Group Title: relationship of the concept of differentiation to the structure of verbal behavior
Title: The relationship of the concept of differentiation to the structure of verbal behavior
CITATION PDF VIEWER THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00097792/00001
 Material Information
Title: The relationship of the concept of differentiation to the structure of verbal behavior
Physical Description: v, 91 leaves. : ill. ; 28 cm.
Language: English
Creator: Davidov, William H., 1936-
Publication Date: 1968
Copyright Date: 1968
 Subjects
Subject: Verbal behavior   ( lcsh )
Psychology thesis Ph. D   ( lcsh )
Dissertations, Academic -- Psychology -- UF   ( lcsh )
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
 Notes
Thesis: Thesis -- University of Florida.
Bibliography: Bibliography: leaves 85-89.
Additional Physical Form: Also available on World Wide Web
General Note: Manuscript copy.
General Note: Vita.
 Record Information
Bibliographic ID: UF00097792
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000570835
oclc - 13755834
notis - ACZ7817

Downloads

This item has the following downloads:

relationshipofco00davirich ( PDF )


Full Text













THE RELATIONSHIP OF THE CONCEPT OF
DIFFERENTIATION TO THE STRUCTURE
OF VERBAL BEHAVIOR










By

WILLIAM H. DAVIDOV


A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY










UNIVERSITY OF FLORIDA
1968













ACKNOWLEDGMENTS


The writer wishes to gratefully acknowledge the

assistance he received from the members of his supervisory

committee: Drs. Benjamin Barger, Chairman; Hugh Davis,

Harry Grater, and Henry Pennypacker of the Department of

Psychology; and Dr. Wilbur Bock of the Department of

Sociology. He also wishes to thank Drs. Philip Costanzo,

Carolyn Hursch, and Paul Satz for assisting with the statis-

tical analysis.

He wishes to express a special word of appreciation

to Dr. Benjamin Barger, who provided a constant source of

encouragement, constructive and creative suggestions and

questions, as well as the willing and consistent patience

and understanding which made the completion of this study

possible.

Most of all, he wishes to thank his lovely wife,

Jane, who worked many long hours, side by side with the

writer, and whose help in editing and typing, as well as

her patience and encouragement throughout, was truly

unmeasureable and deserves a lifetime of appreciation.















TABLE OF CONTENTS


Page


ACKNOWLEDGMENTS . . . . . .

LIST OF TABLES . . . . . .

CHAPTER

I INTRODUCTION . . . . .

The Concept of Differentiation
Verbal Behavior . . . .
Units of Analysis . . . .
Hypotheses . . . . .

II METHOD . . . . . .


a 0 0 0 0 0 0


0 0 . 0 0 v


. . . . 1


* 0


0 0 0 0 0 0 0


Subjects . . . . .
Apparatus . . . . . .
Procedure . . . . . .
Method of Analysis . . . ..

III RESULTS . . . . . . . .

The Discriminant Function Analyses


IV DISCUSSION . . . . . .

V SUMMARY . . . . . . .

APPENDICES . . . . . . . . .

APPENDIX A. Structured Interview .

APPENDIX B. Representative Cards from
Embedded Figures Test .


* 0 0

* 0 0

* 0 0

* 0 0


the
0 0


a 0 0


of Anxiety


APPENDIX C. Two-Way Analyses of Variance
for the Three Verbal Expressions


APPENDIX D. Results of Discriminant Function
Analyses II and III . .





iii


0 0 0 0 0 0 & 0 0 0 0


0
0
0
0


*


* 0








Page

APPENDIX E. Representative Protocols for a
Low-differentiated and a High-
differentiated Individual . . 66

APPENDIX F. Frequencies of Word Categories
Which are Preceded by a Speech
Disturbance . . . . . .76

APPENDIX G. Ratio Scores for All 30 Subjects 78

BIBLIOGRAPHY A . . . . . . . 85

BIOGRAPHICAL SKETCH . . . . . . . . 90












LIST OF TABLES


Table Page

1 Analyses of Variance of the 10 Verbal
Measures for High- and Low-differentiated
Subjects . . . . . . . . 34

2 Mean Ratio Scores and the Differences Between
the Means for the Criterion Groups on the 10
Verbal Measures (Discriminant Function
Analysis I) . . . . . . . . 38

