. . IA..-.-- .g,-.'...

.
I
I
z
'3
v
I
o
C
.
a
Q
I .
‘1
.

MV‘V’I“f :r: 0“,. A”; "v
v I ’ .

f‘ 1'.
,3,.

THE RELATION BETWEEN STIMULUS:DIMENS‘DNAU'W' .—
AND COLOR-FORM MATCHING BEHAVIOR IN.
.PRE-SCHOOL CHILDREN ' '

Thesis for the Degree 0f M. A. ’
MICHIGAN STATE UNIVERSITY .
SANDRA M. JORGENSEN' *
1970 _

_.
V J
.
’.
.‘ 3
‘ p.—
. . .
. ‘ ' .
. , ‘ .
..v ‘
.. . . V
.......
< . ...............
.., .. 1 V . I U ‘ -
u . I I
.' . ..........
.
u'v . .2 , L. '
0.: _ . ‘ ‘ 1 ‘I O
0 .
"-r ...... Q "
. 1/ r," ,5...‘ ‘c‘
.I , .77 . ’-
. l . '- I.
.u .» . .. . k ’
‘ ' . . . --
' ' ._ 5.. ‘ - I
............. .v ’1’.
- . I I I ‘ " ‘O‘ -)—-.
u — 1. ‘7 _"-
‘ ‘ .4 l2.

- - ‘ . ,

......

'A'-'.. .........

gm." "h
_ a Y
LIBRAR
...............................

WWII“IIIHIIIIHHIH‘l1HIUIIHIillHllllllllHUHll L ‘_ vachigmsw

3 1293 00621 8378 U niVCfSltY

- «mm '0"

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A

ABSTRACT

THE RELATION BETWEEN STIMULUS DIMENSIONALITY
AND COLOR-FORM MATCHING BEHAVIOR
IN PRE—SCHOOL CHILDREN

By

Sandra M. Jorgensen

Thirty pre—school children, separated into two groups
of 15, older group (4-8 to 5—5) and younger grOUp (3-5 to
4—7), were each given A8 trials on a color-form matching
task. Four kinds of stimulus forms were used: two-dimensional
geometric, two—dimensional irregular, three-dimensional geo-
metric and three-dimensional irregular. Each child received
12 trials under each of the four stimulus conditions. The
stimuli were constructed in four colors (red, yellow, blue,
green), four geometric shapes (circle, square, triangle,
rectangle) and four irregular shapes.

Contrary to Brian and Goodenough's (1929) research
indicating that younger children make more color matches and
older children make more form matches, the older group made
significantly more color matches than the younger group.

This may be indicative of a later peaking period for the color
response tendency in this sample, possibly attributable to
cultural influences.

Among the four form conditions the only significant
difference between number of color and form matches was on

two-dimensional geometric stimuli. This finding was congruent

Sandra M. Jorgensen

with the results of previous investigations which used only
two—dimensional geometric stimuli.

There was no significant difference between color or
form responses to irregular stimuli. This is consistent
with Clapp and Eichorn's finding (1965) that the effect of
redundancy of figure is negligible.

Brian and Goodenough's finding that children consist—
ently matched three—dimensional forms more frequently on the
basis of color than they matched two-dimensional on the basis
of color was not replicated.

Kagan and Lemkin (I961) speculated that girls do more
implicit verbal labelling than boys, and this labelling
affects their matching behavior. In the present study there
were no significant differences between boys and girls in
mean numbers of verbal tags stated for stimuli. None of the
differences observed in presence of verbal tags was reflected
by a corresponding difference in number of color-form matches.

Girls did not show any observable differences in match-
ing behavior with respect to the four form conditions; how—
ever, boys showed a significantly greater number of form than
color responses for the two—dimensional and three—dimensional
irregular and three-dimensional geometric conditions.

Analysis of variance did not reveal sex—age group or
stimulus condition as accounting for variation in color and
form matching behavior. It appears that color dominance is
a complex mode of responding which is characteristic of pre—

school children (3—6 to 5—5). Although there are differences

Sandra M. Jorgensen

in responding, they cannot be categorized by any age group,
sex, stimulus condition difference, or the ability to state

labels verbally.

THE RELATION BETWEEN STIMULUS DIMENSIONALITY
AND COLOR—FORM MATCHING BEHAVIOR

IN PRE-SCHOOL CHILDREN

By ‘

Sandra M£"J5rgensen

A THESIS

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

MASTER OF ARTS

Department of Family and Child Sciences

1970

ACKNOWLEDGEMENTS

The author wishes to express appreciation to those
people whose assistance has been essential to the completion
of this thesis: to Robert Lance, who recognized the impor-
tance of the questions and assisted in the design and
creation of the stimuli; and to Lauren Harris, who fostered
a theoretical interpretation.

However, I want to express humble gratitude to Frances
Magrabi, whose interest, faith, encouragement, energy, time,
fantastic perseverance and intellect unreservedly given have

brought this project to fruition.

ii

TABLE OF CONTENTS

Page
ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . ii
LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . iv

LIST OF FIGURES. . . . . . . . . . . . . . . . . . . . . V

Chapter
I. INTRODUCTION AND RELATED LITERATURE. . . . . . . 1
Review of the Literature . . . . . . . . . . . 1
Questions for Research . . . . . . . . . . . . 9
Objectives and Hypotheses. . . . . . . . . . . 10
II. METHOD . . . . . . . . . . . . . . . . . . . . . 13
Operational Definitions. . . . . . . . . . . . 13
Apparatus. . . . . . . . . . . . . . . . . . . 13
Subjects . . . . . . . . . . . . . . . . . . . IA
Procedure. . . . . . . . . . . . . . . . . . . 15
III. RESULTS. . . . . . . . . . . . . . . . . . . . . 17
IV. DISCUSSION . . . . . . . . . . . . . . . . . . . 29

V. SUMMARY AND CONCLUSIONS. . . . . . . . . . . . . U1

REFERENCES . . . . . . . . . . . . . . . . . . . . . . . A6

APPENDIX . . . . . . . . . . . . . . . . . . . . . . . . 49

iii

Table

ll.

l2.

13.

LIST OF TABLES

Frequency of Color and Form Matches Distributed
by Age . . . . . . . . . . . . . . . . . . .

Frequency of Irregular Color and Form Matches
Distributed by Dimensional Conditions.

Frequency of Color Matches Distributed by
Dimensional Conditions . . . . . . . .

Frequency of Color and Form Matches in Each
Condition Distributed by Sex of Subjects

Frequency of Color and Form Matches Compared
with Respect to Color and Form Conditions.

Differences Between Mean Numbers of Verbal
Tags Elicited From Subjects Under Each
Form Condition . . . . . . . . . . . .

Mean Number of Verbal Tags Elicited in Each
Condition Distributed by Sex of Subjects

Number of Color and Form Matches Distributed
by Order of Presentation of Stimuli.

Analysis of Variance for Color Matches

Percentages of Form and Color Matches
Distributed by Age of Subjects . . . . . . .

Color Matches for Four Form Stimuli
Distributed by Sex-Age Group .

Color and Form Matches Distributed by
Subjects--Older Group. . . .

