92 1 2
63 1 93 93
64 2 94 1 1 94 1 1
65 95 1 1 95 1
66 2 1 96 2 2 96 1 1
67 3 1 97 1 97 1 1
68 1 98 1 98 1 3
69 1 1 99 1 99 1
7O 2 5 100 1 1 100 2 3
71 2 101 1 101 1
72 2 2 102 4 102 1 2
73 103 1 103 2
74 104 1 104 1
75 105 2 105 2 1
76 106 1 3 106 1
77 1 1 107 107 1
78 1 1 108 108 1
79 1 109 2 109 1
80 1 110 2 110 1 1

111 1 111

112 2 112

113 113

114 1 114

115 115 1

116 116

117 117

118 118

119 119 1

120

121

122

123

124

125

 

26
APPENDIX C
ANALYSIS OF VARIANCE TABLE FOR THE CORRECT RESPONSES SHOWING

DEGREES OF FREEDOM, SOURCES OF VARIATION, MEAN SQUARES AND
F-RATIOS WHICH WERE SIGNIFICANT OR APPROACHED SIGNIFICANCE.

 

 

 

Source DF SS MS F

 

Between:
Group (G) 2 12.0 6.0 2.73*
Learning Condition (L) 1 13.6 13.6 6.2 ***
Sex (S) 1 1.5 1.5 n.s.
Int: G x L 2 8.5 4.3 n.s.
Int: G x S 2 1.3 .6 n.s.
Int: L x S 1 5.1 5.1 n.s.
Int: G x L x S 2 1.2 .6 n.s.
Between SS 108 241.5 2.2

Total Between SS 119 284.7

Within:
Times (T) 1 .4 .4 n.s.
Int: T x G 2 .4 .2 n.s.
Int: T x L 1 1.0 1.0 n.s.
Int: T x S 1 .3 .3 n.s.
Int: T x G x L 2 1.4 .7 n.s.
Int: T x G x S 2 .6 .3 n.s.
Int: T x L x S 1 1.6 1.6 3.72*
Int: G x L x S x T 2 1.3 1.2 2.80*
Int: Within SS Pooled 108 46.5 .43

Total Within SS 120 53.5

Total 239 338.2

 

*(.10< p< .05)
**(.O25<‘p< .01)

27
APPENDIX D

ANALYSIS OF VARIANCE TABLE FOR THE INCORRECT RESPONSES SHOWING
DEGREES OF FREEDOM, SOURCES OF VARIATION, MEAN SQUARES AND F-
RATIOS WHICH WERE SIGNIFICANT OR APPROACHED SIGNIFICANCE.

W

 

Source DF SS MS F

Between:
Group (G) 2 56.1 28.1 3.65**
Learning Condition (L) 1 85.2 85.2 11.05***
Sex (S) 1 3.0 3.0 n.s.
INT: G x L _ 2 9.1 4.6 n.s.
Int: G x s 2 11.9 5.96 n.s.
Int: L x S 1 5.1 5.1 n.s.
Int: G x L x S 2 12.3 6.2 n.s.
Between SS 108 827.3 7.7

Total Between SS 119 1010.0

Within:
Times (T) 1 2.2 2.2 n.s.
Int: T x G 2 3.3 1.6 n.s.
Int: T x L 1 3.5 3.5 3.46*
Int: T x S 1 .1 .1 n.s.
Int: T x G x L 2 1.3 .65 n.s.
Int: T x G x S 2 .1 .05 n.s.
Int: T x L x S 1 4.5 4.5 4.46**
Int: G x L x S x T 2 .1 .5 n.s.
Int: Within SS Pooled 108 109.4 1.01

Total Within SS 120 124.5

Total 239 1134.5

 

*(.10< p< .05)
**(.05< p< .025)
***(.005<p< .001)

28
APPENDIX E

ANALYSIS OF VARIANCE TABLE FOR THE VERBAL RESPONSE MEASURE
SHOWING DEGREES OF FREEDOM, SOURCES OF VARIATION, MEAN
SQUARES AND SIGNIFICANT F-RATIOS

11

 

 

Source DF SS MS F

Between:
Group (G) 2 57.3 28.7 4.7***
Learning Condition (L) 1 64.1 64.1 10.5****
Sex (5) 1 16.1 16.1 n.s.
Int: G x L 2 22.3 11.2 n.s.
Int: G x S 2 2.4 1.2 n.s.
Int: L x S 1 1.6 1.6 n.s.
Int: G x L x S 2 .8 .4 n.s.
Between SS 108 654.7 6.1