3 Classification of Subjects by Use of Discri-
minant Function I . .. . ..... 41

4 Two-Way Analyses of Variance for the Three
Verbal Expressions of Anxiety . . . . 61

5 Results of Discriminant Function Analyses
II and III . . . . . . . . 63

6 Classification of Subjects in the Spontaneous
Speech Task by Use of Discriminant Function
III . . . . . . . . . . . 64

7 Frequencies of Word Categories Which are
Preceded by a Speech Disturbance for a Low-
differentiated Subject in Structured and
Unstructured Speech Conditions . . . . 76

8 Frequencies of Word Categories Which are
Preceded by a Speech Disturbance for a High-
differentiated Subject in Structured and
Unstructured Speech Conditions . . . . 77

9 Ratio Scores for the 10 Verbal Measures for
the Combined Speech Tasks . . . . . 78

10 Ratio Scores for the Three Verbal Expressions
of Anxiety for the Structured Interview . . 82

11 Ratio Scores for the Three Verbal Expressions
of Anxiety for the Spontaneous Speech Task . 83













CHAPTER I


INTRODUCTION



Thepurpose of this study is to investigate the

relationships between cognitive-perceptual style and

verbal behavior.



The Concept of Differentiation

Cognitive-perceptual styles have become highly

important in the field of psychology. The cognitive style

dichotomies,which have been proposed, are many; e.g.,

Differentiated--Non-differentiated (Witkin et al., 1962),

Sharpeners and Levelers (Gardner et al., 1960; 1959)

Yeasayers and Naysayers (Couch and Keniston, 1960), etc.

These dichotomies have been found to be related to per-

ception, intelligence, personality, sex differences,

body concept, and pathology. The most extensively re-

searched work has been that of the continuum of differ-

entiation (Witkin et al., 1962; 1954; Witkin, 1965).

Witkin and his associates have presented causal-etio-

logical factors beginning in childhood which result in

the development of a "specific" cognitive or perceptual

style in an individual. They report evidence that this








style is distinctly related to an individual's perception

of himself, his environment, and the relationship of

himself to his environment (Witkin et al., 1962). Addi-

tionally, Witkin has recently theorized that this style

is, at least in part, genetically determined. "Under

feral conditions, survival may depend on the ability to

separate quickly, from the complex visual and auditory

fields in which the animal lives, those sights and sounds

which signify danger, food, etc." (Witkin, 1965, p. 333).

Witkin and his associates (1962; 1954) have demonstrat-

ed a developmental trend in perceptual style from eight years

of age to young adulthood, progressing from global to a

more analytic approach. His studies have shown that an

individual's "characteristic approach to the world" is usual-

ly determined by the age of eight (Witkin et al., 1962).

The mode of perception associated with early stages of

development is referred to as "field-dependent" (F-D),

"global," "low-differentiated," or "non-differentiated"

(Witkin et al., 1962). Early in development, the child

is believed to experience himself and his environment as

"an amorphous, continuous mass." Formation of the "self"

involves the development of an "inner-core" of experience

and the separation of this core from the environmental

field. Throughout growth, the "self" becomes articulated

from the environment in a more structured manner, During

this period of development, stimulus objects take on








"function and meaning" as a result of continuous, yet

varying, dealings with them. The person who has the

ability to perceive the environment in an articulated

manner such that he is able to perceive items as discrete

from their backgrounds, is referred to as a "field-inde-

pendent," "high-differentiated," or "differentiated" person

(Witkin et al., 1962).

It has been demonstrated that it is possible to

predict an individual's functioning in the areas of percep-

tion, problem solving, intelligence, and personality from

a knowledge of his level of differentiation (Witkin et al.,

1962; 1954).

The concept of increasing articulation has been

applied to the "experience of an immediately present

stimulus configuration," i.e., perception (Witkin et al.,

1962; 1954; Witkin, 1949; Jackson, 1964; Asch and Witkin,

1948). Witkin feels that it also may be applied to "the

experience of symbolic material," i.e., cognition or thinking

(Witkin et al., 1962).