Color and Form Matches Distributed by
Subjects-—Younger Group.

iv

Page

20

21

22

23

2A

25

26

27
28

3A

52

53

5A

LIST OF FIGURES

Figure Page

1. Percentage of Form Responses Distributed
by Age of Subjects (Grouped Data) . . . . . . 33

2. Order of Presentation to Subjects——
Geometric Stimuli . . . . . . . . . . . . . . 50

3. Order of Presentation to Subjects--
Irregular Stimuli . . . . . . . . . . . . . . 51

CHAPTER I

INTRODUCTION AND RELATED LITERATURE

The categorization of objects according to their attri-
butes is a task which our society requires children to master
before embarking on formal conceptual processes of reading
and writing. A pre-school child is required to demonstrate
knowledge of colors and shapes, i.e. forms (Suchman, 1966).
This knowledge of attributes of objects can be termed "con-
cepts" if a child can use them "to produce behavior meeting
an accepted criterion" (Heidbreder, 1946). The ontogeny of
color and form concepts in pre-school children has not been
adequately documented. There were only two studies before
1930 (Descoudres, 1914; Brian and Goodenough, 1929); however,

there has been a renewed interest in the area after 1960.

Review of the Literature

 

In 1929 Brian and Goodenough conducted a study in which
subjects matched stimulus objects either on the attribute of
color or of form. The method used was adapted from Descoudres
(191A). The subject was shown a group of three objects so
selected that if any two objects were similar in form, they
were dissimilar in color,and if they were similar in color

they were dissimilar in form. The subjects were asked to

match one of the two comparison stimuli with the stimulus
object.

Brian and Goodenough used three different groups of
subjects: pre—school children ranging in age from 1“ months
to 5 years and 11 months, elementary school children ranging
in age from 6 to 1A years, and adult women. Three sets of
stimuli were used. The two sets used with pre—school child-
ren were: a three—dimensional set consisting of spheres,
cubes, and cylinders; and a two-dimensional set, consisting
of circles, squares, triangles, diamonds, stars, and oblongs
with rounded ends. The forms in both sets were colored red,
orange, yellow, blue, green, or purple. The third set of
materials used with elementary school children and adults
differed from the two—dimensional stimuli only in that the
forms were pasted on sheets of white paper, thus permitting
group testing.

Brian and Goodenough found that children under three
years of age tended to use form as a basis for matching.
From three to six years color was more often used, and after
six years through the adult years form was more often used.
Brian and Goodenough also found a consistently greater number
of color matches to solid, or three—dimensional, stimuli than
to surface, or two—dimensional, stimuli. No sex differences
were reported.

Corah (196A) presented pre-school children and elemen—
tary school children with stimuli to be matched by either

color or form choices. Corah's stimuli were all two-

dimensional and differed from those used by Brian and
Goodenough. There were three types of figures. One set
consisted of asymmetric colored geometric forms. The color
match was a figure of the same color but of a different
shape, and the form match was a black outline of the stimulus
figure. The second group consisted of a red outline of a
simple form (e.g., circle, rectangle) with complex line detail
inside the figure. The color match was a red outline of the
simple figure with only a suggestion of interior detail, and
the form choice was a black outline replica of the stimulus.
The third set consisted of red outline asymmetrical forms.
The color choice was the same figure reversed left to right,
and the form choice was a black outline replica of the stim-
ulus. Although Corah's stimuli contained more complex vari-
ations in color and form than Brian and Goodenough's, his
results were similar. His nursery school subjects matched
more frequently on the basis of color than form, while his
older subjects matched more frequently on the basis of form
than on color.

Chen and Wong (1965) investigated color—form perception
in a preference study with a sample of two groups of sub-
jects, pre-school children and university students. The sub—
jects were asked to rank colors and forms in order of prefer—
ence. The two age groups did not differ on either color or
form performance. However, pre—school children were more
consistent in their preferences for colors than the university

students. The order of preference for both age groups was

red, blue, green and yellow. The preferences of pre-school
children for forms were less consistent than those of the
university students. The circle was most frequently pre-
ferred to all other forms by both age groups. Next most fre-
quently preferred by the children was the square and by the
university students the rectangle. Chen and Wong observed
that children seemed to form more concrete verbal images

with form stimuli than with color stimuli. The university
students, however, appeared to form distinctive verbal associ~
ations for colors.

Clapp and Eichorn (1965), using 24 pre—school children
and tachistoscopic presentation of stimuli, studied percep-
tual investigatory responses in children.1 Three different
sets of stimuli were used: incongruous pictures, redundant
pictures-geometric figures, and redundant pictures-meaningful
objects (Berlyne, 1957). The set of redundant pictures-
geometric figures consisted of figures and lines in a decreas-
ing order of redundancy, according to Berlyne's definition
(1957) that: "Redundancy varies directly with orderly
arrangement, with symmetry, and with similarity between parts
of a figure, and inversely with changes in contour." The set
of stimuli which contained redundant pictures of meaningful
objects consisted of three modifications of five basic stim-

uli. Type I objects were in color, Type II objects were the

 

l"Tachistoscopic presentation"-—presentation of stim-
ulus objects on a screen by a tachistOSCOpe which allows
the experimenter to control the time of exposure precisely.

same as Type I only in black and white, and Type III were
black and white stick drawings of the same objects. Thus
in Type I similarity between parts of the figure was reduced
by the use of color, and in Type III redundancy was decreased
by eliminating all curvature.

The subject could look at the stimulus if he produced
a one word verbalization; i.e., a verbalization by the sub-
ject would produce a .14 second exposure of one of the above
stimuli. The measure taken was frequency of one-word verbal-
izations. Clapp and Eichorn found that the redundancy attri-
bute had no significant influence on responsiveness. However,
color and incongruity in the pictures affected responsive-
ness in a negative way (i.e., produced fewer requests for
exposure of the pictures) in contrast to the positive effects
obtained by Berlyne (1957) and Minton (1963) with adults.

Coran (1966a), using pre-school children and children
seven to nine years old, investigated the influence of vary-
ing some stimulus characteristics on color-form matching
behavior. The task was similar to the task which Brian and
Goodenough had used with pre—school children. There were
seven sets of stimuli. The first set used three simple forms—-
square, circle, and equilateral triangle—-and three saturated
colors--red, blue, and green. The second set of stimuli was
identical to the first set, except that the comparison figure
of the same color as the standard was not a different figure
but one similar to it. The same—color comparison choice for

the square was a rectangle, an ellipse for the circle, and an

w.

.3 C.
4» a» a1

....

\i

.LJ.

isosceles triangle for .he equilateral triangle. The third

set consisted of asym.,iric forms with five to seven sides.

The comparison figure the same color as the standard
retained the same 1 configuration as the standard,
but was readily df wable from the standard in shape.

All other characteristics of this set were the same as the
first two sets. To test for the effects of the amount of
surface color in the figures, each of the three basic sets

of stimuli was designed as two subsets. In one, the figures
were all solidly colored forms, while the other subset con-
sisted of forms which were only outlined in color. Thus,
there were six sets of stimulus figures with 18 plates in
each set. Since it might be argued that set three figures,
the asymmetric ones, were so difficult that only a color
choice was Open to the subject, a seventh set of figures was
designed. This set was identical to the third set in outline
color, except that both comparison figures on any given plate
were the same color as the standard. Thus, there was only
one correct response to each of these plates: the subject
was forced to make a form discrimination.

Corah found that amount of color present (outline vs.
saturation) and complexity of the figure were related to fre—
quency of color choices by pre-school subjects. Significantly
more color responses were obtained from the pre—school sub-
jects than from the older subjects under all stimulus condi-
tions. However, five of the six pre-school subjects were

form dominant (i.e., more form than color responses). Only

those subjects who were given set three with saturated color
were color dominant in their responses.

