- Mlllll\llllllllllllllllfllllll

 

 

 

I
THEE“
V”, 3 1293 01074 2033
1."? IL":- fh when
325;. .- - my; z!
i T i 4. 4-31.
3 ,- use
i 's" . ‘ . n.
1 "v 3.; $255" 1. a;
5'- "? ~7Frr'r"~'w"'rn-'uv we m f
This is to certify that the

dissertation entitled

THE RELATIONSHIP BETWEEN FIELD DEPENDENCE/INDEPENDENCE
AND CROSS-MODAL SENSORY TRANSFER IN KINDERGARTEN

AND FountyfiggflggbgnlLoREN

Margaret Elizabeth Griffin

has been accepted towards fulfillment
of the requirements for

Doctor of_P_b_i_l_Qs_Qphy_degreein EamflLanthild Ecology

'::Z:;H¢a£917 ‘//62£€£214«u»"

Major professor

Date 40¢?”ny / 2 I. / 75' 2

MS U Lt an Affirmative Action/Equal Opportunity Institution 0-12771

 

 

MSU

RETURNING MATERIALS:

 

 

 

Place in book drop to
remove this checkout from

 

LIBRARIES .
‘12—- your record. FINES Will
be charged if book is
returned after the date
stamped below.
‘32
{EC 0 l. 3.7.33

\O'FV

Cl?

 

 

THE RELATIONSHIP BETWEEN FIELD DEPENDENCE/INDEPENDENCE
AND CROSS-MODAL SENSORY TRANSFER IN KINDERGARTEN
AND FOURTH-GRADE CHILDREN

By

Margaret Elizabeth Griffin

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Family and Child Ecology

1982

m4

xi."

C3\\13<e.3

ABSTRACT

THE RELATIONSHIP BETWEEN FIELD DEPENDENCE/INDEPENDENCE
AND CROSS-MODAL SENSORY TRANSFER IN KINDERGARTEN
AND FOURTH—GRADE CHILDREN

By
Margaret Elizabeth Griffin

The focus of this study was the relationship between field
dependence/independence as measured by the Children's Embedded Figures
Test (CEFT) and cross-modal transfer ability as measured by Cross-modal
Sensory Transfer Tasks l, 2, 3, and Combined Scores. Subjects were
60 children from Clayton County, Georgia, public school system, 30 of
whom were enrolled in kindergarten (15 boys, 15 girls) and 30 in fourth
grade (15 boys, l5 girls). The children's scores on both measures were
compared using Pearson correlations, and a MANOVA procedure was used to
test the effects of grade and gender of subjects on the relationship
between the two measures. Data analysis revealed that there was a
positive and significant relationship between the two measures and that
there was a significant main effect for grade of subject but not for
gender of subject.

The purpose for comparing a measure of field dependence/
independence with a measure of cross-modal transfer was to help to
establish an empirical base for the notion that both abilities are
unanifestations of an underlying function, the ability to isolate and

use essential features of an object. Ability to isolate essential

Margaret Elizabeth Griffin

features has been related to numerous cognitive and personality traits
and has been shown to affect family functioning, group functioning,

and career choice. An understanding of the interrelationships of meas-
ures of ability to isolate essential features and the process of
development of the ability has implications for the study of the
individual, the family, the classroom, and the structure of various

social groups.

ACKNOWLEDGMENTS

The author would like to express appreciation to Dr. Lillian
Phenice, the director of this dissertation research. Her friendship,
advice, and encouragement were invaluable resources. The author would
also like to express thanks to Dr. Linda Nelson, whose kind vigi-
lance and keen eye caught numerous errors and oversights. Apprecia-
tion is also expressed to Dr. Donald Melcer, committee chairperson,
and to Dr. Marjorie Kostelnik and Dr. Lauren Harris, committee members.

Funding for the data analysis was provided by the Department of
Family and Child Ecology, Dr. Eileen Earhart, chairperson. The author
also received support, both financial and personal, through assistant-
ships with the Department of Family and Child Ecology (directed by
Dr. Anne Soderman) and Michigan 4-H-Youth Programs (directed by
Dr. Leah Hoopfer).

Appreciation is expressed to the Clayton County, Georgia, School
Board and Lake Harbin Elementary School staff and students for their
participation in the project. Particular thanks is given to Patricia
Griffin, a teacher at Lake Harbin and my sister, for her efforts in
coordinating the informed-consent process.

Personal thanks is expressed to Corinne Diller and Mary Jane
Flanagan, whose support and care throughout this project have nurtured

the author through injury and health. To Dr. and Mrs. S. A. Griffin,

ii

Mr. and Mrs. C. T. Hammill, and Mr. and Mrs. w. B. Maples, Jr.,
special thanks for the love and tolerance that only a family can

provide.

iii

TABLE OF CONTENTS

LIST OF TABLES .........................
LIST OF FIGURES .........................

Chapter
I. INTRODUCTION ......................

Introduction .....................
Conceptual Framework .................
Rationale .......................
Purpose of the Study .................
Statement of Research Questions and Hypotheses .

Research Questions .................

Hypotheses .....................
Definitions of Terms .................
Assumptions ......................

II. REVIEW OF LITERATURE ..................

Field Dependence/Independence .............
Types of Tests to Measure Field Dependence/
Independence ..................
Research on Children and Field Dependence/
Independence ...................
Age and Field Dependence/Independence ........
Gender and Field Dependence/Independence ......
Relationship of Field Dependence/Independence
and Schooling ...................
Cross-modal Sensory Transfer Ability .........
Theories of Cross-modal Transfer ..........
Research on Children and Cross-modal Transfer . . . .
Training for Cross-modal Transfer ..........
Mode of Primary Presentation ............
Relationship of Cross-modal Ability to Cognitive
Functioning and Schooling .............
Age and Cross-modal Transfer ............
Relationships of Perceptual Style and Cross-modal
Transfer Ability to Other Perceptual and Cognitive
Abilities ......................

iv

Page
vi

vii

III. RESEARCH DESIGN AND PROCEDURES ............

Introduction ....................
Description of the Sample ..............
Instrument Selection and Description ........
Procedure ......................
Ethical Considerations ...............
Data Analysis ....................
Summary .......................

IV. ANALYSIS OF RESULTS . . . . . . . . . .........

Hypothesis 1 ....................
Hypothesis 2 ....................
Hypothesis 3 ....................
Hypothesis 4 ....................
Multivariate Analysis of Variance for Hypotheses 5
and 6 .......................
Hypothesis 5 ....................
Hypothesis 6 ....................
Discussion of Results ................

V. CONCLUSIONS, IMPLICATIONS, AND SUMMARY ........

APPENDICES

A. IN
IN

CH

I'T‘IUOW
000.

Review of the Problem ................
Conclusions .....................
Personal Observations ................
Limitations of the Study ..............
Generalizability .................
Instruments . . . . . . . . . . . . . . . .....
Additional Data on Subjects ............
Assessment of Developmental Trend .........
Implications of the Study ..............
Suggestions for Future Research ...........
Summary .......................

FORMATION TO CLAYTON COUNTY SCHOOL BOARD ......
FORMATION SENT TO SUBJECTS' PARENTS .........

CROSS-MODAL SENSORY TRANSFER TASKS ..........

ILDREN'S EMBEDDED FIGURES TEST ...........

DATA RECORDS .....................

F. CODE BOOK .......................

REFERENCES

Table
4.l.

LIST OF TABLES

Means, Ranges, Variances, and Standard Deviations for
Cross-modal Sensory Transfer Tasks l, 2, 3, and
Combined Scores and Children's Embedded Figures Tests

Pearson Correlation Coefficients for Comparison of

Task l With Task 2 ...................

Pearson Correlation Coefficients for Comparison of

Task l With Task 3 ...................

Pearson Correlation Coefficients for Comparison of

Task 2 With Task 3 ...................

Pearson Correlation Coefficients for Comparison of
Cross-modal Sensory Transfer Tasks l, 2, 3, and
Combined Scores With Children's Embedded Figures

Test ..........................

Multivariate and Univariate Analyses of Grade and
Gender of Subject for Children's Embedded Figures
Tests Scores and Cross-modal Sensory Transfer

Tasks Combined Scores .................

Multivariate and Univariate Analyses of Grade of
Subject for Children's Embedded Figures Test and

Cross-modal Sensory Transfer Tasks Combined Scores . . .

Multivariate and Univariate Analyses of Gender of
Subject for Children's Embedded Figures Test and

Cross-modal Sensory Transfer Tasks Combined Scores . . .

Sample Means and Norming-Data Means for Children's

Embedded Figures Test ................

vi

Page

47

48

49

50

51

53

54

55

63

LIST OF FIGURES

Figure Page

3.l. Design of the Study ................... 37

vii

CHAPTER I

INTRODUCTION

Introduction

 

Perception in human beings is basic not only to the under-
standing of the individual, but also to the understanding of human
social groupings. Perception can be studied as an information-
processing mechanism by which humans gather and translate messages
about the physical and social environment and an intermediate or
transitional stage of information transfer to the cognitive domain.
The perceptual process at its inception is the permeable boundary or
line of demarcation at which the individual human system interfaces
with its physical and social world. As the perceptual systems
interact with the cognitive systems within the human mind, a feedback
system develops by which the individual may selectively scan the
environment and use information collected to build a store of knowl-
edge about the world.

Humans obtain information about the environment through
various sensory modalities such as skin (pressure and heat), eyes
(light and pressure), ears (vibration and pressure), nose (chemicals
and pressure), tongue (pressure and chemicals), and a variety of
proprioceptive cues from muscles, eyes, and ears. Most information

comes to us as a multi-sensory experience. Eating, as an example,

often includes stimulation of an array of sensory organs. Humans,
particularly when they are young, often explore the world in a multi-
sensory fashion-—tasting, smelling, touching, looking, and listening.
Piaget (1969) and other developmental theorists have postulated that
this multi-sensory exploration process, which combines sensory infor-
mation with muscle movement, is basic to the development of the
earliest thought patterns in humans. It is on these patterns that
later thought and information-processing ability builds.

Piaget (1969) stated that early in the sensory-motor exploration
of the world, a child begins to create a sense of "thingness" out of
a series of explorations of any given object. The child, in effect,
has isolated the essential features of that object that carry across
two or more perceptual modes. As the child develops further, language
labels for objects allow the child to have shared meanings with others
for the essential "thingness" of an object. These language labels
incorporate not only the individual's sense of “thingness,” but also
impose familial and cultural meaning onto the individual's unique
interpretation of sensory-motor experience. The blending of indi-
vidual and cultural interpretation of experience via the use of
language is one of the bases for both similarity of reported percep-
tual experience across persons within a social group and individual
differences (Gibson, 1967). This blending is one of many factors that
adds to the difficulty of studying the nature of essential perceptual
features ("thingness") of an object in a group of persons who are
somewhat sophisticated in the use of a common language. The interac-

tion between cultural meanings/language and individual meaning is

often impossible to untangle. Therefore, it is important for students
of human development to understand as much as possible about the early
development of the various cognitive and perceptual processes that
underlie and support the ability to know the essential "thingness"

and isolate its features in order to be able to begin to determine
what facets of our perceptual and cognitive ability are inherent and

what humans must learn to develop and to understand the world.

Conceptual Framework

 

Humans live as individuals, members of a family, and members of
social groups. Depending on the context of the situation in which a
given person may be operating, perception of information about his/her
and others' actions upon and reactions to the environment may differ.
Information based on individual perception and experience is highly
idiosyncratic and not easily communicated to others. Actions and
meanings of those actions within a family group or social group,
although not devoid of individual meaning, reflect more mutually
accepted and understood interpretations. It is the similarity of
understanding of perceptual information that is a factor in binding
the group together into a unit that has an identity of its own beyond
the individual identities of its members. This unit may be a family,
a peer group, or a class of children, among others.

Study of similarities in perceptual ability and interpretation
of perceptual information have revealed that there is, indeed, some
evidence of perceptual similarity among family members (Witkin &

Goodenough, 1981) and among social-class groups (Derevensky, 1977).

It is known that the type of perceptual style of members affects

group structure and function. For example, a number of studies have
shown that some people are more analytic (as opposed to global) in
their interpretation of perceptual stimuli but are less adept in using
information obtained from others (Busch & DeRidder, 1973; Linton,
1955). People who use an analytic perceptual style are also more
autonomous in their interpersonal relationships and are more respon-
sible, self-reliant, and able to think for themselves than those

who use a more global or nonarticulated style of perception (Witkin &
Goodenough, 1977).

The perceptual and cognitive style labeled field dependence/
independence is a part of the global versus articulated or analytic
style. Field independence (analytic) is the ability to cognitively
restructure information from a perceptual experience and manifests
itself in an ability to separate objects from context and align
oneself to the vertical (Witkin & Goodenough, 1981). Field dependence
(global) is manifested in the inability to disembed or align with the
vertical and relates to dependence on social and other external
referents (Witkin & Goodenough, 1981).

In a nursery school setting, it was observed that children who
adapted most easily to school and had the least problem separating
from parents were also those who rated most highly on tests of
perception involving disembedding, a factor in the articulated (field
independent) perceptual style (Olesker, 1978). The more analytic a
person's perceptual style, the more at ease that person will be in a

large, unstructured group setting (Steingart et al., 1975). In group

settings, those groups that have persons manifesting a global per-
ceptual style are more effective in conflict resolution than those
groups without global-style members (Oltman et al., 1975).

Research also showed that for both educational and occupational
choice, the type of perceptual style influenced the choice made by
individuals. Global-style persons tend toward human—services work,
and articulated-style persons tend toward careers emphasizing inde-
pendence in work habits (Witkin et al., 1977). For example, psychi-
atric nurses tended to be more global in perceptual style, while
surgical nurses were more analytic (Quinlan & Blatt, 1972). Pilots
(Cullen, Harper, & Kidera, 1969) and engineers (Barrett & Thornton,
1967) tend to be more articulated in perceptual style, while writers
(Witkin & Goodenough, 1977) tend toward a more global style.

Within the family unit, use of global versus articulated per-
ceptual style also has implications for variation in functioning.
For example, in early studies of perceptual and cognitive functioning,
Witkin and his colleagues (1962) speculated that child-rearing prac-
tices affected the development of perceptual and cognitive styles.
Specifically, they suggested that parents who encouraged autonomous
functioning in their children were also promoting development of a
more articulated style (related to field independence), while more
authoritarian child-rearing practices led to a more global style
(related to field dependence) in the children.

Research involving mothers and sons (Dyk & Witkin, 1965) sup-
ported the early hypotheses about relationships between cognitive and

perceptual style and child-rearing practices. This research also

revealed that parents of global-style children also used more severe
methods of training and disciplining their children than did parents
of analytic-style children. Mothers of children who demonstrated

low disembedding ability gave commands to their children more often
than mothers of children with high disembedding ability (Busse, 1969).
Children with low disembedding ability had parents who were uninvolved
in training their children or did not allow the children to help them-
selves in learning situations (Busse, 1969).

In homes where family interactions are strongly dominated by
parents, children tend to be more global in perceptual style (Witkin &
Goodenough, 1981). Another study revealed that mothers of field-
dependent children were observed to be nonspecific and repetitive in
the comforting of their children during infancy, which was said to
hinder the development in the child of ability to discriminate among
needs and feelings (Dyk, 1969).

The teaching of sex-roles is a primary function of the family,
especially parents.