A primary concern with the perceptual-cognitive

Styles has been the adaptive function of cognitive processes

in the psychological economy of the individual. This

has resulted in a search for consistencies across psycho-

logical areas. A by-product of cognitive style research

has been "its contribution to a more integrated, holistic

view of personality" (Witkin, 1965).







For the low-differentiated person, perception

is strongly dominated by the over-all organization

of the field, and parts of the field are experienced

as "fixed" (Witkin, 1965, p. 318). In a field-in-

dependent mode of perceiving, "parts of the field are

experienced as discrete from organized background"

(Witkin, 1965, p. 318).

The cognitive style of field-dependence is

described as follows, "At one extreme [field-dependence,

low-differentiated] there is a consistent tendency for

experience to be global and diffuse; the organization as

a whole dictates the manner in which its parts are

experienced. At the other extreme [field-independence,

high-differentiated] there is a tendency for experience

to be delineated and structured; parts of a field are

experienced as discrete and the field as a whole organized"

(Witkin et al., 1962, p. 319).

Various techniques have been devised to measure

the articulation dimensions e.g., the Rod and Frame Test,

the Tilting Room Test, and the Tilting Chair Test, etc.

(Witkin et al., 1962; 1954). Witkin (1950) developed the

Embedded Figures Test (EFT) for the measurement of field-

dependence. This test is a modification of the Gottschaldt

figures to which color and increased complexity have been

added. The EFT (Appendix B) has-been found to correlate








significantly with other techniques used for measuring the

field-dependence dimension (Jackson et al., 1964; Witkin

et al., 1962; Young, 1959). Short forms have been devised

for the EFT which have been found to correlate up to +.99

with the original scale (Witkin et al., 1962; Jackson, 1956).

The EFT requires the person to locate a simple

figure in a complex design which is so organized as to

conceal the simple figure. The high-differentiated person

is able to discern the figure easily, whereas others may

not discern it for long (5 minutes) periods of time.

High-differentiated persons give evidence of a

developed sense of identity, i.e., "they have an awareness

of needs, feelings, attributes which they recognize as

their own and which they identify as distinct from those

of others" (Witkin, 1965, p. 320). This implies an experience

of self as segregated and structured. Low-differentiated

individuals reveal greater reliance on external sources

for their feelings, attitudes, and judgments.

From his studies reported in 1954, Witkin found the

following personality components associated with the different

levels of differentiation (Witkin et al., 1954): Low-differ-

entiated individuals display general passivity in dealing

with the environment; they lack a sense of self-awareness;

they have relatively poor control of impulses, with accom-

panying fear of aggressive and sexual impulses, and often,

high anxiety; and they have a sense of low self-esteem








with a low evaluation of their bodies. High-differen-

tiated individuals are active in dealing with their envi-

ronment; they have an awareness of "inner-life" and effective

control of impulses with low anxiety; and they have high

self-esteem, including confidence in their bodies.

Upon replicating some of Witkin's experiments,

Young (1959) found that field-dependence (low-differentia-

tion) does correlate with passive dependence, distrust of

one's own feelings, and lack of introspectiveness. Taft

and Coventry (1958) found that S's who score high extro-

version on Eyesenck's introversion-extroversion scale

were significantly less accurate on measures of field

dependence than were those who scored low. The rationale

here was that extroverts would be relatively unprepared to

handle the environment by means of cues emanating from

themselves. A study by Marlowe (1958) found that field

independence (high-differentiation) was related to "intra-

ception," i.e., the need to be analytic in regard-to the

behavior and motives on one's self and others. Linton

(1955) found that the field dependent personality is associ-

ated with high conformity as measured by the auto-kinetic

effect and attitude change measures.