In a further exploration of color attributes in a
color-form matching task, Corah and Gross (1967) used kinder—
garten children randomly assigned to eight groups. For seven
of the groups, the form alternative differed from the stan-
dard and color alternative in hue, brightness, and saturation,
or some combination of these color characteristics. The
eighth group was a control group in which only a form match
was correct. Difference in brightness produced the greatest
frequency of color matching, while differences in hue had no
significant effect on color matching. Corah and Gross sug—
gested that contrast effects may be an important factor in
focusing a child's attention on color to the exclusion of
form.

Kagan and Lemkin (1961) investigated the effects of
form, color, and size variations in a matching task with 79
children who ranged in age from three years-nine months to
eight years-six months. The stimuli consisted of nine dif—
ferent plates on which were mounted two-dimensional geometric
figures, differing in color, shape and size. The first three
plates had two comparison forms at the top of the paper and
a standard stimulus at the bottom. Stimuli four through nine
had three comparison forms at the top and one standard at the
bottom. For each standard stimulus only two of the three
characteristics, color, form, or size could be matched. Kagan

and Lemkin found that for both boys and girls form was

distinctly preferred to color as a basis for matching, and
that color was preferred over size as a basis for matching
behavior. Older girls were less likely than younger girls

to use color as a basis for matching. Both older and younger
boys preferred form over color and color over size as a basis
for matching; however, older boys were more likely to use
color than were older girls. Kagan and Lemkin, in a post
hog speculation, said that sex differences could be a func-
tion of the precocious verbal skills of girls. The girls

may have implicitly labelled the form stimuli as square,
triangle, or circle, which presumably would facilitate the
use of form characteristics as a basis for matching.

Corah (1966b), in an attempt to clarify a criticism of
the Kagan and Lemkin (1961) study, investigated the effects
of instructional set on performance on a color-form matching
task. Whereas Kagan and Lemkin had used instructions which
stressed the form characteristics of the stimuli, Corah used
three types of instruction: instructions which stressed
color, instructions which stressed form, and neutral instruc-
tions which stressed neither color nor form. Performance set
was varied by presenting pre-test tasks which emphasized
color, form, or neither characteristic. The subjects were
kindergarten children. Corah found that instructional-sets
had no effect on color-form matching, the performance-color—
set induced a high level of color matching, while the
performance-form—set induced a weak tendency toward form

matching. The control groups favored form over color.

Although the subjects responded primarily to form, the ease
with which color matching was reinforced implies that the

tendency to respond to color is strong.

Questions for Research

 

Corah did not demonstrate that the results of Kagan and
Lemkin were invalid because of a prejudice in the instruc—
tions. Rather his work reiterates an important factor in
experimentation with perception and concept formation pro—
cesses: an assumption must always be made that concepts
(whether they are concepts of color, form, or size) are really
the subject matter being investigated in psychological
research. Heidbreder (1946) defined concepts as "the means
by which certain psychological activities are investigated."

A subject is said to use a concept whenever, during the course
of an experiment, he produces behavior meeting accepted cri-
teria. Thus in all of the cited research on color—form match—
ing, from the observed responses it is inferred that the
subject is using concepts of color, form, or size. In fact
the subject may be reacting according to some totally differ-
ent index from the index imposed by the experimenter.

Although in many of the studies reviewed a greater fre-
quency of color matches was found in pre—school children than
in older children, there has not been an investigation of the
differences among specific age levels within the pre-school
age group.

If there is a clear shift from form dominance to color

dominance and then back to form dominance as described by

10

Brian and Goodenough (1929), at what age levels do the shifts
occur? Suchman (1966) raises the question, what conditions
would cause a child to react by matching on the basis of one
stimulus attribute (color or form) if his basal response
level showed predominant use of another stimulus attribute
(color or form)?

Why have so few of the researchers used three-
dimensional stimuli? Would the redundancy attribute of Clapp
and Eichorn (1965) have an effect on performance on a match-
ing task? Most of the experimenters have used different
groups of children for each modification in stimulus presen-
tation. What would be the effect of using the same subjects
in several experimental conditions?

Finally, would it be possible to verify the speculation
of Kagan and Lemkin (1961) that implicit verbal labelling of
stimulus objects facilitated form matching by asking subjects

for verbal tags for the forms and colors used?

Objectives and Hypotheses

 

To partially answer these questions, the present study
was undertaken. It had as its objectives:

1. To observe the performance of pre-school children
on a color—form matching task.

2. To determine the relative frequency of color vs.
form matching at a specific age level.

3. To observe the effects of variation in the form
attributes of dimensionality and regularity of fig-

ures on the frequency of color and form choices.

The

11

To observe the presence or absence of verbal tags
for the stimuli presented and the effect of verbal
tags on matching behavior.

following assumptions were made:

Pre-school children are capable of performing a
matching task with both two- and three—dimensional
stimuli.

Children who show positional methods of responding
are not performing the matching task.

Children who are performing the task are matching on
on the basis of color when they are not matching on
the basis of form and are matching on the basis of
form when they are not matching on the basis of
color; i.e., every response can be classified as

either a color match or a form match.

On the basis of research discussed above, the following

hypotheses were formulated:

1.

There is a difference in the matching behavior of
younger (3—5 to 4—7) pre—school children and older
(4-8 to 5—5) pre-school children when presented
with color—form choice.

Irregularity of forms is related to matching behav—
ior. There is a difference in using color attri—
butes with respect to the regularity—irregularity
form dimension.

Dimensionality affects color matches.

12

Three—dimensional figures would more often be
matched on the basis of color than will two-
dimensional figures. (Brian and Goodenough, 1929).
Form matches are affected by the geometric and
irregular conditions.

Under the geometric condition there should be dif-
ferences between the performance of boys and
girls. (Sigel, 1954; Kagan and Lemkin, 1961).
Presence of verbal labels affects form matching.

Boys and girls exhibit differences in verbal tags
for forms and colors. (Chen and Wong, 1965).

Order of presentation of stimuli (two-dimensional
then three—dimensional or three-dimensional then
two—dimensional) does not affect matching behavior.
There is an interaction between sex-age group and
stimulus in determining number of color and form

matches.

CHAPTER II

METHOD

The methods used in the study will be discussed under
the following headings: Operational Definitions, Apparatus,

Subjects, and Procedure.

Operational Definitions

 

l. Irregularity of figure is defined as having many
unequal straight sides or many unequal curvilinear segments.

2. A positional method of responding is defined as a
response to the same stimulus position six times in succes—

sion.

Apparatus

 

The apparatus consisted of two sets of stimuli, one
two—dimensional and the other three—dimensional. The two—
dimensional stimuli were a set of 24 8 1/2 by 11 inch white
cards on which cut—out colored figures were mounted. The
first 12 cards had groups of three regular geometric figures;
the remaining 12 had groups of irregular forms. On each card
the figures were arranged in an equilateral triangle with the
base closest to the subject (Doerhing, 1960). The geometric
forms used were circle, square, rectangle, and equilateral

triangle. All figures had an area of four square inches.

13

14

The four irregular forms were constructed by cutting either
a circle (for the curvilinear figures) or a square (for the
straight—sided figures) which had an area of four square
inches and recombining the pieces. The procedure yielded
two straight—sided forms and two curvilinear forms. The
stimuli were produced in four colors——red, yellow, blue, and
green.