To the extent that children identify more with the same-sex

than with the opposite-sex parent, sex-role modeling may be

a major factor in the development of sex differences in field

dependence-independence. If so, children (especially boys)

of fathers who do not enter actively into the child-rearing

process are likely to be relatively field dependent, and

children (especially girls) of mothers who do not enter actively
into the child-rearing process are likely to be field indepen-

dent. (Witkin & Goodenough, 1981, p. 87)

For girls, the greater the influence of the father on family decisions,
the more field independent they were, while the greater the influence

of the mother on family decisions, the more field dependent they were

(MacEachron & Gruenfeld, 1978). For boys, families with strong

paternal involvement tend to have sons who are field independent
(Busse, 1969). Father absence has been associated with greater field
dependence for both boys and girls, while the evidence on mother
absence is inconsistent (Schooler, 1972). According to a study by
Holtzman, Diaz-Guerrero, and Swartz (1975), the nuclear-family struc-
ture in the United States tends to produce greater field independence
than does the extended family.
The cognitive development of individuals is influenced by their
ability to disembed.
Although field-dependent and field-independent people do not
appear to differ with regard to immediate percept induced by
most stimuli, field—independent people seem better able to
achieve a different percept--when required to do so by situa-
tional demands or inner needs—-through the restructuring of
their initial perceptual experience, at least with respect to
spatial-configurational material. In contrast, among field-
dependent people the prevailing organization of the perceptual
field is likely to be adhered to as given. A difference in
restructuring ability between field-dependent and field-
independent people, similar to that observed in their perceptual
activities, is also evident in their intellectual functioning.
(Witkin & Goodenough, 1981, pp. 23-24)
Research has demonstrated that field-independent people perform
better than field-dependent people on tests of speed of closure
(Gough & Olton, 1972), conservation (Piaget, 1969), representation of
the horizon (Willemsen et al., 1973), and spatial-visualization tasks
(decentration) (Gough & Olton, 1972). Ability to break a mental set
(Busse, 1968) and ability to use a hypothesis-testing approach in
visual-discrimination tasks (Nebelkopf & Dreyer, 1973) are also
related to field independence. Field independence has been positively

correlated with spatial and performance items on IQ tests such as the

WISC and WAIS but has not been found to correlate with verbal items
(Witkin & Goodenough, 1981).

Field dependence/independence in cognitive and perceptual func-
tioning has been demonstrated to relate to a wide variety of char-
acteristics. A number of personality factors have also been studied
in relation to field dependence/independence. One of the major
differences between field-independent people and field-dependent
people is their relationship to others. Witkin (1978) reported that
field-independent pe0ple are able to function more autonomously of
others. Field-dependent persons are more attentive to social cues
and tend to watch faces more often and more closely than do field
independents (Konstadt & Forman, 1965). Words with social and emo-
tional connotations are more salient to field-dependent persons
(Minard & Mooney, 1969). Field-dependent persons also prefer to be
physically near the persons with whom they are interacting (Witkin,
1977). In contrast, field-independent children spend more time in
solitary play (Witkin & Goodenough, 1981). It was more difficult for
field-dependent persons to display hostile or aggressive feelings
(Dyk & Witkin, 1965).

Research with both clinical and normal populations has revealed
that the global versus articulated style of perceiving is related to
a number of factors associated with psychological development. Witkin
(1965) used figure drawings as a means to assess articulation of body
concept and found that field-independent children produced drawings
that indicated a much more highly developed sense of body concept.

The drawings were more realistic and more detailed. Persons who use

more highly specialized defense mechanisms such as isolation are
generally field independent (Witkin et al., 1962). Field dependence
(global) seems to be related to a number of pathologies such as
alcoholism (Karp et al., 1965), obesity (Karp & Cardes, 1965),
asthmatic children (Fishbein, 1963), enuretic children (Scallon &
Herron, l969),cardiaccfisorders ($011, 1963), hysteria (Zukman, in
Witkin et al., 1971), and catatonics (Janucci, in Witkin et al., 1971).
The Witkin group (1971) reported that types of pathology common to
field-independent (articulated style) persons included delusions,
expansive and euphoric ideas of grandeur, outward direction of aggres-
sion, paranoia, obsessive-compulsive characteristics, and bizarre
attempts to maintain identity.

Based on the information presented above, it is obvious that the
presence of the perceptual and cognitive style labeled field dependence/
independence in each individual has far-reaching consequences for the
development of that individual as a unique human being, a family
member, and a member of various social groups. A vast array of human
functions, such as cognitive, social, and personality, develop in
relation to field dependence/independence. A change in one function
signals a change in many other functions.

Because the field-dependent and field-independent cognitive

styles are different in their adaptive consequences, it need

not be surprising that people develop cognitive styles which

are adaptive to the demands of the life situations with which

they must cope. Conversely, cognitive styles have been shown to

cause people to gravitate toward life situations to which their
styles are suited. Achievement of adaptive correspondence
between cognitive styles and life circumstances is thus a two-

way street. Moreover, reflecting their salience as a develop-
mental force, cognitive styles guide the formation of modes of

10

behavior in people which are compatible with their styles.

And, finally, a person's cognitive style may influence others

to behave toward him in a manner which suits his cognitive style

and is thereby helpful to him in his interpersonal relations.

(Witkin, 1978, p. 30)

One of the other perceptual abilities to which the field
dependence/independence trait is related is cross-modal sensory
transfer. This is the ability to use information gained from one
modality to make a decision about information gained from a second
modality. The ability to disembed visually, one factor of field
dependence/independence, is related quite closely to both tactile and
auditory disembedding and transfer (Witkin et al., 1971). Gibson
(1966), a primary researcher in the field of cross-modal transfer,
has pointed out that the understanding of those mechanisms by which
perceptual information can be transferred between perceptual subsys-
tems is essential to a basic knowledge of learning, attention, and
perception itself. Gibson's theory of cross-modal transfer holds
that information acquired via one sensory modality

is not stored exclusively in a modality-specific form or

"place." Instead, information that is retained by the nervous

system is made "supramodal"; that is, it is transformed into a

configuration or other state that is not specific to the char-

acteristics of any single subsystem. Essential properties and
configural relations in the information are extracted, schema-
tized, and held in schematic or abstract form. (Abravanel,

in Walk & Pick, 1981, p. 72)

This ability to isolate an essential property or relation is hypothe-
sized as being much the same as the ability to disembed found in
field-independent people. The lack of ability to organize and
restructure perceptual information seen in field-dependent people may

be related to relative inability to transfer essential features and

ll

relationships in a cross-modal manner. As Abravanel (1981) indicated,

In large measure, perceptual development consists of learning

to attend to the most relevant features and relations of objects

< events; of learning how to explore in order to canvass global

information and to pick up more precise data. As these func-
tions of perception develop, the organism becomes a more pre—
cise, astute, and specialized perceiver who is capable of making

more differentiated and accurate judgments. (p. 103)

The development of cognition is said to parallel the development
of perception. Piaget (1969) has postulated that perception and cog-
nition are interrelated processes that work in tandem to assist the
individual in achieving an understanding of a given experience.
Piaget called this process equilibration. Just as Piaget's levels
of cognitive development are organized in a hierarchical fashion, so
are the accompanying perceptual stages. The interaction between
perception and cognition leads to intellectual development in the
individual. Piaget (1969) noted that the study of interrelationships
between perceptual and cognitive phenomena and the changes in the
nature of those interrelationships across development of individuals
was crucial to the understanding of the development of the intellect.

Therefore, to study the nature of the structures that the indi-
vidual perceptual and cognitive system creates to deal with early
multi-sensory experiences with objects, it is wise to choose children
as subjects for study. Language as a tool for communication and
mental organization exerts the least influence on young children who
still rely heavily on individual sensory-motor schemes to organize
their responses to the world (Piaget, 1969). However, it is also

difficult to communicate lengthy instructions for the purpose of

eliciting a standard response to children between birth and 4 years

12

of age. Early-elementary—school-aged children have attained a level
of language development that is sufficient to allow them to follow
instructions, yet they have not lost the idiosyncratic language asso-
ciated with early childhood. Kindergarten-aged children, in general,
are at a point in their development during which they are making the
fullest possible use of sensory-motor information combined with the
idiosyncratic use of cultural labels (end of preoperational stage)
(Piaget, 1969).

By fourth grade, most children are able to represent mentally
an event in the absence of a real-world referent, although they are not
able to deal with the abstract (concrete operations) (Piaget, 1969).
They are able to use cultural labeling systems to generate names for
categories of objects and are able to articulate the essential "thing-
ness" or features of an object (Gibson, 1967). Their cognitive abili-
ties are qualitatively different from the kindergarteners', but they
have not yet reached the sophistication of adult thought.

Due to the above-mentioned characteristics of the kindergarten
and fourth-grade age groups, these two groups of children seem to
represent populations that demonstrate age-group characteristics
that bear directly on the study of the development of perception of
essential features of objects. The examination of differences between
these groups in their ability to isolate and manipulate "thingness"
of an object will also shed light on the process of development of
this perceptual and cognitive ability.

The framework upon which this study is designed is a combina-

tion of the notions about perceptual development put forth by the

l3

Witkin group, Gibson, and Piaget. The author is seeking to estab-
lish (a) that field dependence/independence and cross-modal transfer
ability in school-aged children are related abilities based on the
idea that both abilities depend on the skill with which an indi-
vidual can isolate and use essential features of an object or event,
and (b) that the development of both abilities parallels that of other
types of cognitive and perceptual processes as set forth in the

developmental stages of Piaget.

Rationale

For a number of years, researchers have been intrigued by the
ability of most humans to generalize from perceptual experience and
use that generalization to solve a second perceptual problem, i.e.,
to use the ability to isolate essential features of an object. Witkin
and his colleagues (1965) have studied a perceptual phenomenon they
called field dependence/independence. One of the abilities in the
cluster associated with field dependence/independence is the ability
to separate object from field or isolate an object from its context.
This type of behavior is measured by one of several "embedded figures
tests" developed by the Witkin group. A second perceptual phenomenon,
labeled cross-modal transfer, has been studied by a number of research-
ers including Gibson (1967). Cross-modal transfer involves the ability
to experience an object with any single sensory modality and subse-
quently to identify that object using a second sensory modality. It
appears that both embedded figures tests and cross-modal sensory

transfer tasks are tapping the same perceptual and cognitive

14

process-~that of isolating the essential features of an object and
using them to judge other perceptual stimuli as same or different
(Witkin et al., 1971). Therefore, it seems logical to use these two
well-documented processes for tapping the perceptual and cognitive
system in tandem to study the development of the ability to isolate

and use essential features of objects.

Purpose of the Study'

 

The purpose of this study was threefold. First, the researcher
examined the relationship between field dependence/independence and
cross-modal sensory transfer using a comparison of scores of subjects
on tests of each trait. Two questions at this point were: did a
relationship between the two traits exist and if so, what was the
nature of the relationship? As a relationship did exist between the
two traits, a second purpose of the study was to examine the possi-
bility of a developmental trend in the relationship by comparing
age-group scores on the measures of both traits. The possibility
of a developmental trend was suggested by previous research in
which perceptual ability increased with age (Kleinman, 1979; Witkin
et al., 1967). Also based on previous research, a third focus of the
study was designed to examine the possibility of any gender-related
differences that might appear across age-groups, within age-groups,
across measures of traits, and/or within the relationship between the
two sets of measures, as gender differences in ability in field
dependence/independence and cross-modal sensory transfer have been

found in some populations (Vaught, 1965).

15

Statement of Research Questions and Hypotheses

The primary purpose of this study was to examine the relation-
ships and nature of those relationships between scores on measures
of field dependence/independence and cross-modal sensory transfer
for kindergarteners and fourth graders. Research in both areas
seemed to indicate that the mechanism allowing both processes to take
place represents a level of ability to discriminate the essential
features of an object (Gibson, 1967; Witkin, 1965).

Questions raised by this type of inquiry were: (a) speculation
as to the relationship(s) between measures of performance on tests of
field dependence/independence and cross-modal sensory transfer ability,
(b) the relationship of age and/or gender of subjects to scores on
measures of ability on tests of field dependence/independence and
cross-modal sensory transfer, and (c) the relationship of age and/or
gender of subjects to the suspected relationship between cross-modal

sensory transfer ability and field dependence/independence.

Researchgguestions

 

The following three questions represent the main directions of
the study.

1. Are scores of kindergarten and fourth-grade children on tests
of field dependence/independence and tactile-visual cross-modal sensory
transfer, as measured by the Children's Embedded Figures Test and
three Cross-Modal Sensory Transfer Tasks, respectively, related,

and if so, what is the nature of that relationship?

2.

16

Is there a developmental trend in the above-mentioned scores

and/or their relationships, and if so, what is the direction of that

trend?

3.

Is gender of subject a factor in the scores obtained on

measures of either or both field dependence/independence and tactile

and visual cross-modal sensory transfer or in the relationship between

scores?

Hypotheses

 

The following hypotheses were generated to guide the direction

of the study.

1.

Scores for Cross-modal Sensory Transfer Task 1 (visual-to-
visual matching) are positively related to scores for Cross-
modal Sensory Transfer Task 2 (visual-to-tactile matching).

Scores for Cross-modal Sensory Transfer Task 1 (visual-to-
visual matching) are positively related to scores for Cross-
modal Sensory Transfer Task 3 (tactile-to-visual matching).

Scores for Cross-modal Sensory Transfer Task 2 (visual-to-
tactile matching) are positively related to scores for Cross-
modal Sensory Transfer Task 3 (tactile-to-visual matching).

Scores for the Children's Embedded Figures Test are posi-

tively related to:

a. Cross-modal Sensory Transfer Task 1 (visual-to-visual
matching) scores

b. Cross-modal Sensory Transfer Task 2 (visual-to-tactile
matching) scores

c. Cross-modal Sensory Transfer Task 3 (tactile-to-visual
matching) scores

d. Cross-modal Sensory Transfer Combined Tasks scores

Fourth graders, as a group, will perform better than kinder-
garteners on:

a. Cross-modal Sensory Transfer Combined Tasks scores

b. Children's Embedded Figures Tests scores

Males, as a group, will perform better than females on:
a. Cross-modal Sensory Transfer Combined Tasks scores
b. Children's Embedded Figures Tests scores

17

Definitions of Terms

 

Field dependence/independence--a hypothetical continuum of
behavior used to label skill in the ability to separate objects from
context or to see objects as discrete from background. Field depend-
ence is the lesser ability. Field independence is the greater ability.

Tactile-visual (visual-tactile) cross-modal sensory transfer--
the process involved in the intake of information about an object
tactilely (visually) and the use of that information to make a deci-
sion about the object using the visual (tactile) mode.

Perception--the process by which information about the environ-

 

ment is moved through the human nervous system. This process is
initiated in the sensory organs and combines with previously acquired
mental structures in the brain to produce awareness of the environ-
ment.

Proprioceptive,perception--reaction to stimuli originating within

 

the organism (e.g., muscle tension, balance).

Assumptions

 

The following assumptions gave implicit direction to the study.

1. Perception and cognition are interrelated processes.

2. Perception is one of the primary mechanisms of information
exchange for the human system.

3. The two instruments (Children's Embedded Figures Test and
the three Cross-modal Sensory Transfer Tasks) chosen to measure
perceptual and cognitive ability do, actually, measure those abili-

ties.

18

4. The kindergarten children in the sample were actually
operating on a preoperational level of cognitive development.
5. The fourth graders in the sample were actually operating on

a concrete operational level of cognitive development.