Total Between SS 119 819.3

Within:
Times (T) 1 19.3 19.3 16.1****
Int: T x G 2 1.1 .6 n.s.
Int: T x L 1 1.3 1.3 n.s.
Int: T x S 1 .2 .2 n.s.
Int: T x G x S 2 1.8 .9 n.s.
Int: T x G x L 2 5.4 2.7 n.s.
Int: T x L x S 1 7.4 7.4 6.2***
Int: G x L x S x T 2 1.6 .8 n.s.
Int: Within SS Pooled 108 124.9 1.2

Total Within SS 120 163.0

Total 239 982.3

 

***(.025< p< .01 V5
****(.005<p< .001)

29

APPENDIX F

RAW SCORE SUMS AND SUMS OF SQUARES FOR THE CORRECT RESPONSES
BY GROUP, LEARNING CONDITIONS, SEX, AND REPLICATION. ,N = 10
IN EACH CELL. .

 

 

 

Normals Normals Retardates
(M.A. Match) (C.A. Match)
INT INC INT INC INT INC

EX EX2 EX EX2 EX EX2 Ex EX2 Ex Ex2 EX Ex2

 

Females
Test 1 36 136 26 80 37 143 32 108 37 165 20 50
Test 2 34 120 28 9o 32 126 33 115 35 149 26 78

Males
Test 1 32 116 30 100 36 136 40 168 35 135 29 97
Test 2 35 133 27 79 35 137 37 145 30 108 29 101

 

30

APPENDIX G

RAW SCORE SUMS AND SUMS OF SQUARES FOR THE INCORRECT RESPONSES
BY GROUP, LEARNING CONDITION, SEX, AND REPLICATION. N z 10
IN EACH CELL.

W

 

 

Normals Normals Retardates
(M.A. Match) (C. A. Match)
INT INC INT INC INT INC
2x xx? 1.x :x2 2x 2X2 5x 3x2 2x 2x2 5x 2x2
Females

Test 1 38 188 53 307 34 154 45 231 36 188 70 562
Test 2 44 206 51 293 43 245 46 226 42 234 61 401
Males
T at 1 46 262 54 312 29 91 38 188 43 249 52 310
Test 2 45 247 57 349 34 146 43 213 44 226 51 339

 

31

APPENDIX H

RAW SCORE SUMS AND SUMS OF SQUARES FOR THE VERBAL RESPONSES
BY GROUP, LEARNING CONDITION, SEX, AND REPLICATION. N = 10
IN EACH CELL.

 

 

 

 

Normals Normals Retardates
(M.A. Match) (C.A. Match)
ZINT INC INT INC INT INC
xix? 2x §X2 :3 xx? xx xx? 2x 2x2 2x 2x2
Females

Test 1 38 182 19 61 37 173 31 107 32 142 16 52
Test 2 31 145 19 61 29 121 28 126 17 83 19 81

Males
Test 1 39 189 16 36 30 108 32 132 19 89 1O 26
Test 2 31 139 10 42 28 104 15 57 16 64 8 16

 

5.

7.

10.

11.

REFERENCES

Akugatawa, D. and Benoit, E. The effect of age and
relative brightness on associative learning in children.
Child Deve10pment, 1959. 39. 229-238.

Annett, J. The information capacity of young mental
defectives in an assembly task. J. ment. Sgi., 1957,
103. 621-631.

Badit, M. Levels of abstraction in vocabulary defin-
itions of mentally retarded school children. Amer.
J. ment. Defic., 1958, 62, 241-246.

Benoit, E. Relevance of Hebb's theory of the organ-
ization of behavior to educational research on the
mentally retarded. Amer,J,:ment. Defic., 1957. 91.
497-507 e

Bensberg, G. Concept learning in mental defectives
as a function of appropriate and inappr0priate atten-
tion sets. J. educ._§sychql., 1958, 4 , 137-143.

Berkson, G. and Cantor, G. A study of mediation in
mentally retarded boys as compared with normal school
children. J. educ, Psychol., 1960, 51, 82-86.