The three-dimensional stimuli consisted of 12 geometric
figures—-sphere, cube, rectangular solid, and equilateral
prism. These figures were constructed of posterboard with
the exception of the spheres, for which rubber balls were
used. The three-dimensional irregular figures were so con-
structed that they would be three-dimensional analogues of
the two—dimensional irregular figures, in the sense that the
frontal plane of the three—dimensional figure was the same
as its two-dimensional correlate. Each irregular three-
dimensional figure was constructed from eight cubic inches of
clay. Rubber molds were constructed from the clay models and
the figures were cast in plaster. All three-dimensional fig-
ures were sprayed with a fixative and painted in the four
selected colors. The three-dimensional stimuli were mounted
on white cardboard (8 1/2 by 11 inches), supported by press-
board. The colors used were nearly identical in hue to those

used for the two—dimensional figures.

Subjects

The subjects were 30 pre-school children in two groups

of 15 each. The older group ranged in age from 4 years

15

8 months to 5 years 5 months and had a mean age of 5 years

2 months. The older children (8 boys, 7 girls) were enrolled
in the summer session of the Michigan State University Lab-
oratory Pre—School. The younger children (7 boys, 8 girls)
ranged in age from 3 years 5 months to 4 years 7 months
(i=4-2). These children had attended Laboratory Pre—School
during the previous academic year. The older subjects were
selected because of their availability for research and pre-
kindergarten age range. The younger subjects were selected
because of their similarity in general background to the
older group and because they represented young children

within the pre—school age range.

Procedure

 

Each child was given 48 successive trials on which he
was required to point to the one on the comparison figures
which "goes with" or "belongs with” the stimulus figure at
the top of the card (Corah, 1966b). The responses were 1
recorded by the experimenter as the subject made them. At
the conclusion of each section, two-dimensional and three-
dimensional, each subject was asked to name the colors and
forms in the set as the experimenter pointed to them. The
children's statements about either colors or forms were
recorded. In the presentation order, form, color, and posi—
tion of the conparison stimulus were randomized for the first

12 presentations. The order for the second set of 12 trials

was generated from the order of the first 12 trials by

16

reversing the 1 to 12 sequence. Thus the order of the first
24 trials was 1 to 12, 12 to 1. The order for the first 24
trials was repeated unchanged for the second 24 trials (see
Appendix, Figures 2 and 3). Half of each age group received
the three-dimensional stimuli first; the other half was

given the two—dimensional stimuli first.

CHAPTER III

RESULTS

In order to clarify results this section has been sub-
divided into eight sections. Each of the following sections
refers to the corresponding numbered hypothesis stated on
pages 11 and 12.

l. A z-score for difference between uncorrelated pro-
portions was obtained. The proportions were: color matches/
total number of matches made by the younger group and color
matches/total number of matches made by the older group. The
z-score was not significant at the .05 level (z= -l.08;

Table 1).

2. In the irregular form condition (two-dimensional
and three-dimensional stimuli combined) the difference between
the proportion of color matches/total matches and the propor-
tion of form matches/total matches was not significant at the
.05 level (z=.96; Table 2).

3. The difference between proportion of two-dimensional
color matches/total number of two-dimensional matches and pro-
portion of three-dimensional color matches/total number of
three—dimensional matches was not significant at the .05

level (z= -.054; Table 3).

l7

18

4. Chi square tests were performed to determine the
differences in the number of form matches in the geometric
and irregular conditions (two—dimensional and three-
dimensional) by boys and girls. Significant difference was
found between girls and boys on two—dimensional irregular

(x2

=10.10, p=.01, df=l); three-dimensional geometric
(X2=7.48, p=.01, df=l); three-dimensional irregular (x2=5.64,
p=.05, df=1; Table 4).

5. Chi square tests were performed to estimate dif—
ferences between number of color and number of form matches
in each of the four form conditions (two—dimensional geo-
metric, two—dimensional irregular, three-dimensional geo-
metric, three-dimensional irregular). The results are
summarized in Table 5. Only the difference between two-
dimensional geometric color matches and two—dimensional geo-
metric form matches was significant (x2=l6.8, p‘(.01, df=l;
Table 5).

T tests were performed on the mean numbers of verbal
tags elicited from girls and boys under each of the four
stimulus conditions: two—dimensional geometric, two-
dimensional irregular, three—dimensional geometric, three—
dimensional irregular. The t's are shown on Table 6. None
of the t's were significant under the two-dimensional con-
ditions. Under the three-dimensional conditions the only
significant differences were between mean number of three-
dimensional color tags and mean number of three-dimensional

form tags (t=3.44, p=.01, df=29) and the difference between

19

mean number of verbal tags for three-dimensional form geo-
metric and three—dimensional form irregular (t=3.l9, p=.01,
df=29; Table 6).

6. T tests were performed to compare mean number of
verbal tags elicited for colors and geometric and irregular
forms in both two—dimensional and three-dimensional condi-
tions. None of the t's were significant at the .05 level
(Table 7).

7. The effect of order of presentation of stimuli
(two-dimensional then three-dimensional or three-dimensional
then two-dimensional) on matching behavior was examined by
computing a chi square for the total number of color matches
and form matches made by each group (i.e., those who received
the two—dimensional then the three-dimensional, and those
who received the three-dimensional then the two-dimensional).
The results are shown in Table 8 (x2=l.l7, p)~.05, df=1).

8. An analysis of variance was done to test for the
age—sex group and stimulus interaction effect on number of
color—form matches. No significant F ratios were found.

See Table 9.

20

Table l.—-Frequency of color and form matches distributed

 

 

 

by age.
Older Group Younger Group Total 2
Color Matches 440 339 799
Total Matches 72o 717a 1437 1.08

 

8Total for younger group reflects no response on three

trials for one subject.

21

Table 2.--Frequency of irregular color and form matches
distributed by dimensional condition.

 

 

Two—Dimensional Three—Dimensional

 

Stimuli Stimuli Total 2
Color Matches 197 186 383
Form Matches 161 174 335

Total 718 .96

 

22

Table 3.-—Frequency of color matches distributed by
dimensional conditions.

 

 

 

Two-Dimensional Three-Dimensional 2
Color Matches 414 365
Total 717 720 -.054

 

23

Table 4.-—Frequency of color and form matches in each
condition distributed by sex of subjects.

 

 

 

 

Matches by Matches by Chi
Condition Boys Girls Square
Two Dimensional:
Geometric,
Color 100 118
Form 80 62 1.62
Irregular
Color 83 114
Form 95 66 10.10a
Three Dimensional:
Geometric
Color 71 108
Form 109 72 7.48a
Irregular
Color 77 109
Form 103 71 5.614b
a.01 level.
b.05 level.

Degrees of freedom in all above tests equaled one.

24

Table 5.-—Frequency of color and form matches compared with
respect to color and form conditions.

 

 

 

 

Total Chi
Two—Dimensional Forms Matches Square
Color Matches- Form Matches—
Geometric Geometric
219 141 360 16.8a
Color Matches- Form Matches—
Irregular Irregular
195 162 357 3.43
Form Matches- Form Matches-
Geometric Irregular
141 162 303 1.45
Total Chi
Three-Dimensional Forms Matches Square
Color Matches— Form Matches-
Geometric Geometric
179 181 360 0.01
Color Matches— Form Matches-
Irregular Irregular
185 175 360 0.43
Form Matches— Form Matches—
Geometric Irregular
181 174 355 0.14

 

a . . ,
Significant at .01 level.