CHAPTER II

REVIEW OF LITERATURE

This review consists of three parts. The first is a description
of the construct of field dependence/independence and a summary of
selected research dealing with a particular test (Children's Embedded
Figures Test), which is purported to measure children's capacity to
disembed objects from context, one of the facets of field dependence/
independence. The second section covers research dealing with ability
of children to perform Cross-modal Sensory Transfer Tasks involving
visual and tactile sensory modalities. The third section of this
review presents research and commentary concerned with the interrela-
tionships of perceptual abilities.

Research presented in this review deals specifically with those
areas of research in field dependence/independence and cross-modal
transfer that bear directly upon the study. The following review of
literature represents an attempt to present background information
about the two measures (Children's Embedded Figures Test and Cross-
modal Sensory Transfer Tasks l, 2, and 3 and Combined) used in this
study, particularly the use of the measures with early-elementary-aged
children and the results of those studies. Previous research involving
the use of these two measures has provided direction in developing both

the research questions and the design of this study. Conclusions of

19

20

previous research into cognitive and perceptual development led to
the use of the Children's Embedded Figures Test and the Cross-modal
Sensory Transfer Tasks l, 2, and 3 and combined in tandem to explore

interrelationships in cognitive and perceptual functioning.

Field Dependence/Independence

In 1954, Witkin and his colleagues introduced the field
dependence/independence dimension of cognitive and perceptual style
as part of a theory of individual differences. The primary interest
of the Witkin group (1954) was in the psychology of perception, par-
ticularly in distinguishing between global and analytic ways of
perceiving objects and events. The field-independent person, as
described by Witkin, is likely to be analytic and tends to separate
elements from background or context. Such persons also are able to
relate to the physical upright despite a tilt in the field or in body
orientation. The field-dependent person tends toward a global approach,
attending to wholes rather than parts and being more sensitive to social
cues. Witkin defined the field dependence/independence continuum as
"an analytic, in contrast to global way of perceiving [which] entails
a tendency to experience items as discrete from their backgrounds and
reflects ability to overcome the influence of an embedding context"
(Cross, 1976, p. 114).

There are two primary foci forresearchinvolving the ability
to disembed: "(1) the relation between disembedding ability in per-
ception and disembedding ability in intellectual functioning; and

(2) the relation between disembedding ability and structuring

21

ability" (Witkin & Goodenough, 1981, p. 17). The ability to disembed
was seen as relating to cognitive tasks in which the problem to be
solved involved separation of elements from a context. It was also
seen to relate to perceptual tasks. Field-dependent subjects also had
difficulty with "that particular class of problems in which the solu-
tion depends on taking an element critical for solution out of the
context in which it is presented and restructuring the problem material
so that the element is now used in another context" (Witkin & Good-
enough, 1981, p. 17). This evidence seems to indicate that the
ability to disembed spans both perceptual and cognitive domains in

an individual.

The relationship between disembedding and structuring is based
on the hypothesis that there is a tendency to deal in a more active
or more passive way with the field in each individual (Witkin &
Goodenough, 1981). The field-independent person would demonstrate
ability to impose a structure on a field that lacked an inherent
organization, while the field-dependent person would experience lack
of organization.

Types of Tests to Measure
Field Dependence/Independence

Three types of tests were developed by Witkin and his colleagues
(1954) to measure the field dependence/independence dimension. The
common element in all the tests is the extent to which people are
influenced by the surrounding visual field (Cross, 1976). The tests
are the Body Adjustment Test (BAT), the Rod—and-Frame Test (RFT), and
the Embedded Figures Test (EFT). The BAT and RFT involve alignment

22

of the body and the alignment of a luminous rod, respectively, to

the vertical. The EFT involves the isolation of a simple geometric
form from a complex one. The first two tests (BAT and RFT) involve
the use of both proprioceptive and visual cues, while the EFT involves
only the use of visual cues. Other researchers have studied field
dependence/independence using various other sensory modalities than
those studied by the Witkin group and have found that field inde-
pendents are able to isolate melody line from a complex composition
and that field dependents are more aware of social cues (Cross, 1976).

Several other researchers have developed tests based on figure/
ground relationships. Frostig and her colleagues developed the
Marianne Frostig Developmental Test of Perception in the early 19605
as a battery to test perceptual skills of children ages 3-8 years
and as a diagnostic screening test (Maslow, Frostig, Lefever, &
Whittlesey, 1964). Subtest Z, Figure-Ground, is the only subtest of
the five included in the battery that relates directly to ability to
disembed. Reliability of the test is low, cited in the .30-.50
range, and validity is moderate with correlations of .40-.50 to
reading scores (Buros, 1972).

Another test, the Southern California Figure-Ground Visual Per-
ception Test, was developed by Ayres in 1966 to test visual percep-
tion of 4-10 year olds of a figure against a background (Ayres, 1966).
The test consists of 18 paired items, 9 overlapping-outlines pairs,
and 9 embedded-figures pairs. Reliability of the test is low, at

.37 to .52 on test-retest evaluations (Buros, 1972).

 

23

The Children's Embedded Figures Test (CEFT) used in this study
was developed by Karp and Konstadt in 1963 and revised in 1971. It
consists of two subtests, one with a "tent" as the stimulus figure
(11 items) and one with a "house" as the stimulus figure (14 items).
The score for the test is the total of the two subtests. (See Chap-
ter III, Instruments.) The test stimuli are cardboard geometric
shapes and geometric drawings in which the child must locate the
embedded shape that corresponds to the test stimuli. The CEFT is
administered individually and is designed for children between the
ages of 5 and 12 years. The CEFT is the most widely used of the
embedded-figures tests for children due to handy construction (small
cards as opposed to puzzles) and ease of transportation of materials.

Several children's versions of the Embedded Figures Test have
been developed. Goodenough and Eagle (1963) developed a children's
form of the EFT called the Children's version of the Embedded Figures
Test (CHEF). This test is appropriate for subjects as young as
5 years. Familiar objects such as a man, boat, and car are substi-
tuted for simple geometric forms and are embedded in knob puzzles.
This test is not used currently because of its bulky construction and
the difficulty associated with transporting the test.

Research on Children and
Field’DependencelIndepgndence

 

 

The majority of research involving the use of the Witkin group's
three tests has been conducted using subjects over 10 years of age.
The field dependence/independence dimension is difficult to study in

young children due to the egocentric nature of their relationships

24

to the environment. Very little research with children using the
BAT or the RFT has been conducted.

Ghuman (1977) used the Children's Embedded Figures Test as
part of a battery of tests assembled to explore social and cognitive
development. He found that middle-class children performed better
than working-class children on the CEFT. CEFT was positively related
to cognitive measures (Piagetian tasks) but not to the personality
measures used. There was no difference between performance of boys
and girls (ages 11-12 years) on the CEFT.

Connor, Schackman, and Serbin (1978) found that first-grade
children who received training in visual-spatial disembedding could
improve their performance on the CEFT. The researcher used half of
the test items as a pretest and half as a posttest. The girls' per-
formance improved significantly more than did the boys' as a result
of training.

Gildemeister and Friedman (1978) studied first graders of excep-
tionally high and low verbal ability in relation to visual-analysis
ability. The high-verbal-ability group performed better on the
Spatial Relations subtest of the Primary Mental Abilities Test but
did not differ significantly from the low-verbal-ability group in
performance on the CEFT. The researchers felt that this finding
gave more support for the hypothesis that the CEFT measures abilities
that are not related to verbal ability.

Damusis and Desjarlais (1977) reviewed literature on field
dependence/independence and educational policy. They found that the

literature supports the ideas that field-independent children can

25

analyze and restructure problems and can take another's point of
view. This leads to greater social ability. The researchers con-
cluded that in an educational setting, children high in field inde-
pendence will excel in spatial relations and that these children will
be more intrinsically motivated. On the other hand, research on
field-dependent children showed that they will be more inclined to

respond to contingencies of social reward.

Age and Field Dependence/Indepencence

 

Witkin, Goodenough, and Karp (1967) demonstrated in two longi—
tudinal studies in which subjects ranged in age from 8 to 24 years
that the field dependence/independence dimension was stable across
time. There is a trend toward increased performance (i.e., toward
field independence) on all three types of tests up to age 17, with
a slight decline in performance through age 21 and then stability
through at least age 24. There also seems to be a sharp decline in
field independence in post-retirement years (Schwartz & Karp, 1967).

Studies on children under 8 years presented contradictory findings.

Gender and Field DependenceLIndependence

Although sex differences in performance on the field dependence/
independence measures have been noted for older children and adults
(Vaught, 1965), sex differences in performance have not been noted
for children in the 4-8 year range (Maccoby et al., 1965) or in
geriatric populations (Schwartz & Karp, 1967). Gender differences
in performance on BAT and RFT may be related to ability to perform

well on spatial-relationships tests (Gardner, Jackson, & Messick,

 

 

26

1960). Sherman (1967) postulated that the reason for superior male
performance on tests of spatial ability and on BAT and RFT is due to
a sex bias built into toys that allows males more practice in per-
ceiving spatial relationships.

Relationship of Field Dependence/

Independence and Schooling

It has been suggested by researchers that ability in art and
music, both disciplines requiring skill in restructuring, may be
enhanced by training that encourages analytic processes (Witkin
et al., 1977). First-grade children who received 7 months of musical
training by the Kodaly method showed significantly more field inde-
pendence than the control group, who received no training (Hurwitz,
Wolf, Bortnick, & Kokas, 1975).

Preschool-age children showed significant improvement in field
independence after a year's viewing of games involving perceptual
analysis that are part of the Sesame Street television series (Ball &
Bogatz, 1970). Children who participated in Project SEE, a percep-
tual, analytical training program for kindergarteners and first
graders, showed a generalized enhancement of restructuring ability
and improved performance on embedded-figures tests after 1- and 2-year

training programs (Dolecki, 1976).

Cross-modal Sensory Transfer Ability,
Gibson (1967) wrote that "cross-modal transfer carries the impli-
cation that some discrimination has been learned by presentation in

one modality and that the discrimination can be made, without specific

7
n

27

practice, when the objects are presented in a different modality"
(p. 220). Gibson (1967) also classified "higher orders of proper-
ties of stimulation which are not sensation specific" (p. 219) as
amodal and asserted that "many distinctive features of objects and
events (corners, motions, temporal patterns, and transitions)"
(p. 219) may very well be products of amodal or cross-modal percep-
tions as opposed to a single mode of perception. Although cross-
modal or intermodal transfer has been demonstrated to exist in
tactile-auditory, auditory-visual, and visual-tactile pairs, it is
only the visual-tactile transfer that bears directly on this study.
In studies of visual-tactile matching of form it has been demon—
strated that for all age groups and for both sexes visual matching
is superior to tactile matching and that "discrimination of form is
more efficient when standards and comparisons are presented visually"
(Walk & Pick, 1981, p. 118). It has been noted that visual and
tactile (haptic) exploratory patterns contain elements that are
similar (two eyes, two hands exploring in tandem) and different
(hands are not required to work as a pair) and unique (hue perception
is visual only). This points to the fact that only certain types of
information can be perceived intermodally. Study of information
transfer between sensory modalities--both the types of information
available for cross-modal transfer and the processes involved in the
transfer-~is basic to the understanding of both the nature of inter-
sensory integration and the development of intersensory integration

(Walk & Pick, 1981).

28

Theories of Cross-modal Transfer

One of the early conceptual frameworks for the study of the
development of cross-modal matching was developed by Birch and Lefford
(1967). They argued that the level of cross-modal matching ability
at a particular age level corresponds to the degree of intersensory
organization at that age level. Movement from dependence on proximal
stimulation (dominance in the tactile mode) to distal stimulation
(dominance in the visual mode) was the process of perceptual develop-
ment. This shift from tactile to visual dominance was accompanied
by an increase in intersensory integration and in the ability to use
intersensory transfer.

Piaget's theory of cross-modal transfer is a mediation theory.
The build-up of representative schemata with increased age is accomp-
lished through search for significant objects in the environment.

The mediation of schemes common to both systems precipitates the
interrelationships of visual and tactile perceptual systems. "The
sensory scheme is a kind of referent to two modalities, making cross-
modal transfer possible" (Gibson, 1967, p. 230).

J. J. Gibson's (1966) theory of perceptual development is an
information-based theory. In Gibson's theory, no one sense presents
more truthful or accurate information about a stimulus than any
other; each is an information-seeking system that has equal potential
for supplying the individual with knowledge of properties of objects.
Gibson's theory is not developmental ; he maintained that structure

in perceptual experience is discovered, not constructed. Gibson

29

coined the term "amodal" to describe those experiences and informa-
tion that are the products of cross-modal or multi-sensory processes.
At present, the theory of Birch and Lefford is no longer used
as a basis for research. The ideas of both Piaget and Gibson have
been springboards for much research into the nature of cross-modal
transfer, and both theories have empirical support. It is impossible
at present to determine which of the theories, that of Piaget or that
of Gibson, presents a more accurate description of the development of
perception and, more specifically, the nature and development of
cross-modal transfer.

Research on Children and
Cross-modal’Transfer

 

 

Walk (1965) found in comparing visual and tactile modes that
symmetrical forms were differentiated more easily in the visual mode
than were asymmetrical forms. In the tactile mode, neither type of
form was discriminated more easily. He also found that by having
subjects use both hands to explore figures the tactile experience
became more similar to the visual experience in accuracy of percep-
tion. Rudel and Teuber (1964) found that in cross-modal matching
tasks the number of correct visual-visual and visual-tactile matches
was about equal for children ages 5 and 6 years old, but the visual-

visual matches were favored by children ages 3 and 4 years.

Training for Cross-modal Transfer
Pick, Pick, and Thomas (1966) used printed and raised letter-

like stimuli in a training study involving visual-tactile cross-modal

30

matching. First graders were trained to discriminate standard forms
by sight or by touch and then were trained for transfer to the other
condition. Those groups of children who were trained to attend to
differences in forms showed more cross-modal transfer.

Butler (1979) trained third and fourth graders in visual and
haptic cross-modal transfer tasks. All subjects in this study scored
high on measures of impulsivity. Haptic training decreased errors
on visual measure, while visual training decreased errors on both the
haptic and the visual measures. It was noted that presentation in
the visual mode was more salient than presentation in the tactile
mode for training in cross-modal transfer.

Blank and Bridger (1964) compared cross-modal concepts and
cross-modal equivalence in children ages 3 to 6 years. They assumed
that the children would have to be able to verbalize the transfer
concept in order to solve the cross-modal transfer problems. Children
as young as 4 years could make the transfers (visual-tactile matching
and discovery of the concept of succession) required of them at a
greater-than-chance level but could not verbalize the rules that
governed the transfer. The children also either did not or could not
make use of the rules for transfer when they were trained in the use
of the rules. The researchers concluded that verbal mediation was

not a factor in the transfer process.

Mode of Primary Presentation

 

Davidson, Pine, Wiles-Kettenmann, and Appelle (1981) compared

retarded (mental ages of 5.5 years to 8.5 years) and nonretarded

31

(nursery school, elementary school, and junior high) children on
haptic and visual-matching tasks. Results showed that for both
groups, stimulus complexity, presentation modality, and mental age
affected accuracy of match. Presentations to the visual mode
increased accuracy in matching.