Buss, A. Rigidity as a function of reversal and non-
reversal shifts in the learning of successive discrim-
inations. J. exp. Psychol., 1953. 35, 75-81.

Cassel, R. Serial verbal learning and retroactive
inhibition in aments and children. J. clin. Psychol.,
1957’ u! 369-372e

Clarke, A. and Hermelin, B. Adult imbeciles: Their
abilities and trainability. Lancet, 1955. 2, 337-340.

Cruickshank, W. and Blake, K. A comparative study of

the performance of mentally handicapped and intellectually
normal boys on selected tasks involving learning and
transfer. Syracuse University Research Institute,

U.S. Office of Education, Department of Health, Education
and Welfare, Contract #SAE 6414, 1957.

Denny, M. Mimeographed memorandum report to Jewish
Vocational Research Service, 1961.

32

33

12. Edwards, A. Exgerimental design in psychological research.
New York: Holt, Rinehart and Winston, 1962.

13. Eisman, B. Paired associate learning, generalization
and retention as a function of intelligence. Ame .
J, ment, Defic., 1958, 6 , 481-489.

14. Ellis, N. Object-quality discrimination learning sets
in mental defectives. J; comp. physiol, Psychol.,
19589 El, 79-81e

15. Ellis, N. and Pryer, M. Primary vs. secondary rein-
forcement in simple discrimination learning of mental
defectives. Psychol. Rep., 1958, 4, 67-70.

16. Ellis, N., Pryer, R., Distefano, M. and Pryer, M.
Learning in mentally defective, normal and superior
subjects. Amer. J, ment, Defic., 1960, 64, 725-734.

17. Goldfarb, W. Psychological privation in infancy and
subsequent adjustment. Amerg_J. Ortho-psychiat., 1945,
15, 247.

18. Goldstein, H. and Kass, C. Incidental learning of
educable mentally retarded and gifted children. Amer.

19. Gordon, 8., O'Connor, N., and Tizard, J. Some effects
of incentives on the performance of imbeciles. Brit.
J. PSXCth... , 1954, 4 , 277-2870

20. Harlow, H. The formation of learning sets. Psychol.
Rev. , 1949, fig, 51-65.

21. Hebb, D. O. The organization of behavior. New York:
Wiley, 1949.

'22. House, B. J. A comparison of discrimination learning
in normal and mentally defective children. Child
Develogm., 1958, 2 , 411-415.

~23. House, B., Orlando, R., and Zeaman, D. Role of positive
and negative cues in the discrimination learning of
mental defectives. Percent. mot. Skills., 1957. 1.
73‘79 e

24. House, B., Zeaman, D., Orlando, R. and Fisher, W. A
comparison of discrimination learning. Progress report
No. 1, Research Grant M 1099. National Institute of
Mental Health, U.S. Public Health Service.

25. Kendler, H. and D'Amato, M. A comparison of reversal
shifts in human concept formation behavior. J. exp.
PBEChOIe, 1955, .112, 165-174.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

34

McCullough, T., Reswick, T., and Roy, I. Studies of
word learning of mental defectives: I. Effects of
mental level and age. Amer. J. ment.gDefic., 1955-56,

' 9.2. 133-139.

Plenderleith, M. Discrimination learning and discrim-
ination reversal learning in normal and feeble-minded
children. J. genet. Psychol., 1956, 88, 107-112.

Sarason, S. Psychological problems in mental deficiency.

New York: Harper Brothers, 1953.

Siegel, S. N ar me c statis fo the beh v r
sciences. New York: McGraw-Hill, 1956.

Sloan, W., and Berg, I. A comparison of two types of
learning in mental defectives. Amer, J. ment. Defic.,
1957. 61. 556-566.

Spitz, R. Hospitalism: An inquiry into the genesis
of psychiatric conditions in early childhood. The
psychoanalytic studygof the chi; , 1945, l.

Stephenson, H. and Iscoe, I. Transposition in the
feeble-minded. J. exp, Psychol., 1955. 49, 11-15.

Stephenson, H. and Zigler, E. Discrimination learning
and rigidity in normal and feeble-minded individuals.
J. Pers., 1957, 25, 699-671.

Walker, H. and Lev, S. Statistical inference. New
York: Holt, 1953. '

Zigler, E., Hodgen, L. and Stephenson, H. The effect
of support on the performance of normal and feeble-
minded children. J. Pers., 1958, 26, 106-122.