Degrees of freedom in all above tests equals one.

25

Table 6.-—Differences between mean numbers of verbal tags
elicited from subjects under each form condition.

 

 

 

 

Two—Dimensional Standard Error Mean a
Mean Number of Tags of the Difference Difference t
Form
Color Geometric
3.30 3.17 .16 .13 .81
Form
Color Irregular
3.30 2.77 .63 .53 .84
Form Form
Geometric Irregular
3.17 2.77 .18 .40 2.22
Three—Dimensional Standard Error Mean a
Mean Number of Tags of the Difference Difference t
Form
Color Geometric
3.27 3.13 .24 .13 .54
Form
Color Irregular
3.27 2.43 .27 .83 3.44b
Form Form
Geometric Irregular
3.13 2.43 .22 .70 3.19b

 

aDegrees of freedom in each of the above tests equaled
29.

bSignificant at the .01 level.

26

Table 7.—-Mean number of verbal tags elicited in each
condition distributed by sex of subjects.

 

 

 

Condition Boys Girls a
Two—Dimensional
Color Tags 2.93 3.33 .79
Geometric Form
Tags 3.00 3.33 .70
Irregular Form
Tags 3.13 2.67 .84
Three-Dimensional
Color Tags 2.87 3.67 .68
Geometric Form
Tags 3.40 2.87 .77
Irregular Form
Tags 2.60 2.40 .36

 

aDegrees of freedom in

all above tests equaled 29.

27

Table 8.—-Number of color and form matches distributed by
order of presentation of stimuli.

 

 

 

Color Form Chi
Order Matches Matches Total Square
Two-dimensional to
Three—dimensional 412 356 769
Three-dimensional to
Two—dimensional 391 278 669
Total 803 634 1437 1.17a

 

aDegrees of freedom equaled l.

28

 

 

 

Table 9.-—Analysis of variance for color matches.a
Source of Degrees Mean
Variation of Freedom Square F
Sex-Age 3 108.18 4.70b
Stimulus 3 5.07 .22b
Interaction 9 .45 .02b
Within 104 23.02

Total 119

 

aSee Table 11, page 52 for raw data.

b(p>.05).

CHAPTER IV

DISCUSSION

The discussion has been subdivided into individual
treatments of the eight hypotheses stated on pages 11 and
12. Each is restated in each sub-section.

Hypothesis l.--There is a difference in the matching

behavior of older and younger pre-

school children when presented with a
color-form choice.

 

Although the subjects in the present study were all
of pre—school age (ranging from three years and six months
to five years and five months), it was expected that there
would be differences in performance between the younger
group (ranging from three years and six months to four years
and seven months) and the older group (ranging from four
years and eight months to five years and five months). Since
four years six months represents the peak of the color
response (Brian and Goodenough, 1929), it would be expected
that the younger children in this sample would show more color
responses and that the older children would show more form
responses. However, the results were in the opposite direc-
tion. The older group made significantly more color
responses than the younger group, indicating that the color

response tendency is stronger above four years six months

29

30

for this sample than below four years six months. The
samples are different in age range. Brian and Goodenough's
subjects ranged from 14 months to 5 years 11 months, where
those in the present study ranged from three years six months
to five years five months. Serious question arises concern-
ing the capability of a 14-month-old child choosing a color
or form. It seems strange that children in 1929 at age four
years six months would have already shifted to adult form
dominance, whereas children of the same age in 1967 had not
yet made this shift from color dominant to form dominant.

After examining the results of the present study it
appears that children at age four years six months, in the
process of learning colors and names for colors, seem to have
at this time more interest in colors than shapes. The child-
ren in the present study had attended or were attending a
nursery school, an environment in which colors are more
emphasized than forms. That is, a teacher of a three— and
four—year-old group would expect children to "learn" colors,
probably, before forms (shapes) which are usually emphasized
with five—year—olds (kindergarten).

Hypothesis 2.-—Irregularity of forms is related to
matching behavior. There is a differ-
ence in using color attributes with

respect to the regularity—irregularity
form dimension.

 

It was expected that the form attribute of irregularity
would be unfamiliar and, therefore, difficult to use in
matching behavior. If the form attribute could not be readily

used, the subject would be forced to use the only other

31

attribute present, color, resulting in a greater number of
color matches on those trials involving irregular forms.
However, there was no significant difference between the num—
ber of color and the number of form matches on the irregular
stimuli. About equal proportions of each were elicited
(Table 2 summarizes the response to irregular stimuli). This
finding is consistent with Clapp and Eichorn's (1965) finding
that the effect of redundancy, or relative uncertainty, of
figure on the responsiveness of pre-school children was
negligible.

Apparently many of the children did use the irregular
form dimension as a basis for matching since there was no
time limit on the child's response; many children examined
the irregular forms before making either a color or form
choice. Since there were only four irregular stimuli it is
possible that the forms became very familiar in the course
of 48 trials and could be used as readily as geometric shapes

in matching.

 

Hypothesis 3.--Dimensionality affects color matches.
Three—dimensional figures would more often be matched
on the basis of color than will two—dimensional
figures. (Brian and Goodenough, 1929).

From Brian and Goodenough's finding that children con—
sistently matched three-dimensional forms more frequently on
the basis of color than with two-dimensional forms, it was
expected that the three-dimensional stimuli used in the cur-

rent study would elicit more color responses than the two-

dimensional stimuli. However, no significant difference was

32

found between the number of color responses to two-dimensional
and to three—dimensional stimuli (See Table 3). Brian and
Goodenough did not perform any statistical test, but grouped
their data by half-year intervals in the form of calculated
percentages. Similar treatment of the data from this study
is shown by Figure 2. It can be observed that the curves
from the data of the present study seem to resemble the
curves of the Brian and Goodenough data in general shape.
However, the differences as assessed by the 2 test between
the number of color responses to two-dimensional and number
of color responses to three-dimensional stimuli were not sig—
nificant. Therefore, although there may be an observable
difference when the data are plotted, the difference was not
statistically supported.

Hypothesis 4.--Form matches are affected by the geo-
metric and irregular conditions.

 

Under the geometric condition there should be dif-
ferences between the performance of boys and girls.
(Sigel, 1954; Kagan and Lemkin, 1961).

The sample was divided into the two groups of 15 boys
and 15 girls. When the number of color and form matches in
each category were compared, several differences between
girls and boys were evident. The results of chi-square tests
between boys and girls on all four stimulus conditions are
summarized in Table 4.

Under the two—dimensional geometric condition both boys

and girls showed more color than form responses, and the dif-

ferences between boys and girls in number of color matches

33

Figure l.-—Percentage of form responses distributed by age
of subjects (grouped data).

 

 

 

 

70—
60-
50-
4
go' Solids (3D)
Percentage _ , , \ \
of Form ‘\ z ‘ \ Surfaces (2D)
Responses \I/ \ ‘.Brian and
30- b \\ Goodenough
\
fl \
Solids (3D)
\Brian and
20- Goodenough
Surfaces (2D)
.4
10-
0 _l_

 

Age in Years

34

Table lO.--Percentages of form and color matches distributed
by age of subjects.