In a study of cross-modal ability in first, second, and fourth
graders, Fishbein, Decker, and Wilcox (1977) used a task involving
location of a group of three geometric objects. The initial sensory
input could be either visual or tactile, and the choice stimuli were
photographs of the groups of objects. The choice stimuli were pre-
sented either simultaneously or successively with the photographs.
Older children performed better than the younger ones. No difference
was found between intramodal and cross-modal tasks, but performance
was better for all age groups for the simultaneous-presentation con-
dition than it was for the successive-presentation condition for both
visual and tactile initial inputs. However, presentation of an
initial visual-input stimulus increased performance for all ages.

The researchers concluded that visual perceptions mediate performance
in tasks where there are both visual and tactile inputs.

Jain and Sinha (1978) studied the effects of mode of presenta-
tion on accuracy of cross-modal matching. They found a significant
difference between visual-tactile and tactile-visual conditions of
presentation with the visual-tactile presentation yielding the greater
number of accurate matches. The researchers suggested that initial

information gained through the use of the visual mode is more useful

32

in accomplishing tasks of cross-modal matching than is information
gained through use of the tactile mode.

Relationship of Cross-modal Ability

to Cognitive Functioning and SChooling

In a review of literature on cross-modal functioning, Derevensky
(1977) found support for the notion that cross-modal functioning and
the resultant sensory integration are related to reading achievement.
For children beyond second grade, cross-modal ability was more highly
related to high reading scores than was visual discrimination.
Derevensky noted that results of studies showed that individual
development of the child, sex of the child, socioeconomic status,
and types of tasks used to measure cross-modal ability all affected
the outcome of the studies reviewed.

Jarman (1978) studied fourth graders of high, normal, and low IQ
in relation to their performance in cross-modal and intramodal match-
ing ability in tasks with visual and auditory content. He found that
therewas an interaction between tasks and child's strategy for
problem solving that determined competence in both cross-modal and
intramodal tasks and that the interaction was positively related to
IQ of the child.

Using blind, partially sighted, and sighted 4 to 7 year olds,
Simpkins (1979) studied the effect of school experience on tactile
discrimination. No significant differences were found on the basis
of vision or gender; however, school experience had a positive effect

on the ability to discriminate tactually in all groups.

33

Age and Cross-modal Transfer

In a study of developmental changes in haptic matching ability,
Kleinman (1979) found that there was an increase in ability to match
stimuli using haptic clues between kindergarten and second grade and
then again between second grade and fourth grade. He cited increased
exploration time and more efficient use of scanning strategies as
factors in the improved ability across age groups.

A battery of tests for measuring cross-modal ability across
three modalities was developed by Temple, Williams, and Bateman
(1979). They found that age and developmental level were factors in
determining a child's performance on cross-modal and intramodal tasks
as represented by the tests selected for inclusion in the battery.
The younger child and the learning-disabled child were less able to
perform well on the measures tested for use in the battery. The
sample included both normal and learning-disabled 6- and 8—year-old
children.

Relationships of Perceptual Style and Cross-

modal Transfer Ability to Other Perceptual
and Cognitive Abilities

 

 

The study of field independence and cross-modal transfer has
involved much research into the relationships between these constructs
and their respective behavior correlates and other postulated cognitive
and perceptual abilities. Such research helps to establish a framework
for understanding the basic processes that underlie those behaviors
labeled cognitive or perceptual and helps to point out gaps in the
patterns of constructs that have been developed to explain mental

processes .

34

Various studies have demonstrated that the Children's Embedded
Figures Test scores used to measure field dependence/independence in
children are related to a number of other measures of cognitive and
perceptual functioning. Goodenough and Karp (1961) found the measures
on the Embedded Figures Test to load the same factor as the WISC
Block Design, Object Assembly, and Picture Completion subtests for
10 and 12 year olds. Pascual-Leone (1969) f0und that the CEFT also
related to the performance tests of the WISC but not to the verbal-
comprehension subtests. Field-dependent persons have been found to
have better memory for faces (Konstadt & Forman, 1965). Field-
independent persons have been found to be more creative (Stevens,
1969). Personality also seems to have some relation to field
dependence/independence in that field-dependent children have a
greater tendency to mental illness (Karp et al., 1965; Scallon &
Herron, 1969) and are more easily influenced by social criticism and
show less autonomous play (Crandall & Sinkledam, 1964).

Cross-modal transfer abilities have been shown to be related to
reading ability, particularly in children over second-grade level
(Derevensky, 1977). In children over second-grade level, spelling,
language skills, and the Iowa Test of Basic Skills have been found
to relate to cross-modal transfer ability in children (Early, 1976).
Children with normal to high IQ were found to be better able to use
training in cross-modal transfer skills (Jarman, 1978). Improvement
in cross-modal transfer ability with age has been hypothesized to
relate to brain development, specifically myelination of forebrain

commissural neurons (Galin, Diamond, & Herron, 1977). Thus, it can

35

be seen that field dependence/independence and cross-modal sensory
transfer ability have been studied in relation to varied and numerous

other aspects of perceptual and cognitive functioning.

CHAPTER III

RESEARCH DESIGN AND PROCEDURES

Introduction

 

The design for this quasi-experimental study was a combination
of several simple correlational designs and a 2 x 2 factorial design
(Campbell & Stanley, 1966). This combination of designs was chosen
because of the flexibility it allowed in comparing measures of
perceptual and cognitive ability with other traits of the subjects.
The independent variables in this study were scores on the Children's
Embedded Figures Tests (see Appendix 0); scores on Cross-modal
Sensory Transfer Tasks 1 (visual to visual matching), 2 (visual to
tactile matching), and 3 (tactile to visual matching); scores on
Cross-modal Sensory Transfer Combined Tasks (see Appendix C); gender
of the subjects; and grade of the subjects. Gender of the subjects
and grade of the subjects were the main effects tested in the multi-
variate analysis of variance (MANOVA). These variables were selected
in order to determine whether a developmental trend existed in the
data and/or whether there were gender-related differences in the data.

(See Figure 3-1.)

Description of the Sample
The subjects were 60 children (55 white, 5 black) enrolled in
the Lake Harbin Elementary School, Clayton County, Georgia, public

36

37

om

cm

:I

:I

 

 

 

 

.zuzum ms» co cmwmmo .Pum mczmwd
Am_ u z .m_ u av cm H em
Amp u 2 .mp u my on u.w:
om u z
mcoum cmcmneou ememcmeu zgomcmm pmuoe-mmocu u hmzu
pmmh mmcammd cucumnsm m.:meupw;u u Puma
om n d cm H d
I I M .520
mp u : mp u : cacao;
ix Emu
3825
o m we
... 1 w 55 c e w
m_ u c m_ n c cmpgmmcmccwg
.M puma
mmez mmpwsmd

 

 

 

pumwnsm we cmccmo

 

 

38

school system. (See Appendix A.) Clayton County is a suburban county
to the south of Atlanta, Georgia. The residents of the county are
primarily lower middle class or working class. The primary source of
employment within the county is locally owned small businesses. Many
of the residents commute to Atlanta to work.

Four kindergarten classrooms and four fourth-grade classrooms
were chosen as the population from which subjects could be selected.
There was a total of 218 children enrolled in the eight classrooms.
Thirty children (15 males, 15 females) from both the kindergarten and
the fourth grade were selected as the sample. The sample was selected
randomly from the 106 children who had parental permission to partici-
pate in the study, for a total sample of 60 children. The children
were grouped by grade and sex before selection in order to insure
equal numbers in each age/sex group.

All subjects in the study were enrolled in regular classrooms,
and none were involved in remedial or special-education classes. As
determined by test data, all children in the study were performing
at grade level or above in their academic work. The tests given by
the school to the children were the Metropolitan Readiness Test (kin-

dergarten) and the Metropolitan Achievement Test (fourth grade).

Instrument Selection and Description
The Children's Embedded Figures Test (CEFT) (Karp & Konstadt,
1971) was used to obtain a field dependence/independence measure for
each subject. The CEFT was administered and scored in accordance

with standardized procedures outlined in the test manual. (See

39

Appendix D.) The CEFT is administered individually and consists of

a series of 25 geometric drawings of familiar objects in which the
child is to locate and trace around a specified geometric figure.
(Only one exact matching figure is embedded in each of the 25 geometric
drawings.) The CEFT is divided into two parts, tent (11 items) and
house (14 items). The division is based on the stimulus figure that
the child must locate in the drawings. The tent figure is an equi-
lateral triangle colored red and black. The house figure is a combi-
nation of a rectangle and triangle and is blue. Scores for the test
are the number of correct matches out of 25.

The CEFT was chosen because it is the most reliable standard
measure of its type designed specifically for children in the 5—12
year age range. Internal consistency reliability coefficients for
the CEFT rangefrom .84 to .90 (Witkin et al., 1971). Correlations
between the CEFT and the Embedded Figures Test, a similar measure
for children over 12 and adults, yield validity coefficients of .71
at 9 years and .85 at 11 years (Witkin et al., 1971).

The Cross-modal Sensory Transfer Tasks consisted of three
lO-item identification tasks. The objects used for stimuli in all
three tasks were Milton Bradley geometric solids (ordering number
MB-8064), a series of wooden figures averaging 3 inches in height.
(See Appendix C.) Rational for selection of these figures as stimuli
was threefold. First, the literature reveals that geometric figures
are one of three types (geometric, biomorphic, alphabet) of stimuli
used to test children's tactile-visual cross-modal sensory transfer

ability (Pick, Pick, & Thomas, 1966; Walk, 1965). Second, the Milton

40

Bradley figures are widely available for use in replication of the
study and are of convenient size and variety for use with children.

A third reason for selection of the geometric figures as a stimuli set
was that stimuli in the Children's Embedded Figures Test are also geo-
metric. The relative consistency across measures of stimulus type
helped to eliminate confounding that might have been added by a total
nonequivalence of stimuli sets across measures, despite the fact that
the CEFT involved use of plane geometric shapes and the cross-modal
tasks used solid figures.

Task 1 of the Cross-modal Sensory Transfer Tasks was a visual-
to-visual matching task (intramodal). This task served to bridge
the gap between the visual matching tasks of the Children's Embedded
Figures Test and the cross-modal matching of Tasks 2 and 3. In
Task 1, the subject was asked to choose from among five geometric
figures the object that matched a single standard figure. The stan—
dard and the choices were changed for each trial. (See Appendix C for
details on administration and stimuli-set ordering.) Task 1 consisted
of a practice trial followed by 10 scored trials. Each trial was
scored 1 for a correct match and 0 for an incorrect response.

Task 2 of the Cross-modal Sensory Transfer Tasks was a visual-
to-tactile matching task. In this task, the subject viewed the
standard without being allowed to touch it and then was asked to find
its match from an array of five geometric figures placed within a
"feeling box" and out of the subject's sight. (See Appendix C.)

Task 2 consisted of a practice trial followed by 10 scored trials.

.2

41

Each trial was scored 1 for a correct match and 0 for an incorrect
response.

Task 3 of the Cross-modal Sensory Transfer Tasks was a tactile-
to-visual matching task. In this task, the subject was asked to feel
the standard, which was presented inside the "feeling box" out of sight
of the subject. The subject then visually chose a match for the
standard from an array of five geometric figures placed on top of the
"feeling box." (See Appendix C.) Task 3 consisted of a practice
trial followed by 10 scored trials. Each trial was scored 1 for a
correct match and 0 for an incorrect response.

The Combined Task Score was found by adding the scores of Tasks 1,
2, and 3. Each task had a maximum of 10 possible points and a minimum
of 0 points. The maximum Combined Score was 30 points.

In Tasks 2 and 3, figures presented to the tactile mode for
inspection were hidden from view of the subject by placing them within
a "feeling box" of the type used in early childhood education settings
to enhance tactile-discrimination skills. The box used measured
12" x 18" x 12" and had two 6-inch circular openings equipped with
felt covers into which the subjects placed their hands. The back of
the apparatus was open to the view of the researcher. Subjects were
encouraged to use both hands to explore the figures presented for
manipulation. The reasoning behind this was that two-handed explora-
tion has been found to more closely resemble visual inspection of
objects (Walk, 1965). Both tactile and visual stimuli were placed in
a standardized order and position assigned to each trial of Tasks l,

2, and 3. (See Appendix C for details related to the ordering of the

42

stimuli.) None of the cross-modal tasks was timed. This procedure
was adopted in order to match the nontimed procedures developed for

the CEFT.

Procedure

The researcher assumed responsibility for administration of the
Children's Embedded Figures Test and the Cross-modal Sensory Transfer
Tasks. All testing sessions were held in the school library. Test-
ing sessions were approximately 40 minutes in length. All subjects
were given the Children's Embedded Figures Test first and then the
Cross-modal Sensory Transfer Tasks 1, 2, and 3, in that order. It
has been demonstrated that primary visual presentation enhances per-
formance on visual and tactile cross-modal tasks (Jain & Sinha, 1978).
Therefore, all children were given the stimuli in the same order and
Task 1, Task 2, and Task 3 in the same order so as to minimize order
effects. All testing followed the procedures described in the instru-
ments section of this chapter and in Appendices C and D. Every attempt
was made to provide a standard testing situation across subjects.

The testing period lasted 2 weeks. The kindergarten children
were tested the first week, and the fourth graders were tested the
second week. Fridays of both weeks were scheduled as make-up days
for subjects who missed their assigned test periods on Monday through

Thursday.

Ethical Considerations
All children participating in the study had a signed letter of

informed consent on file with the researcher. (See Appendix B for

43

letter.) No child was forced to participate in a testing situation
against his/her own will, nor was any attempt made to coerce any child
to participate. All 60 of the children in the original sample par-
ticipated willingly in the study and were present on their assigned
testing date. Scheduling of testing dates was made with the coopera-
tion of the classroom teacher to insure that no child missed important
course work or was deprived of a favorite activity. The researcher
escorted the kindergarten children to and frbm the testing site and
waited at the door of the testing room for the fourth graders, who
walked from their classrooms. Every attempt was made to insure that
a minimum amount of class time was missed by each subject.

All data were collected and stored using a four-digit code number
(selection number, grade, and sex). (See Appendix E.) Children's
names were not used in reporting the study. Parents of the subjects
received a copy of the child's scores at the completion of the data-
collection period. A written summary of results was made available

to school staff and to parents upon completion of the project.

Data Analysis
The data were analyzed using a series of Pearson's correlations
and a 2 x 2 multivariate analysis of variance (MANOVA). Level of
significance was set at .05. The Statistical Package for the Social
Sciences (SPSS) Version 8 was the statistical package used for the
analysis. The Michigan State University Computer Center provided

assistance in the analysis. (See Appendix F.)

44

Summar

This study used a series of correlational designs and a 2 x 2
factorial design to test relationships between two measures of
perceptual and cognitive abilities in kindergarteners and fourth
graders. Instruments used to measure perceptual and cognitive abili-
ties were the Children's Embedded Figures Test and Cross-Modal
Sensory Transfer Tasks 1, 2, and 3. Sixty subjects, 30 each in kin-
dergarten (males = 15, females = 15) and fourth grade (males = 15,
females = 15), participated in the study. Independent variables were
scores on the Children's Embedded Figures Test and the three Cross-

modal Sensory Transfer Tasks, and a combined score of all three Cross-

modal Sensory Transfer Tasks, grade of subjects, and gender of subjects.