 

 

 

 

 

 

 

 

Age Groups
3-6 to 4-0 to 4-6 to 5—0 to
3-11 4-5 4—11 5-11

Data from
Present Study
Two—Dimensional

Condition

Form .47 .30 .46 .20

Color .53 .70 .54 .80
Three-Dimensional

Condition

Form .42 .39 .48 .42

Color .58 .61 .52 .58
Brian and
Goodenough Data
Two—Dimensional

Condition

Form .34 .25 .36 .29

Color .66 .75 .64 .71
Three—Dimensional

Condition

Form .24 .17 .29 .22

Color .76 .83 .71 .78

 

35

and number of form matches was not significant. However, on
all of the other three conditions—«two-dimensional irregular,
three-dimensional geometric, and three—dimensional irregular--
girls made a significantly larger proportion of color
responses than the boys (Table 4). It also should be noted
that the ratio of color to form responses holds through all
four conditions for the girls. However, the boys showed more
color than form responses on the two—dimensional geometric
stimuli only, and for all other conditions showed a ratio of
more form than color responses.

As the color attributes were constant over all four
stimulus conditions, the form attributes of dimensionality
and regularity-irregularity were the only attributes of the
stimuli which were varied. Previous research has generally
used only two—dimensional geometric stimuli. There has been,
with the exception of the Kagan and Lemkin study (1961), a
predominance of color over form matches. The findings of the
present study with respect to the two—dimensional geometric
stimulus condition are consistent with those of the previous
studies. However, in the conditions in which there was a
modification of the form attributes of either dimensionality
or regularity, the boys reacted to the change by showing more
form responses. The girls apparently did not react to the
change in the form dimension, since they consistently showed
a preponderance of color rather than form responses over all

four stimulus conditions.

36

Hypothesis 5.--Presence of verbal labels affects
form matching.

 

On the basis of Sigel's (1954) finding that semantic
meaning dominates other bases of conceptualization and on the
basis of Kagan and Lemkin's (1961) speculation that implicit
verbal labelling could account for differences in performance
on a matching task, it was expected that ability to give ver-
bal labels for the stimuli used would affect matching perform-
ance. If verbal labels do carry relevant semantic meaning
for the pre—school child, and if the color-form matching task
is an example of conceptualization (Heidbreder, 1946), then
the number of verbal tags present for colors and forms should
be directly related to the number of matches made using the
colors and forms for which the tags exist. Pre-school child-
ren probably would have verbal tags for the four colors and
four geometric forms; however, they would not be expected to
have verbal tags for the irregular forms.

The data were analyzed for differences in frequency
of color and form matches under each condition:

color matches—geometric compared with form matches-
geometric

color matches-irregular compared with form matches-
irregular

form matches—geometric compared with form matches-
irregular

The last comparison was made to determine the differ-
ence in the number of form matches attributable to the
irregularity form component of the stimuli. The above analy-

ses were done for both two- and three-dimensional stimuli.

37

The only significant difference in number of color and form
matches was under the two—dimensional, geometric condition.
There were significantly more color matches than form
matches (Table 5).

In the analysis for difference between mean number of
verbal tags present for forms and colors, none of the differ-
ences are significant under the two-dimensional conditions
(Table 6). However, under the three-dimensional conditions
the difference between mean number of color verbal tags and
form verbal tags under the irregular condition is significant.
Under three—dimensional condition the difference between mean
number of form tags for geometric forms and form tags for
irregular forms is also significant. More verbal tags were
expected for both colors and geometric forms than for the
irregular forms. However, this was supported by the results
from the tests on the three—dimensional stimuli but not by
the test on the two-dimensional stimuli. Apparently, child-
ren assign labels differently to colors, geometric forms,
and irregular forms only when three-dimensional objects are
used. When the stimuli were familiar to the children they
may have regarded the two—dimensional irregular forms as cut-
outs or designs in a general class of two-dimensional pieces
of paper. However, the three-dimensional objects were not
easily classifiable into a larger class of three-dimensional
objects. Thus the two-dimensional objects could be assigned
verbal tags in a similar way to their geometrical class mem-

bers. The three-dimensional irregular stimuli, however, were

38

quite unfamiliar and the children lacked pre—established
modes of assigning verbal labels to them. It was observed
that the children either named the three-dimensional geo-
metric stimuli accurately or did not name them at all;
whereas they gave the three-dimensional irregular stimuli a
greater variety of names.

The implicit verbal labelling described by Kagan and
Lemkin (1961) appears to be present and in some instances
explicit. The effect of verbal labelling appears to be more
complex than originally was thought by Kagan and Lemkin and
its importance changes with variations in the dimension (two-
dimensional or three-dimensional) of the stimuli. The data
lend some support for Chen and Wong's (1965) findings that
children form more specific verbal tags to colors than to
forms (Table 5). However, colors appear to have more verbal
tags only when compared with irregular three—dimensional
forms and not when compared with geometric forms. Therefore,
the data do not support the expectation that children who
form more concrete verbal tags to colors than to forms will
also show more color than form matches.

There may be an explanation for the lack of influence
of presence of verbal tags on matching behavior in the work
of Suchman (1966) with children in Zaria, Nigeria, West
Africa. Koranic education in Zaria requires that form dif-
ferences be discriminated in terms of producing spoken sounds
and written symbols, but comprehension is not required.

These children (range 3.0-15.6) in Nigeria did not show the

39

classic color-to-form shift found in Euro—American cultures,
but were clearly color dominant well into adolescence.
Suchman raises the question, "Is it possible that perceptual
discrimination skill is a component of some concept or
organizer meaningful to the individual?" Thus in the present
study it was possible for children to possess verbal tags for
the color and form stimuli which may or may not have had much
meaning for the individual. If a verbal tag, e.g., "ball,"
had meaning for the individual child he may have used that
concept for his matching behavior. However, if a tag, e.g.,
"rectangle,” had very little meaning for the individual he
may have used a more salient tag, probably one of color, which
would have more references in his conceptual organization.
This would also happen with the irregular stimuli; e.g.,
"duck,” "hat" or ”holey" may have had more meaning for a
child than color, thereby producing a form match under the
irregular condition rather than a color match. The difficulty
with this explanation lies in delineating "meaningfulness"
of verbal labels for the individual.

Hypothesis 6.——Boys and girls exhibit differences in

verbal tags for forms and colors. (Chen
and Wong, 1965)-

 

The t-tests performed between mean number of verbal
tags for boys and mean number of verbal tags for girls showed
no significant differences under any of the four conditions
(Table 7). These results are not in accordance with Kagan
and Lemkin's speculation that girls label forms implicitly

more often than boys. However, the subjects in the present

40

study were asked to state labels for the stimuli. It is pos-
sible that implicit labels which cannot be stated by children
influence matching behavior but there was no way for implicit
labelling to be recorded in this present study except by the
child's verbal response to a direct question.
Hypothesis 7.--Order of presentation (two—dimensional
then three—dimensional or three-

dimensional then two—dimensional) does
not affect matching behavior.

 

There was no significant difference in the frequencies
of color and form matches for subjects who received the two
different orders of presentation. The two different orders
of stimuli were introduced to control for inducing a "set to
respond." However, the initial type of stimuli (two-
dimensional or three—dimensional) presented did not affect
matching behavior.

Hypothesis 8.-—There is an interaction between sex—age

and stimulus in determining number of
color and form matches.

 

An analysis of variance was performed to explore the
relationship of age-sex stimulus interaction. No significant
F—ratios were found (Table 9). In this sample, although
there are isolated sex and age differences in performance,
there is no consistent performance relationship between sex-

age group and stimulus condition.

CHAPTER V

SUMMARY AND CONCLUSIONS

Thirty pre-school children, separated into two groups of
15 each—-older group (4-8 to 5-5) and younger group (3—5
to 4—7)--were each given 48 trials on a color-form matching
task. Four kinds of stimuli were used: two-dimensional
geometric, two—dimensional irregular, three-dimensional geo-
metric, and three-dimensional irregular. Each child received
12 trials under each of the four stimulus conditions. The
stimuli were constructed in four colors (red, yellow, blue,
green), four geometric shapes (circle, square, triangle,
rectangle), and four irregular shapes.

Contrary to Brian and Goodenough's finding that younger
children would make more color responses and older children
would make more form responses, the older group made signifi-
cantly more color responses than the younger group. This
result may be indicative of a later peaking period for the
color response tendency in this sample, seemingly attribut—
able to cultural influences.

The form attribute of irregularity was expected to be
unfamiliar to the children and therefore difficult to use in
matching. However, examination of the responses to irregular

stimuli indicated no significant difference between color or

41

42

form responses to irregular stimuli. This finding is con—
sistent with Clapp and Eichorn (1965) that the effect of
redundancy of figure was negligible. It was inferred that
children could have become sufficiently familiar with the
irregular forms to use them as well as geometric forms in
matching behavior.

Brian and Goodenough's finding (1929) that children
consistently matched three-dimensional forms more frequently
on the basis of color than they matched two—dimensional on
the basis of color was not replicated by the present study.

It was expected that children would possess more verbal
labels for both colors and geometric forms than for irregu-
lar forms. However, there were differences in assigning
labels to colors, geometric forms, and irregular forms only
with three-dimensional stimuli. A possible explanation was
conceptual grouping with respect to the familiarity component
of the stimuli. None of the differences observed in verbal
tags present was reflected by a corresponding difference in
number of color-form matches in corresponding conditions.

In the present study verbal tags had no significant effect

on matching behavior. There were no significant differences
in mean numbers of verbal tags for boys and girls. This
finding did not support the speculation of Kagan and Lemkin
(1961) that girls do more implicit verbal labelling than boys
and this labelling affects their matching behavior. However,
there is possible support for Suchman's (1966) inference that

perceptual discrimination skills are used only when the skill

43

is a component of some concept or organizer meaningful to

the individual. In order for the child to make use of a
verbal label in matching behavior that verbal label must have
meaning for him in his own conceptual organization.

Among the four form conditions the only significant
difference between number of color and form responses was on
two—dimensional geometric stimuli. This finding was con-
gruent with the results of previous investigations which used
only two-dimensional geometric stimuli.

The children in the present study exhibited more verbal
tags for colors than for irregular forms but not significantly
more verbal tags for colors than for geometric forms.

When the sample was divided into two groups of 15 boys
and 15 girls several differences in number of color and form
matches were evident. Girls showed a significantly greater
number of color than form responses throughout all four stim-
ulus conditions. However, boys showed a significantly greater
number of color than form responses on only the two-dimensional
geometric condition. On the remaining three stimulus condi—
tions the boys showed a significantly greater number of form
than color responses. Therefore, when there was a change in
the form attribute from geometric to irregular or two-
dimensional to three-dimensional, girls did not show any
observable change in matching behavior; however, boys reacted
by exhibiting a significantly greater number of form responses.
The sex differences were evaluated with respect to the verbal

tag hypothesis of Kagan and Lemkin (1961); however, differences

44

in matching behavior could not be explained by corresponding
differences in stated verbal tags.

Order of presentation of stimuli (two-dimensional then
three—dimensional or three-dimensional then two-dimensional)
did not produce a significant difference in frequency of
color and form matches. Therefore, the order of presentation
did not set the subjects to respond on either the color or
form dimension. There was no consistent color and form
matching behavior shown by sex—age group of subjects or under
any stimulus condition. There was no significant inter-
action between sex—age group and stimulus condition.

The present study lends support for Suchman's (1966)
conclusion that the shift from color to form dominance in
American children is culturally induced rather than the
result of a develOpmental process. No significant shift was
evident in the sample of children (3—5 to 5—5) who had not
been forced to attend to form difference in formal reading
and arithmetic instruction. Apparently, with pre—school
children change in dimensionality and regularity of the form
of the stimulus do not produce corresponding changes in match-
ing behavior. The findings of the present study are not con»
sistent with respect to dimensionality with those of Brian
and Goodenough's (1929) findings concerning dimensionality
but are consistent with Clapp and Eichorn's (1965) findings
concerning irregularity of form.

The sex difference between the reactions of boys and

girls in ratio of color to form matches under the four form

45

conditions was eXpected from the data of Kagan and Lemkin
(1961). Boys reacted to a change in stimulus form by show—
ing more form matches; girls did not change their ratio of
color to form matches. However, Kagan and Lemkin's explana-
tion of verbal labelling by girls was not supported by the
data collected on verbal tags.

In the absence of direct relationship between presence
of verbal labels and matching there is some support for the
findings of both Sigel (1954) and Suchman (1966) that in
conceptual processes meaning is the most salient dimension
over color, form, size, or ability to label verbally.

No clear sex-age stimulus interaction was evident. It
appears the color dominance is a complex mode of responding
which is characteristic of pre-school children (3-6 to 5-5).
Although there are differences in responding they cannot be
categorized consistently by any age group, sex, or stimulus
condition difference. The differences are not attributable
to the ability to state labels verbally for the stimuli.

In subsequent research, measurement of meaningfulness
of the verbal labels for the stimuli might be incorporated
into the color—form matching task, and the role of meaning-
fulness could be clarified. Important aspects of form which
cause a child to shift from one index of classification
(color or form) might be explored with individual children

rather than sex or age groups.

REFERENCES

46

REFERENCES

Aikawa, T. Some experimental studies on the "form-color"
problem; on the relation between instruction and its
response patterns. Jan. J. Psychol., 1964, 35,
70-81.

Berlyne, D. E. Conflict and information-theory variables
as determinants of human perceptual curiosity. J.
Exp. Psychol., 1957, 53, 399-404.

 

Brian, C. R. and Goodenough, F. L. The relative potency of
color and form perception at different ages. J. Exper.

Psychol., 1929, 12, 197-213.

Chen, L. and Wong, A. S. Color and form preferences. Acta
Psychologica Sinica, 1965, 32(3), 265-269.

 

 

Corah, N. L. Color and form in children's perceptual behav-
ior. Percept. and Mot. Skills, 1964, 18(1), 313—316.

 

Corah, N. L. The effect of instruction and performance set
on color-form perception in young children. J. Genet.

Psychol., 1966, 108, 351-356. (a)

Corah, N. L. The influence of some stimulus characteristics
on color and form perception in nursery-school child-
ren. Child Dev., 1966, 37, 205-211. (b)

 

Corah, N. L. and Gross, J. B. Hue, brightness and satura-
tion variables in color-form matching. Child Dev.,
1967, 38, 137-142.

 

Clapp, W. E. and Eichorn, Dorothy. Some determinants of
perceptual investigatory responses in children. J.
Exp. Child Psych., 1965, 2, 371-388.

 

 

Descoeudres, A. Couleur, forme ou nombre? Arch de Psychol.,
1914, 14, 305-341.

DeLanney, J. D. Perception de la form elle-meme. J.