“I

CHAPTER IV

ANALYSIS OF RESULTS

Data for this study were analyzed using Pearson's correlations
for those comparisons involving only two variables and a multivariate
analysis of variance (MANOVA) for comparison of Children's Embedded
Figures Test scores means and Cross-modal Sensory Transfer Tasks com-
bined scores means both across and within age/gender groups. A level
of .05 was chosen as the lowest possible level of significance accept-
able in this study.

Correlation is the statistic of choice when there is "no clear-
cut distinction between two kinds of scores as to which is the inde-
pendent or predictor variable" (Hays, 1973, p. 619), which was the
case in comparing the variables in this study. A correlation proce-
dure was selected for analysis of pairs of variables in this study
because this type of statistic not only allows for measurement of the
directionality and strength of a relationship between two variables
but also provides a basis for establishing a means of predicting the
direction of a second variable when the direction of the first vari-
able is known (Hays, 1973). The statistical procedure yielding the
Pearson product-moment correlation coefficient was chosen because the
process results in correlation coefficients that are independent of

both the means of the scores tested and of the standard deviations of

45

46

groups of scores (Glass & Stanley, 1970). This allows for correlation
coefficients to be converted to Z_scores and a level of significance
to be found for the relationships.

The multivariate analysis of variance (MANOVA) is used to test
main effects in analysis involving more than one variable per cell.
The MANOVA procedure also allows for the testing for interaction
between main effects, which in this study were grade and gender of
subject. Interaction between main effects precludes any further
analysis (Hays, 1973). The MANOVA procedure contains univariate analy-
ses within it so that effects of single main effects may also be
tested.

Analysis of data for this study involved comparison of scores
for Cross-modal Sensory Transfer Tasks l, 2, 3, and Combined Scores
with scores for the Children's Embedded Figures Test for boys and
girls in kindergarten and fourth grade. The analysis was structured
so that (a) relationships among the Cross-modal Sensory Transfer
Tasks could be determined, (b) relationships between the Cross-modal
Sensory Transfer Tasks and the Children's Embedded Figures Test could
be determined, and (c) effects of grade of subject and/or gender of
subject on the relationship between the Cross-modal Sensory Transfer
Tasks and the Children's Embedded Figures Test could be determined.
The means and standard deviations used for the analyses presented

later in this chapter appear in Table 4-1.

47

Table 4+1. Means, Ranges, Variances, and Standard Deviations for
Cross-modal Sensory Transfer Tasks l, 2, 3, and Combined
Scores and Children's Embedded Figures Tests (N_= 60)

 

Maximum Sample

 

X: Range Range Variance §9-
Combined Scorea 26.33 0-30 16-30 8.53 2.92
Task 1a 9.93 0-10 8-10 .09 .31
Task 2a 8.28 0-10 4-10 2.31 1.52
Task 3a 8.18 0-10 3-10 2.17 1.47
CEFTb 10.81 0-25 1-23 44.42 6.67

 

aCross-modal Sensory Transfer Task.

bChildren's Embedded Figures Test.

Hypothesis 1

 

Null Hypothesis: Scores for Cross-modal Sensory Transfer Task 1
(visual-to-viSual matching) are not related to scores for Cross-
modal Sensory Transfer Task 2 (visual-to-tactile matching).

 

Alternative Hypothesis: Scores for Cross-modal Sensory Transfer
Task 17(v15ual-to-visual matching) are positively related to
scores for Cross-modal Sensory Transfer Task 2 (visual-to-tactile
matching).

 

Comparisons between Task 1 and Task 2 were analyzed using Pearson
correlations and were based on the following groupings: total group
scores (N.= 60), kindergarteners (g_= 30), fourth graders (g_= 30),
females (g_= 30), and males (g,= 30). Results were significant for
total group, kindergarteners, and females at a .05. Results were

nonsignificant for males. (See Table 4-2.)

48

Table 4-2. Pearson Correlation Coefficients for Comparison of
Task 1 With Task 2

 

 

Variable Correlation Coefficient Level of Significance
Total group r_= .43 p_= .001*
Kindergarteners .r = .49 p_= .002*
Fourth graders uncomputablea ..

Females r_= .52 p_= .002*
Males r_= .24 p_= .096

 

a . . .
No variation in Cross-modal Sensory Transfer Task 1 scores.

*Significance level set at .05.

Hypothesis 2

Null Hypothesis: Scores for Cross-modal Sensory Transfer Task 1
(Visual-to-visual matching) are not related to scores for Cross-
modal Sensory Transfer Task 3 (tactile-to-visual matching).

 

Alternative Hypothesis: Scores for Cross-modal Sensory Transfer
Task 1 (Visual-t04Visual matching) are positively related to
scores for Cross-modal Sensory Transfer Task 3 (tactile-to-
visual matching).

 

Comparisons between Task 1 and Task 3 were analyzed using Pearson
correlations and were based on the following groupings: total group
scores (N_= 60), kindergarteners (n'= 30), fourth graders (n_= 30),
females (n_= 30), and males (fl_= 30). Results were significant for
total group, kindergarteners, females, and males at a = .05. (See

Table 4-3.)

49

Table 4-3. Pearson Correlation Coefficients for Comparison of Task 1

 

 

With Task 3
Variable Correlation Coefficient Level of Significance
Total group r.= .49 p_= .001*
Kindergarteners r_= .56 p_= .001*
Fourth graders uncomputablea ..
Females r_= .56 p_= .001*
Males r_= .34 p_= .034*

 

a . . .
N0 variation in Cross-modal Sensory Transfer Task 1 scores.

*Significance level set at .05.

Hypothesis 3

Null Hypothesis: Scores for Cross-modal Sensory Transfer Task 2
(visual-to-tactile matching) are not related to scores for Cross-
modal Sensory Transfer Task 3 (tactile-to-visual matching).

 

Alternative Hypothesis: Scores for Cross-modal Sensory Transfer
Task 2 (visual-to-tactile matching) are positively related to
scores for Cross-modal Sensory Transfer Task 3 (tactile-to-visual
matching).

 

Comparisons between Task 2 and Task 3 were analyzed using Pearson
correlations and were based on the following groupings: total group
scores (N_= 60), kindergarteners (g_= 30), fourth graders (g_= 30),
females (g_= 30), and males (g_= 30). Results were significant for

all comparison groups at a = .05. (See Table 4-4.)

50

Table 4-4. Pearson Correlation Coefficients for Comparison of Task 2
With Task 3

 

 

Variable Correlation Coefficient Level of Significance
Total group r_= .69 p_= .001*
Kindergarteners r_= .69 p_= .001*
Fourth graders r_= .55 = .001*
Females r_= .65 = .001*

Males r_= .75 = .001*

 

*Significance level set at .05.

Hypothesis 4

 

Null Hypothesis: Scores for the Children's Embedded Figures Test
are not related to:
a. Cross-modal Sensory Transfer Task 1 (visual-to-visual

 

matching)

b. Cross-modal Sensory Transfer Task 2 (visual-to-tactile
matching)

c. Cross-modal Sensory Transfer Task 3 (tactile-to-visual
matching)

d. Cross-modal Sensory Transfer Combined Tasks scores

Alternative Hypothesis: Scores for the Children's Embedded
Figures Tests are positively related to:
a. Cross-modal Sensory Transfer Task 1 (visual-to-visual

 

matching)

b. Cross-modal Sensory Transfer Task 2 (visual-to-tactile
matching)

c. Cross-modal Sensory Transfer Task 3 (tactile-to-visual
matching)

d. Cross-modal Sensory Transfer Combined Tasks scores
For Hypothesis 4a, comparisons between CEFT and Tasks 1, 2, 3,
and Combined Scores were analyzed using Pearson correlations and

were based on the following score groupings: total group (N_= 60),

kindergarteners (n_= 30), fourth graders (3" 30), females (g_= 30),

and males (n_= 30). Results were significant for total group and

51

kindergarten at a .05. Results were nonsignificant for males and

for females. (See Table 4-5.)

Table 4-5. Pearson Correlation Coefficients for Comparison of

Cross-modal Sensory Transfer Tasks l, 2, 3, and
Combined Scores With Children's Embedded Figures Test

 

Variable Correlation Coefficient Level of Significance

 

Total Group

Task 1 r_= .28 p_= .015*
Task 2 r_= .71 p_= .001*
Task 3 r_= .61 p_= .001*
Combined Scores .5 = .70 p_= .001*
Kindergarteners
Task 1 r_= .69 p_= .Ol4*
Task 2 r_= .66 p_= .001*
Task 3 r_= .51 p_= .002*
Combined Scores r_= .63 p_= .001*
Fourth Graders a
Task 1 uncomputable .
Task 2 r.= .85 p_= .001*
Task 3 r_= .73 p_= .001*
Combined Scores r.= .90 p_= .001*
Females
Task 1 r_= .29 p_= .054
Task 2 r_= .62 p_= .001*
Task 3 r_= .49 p_= .003*
Combined Scores r_= .59 2.: .001*
Males
Task 1 r_= .25 p_= .088
Task 2 r_= .81 p_= .001*
Task 3 r_= .75 p_= .001*
Combined Scores r_= .83 p_= .001*

 

aN0 variation in Cross-modal Sensory Transfer Task 1 scores.

*Significance level set at .05.

52

For Hypothesis 4b, comparisons between measures were analyzed
using Pearson's correlations and were based on the following score
groupings: total group (N_= 60), kindergarteners (n_= 30), fourth
graders (n_= 30), females (n_= 30), and males (n_= 30). Results were
significant for all score groupings. (See Table 4-5.)

For Hypothesis 4c, comparisons between measures were analyzed
using Pearson correlations and were based on the following score

groupings: total group (N_= 60), kindergarteners (g_= 30), fourth

graders (n_- 30), females (n_= 30), and males (g_= 30). Results were
significant for all score groupings. (See Table 4-5.)

For Hypothesis 4d, comparisons between measures were analyzed
using Pearson correlations and were based on the following score
groupings: total group (N.= 60), kindergarteners (g_= 30), fourth
graders (g_= 30), females (g_= 30), and males (g_= 30). Results
were significant for all score groupings. (See Table 4-5.)

Multivariate Analysis of Variance
for Hypotheses 5 and 6

 

A multivariate analysis of variance (MANOVA) procedure was used
to test the combined main effects of grade of subject and gender of
subject on the scores of Children's Embedded Figures Tests and Cross-
modal Sensory Transfer Tasks Combined Scores. Both the multivariate
analysis of variance and the subsequent analyses of variance (ANOVA)
yielded no significant interaction effect between the two main
effects. (See Table 4-6.) As there was no significant interaction
found between the two main effects, analysis of each main effect

could be carried out to determine the individual effect of each main

 

53

effect on the scores of the two measures. Had a significant inter-
action been found, analysis would have been terminated at this point,

and it would have been impossible to test Hypotheses 5 and 6.

Table 4-6. Multivariate and Univariate Analyses of Grade and Gender
of Subject for Children's Embedded Figures Tests Scores
and Cross-modal Sensory Transfer Tasks Combined Scores

 

 

 

 

Variable gf_ §§_ 5. BF
Multivariate Analysis A
Gender & grade 2,55 .03 .43

 

Univariate Analyses

 

CMSTa 1,56 1.66667 .24 .63
CEFTb 1,56 6.01667 .35 .56

 

aCross-modal Sensory Transfer Tasks Combined Scores.

bChildren's Embedded Figures Test scores.

Hypothesis 5

 

Null Hypothesis: Fourth graders, as a group, will perform no
better than kindergarteners on:

a. Cross-modal Sensory Transfer Tasks Combined Scores

b. Children's Embedded Figures Test scores

 

Alternative Hypothesis: Fourth graders, as a group, will perform
better than kindergarteners on:

a. Cross-modal Sensory Transfer Tasks Combined Scores

b. Children's Embedded Figures Tests scores

 

Comparisons for Cross-modal Sensory Transfer Tasks Combined
Scores and Children's Embedded Figures Tests scores were analyzed

using a multivariate analysis of variance (MANOVA) and several

54

univariate analyses of variance (ANOVA) procedures. Results of the
analysis indicate that for a combined analysis of both measures (CEFT
and Cross-modal Tasks) grade is significant as a main effect (E_=
.66133, p_= .00001). Univariate analyses yielded the following results:
For Children's Embedded Figures Tests, grade was a significant main
effect (f_= 92.67429, p_= .00001); for Cross-modal Sensory Transfer
Tasks Combined Scores, grade was a significant main effect (E_=

11.62712, E.‘ .00121). (See Table 4-7.)

Table 4-7. Multivariate and Univariate Analyses of Grade of Subject
for Children's Embedded Figures Test and Cross-modal
Sensory Transfer Tasks Combined Scores

 

Variable df SS f_ Bf

 

Grade Multivariate Analys1s

Grade 2,55 .66 .001*

 

Univariate Analyses

 

CMSTa 1,56 81.66667 11.63 .001*
CEFTb 1,56 1591.35 92.67 .001*

 

aCross-modal Sensory Transfer Tasks Combined Scores.
bChildren's Embedded Figures Tests scores.

*Significance level set at .05.

Hypothesis 6

 

Null Hypothesis: Males, as a group, will perform no better than

females on:
a. Cross-modal Sensory Transfer Tasks Combined Scores

b. Children's Embedded Figures Tests scores

 

55

Alternate Hypothesis: Males, as a group, will perform better
than females on:

a. Cross-modal Sensory Transfer Tasks Combined Scores

b. Children's Embedded Figures Tests scores

 

Comparisons for Cross-modal Sensory Transfer Tasks Combined
Scores and Children's Embedded Figures Test scores were analyzed
using a multivariate analysis of variance (MANOVA) and several univari-
ate analyses of variance (ANOVA) procedures. Results of the analysis
indicate that for a combined analysis of both measures (CEFT and
cross-modal tasks) gender is not significant as a main effect (f_=
.07347, p_= .12264). Univariate analyses yielded the following
results: For Children's Embedded Figures Tests, gender was not a
significant main effect (f_= 3.66162, p_= .06253); for Cross-modal
Sensory Transfer Tasks Combined Scores, gender was not a significant

main effect (f_= 3.77661, p_= .05637). (See Table 4—8.)

Table 4-8.--Multivariate and Univariate Analyses of Gender of Subject
for Children's Embedded Figures Test and Cross-modal
Sensory Transfer Tasks Combined Scores

 

Variable df SS f_ 25

 

Multivariate Analysis

Gender 2,56 .07 .122

 

 

Univariate Analyses

 

CMSTa 1,56 26.66667 3.79 .056
CEFTb 1,56 62.01667 3.66 .063

 

aCross-modal Sensory Transfer Tasks Combined Scores.

bChildren's Embedded Figures Tests scores.

56

Discussion of Results

 

Analyses associated with Hypotheses l, 2, and 3 (see Tables 4-2,
4-3, and 4-4) indicated that all three of the Cross-modal Sensory
Transfer Tasks were significantly and positively related to one
another. The results also showed that the two cross-modal tasks
(Task 2--visual-to-tacti1e matching and Task 3--tactile-t0-visual
matching) were more closely related to each other than they were to
the intramodal task (Task 1--visual-to-visual matching). Although
the Cross-model Sensory Transfer Tasks were based on a standard
experimental procedure, they were not a standardized measure. There-
fore, it was important to determine if the three tasks were positively
related in order to collapse the scores of the three tasks into the
Cross-modal Sensory Transfer Tasks Combined Scores used to compare
cross-modal ability with a measure of field dependence/independence,
the Children's Embedded Figures Test.

Comparison of field dependence/independence with cross-modal
ability was accomplished through the correlation of scores on the
Children's Embedded Figures Test with scores for Cross-modal Sensory
Transfer Tasks l, 2, and 3, and Combined Scores (see Table 4-5). As
it was not known before analysis of the data that Tasks 1, 2, and 3
were positively related, it was necessary to structure this part of
the analysis to include both the subtests (Tasks l, 2, and 3) and
the collapsed scores (Combined Scores) of cross-modal ability. The
results of these analyses indicate that scores for Tasks 1, 2, and 3
and Combined Scores are all positively and significantly related to

scores on the Children's Embedded Figures Test for subjects in this

57

study. These results indicate that at least for behaviors measured

by the Children's Embedded Figures Test and the Cross-modal Sensory
Transfer Tasks, field dependence/independence and cross-modal transfer
ability are related abilities.

To determine gender effects and grade-level effects on the
scores of the two measures used in this study (Children's Embedded
Figures Test and Cross-modal Sensory Transfer Tasks), a series of
multivariate analyses of variance (MANOVA) and subsequent analyses of
variance (ANOVA) were used to analyze the data. It was determined
that for this sample there was no interaction between the main effects
of gender of subject and grade level of subject. Therefore, the
effects of the variables grade and gender could be determined indi—
vidually. (See Tables 4-6, 4-7, and 4-8.) The main effect of grade
level of subject was highly significant for both measures combined
(p.= .00001) and for each measure individually (CEFT: p_= .00001;
CMST: p_= .00121). These findings are consistent with the hypothe-
sized developmental trend and with findings of other studies of
measures of field dependence/independence and cross-modal transfer
ability (Derevensky, 1977).

The main effect of gender of subject was not significant for
either both measures combined (p_= .12264) or for each measure indi-
vidually (CEFT: p_= .06253; CMST: p_= .05637). These findings are
also consistent with reports of other studies of field dependence/
independence and cross—modal transfer ability children (Maccoby et al.,

1965).

58

In summary, the analysis of data revealed that Cross-modal
Sensory Transfer Tasks l, 2, and 3 were all significantly and posi-
tively related to each other. Cross-modal Sensory Transfer Tasks
l, 2, 3, and Combined Scores were significantly and positively related
to scores on the Children's Embedded Figures Test (CEFT). There was
no interaction effect between gender of subject and grade of subject
in the comparison of Cross-modal Sensory Transfer Tasks Combined
Scores and Children's Embedded Figures Tests scores. Grade of sub-
ject was significantly and positively related to scores on both

measures, whereas gender of subject was not.

CHAPTER V

CONCLUSIONS, IMPLICATIONS, AND SUMMARY

Review of the Problem

 

In this study, the researcher was concerned with the nature of
the relationship between field dependence/independence as measured by
the Children's Embedded Figures Test (CEFT) and cross-modal transfer
ability as measured by Cross-modal Sensory Transfer Tasks l, 2, 3,
and Combined Scores. Subjects in the study were 30 kindergarteners
(15 girls, 15 boys) and 30 fourth graders (15 girls, 15 boys). The
children's scores on both measures were compared and the effects of
grade and gender of the subject were determined. Data analysis
revealed that there was a positive and significant relationship between
the CEFT and Cross-modal Sensory Transfer Tasks l, 2, 3, and Combined
Scores. (See Tables 4-2, 4-3, 4-4, 4-5.) There was a significant main
effect for grade (see Table 4-7) but not for gender of subject (see
Table 4-8), indicating that for both measures there were age-related
differences but no gender-related differences.

The purpose for comparing a measure of field dependence/
independence with a measure of cross-modal transfer was to help
establish empirically that both abilities are manifestations of an
underlying perceptual and cognitive function, the ability to isolate

and use essential features of an object. Ability to isolate essential

59

60

features has been related to numerous cognitive and personality
traits and has been shown to affect family functioning, group func-
tioning, and career choice. An understanding of the interrelation-
ships of measures of ability to isolate essential features and the
process of deve10pment of the ability has implications for the study
of the individual, the family, and the structure and function of

various social groups such as the classroom.

Conclusions

 

Results of this study seem to support ideas concerning the
interrelationships of those abilities associated with field dependence/
independence and cross-modal sensory transfer found in the literature.
A developmental trend throughout childhood has been noted in research
dealing with both constructs (Derevensky, 1977; Witkin, Goodenough,

& Karp, 1967). In this study, both girls and boys in the fourth
grade exhibited more accuracy on both measures (Cross-modal Sensory
Transfer Tasks 1, 2, 3, and Combined Scores and Children's Embedded
Figures Tests) than did the kindergarteners.

Although the question of causality has not been resolved in the
literature, it is firmly established that gender differences in per-
formance in cross-modal transfer ability and field independence found
in adults' scores are not a factor in children under 12 years of age
(Vaught, 1965). Results of this study add to the growing body of
literature that describes abilities of boys and girls on perceptual/
spatial tasks as equivalent in the childhood years. However, it

should be noted that while gender differences noted in this study

61

were not significant, they were present and several were close to the
established level for significance (see Table 4-8). Apparently,
differences in spatial ability between boys and girls that reach a
level of significance by adolescence are present in younger children
to some extent.

Many researchers have noted that presentation of stimuli to the
visual mode increases performance levels in visual/tactile cross-
modal tasks and that visual-to-visual tasks result in higher perform-
ance levels than do cross—modal tasks (Fishbein, Decker, & Wilcox,
1977). A11 scores on Cross-modal Tasks l, 2, and 3 correspond with
the above findings. All the children performed better on Task 1
(visual-to-visual matching) than on Task 2 (visual-to-tactile matching)
or on Task 3 (tactile-to-visual matching). Performance was better on
Task 2 than on Task 3. Although total group scores for Task 1 corre-
1ated significantly with scores for both Tasks 2 and 3, the correla-
tion for Task 2 with Task 3 was substantially higher. (See Tables
4-2, 4-3, and 4-4.) This finding seems to follow the logic that
cross—modal tasks by the nature of their use of two modalities are
more similar to each other than a cross-modal task would be to an
intramodal or single-mode task (Walk & Pick, 1981).

It should be noted, however, that for boys, the correlation
between Task 1 and Task 2 was low and nonsignificant (r_= .25,
p.= .096) and for Task 1 with Task 3 was low and barely within the
acceptable level of significance (r_= .37, p_= .034). This was due
to less variation in Task 1 scores for boys and might be an area for

further research into the nature of gender differences in spatial

62

abilities that appear in adolescents and adults. Another finding

of note along these same lines is that Task 1 was significantly
correlated to the Children's Embedded Figures Test (see Table 4-5)
only for kindergarten and total group scores. The total group score
correlation seemed to be biased by the kindergarten scores. It was
impossible to calculate correlation coefficients for the fourth
grade due to no variation in Task 1 scores. It seems that by fourth
grade, intramodal ability is firmly established and that cross-modal
ability is a much better indicator of an individual child's ability
to disembed.

Other researchers have established that field dependence/
independence and cross-modal transfer ability are related to scores
on performance subtests of the WISC and t0 socioeconomic status
(Derevensky, 1977; Ghuman, 1977). As noted earlier, children in this
study sample were from primarily lower-middle and working-class fami-
lies. The children in the sample were all of average to above-
average intellectual ability as evidenced by school achievement test
scores and placement in normal classrooms. The mean scores on the
Children's Embedded Figures Test for this study sample were similar
for all gender/grade groups (except kindergarten boys) to the norms
presented in the test manual (Witkin et al., 1971). (See Table 5-1.)
The difference between the sample kindergarten boys and the norming-
data kindergarten boys can be attributed to two extremely high scores
(15 and 19 out of a possible 25) on the Children's Embedded Figures

Test by two kindergarten boys in the sample.

63

Table 5-1. Sample Means and Norming-Data Means for Children's Embedded
Figures Test

 

 

Subject Norming-Data Means Sample Means
Kindergarten girls 4.2 4.23
Kindergarten boys 3.8 6.33
Fourth-grade girls 16.3 15.26
Fourth-grade boys 16.6 16.66

 

As noted above, the research findings from this study are sup-
ported by other literature concerned with the development of cross-
modal sensory transfer ability and field dependence/independence.
Results of this study seem to fall within the expected range of pos-
sibilities established over the past 30 years for field dependence/
independence and cross-modal sensory transfer ability in kinder-
garteners and fourth graders. This study has helped to demonstrate
that cognitive and perceptual abilities associated with ability to

disembed in children are interrelated and developmental in nature.

Personal Observations

 

As this study was conducted in the library of an elementary
school, there were numerous opportunities to observe interactions of
children with the test materials outside of actual data collection.
Approximately 20 children in Grades 5 and 6 served as library aides.
These children helped the librarian for a l-hour period several
times during the week. Most of these children were enrolled in the
school's gifted program. After watching the testing for several

days, the library aides asked to try the cross-modal tasks. All

 

64

these children received perfect scores on all three tasks. The
researcher then modified the two cross-modal tasks (Tasks 2 and 3)
so that orientation of stimulus figure did not match orientation of
matching figure (e.g., rectangular prism on end versus rectangular
prism on side). This addition of a mental-rotation component to the
tasks seemed to increase difficulty, and scores dropped to approxi-
mately 7 correct responses out of 10.

There has been some speculation that verbal mediation plays a
role in cross-modal transfer (Piaget, 1969). The researcher observed
that for this sample, verbal mediation, at least knowledge of names of
geometric solids, did not seem to be necessary for cross-modal
transfer. Almost all the kindergarteners and many of the fourth
.graders asked to examine both sets of the geometric solids used as
stimuli for the cross-modal tasks. While the children played with
the solids, the researcher asked each child to name the solids. Very
few children could name the solids, although several children in
both grades called the solids by the name of the proper corresponding
plane figure. For example, children were presented spheres and called
them circles. A few other children named the solids after objects
in the environment. For example, children were presented circles
and called them balls. Apparently, some of the children had a label-
ling scheme, but not for all the solids in the stimuli sets.

The researcher also had an opportunity to talk informally on
several occasions with teachers in the school. During one of these
discussions it was learned that the fathers of both of the kinder-

garten boys whose scores were unusually high on the Children's

 

65

Embedded Figures Test were commercial airline pilots. This informa-
tion is most interesting in light of the research indicating high
levels of field independence in airline pilots (Cullen et al., 1969).
This might suggest a possible genetic and/or family-training component
in the development of ability to disembed. This researchable question

needs further study.

Limitations of the Study

 

Generalizability

 

Conclusions based on the results of this study are limited in
their ability to apply to the general population. The sample was
fairly homogeneous, and assignment procedures were not totally random.
(See Chapter II, Subjects.) A wide range of geographic areas, socio—
economic groups, and racial/ethnic groups was not included in the
sample, so results cannot be generalized beyond southern, suburban

white children.

Instruments

 

Cross-modal Sensory Transfer Tasks 1, 2, and 3 were too simple
and did not allow for enough sample variation in scores, particularly
at the fourth-grade level. All fourth-grade children made perfect
scores on Task 1 (visual-to-visual matching). It is suggested that
only cross-modal measures be used in future studies or that separate
hypotheses be developed for the testing of intramodal tasks. It is
also suggested that stimuli with more subtle variation such as
irregular or combination geometric solids be used in subsequent

studies.

66

Additional Data on Subjects

Actual measures of intelligence were not given to the subjects
in this study because of resistance to this idea by school officials.
These data would have been helpful in explaining some of the variance
from the norm of scores on the Children's Embedded Figures Tests and
in establishing the actual mean level of intelligence of the groups
tested in order to determine how closely the sample groups matched
groups from other studies. Actual income and educational level of
the family of each child in the study was not available because of
concern by the school board for family confidentiality. This infor-
mation would also have been helpful in explaining differences in the
study sample and in relating data from this study to data in other
studies. It is suggested that this information be collected if the

study is replicated.

Assessment of Developmental Trend

 

Only two grade levels were used in this study. It can be demon-
strated that there was an age-related change between the groups, but
developmental change beyond that cannot be inferred. It is suggested
that several grades, possibly kindergarten, second, fourth, and
sixth, be tested in order to understand more fully the nature of the
development of ability in cross-modal transfer and in field inde-

pendence.

Implications of the Study
Numerous theorists have postulated linkages between perceptual

and cognitive abilities (Piaget, 1969; Witkin & Goodenough, 1981).

67

This study demonstrated that there was, indeed, a positive relation-
ship between field dependence/independence and cross-modal sensory
transfer ability in elementary-school-aged children and that the
relationship was stronger for the older children. This type of
research adds to the empirical base supporting cognitive and percep-
tual theories of Witkin, Gibson, and Piaget. If field dependence/
independence and cross-modal sensory transfer ability do tap the
underlying cognitive and perceptual ability to isolate essential
features of objects, then knowledge of the relationship between the
measures can be used as a tool for further exploration of how humans
come to know objects in their environment and how behavior is
affeCted by perceptual and cognitive style.

Other research on field dependence/independence and cross-modal
sensory transfer ability has demonstrated a relationship of these
traits to intelligence (Ghuman, l977), creativity (Damusis &
Desjarlais, l977), age (Gibson, 1967), and ability of parent in simi-
lar tasks (Schaffer, l969). The interrelationship of the two traits
investigated in this study and their relationships to other cognitive
and perceptual traits point quite sharply to the systemic nature of
cognitive and perceptual processes in human beings. Further study
of the relationships between aspects of the cognitive and perceptual
process will, no doubt, add to the body of knowledge concerned with
how humans relate to their environments and how they use information
from their environments to make decisions.

The results of this study are particularly applicable to the

development of screening materials for special education. Batteries

68

of tests for assessing cognitive and perceptual performance have been
developed by Special educators to help determine nature and direction
of perceptual problems in children with reading difficulties and
learning disabilities (Temple, Williams, & Bateman, l979). Knowledge
of the interrelationships between cognitive and perceptual abilities
and their related measures will help to reduce the number of tests
needed for general testing by providing information about which tests
will yield the same or similar information and can, therefore, be
combined or eliminated. It will also increase the number of asso-
ciated measures for testing for a specific disability. This will
lead to less stress in testing situations for both the child and the
examiner and more accurate assessment of the child.

General knowledge of organization of cognitive and perceptual
learning schemes is of utmost importance in the continual develop-
ment of appropriate educational materials and methods for all children.
It has been noted that cross-modal transfer ability is strongly
related to reading skill in children beyond second grade (Derevensky,
l977) and that the Children's Embedded Figures Test relates to per-
formance subtests on the WISC (Damusis & Desjarlais, l977). The
visual/spatial component found in both field independence and cross-
modal transfer ability seems to be an essential factor in creativity
and independence in children, and it has been shown that training in
these areas can increase performance in visual/spatial abilities,
especially for girls (Butler, 1979). Knowledge that tasks involving
cross-modal transfer ability are related to field independence can

help educators in training efforts in visual/spatial skills and may

 

69

aid in the development of tasks and methods that can help to increase
both reading skills and visual/spatial skills in children. More
information about the assessment of ability to disembed would be
helpful in identification of children who show talent in art and
music (Witkin, Moore, Goodenough, & Cox, 1977).