Psychol. Norm. Pathol., 1962, 59(1—2), 59—74.-

Doehring, D. Color-form attitudes of deaf children. J. Sp.
g Hear. Res., 1960, 3(3), 242-248.

 

47

48

Guilford, J. P. Fundamental statistics in psychology and

education. New York: McGraw-Hill, 1956T7

 

 

 

Hays, W. L. Statistics for psychologists. New York:
Holt, Rhinehart, Winston, 1963.

 

Heidbreder, E. Toward a dynamic psychology of cognition.
Psychol. Rev., 1945, 52, 1-22.

Heidbreder, E. The attainment of concepts. (Trans.),
New York Acad. Sci., 1945, 7, 171-188.

 

Heidbreder, E. The attainment of concepts: I, Methodology
and terminology. J. Gen. Psychol., 1946, 35, 173-189.

 

Heidbreder, E. The attainment of concepts: III, The process.

J. Psychol., 1947, 24, 93-138.

Heidbreder, E. The attainment of concepts: IV, Regularities
and levels. J. Psychol., 1948, 25, 299-329.

Honkavara, S. A. A critical evaluation of the color and
form reaction, and disproving of the hypothesis con-
nected with it. J. Psychol., 1958, 46, 39-51.

Kagan, J. and Lemkin, Judith. Form, color, and size in
children's conceptual behavior. Child Dev., 1961,
32, 25-28.

 

Kerpleman, L. C. and Pollack, R. H. Developmental changes
in the location of form discrimination cues. Percept.
and Mot. Skills, 1964, 19(2), 375-382.

Minton, H. L. A replication of perceptual curiosity as a
function of stimulus complexity. J. Exp. Psychol.,
1963, 66, 522-524.

Sigel, I. The dominance of meaning. J. Genet. Psychol.,
1954, 85, 201-207.

Suchman, Roslyn G. and Trabasso, T. Stimulus preference and
cue function in young children's concept attainment.
J. Exp. Child Psych., 1966, 3, 188-189. (a)

 

Suchman, Roslyn G. and Trabasso, T. Color and form prefer-
ence in young children. J. Exp. Child Psych., 1966,
3, 177-187- (b)

 

 

Suchman, Roslyn G. Cultural differences in children's color
and form preferences. J. Soc. Psychol., 1966, 70,
3—10.

 

APPENDIX

49

50

Figure II.--Order of presentation to subjects--geometric

 

 

 

 

stimuli.
Circle Triangle Square Rectangle
rybs rybs rybs r‘ybe;
Order
1 L S R
2 R S L
L S R
L S R
5. R S L
6. L R S
7. S R L
8. R L S
9. R S L
10. S L R
11 R S L
12 L S R
Key:
S = stimulus r = red
L = left comparison y = yellow
R - right comparison b = blue
g = green

51

Figure III.-—Order of presentation to subjects--irregu1ar

 

 

 

 

stimuli.
Curvi- Curvi—
linear Straight linear Straight
Few- Few- Many- Many-
Sided Sided Sided Sided
r y b g r y b g r y b g r y b g
Order
13. L R S
14. S R L
15. L S R
16. R L S
17. R S L
18. R S L
19. L S R
20. R L S
21. S L R
22. S L R
23. S R L
24. L R S
Key
S = stimulus r = red
L = left comparison y = yellow
R = right comparison b = blue
g = green

52

Table ll.-—Color matches for four form stimuli distributed
by sex-age group.

 

Stimulus Condition

 

 

Sex-Age Group 2DG 2DI 3DG 3DI Total
Older Girls 2 8 8 l2
4 5 4 15
12 1 O O
12 12 2 1
12 12 12 12
7 9 10 4
11 12 11 12
Sum 60 59 47 56 222
Mean 8.6 8.4 6.7 8.0 7.9
Older Boys 9 l2 l2 12
12 12 12 12
12 12 12 9
11 6 1 1
1 O O O
0 0 4 3
12 12 6 11
10 4 O 0
Sum 67 58 47 48 220
Mean 8.4 7 2 5.9 6.0 6.9
Younger Girls 5 4 0 6
7 11 12 12
12 12 9 1
12 12 12 12
2 4 6 7
12 12 11 12
O O 10 12
O O 1 1
Sum 50 55 61 63 229
Mean 6.2 6.9 7.6 7.9 7 2
Younger Boys 1 0 0 l
5 4 2 8
6 0 0 0
l 0 4 1
O O 2 O
7 9 4 7
12 12 12 12
Sum 32 25 24 29 110
Mean 4.6 3.6 3.4 4.1 3.9
Sums for Stimuli 209 197 179 196 781
Means for Stimuli 7.0 6.5 5.9 6.5 6 5

 

53

 

NNN Na: NN :N NN am NN NHH. Na mmH_ . Hspoe
:m 3H NH o NH o m z m OH. mum z. pmH
N N: o NH H HH 0 NH H HH Nam a .HH
2 om m a m 0H m m m N. mum NH .MH
0 N: o NH o NH o NH o NH mum m .NH
N H: H HH N N o NH o NH mum 2 .HH
H: N N m N N NH o NH o Nnm 2 .0H
N: H NH o NH 0 NH. 0. HH H N.: z .N
NN NH HH H HH H N N H HH NI: 2 .N
HN NN HH . H OH .N o NH o NH HHus m .N
mm NH NH o NH 0 HH H o NH N-N m .N
m N: m N o NH o NH o NH an: 2 .m
o N: o NH o NH o NH o NH cum 2 .:
om NH N .N N a N N N a N-N N .N
N N: o NH o NH o NH m N NHua : .N
NH Nm 0 NH 3 N a N N NH Hum _m .H

 

Esom HOHoo Epom HOHoo Epom HOHoo atom HOHoo Show HOHoo ow< xmm monasz
Hmpoe HNstoHHH oHHpoEooo HmazonHH OHHpoEoou pomfinzm
HNCOHmcmEHQIomHEB HmconcoEHoloze

 

 

.dsopw HooHOIImuoonnzm mo ompznappmfio monopme EH09 ocm poHooII.mH oHnt

54

 

 

NNN Nmm NN NN NN NN NN ON NN NN Hopoe
o N: o NH o NH o NH. o NH. Num 2, .om
HN NN N N N : m N N N N-m : .NN
N: N HH H HH H NH o NH 0 Nu: m .NN
N: N NH o NH N NH o NH 0 N.: z .NN
NN NN o NH N OH NH o NH 0 N7: m .NN
H N: o NH H HH 0 NH o NH Num m .NN
N: N HH H N : NH O HH H Num 2 .:N
N: N NH o NH o NH. o N N N.: 2 .NN
NN NH N N N N w : 0H N Num m .NN
o ,N: o NH o NH o NH o NH 0.: m .HN
:H :m HH H m N o NH o NH Nu: m .oN
N N: o NH o NH H HH N N N.: m .NH
NN NH : N NH N N : N N Num 2 .NH
mm NH N N NH o w : N N HHum m .NH
N: N. HH H NH o NH 0 .HH H N.: 2 .NH
Epom HOHoo Epom HOHoo EHom HOHoo Epom HOHoo Epom HOHoo mwg xmm Honezz
Hmpoe HmHsmoHHH oHHpoEoou LNstoHHH oHHpoEoou poownzm

HNCOHmcoEHQIooHnB

HNCOHmcoEHmnozB

 

 

.QSOHw nowadoztlmpoanSN an oopanNpmHo monouwe Show ocm H0H00I|.mH oHnme

"I7'1111111114144