It has been demonstrated that field dependence/independence is
related to numerous aspects of child-rearing practice and family
functioning (Witkin & Goodenough, 1981). The understanding of the
ability to use perceptual information to solve problems and how it
develops in individuals and within a family context has widespread
implications for study of the family and of child rearing. Knowledge
of how perceptual information is transferred and how that process is
taught within the family could be used to enhance the study of decision
making in families. A particular emphasis in this area might be
parent-child relationships and the interaction of inherited percep-
tual ability with learned behavior. Communication patterns within
the family also depend on perception of information, and this, in
turn, has bearing on personality and role development of family mem-
bers. Studies such as the one reported in this paper help to provide
information about the linkages between perceptual abilities and lay
the groundwork for an understanding of the interplay among numerous
abilities that comprise an individual's cognitive and perceptual
style and how that style affects interactions with other human

beings.

70

Suggestions for Future Research

An initial suggestion would be to repeat this study correcting,
as far as possible, those problems listed as limitations to the
study, i.e., obtain a more heterogeneous sample; obtain information
about subjects, such as intelligence test scores and socioeconomic
status of subjects' families; select more challenging geometric
solids for stimuli sets in cross-modal tasks; test several grades
rather than two; and separate cross-modal frOm intramodal tasks for
data analysis. It might also be interesting to determine the rela-
tionship of other figure-ground tests to the Children's Embedded
Figures Test and Cross-modal Sensory Transfer Tasks to determine
how strong the relationship among tests might be. The CEFT has been
related to a number of diverse human functions; however, cross-modal
transfer ability has been studied primarily in relation to school
learning. Given that CEFT has been shown to relate to cross-modal
ability in school-age children, it would be useful to know if and how
cross-modal transfer ability is related to personality development,
family functioning, and child-rearing practices. Evidence such as
this would further solidify the linkage between field independence

and cross-modal transfer ability.

Summary

This study demonstrated that for children in kindergarten and
fourth grade, there was a positive and significant relationship
between measures of field dependence/independence and cross-modal

transfer. This relationship showed grade-related patterns of

 

 

71

improvement, and there was no significant difference between per-
formances of boys and girls.

The results of this study and others like it have both theoreti-
cal and practical implications. In the realm of theory, this study
has added to the empirical support for a hypothesized cognitive and
perceptual system that governs intellectual and family functioning.
The results of this study can also be used to help create both special-
and general-education materials and teaching methods for areas such
as reading and visual/spatial tasks. The study of interactions of
individuals within a family can also be enhanced through continual
discovery of the relationships of perceptual and cognitive functions

that affect human social behavior.

 

APPENDICES

72

APPENDIX A

INFORMATION TO CLAYTON COUNTY SCHOOL BOARD

73

MICHIGAN STATE UNIVERSITY

 

COI IAIEGE OF HUMAN ECOLOGY EAST LANSING ° MICHIGAN '

DFPARTMENT OF FAMILY AND CHILD ECOLOGY

431 Abbott Road, Apt. 3
East Lansing, Michigan 48823

February 19, 1982

Mr. Emmett Lee

Assistant Superintendent of Schools
Board of Education-Clayton County
Jonesboro, Georgia 30236

Dear Mr. Lee:

In response to our telephone conversation of February 17, I am sending you
the enclosed information concerning the research project I am proposing for
kindergarteners and fourth graders at Lake Harbin Elementary School. This
research is being prOposed as part of the requirements for my Ph.D. in
Child DevelOpment from Michigan State University. A full prOposal will be
sent separately.

Should you decide to allow me to conduct this research project at Lake
Harbin Elementary School, the weeks of March 15-19 and March 22-26 would be
my first choice of times to gather the data. Of course, I am very willing
to work around a schedule that is convenient for you and for the school
personnel.

I can be reached at the above address or at (517) 332-3337 if you have any
questions. My dissertation director, Dr. Lillian Phenice, can be reached
at (517) 353-5248. Thank you for your consideration.

Sincerely,

Margaret E. Griffin

MSl.’ u an AI/imulwe Adina/Fund! ()Donrlunir. !: :!:.':;.':r.-:

43824

 

 

75

Proposal Abstract

This research project is being conducted to determine the relationship
between field dependence/independence and visual/tactile cross—modal
sensory transfer. In practical terms, what we are looking for are
links in children's abilities to determine essential features of
objects and to use those features to make decisions about similar
objects. Much research has been conducted on children's abilities in
field dependence/independence tasks and in cross-modal sensory transfer.
However, very few studies have been conducted on wide-scale ability

to determine essential features. We believe that the mechanism under-
lying both abilities is the same and hope to gain evidence to bolster
this idea through this research.

The procedure for testing the two abilities is relatively simple. The
measure for field dependence/independence is a standardized test called
Children's Embedded Figures Test (CEFT). The test was developed by
Karp and Konstadt and is published by Consulting Psychologists Press,
Inc. The CEFT is administered individually and consists of a series of
pictures in which the child is to locate a specified geometric figure.
There is no standardized test of cross-modal transfer ability; however,
there is a standard methodology for measurement of skill in that area.
The methodology calls for the child to select from an array of figures
a standard geometric stimulus. If the stimulus is presented visually,
the child will select its match tactilely and vice-versa. The object
is to match what the child sees with what he/she touches. The cross-
modal transfer tasks are also administered individually. Total test-
ing time for each child is estimated at 30 minutes. The researcher,
Margaret Griffin, will take responsibility for all testing.

This proposal has been submitted to all appropriate college and uni-
versity committees.

Subjects Needed

 

3O Kindergarteners--l5 males, l5 females
30 4th graders —-l5 males, l5 females

 

76

I will need cooperation of the school in the following:
l. Distribution and collection of informed consent forms.

2. A room at the school in which to conduct testing sessions.

I will provide the following to the school:
l. Informed consent forms for each child's parent(s).
2. Orientation meetings for parents, teachers, and children.

3. Proper supervision of children to, from, and during the
testing sessions.

4. A report of the results of the study.

5. Library materials appropriate for kindergarten and fourth
grade (to be selected by school personnel).

77

Ethical Considerations

 

All children participating in this study will have a signed letter of
informed consent on file with the researcher. No child will be asked
to participate in a testing session against his/her own will, nor will
any attempt be made to coerce the child to participate. Should any
child show signs of distress during any session, the session will be
terminated and the child returned to home room. Every effort will be
made to cooperate fully with teachers in order to minimize any dis-
ruptive effect that removal of children from the classroom might have.
It will be the responsibility of the researcher to escort each child
to and from the testing site. It is not expected that participation
in this study will be either harmful or beneficial to children in any
way.

All data will be collected, stored, and reported using a four-digit
code number (grade, sex, selection number) to identify the individual
child. No names of children will be recorded in any phase of the study.

APPENDIX B

INFORMATION SENT TO SUBJECTS' PARENTS

78

79

Letter of Personal Information

Dear Parents:

My name is Margaret (Meg) Griffin. I am a graduate student at Michigan
State University in the field of Child Development. I have lived in
Michigan for 3-l/2 years now, but I am originally from Tennessee. I
have been a teacher for four years and am presently working part-time
for Michigan 4-H while completing my degree. I learned about your
school district through my sister, Patty Griffin, who is a physical
education teacher at Lake Harbin Elementary School.

My major research interest is centered around how the way people see
affects the way they think. The research that I plan to conduct at
your child's school fits into this type of research plan. In particu-
lar, your child if he or she participates in the study will be helping
me to find out if children can look at a simple geometric figure and
then locate that figure in a picture or by feeling for it inside a

box. Collection of this type of simple information helps us to develop
a more complete idea of how children organize their actions and thoughts.
That information will ultimately help us to be better teachers of chil-
dren.

I am looking forward to meeting you at the parent/staff information
meeting to discuss the research with you and answer any questions.
See you on

 

Thank you,

Margaret Griffin

80

Letter of Informed Consent

Dear Parent:

The classroom in which your child is enrolled has been chosen as one

of several kindergarten and fourth-grade classrooms from which children
will be selected to participate in a study of perceptual styles used

by children. Each child in the study will be asked to complete two
sets of exercises, the children's form of the Embedded Figures Test

and three tasks involving matching blocks by sight and touch. An
observer will watch each child complete the exercises and will record
number of correct responses. Each child participating in the study
will be identified by a code number, and no use of names will be made
in any reporting of results. All materials that will be used in the
study will be available at the school on at
for your inspection.

 

 

If you choose to allow your child to participate in this study, you
may review all data concerning your child at the close of the study.
You are also free to withdraw consent for your child's participation
at any time during the study. To do so, please inform the child's
teacher in writing that you are withdrawing from the study. The
teacher will inform the experimenters of your decision.

Any child participating in the study has the freedom to choose not to
participate on his/her scheduled day. If this occurs, the child may
choose to do the task on another day. If a child does not want to par-
ticipate in the study, he/she will be dropped from the study. A deci-
sion to withdraw from the study by either parent or child will not
affect the child's standing in his/her class.

A four-digit code number indicating grade, gender, and selection number
of the child will be assigned to each child participating in the study.
This number will be used to identify data associated with each child.
There will be no record of children's names kept. No names of chil-
dren will be used in reporting the study.

All phases of the study will be carried out at the school. It is
estimated that the amount of time missed from class will be approxi-
mately 30 minutes. It is not expected that participation in this study
will be either harmful or beneficial to the children in any way.

 

81

If you give consent for your child to participate, please read and
sign the attached sheet and return to your child's teacher by

 

If your child is chosen by random drawing from those having parental
consent to participate, you will be informed by

 

Please keep this letter for your information.

Thank you,

Margaret Griffin

**To give permission: detach, sign, and return this form to the teacher*

I have read the information sent to me concerning the study of chil-
dren's perceptual styles being conducted at Lake Harbin Elementary
School by Margaret Griffin. I give permission for my child

to participate in the study.

 

(name)

Signed
Relationship to child
Date

Witness

 

 

82

Dear Parent:

Your child has been selected by a random drawing
from those children having parental permission to participate inthe
study of children's perceptual/cognitive abilities being carried out
by Margaret Griffin. Your child's code number is . You will
receive a copy of the task scores obtained by your child at the end of
the data-collection period. A complete report of the project results
will be sent to the school in early summer.

 

Thank you for your cooperation.

Sincerely,

Margaret Griffin

Dear Parent:

The children have been randomly selected for the study of perceptual/

. cognitive abilities being carried out by Margaret Griffin. Your child
was not one of those whose name was drawn. The selection process was
a blind, random drawing from the names of all children who had parental
permission to participate in the study. Selection was based on the
luck of the draw and had nothing to do with the personal characteris-
tics of your child.

Thank you for your c00peration.

Sincerely,

Margaret Griffin

APPENDIX C

CROSS-MODAL SENSORY TRANSFER TASKS

83

 

84

Numbering System1 for Milton Bradley Geometric Solids
Used for Stimuli in Cross-modal Tasks

l. equilaterial triangular prism

2. right-angled triangular prism

3. square prism

4. hexagonal prism

5. octagonal prism2

6. cylinder

7. truncated cylinder

8. cone

9. frustum of cone

10. square pyramid

ll. triangular pyramid

l2. truncated cone

l3. cone, plane parallel axis

l4. cone, plane parallel one element
l5. cube
l6. cylinder, height equal to diameter
l7. hemisphere

l8. sphere3

l9. rectangular solid, three dimensions different
20. oviod

2l. ellipsoid

22. oblate spheroid

 

1This is the system devised by the Milton Bradley Company.

2This solid was not used because of its similarity to #4.

3This solid was not used because it rolled out of the "feeling
box" too easily.

85

Positioning and Ordering of Stimuli Sets--Cross-modal Transfer Tasks

Task l

  
 
  

Subject

Researcher

x = stimulus A-E = matching set

 

 

 

 

 

 

 

 

Task 2
Subject ___
X
hand
Researcher
x = stimulus A-E = matching set
Task 3
Subject e—~

 

 

 

Jhand
\ x
Researcher

x = stimulus A—E = matching set

 

 

 

 

 

Task I

stimulus

D)

SOQVONU‘I-wa—J

3

4

oo

dd
\l-h

Task 3

stimulus

.4
O

O)

\DmNOUIth-J

__a
O

_.1
(.10

_J
01

N
C)

d
d

._I
A

86

Positioning and Ordering of Stimuli Sets--

Cross-modal Transfer Tasks

 

 

Task 2
matching set stimulus
3, ll, l3, 19, l a. 2
7, 4, 17, 20, 9 6
2, 7, 20, 8, l6 __§L__

 

11, 6, 14, 4, 22

_;
N

 

21, I, 9, 14, 16

_a
U"

 

6, 17, 8, 13, IO

 

7, 16, 12, 11, 21

N
N

 

 

 

\OQVO‘U’l-QOON—J
._a
to

 

10,14, 2,8, 20 ._3__
6, 13, 22, 19, 1 ._1_
1,11,17,9, 12 _1_o_
3, 12, 14, 2o, 9 1o 13

 

matching set
6, 12, 3, 22, l9

 

I, 13, 21, 10, 7

 

II, 10, 12, 17, I

 

7, 8, 13, 2, 15

 

16, 17, 19, 4, 7

 

22, 6, I, 20, 15

 

9, I, 2, 3, II

 

6, 2, 12, 8, 4

 

21, 14, 8, 17, 3

 

11, 8,16, 4,12

 

22, 15, 19, I4, 6

 

matching set
l4, 4, 2l, 2, l6
6, 8, 15, 10, l7
22, 15, 9, 14, ll

 

20, 12, I9, 3, 9
2,10,12, 20,15
20, 14, 19, 4, 3

 

22,19,17,13, 4

 

21, 10, 15, 3,,7
20, 7,11J 9, 16
2, 7,13,15,10
21, 19, 13, 12, I

 

87

Cross-modal Transfer Tasks--Scoring Sheet

Subject code number

(grade) (sex) (i.d.) (i.d.)

ESL}. E‘s—k].
a ______ a. ______
1 __ 1.__
2.__ 2._
3.___ 3._
4._ 4.__
5.__ 5,____
6.____ 6.______
7._ 7.__
8._ 8.______
9._ 9.__
1o.___ 1o.__
1 = /10 1 = /10

Total Cross-modal Score = /30

‘0 oo \1 01 01 «b on N
o o o o o o o o

._..a
O

APPENDIX D

CHILDREN'S EMBEDDED FIGURES TEST

88

 

SCORE SHEET FOR

“m“ NAME

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

CHILDREN'S CLASS
EMBEDDED
FIGURES TEST WITH DATE SEX: 44—F—
m DATE EXAMINER
TENT DESCRIPTION SCORE HOUSE DESCRIPTION SCORE
P1 P3
92 H1
T1 H2
72 "3
T3 ”4
T4 ”5
Ts ”a
T6 l"7
T7 H8
Ts H9
Ts H1o
T10 ”11
T11 H12
”13
"14
TotalSoon TENT 1,___.__.1 ToulScoroHOUSE 1____11
TOTAL TEST SCORE T;____Jl

 

 

".
O. ‘5

9°. cowsumws PSYCHOLOGISTS Pnsss,1wc.

  

I
o" "'1

577 College Avenue, Palo Alto, California 94306

a
0.3‘

APPENDIX E

DATA RECORDS

9O

 

91

INDIVIDUAL DATA RECORD

Code number
(grade) (Sex) (1.61) (i.d.)

 

 

Birthdate

 

Racial/ethnic group

 

CROSS-MODAL TRANSFER TASKS CHILDREN'S EMBEDDED FIGURES TEST
1. Combined score ______ 1. Tent score

2. Task 1 score ______ 2. House score

3. Task 2 score ______ 3. Total score

4. Task 3 score

REMARKS

 

 

 

PARENT REPORT SHEET--INDIVIDUAL CHILD DATA

Child's Code Number
(grade) (sex) (1.d.) (i.d.)

CROSS-MODAL TRANSFER TASKS CHILDREN'S EMBEDDED FIGURES TEST
1. Combined score ‘_____ 1. Tent score ._____

2. Task 1 score ______ 2. House score ______

3. Task 2 score 3. Total score ______

4. Task 3 score

APPENDIX F

CODE BOOK

92

93

 

CODE BOOK
Vaagggle va{;32}e Value Label Columns Format
ID Identification XXX l- 2 F2.0
Vl Grade l = kindergarten 3 Fl.O
2 = fourth
V2 Gender 1 = female 4 Fl.0
2 = male
V3 Age XXX, 5- 7 F3.0
V4 Race 1 = white 8 Fl.O
2 = black
V5 Combined score XXX 9-lO F2.0
V6 Task 1 score XXX 11-12 F2.0
V7 Task 2 score XXX l3-l4 F2.0
V8 Task 3 score XXX 15-16 F2.0
V9 Tent score XXX 17-18 F2.0
VlO House score XXX l9-20 F2.0
Vll Total score XXX 2l-22 F2.0

REFERENCES

94

REFERENCES

Abravanel, E. Integrating the information from eyes and hands: A
developmental account. In H. L. Pick, Jr., & R. D. Walk (Eds.),
Intersensoryperception and sensory integration. New York:
Plenum, 1981.

Ayres, A. J. Interrelations among perceptual-motor abilities in a
group of normal children. American Journal of Occupational
Therapy, 1966, 29, 288-292.

Ball, 5., & Bogatz, G. A. The first year of Sesame Street: An
evaluation (ETS PR 70-15). Princeton, N.J.: Educational
Testing Service, 1970.

Barrett, G. V., & Thornton, C. L. Cognitive style difference
between engineers and college students. Perceptual and
Motor Skills, 1967, 25, 789-793.

 

 

Birch, H. G., & Lefford, A. Visual differentiation, intersensory
integration, and voluntary motor control. Mono ra hs of the
Society for Research in Child Development, 1967, EEIZI, 1-87.

Blank, M., & Bridger, W. H. Cross modal transfer in nursery school

children. Journal of Comparative Physiological Psychology, 1964,
§§, 272-82.

 

Buros, O. K. The seventh mental measurements yearbook. Highland
Park, N.J.: Gryphon Press, 1972.

Busch, J. C., & DeRidder, L. M. Conformity in preschool disadvantaged
children as related to field-dependence, sex, and verbal rein-
forcement. Psychological Reports, 1973, 22, 667-673.

 

Busse, T. V. Establishment of the flexible thinking factor in fifth
grade boys. Journal of Psychologx; 1968, 62, 93-100.

 

Busse, T. V. Child-rearing antecedents of flexible thinking.
Developmental Psychology. 1969, 1, 585-591.

Butler, E. J. Visual and haptic training and cross-modal transfer

reflectivity. Journal of Educational Psychology, 1979, 21(2),
212-219.

95

96

Campbell, D. T., & Stanley, J. C. Qperimental and quasi-experimental
designs for research. Chicago: Rand McNally, 19661

Connor, J. M., Schackman, M., & Serbin, L. A. Sex-related differences
in response to practice on a visual-spatial test and generaliza-
tion to a related test. Child Development, 1978, 49(1), 24-29.

Crandall, V. J., & Sinkeldam, C. Children's dependent and achievement
behaviors in social situations and their perceptual field depend-
ence. Journal of Personality, 1964, 22, 1-22.

Cross, P. Accent on learning, San Francisco: Jossey-Bass, 1976.

Cullen, J. F., Harper, C. R., & Kidera, G. J. Perceptual style dif-
ferences between airline pilots and engineers. Aerospace
Medicine, 1969, 4, 407-408.

Damusis, V., & Desjarlais, L. Field dependence-independence as
cognitive style: Implications for instruction (French). Revue
de Psychologie Appliquée, 1977, 22(3), 211-229.

Davidson, P. W., Pine, R., Wiles-Kettenmann, M., & Appelle, S.
Haptic-visual shape matching by mentally retarded children:
Exploratory activity and complexity effects. American Journal
of Mental Deficiency, 1981, 84(5), 526-533.

 

Derevensky, J. Cross modal functioning and reading achievement.
Journal of Reading Behavior, 1977, 2(3), 233-251.

Dolecki, P. G. The effects of a perceptual training program on
conservation task performance and field independence in first
grade children (Doctoral dissertation, University of Utah,
1976). Dissertation Abstracts International, 22, 878A.
(University Microfilms No. 76-17, 311)

Dyk, R. B. An exploratory study of mother-child interaction in
infancy as related to the development of differentiation.
Journal of the American Academy of Child Psychiatry, 1969, 8,
657-691.

 

Dyk, R. B., & Witkin, H. A. Family experiences related to the
development of differentiation. Child Development, 1965, 29,
21-55.

Early, G. Intermodal abilities as predictors of academic achievement.
Academic Therapy, 1976, 12(2), 163-169.

 

Fishbein, G. M. Perceptual modes and asthmatic symptoms: An applica-
tion of Witkin's hypothesis. Journal of Consult1ng Psychology,
1963, 22, 54-58.

97

Fishbein, H. D., Decker, J., & Wilcox, P. Cross-modal transfer of
spatial information. British Journal of Psychology, 1977, 48(4),
503-508.

Galin, D., Diamond, R., & Herron, J. Development of crossed and
uncrossed tactile localizations on the fingers. Brain and
Language, 1977, 4(4), 588-590.

Gardner, R. W., Jackson, D. N., & Messick, S. J. Personality organi-
zation in cognitive controls and intellectual abilities. Psycho-
logical Issues, 1960, 2(4).

Ghuman, P. A. An exploratory study of Witkin's dimension in relation
to social class, personality factors, and Piagetian tasks.
Social Behavior and Personality, 1977, 5(1), 87-91.

Gibson, E. Principles of perceptual learning and development. New
York: Appleton-Century-Crofts, 1967.

 

Gibson, J. J. The senses considered as perceptual systems. Boston:
Houghton-Mifflin, 1966.

 

Gildemeister, J. E., & Friedman, P. Cognitive style and visual analy-
sis in first graders of hi h and low verbal ability. Perceptual
and Motor Skills, 1978, 4223 pt. 1), 759-766.

 

 

Glass, G. V., & Stanley, J. C. Statistical methods in education and
psychology. Englewood Cliffs, N)J.: Prentice-Hall, 1970.

 

Goodenough, D., & Karp, S. A. Field dependence and intellectual
functioning. Journal of Abnormal and Social Psychology, 1961,
63, 241-246.

Goodenough, D. R., & Eagle, C. J. A modification of the embedded
figures test for use with young children. Journal of Genetic
Psychologx; 1963, 192, 67-74.

 

Gough, H. G., & Olton, R. M. Field independence as related to non-
verbal measures of perceptual performance and cognitive ability.
Journal of Consulting and Clinical Psychology, 1972, 22, 330-342.

Hays, W. L. Statistics for the social sciences. New York: Holt,
Rinehart, & Winston, 1973.

Holtzman, W. H., Diaz-Guerrero, R., & Swartz, J. D. Personality
development in two cultures: A cross-cultural longitudinaI stugy,
of sChoolFChderen in Mexico and the United States. Austin:
University of Texas Press, 1975.

Hurwitz, 1., Wolff, P. H., Bortnick, B. D., & Kokas, K. Nonmusical
effects of the Kodaly music curriculum in primary grade ch1ldren.
Journal of Learning Disabilities, 1975, 2, 167-174.

98

Jain, R., & Sinha, D. Inter-modality transfer and kinesthetic figural
ggtggeffect. Journal of Psycho1pgical Researches, 1978, 22(2),

Jarman, R. F. Cross-modal and intramodal matching: Relationships to
simultaneous and successive syntheses and levels<rfperformance
among three intelligence groups. Alberta Journal of Educational
Research, 1978, 24(2), 100-112.

 

Karp, S. A., & Konstadt, N. L. Manual for the Children's Embedded
Figures Test. Baltimore: Authors, 1963.

 

 

Karp, S. A., & Pardes, H. Psychological differentiation (field
dependence) in obese women. Psychosomatic Medicine, 1965, 22,
238-244.

 

Karp, S. A., Witkin, H. A., & Goodenough, D. R. Alcoholism and
psychological differentiation: The effect of alcohol on field
dependence. Quarterly Journal of Abnormal Psychology, 1965, 29,
262-265.

Kleinman, J. M. Developmental changes in haptic exploration and
matching exploration. Developmental Psychology. 1979, 19(4),
481-482.

 

Knudson, K. H., & Kagan, S. Visual perspective role-taking and
field-independence among Anglo American and Mexican American
children of two ages. Journal of Genetic Psychology, 1977, 121
(2), 243-253.

 

Konstadt, N., & Forman, E. Field dependence and external directed-
ness. Journal of Personality and Social Psyphology, 1965, 1,
490-493.

Linton, H. B. Dependence on external influence: Correlates in
perception, attitudes, and judgment. Journal of Abnormal and
Social Psychology, 1955, 91; 502-507.

 

Maccoby, E. E., Dowley, E. M., Hagen, J. W., & Degerman, R. Activity
level and intellectual functioning in normal preschool children.
Child Develppment, 1965, 29, 761-770.

 

MacEachron, A. E., & Gruenfeld, L. W. The effects of family authority
structure and socioeconomic status on field independence.
Journal of Social P§ychology, 1978, 194; 49-56.

Maslow, P., Frostig, M., Lefever, D. W., & Whittlesey, J. R. B. The
Marianne Frostig Developmental Test of Visual Perception, 1963
standardization. Perceptual and Motor Skills, 1964, 19, 463-
499.

99

Minard, J. G., 8 Mooney, W. Psychological differentiation and per-
ceptual defense: Studies of the separation of perception from
gmotion. Journal of Abnormal and Social Psychology, 1969, 74,

31-139. __'

Nebelkopf, E. B., 8 Dreyer, A. S. Continuous-discontinuous concept
attainment as a function of individual differences in cognitive
style. Perceptual and Motor Skills, 1973, 29, 655-662.

 

Olesker, W. Cognition and the separation-individuation process: A
study of three boys at nursery school. B§ychoana1ysis and
Contemporary Thought, 1978, 1, 237-268.

 

Oltman, P. K., Goodenough, D. R., Witkin, H. A., Freedman, N., 8
Friedman, F. Psychological differentiation as a factor in
conflict resolution. Journal of Personality and Social Psy-
chology, 1975, 22, 730-736.

 

Pascual-Leone, J. Cognitive development and cognitive style: A
general psychological integration. Unpublished doctoral disser-
tation, York’University, 1969}

 

 

Piaget, J. The mechanisms of perception. London: Routledge 8
Kegan Paul, 1969.

Piaget, J., 8 Inhelder, B. The child's conception of space. New
York: Humanities Press, 1956.

Pick, A. 0., Pick, H. L., 8 Thomas, M. L. Cross-modal transfer and
improvement of form discrimination. Journal of Experimental
Child Psychology, 1966, 2, 279-288.

 

 

Quinlan, D. M., 8 Blatt, S. J. Field articulation and performance
under stress: Differential predictions in surgical and psychi-
atric nursing training. Journal of Consultingiand Clinical
Psychology, 1972, 29, 517.

Rudel, R. G., 8 Teuber, H. L. Decrement of visual and haptic Muller-
Lyon illusions on repeated trials: A study of cross-modal transfer.
Quarterly Journal of Experimental Psychology, 1963, 19, 125-131.

Rudel, R. G., 8 Teuber, H. L. Crossmodal transfer of shape discrimi-
nation by children. Neuropsychologia, 1964, 2, 1-8.

 

Scallon, R. J., 8 Herron, W. G. Field articulation of enuretic boys
and their mothers. Perceptual and Motor Skills, 1969, 22,
407-413.

Schaffer, M. C. Parent-child similarity in psychological differen-
tiation. UanbTiShed doctoral dissertation, Purdue University,

 

lOO
Schooler, C. Childhood family structure and adult characteristics.

Sociometry, 1972, 29, 255-269.

Schwartz, D. W., 8 Karp, S. A. Field dependence in a geriatric
population. Perceptual and Motor Skills, 1967, 24, 495-504.

 

Sherman, J. A. Problem of sex differences in space perception and
aspects of intellectual functioning. Psychological Review, 1967,
74, 290-299.

 

Simpkins, K. E. Tactual discrimination of shapes. Journal of Visual
1mpairment and Blindness, 1979, 22(3), 93-101.

 

 

5011, J. The effect of frustration on functional cardiac disorder
as related to fieldioriEntatiBn. Unpublished doctoral disserta-
tion,’Ade1phi'University, 1963.

Steingart, I., Freedman, N., Grand, 5., 8 Buchwald, C. Personality
organization and language behavior: The imprint of psychological
differentiation on language behavior in varying communication
conditions. Journal of Psycholinguistic Research, 1975, 4,
241-255.

 

Stevens, J. C. A study of the relationships between field independence,
dogmatism, and creatiVity in rural seventh grade students. Unpub-
1i§hed doctoral dissertation, UniVersity afiGeorgia, 1969.

Temple, 1. G., Williams, H. G., 8 Bateman, N. J. A test battery to
assess intrasensory and intersensory development of young chil-
dren. Perceptual and Motor Skills, 1979, 49(2), 643-659.

 

Vaught, G. M. The relationship of role identification and ego strength
in the rod-and-frame test. Journal of Personality, 1965, 22,
271-283.

Walk, R. D. Tactual and visual learning of forms differing in degree
of symmetry. Epychonomic Science, 1965, 2, 93-95.

Walk, R. 0., 8 Pick, H. L. Intersensory perception and sensory inte-
gration. New York: Plenum, 1981.

Willemsen, E., Buchholz, A., Budrow, M. S., 8 Geannacopulus, N.
Relationship between Witkin's rod-and-frame task and Piaget's
water-line task for college women. Perceptual and Motor Skills,
1973, 29, 958.

Witkin, H. A. Psychological differentiation and forms of pathology.
Journal of AbnormalPsychology, 1965, 29, 317-336.

101

Witkin, H. A. Cognitive styles in personal and cultural adaptation.
Worcester, Mass.: Clark University Press, 1978.

Witkin, H. A., Dyk, R. B., Faterson, H. F., Goodenough, D. R., 8

Karp, S. A. Psychological differentiation. New York: Wiley,
1962.

Witkin, H. A., 8 Goodenough, D. R. Field dependence and interper-
sonal behavior. Esychological Bulletin, 1977, 94, 661-689.

Witkin, H. A., 8 Goodenough, D. R. Cognitive sty1es: Essence and
origins--Field ggpendence and fieldFindePendence. New York:
International Universities Press,"Inc., 1981.

Witkin, H. A., Goodenough, D. R., 8 Karp, S. A. Stability of cognitive
style from childhood to young adulthood. Journal of Personality_
and Social Psychology, 1967, 2, 291-300.

 

Witkin, H. A., Moore, C. A., Goodenough, D. R., 8 Cox, P. W. Field-
dependent and field-independent cognitive styles and their
educational implications. Review of Educational Research, 1977,
47: 1'64.

Witkin, H. A., Oltman, P. K., Raskin, E., 8 Karp, S. A. A manual
for the embedded figures tests. Palo Alto: Consulting Psycholo-
gists Press, 1971.