There are distinct sex differences regarding

degree of field-dependence, with females showing a con-

siderably lesser degree of differentiation than males (Witkin

et al., 1962; 1954; Bennet, 1956; Young, 1959). "The sex








differences that have been observed are clear-cut and

pervasive, but they are relatively slight, compared to the

range of individual differences within each sex (Witkin

et al., 1962, p. 221). These stylistic tendencies mani-

fest themselves in the individual's intellectual tendencies

as well. Low-differentiated people do less well in solving

problems which require isolating essential elements from

the presented contexts and using them in different contexts

(Witkin et al., 1967). Jackson (1957) found that time

taken on the EFT correlated -.57, for males and females

combined, with undergraduate scores on the ACE. Goode-

nough and Karp (1961) using the EFT, found it was signi-

ficantly correlated with the Block Design, Figure Completion,

and Object Assembly for two groups of children (ages 10

and 12). Messick and Damarin (1964) found that low differ-

entiated S's showed greater incidental learning when the

material consisted of human faces, as compared to high

differentiated S's. The reverse was found when non-

human incidental material was used.

At any level along the continuum of differentia-

tion, different modes of integration are possible, although

more complex modes are expected with greater differentia-

tion. Adjustment is a direct function of the effective-

ness of integration, i.e. "a: more or less harmonious

working together of parts of the system as a whole with

its environment. Adequate adjustment is found at any








level of differentiation, resulting from integration

effective for that level, although the nature of adjust-

ment that may be considered adequate varies from level to

level" (Witkin, 1965, p. 324). Also, faulty integration

can occur at all levels of differentiation; however, the

kind of impairment will most likely be different. There

is evidence for greater pathology at the extremes of the

differentiation continuum than at the middle (Witkin et al.,

1962). The pathology takes on quite different forms at

the two extremes.

Low-differentiated persons (at the extreme) usually

have severe identity problems and exert little effort

to maintain their identity. They show deep-seated pro-

blems of dependency, inadequately developed controls

which result in chaotic functioning, and are passive with

strong feelings of helplessness (Witkin, 1965; Witkin et al.,

1962). They are prone to feelings of shame, hostility

directed inward, and diffuse anxiety, and their major

defenses are repression and denial (Witkin, personal

communication).

The types of pathology associated mainly with low-

differentiation are alcoholism (Karp et al., 1963; Bailey

et al., 1961; Witkin et al., 1959); ulcer problems, asthma,

obesity, and hysteria (all reported in Witkin, 1965;

Witkin et al., 1962).

The pathology associated with high-differentiated








persons usually involves delusions, expansive and euphoric

ideas of grandeur, outward direction of aggression, over

ideation, and struggle for identity and self-aggrandisement

(Witkin, 1965; Witkin et al., 1962). They are more like-

ly to have feelings of guilt and hostility directed outward

and they are apt to experience anxiety as having a specific

source and occurring under specific circumstances (Witkin,

personal communication). Most paranoid disorders occur

in high-differentiated persons (Witkin, 1965). Projection

is quite specialized and selective regarding persons and

situations. This selectivity necessitates an ability to

articulate. The major defenses of the differentiated

person are isolation and over-intellectualization. This

enables them to maintain the discreteness of feelings

and ideas, with the possibility of feeling components being

"split-off."

"The global-analytical dimension of cognitive

functioning appears not to relate to presence or absence

of pathology, adequacy of adjustment, or some of the con-

ventional psychiatric nosological categories. However,

the kinds of problems, symptoms, and maladaptations

found in children and adults with contrasting modes of

field approach appear to be very different" (Witkin et al.,

1962, p. 213).

Differentiated persons show a greater degree of

"activity level" (Witkin et al., 1962). However, this








does not refer to expenditure of motoric energy. "Activi-

ty level" refers to an activity attitude, i.e., a quality

of directness which involves a definition of goals and

the means of reaching the goals. It involves assertiveness

or striving. "The weight of the evidence from the studies

presented is that people with an analytic approach are

more likely to show an active attitude than people with

a global approach (Witkin et al., 1962, p. 189).

The differentiated individual utilizes structured

controls which lead to regulation of attention. This

regulation of attention is related to directness of activi-

ty. This "type" of person can better rely on internal

frames of reference as guides for action. However, the

opposite is true of the low-differentiated person whose

diffuse anxiety and weakly developed boundaries of inter-

action often preclude the formation of a delineated

directedness towards specific goals.

Therefore, the numerous studies on the dimen-

sion of differentiation have shown it to be a stable

phenomenon involving many psychological processes.

However, the evidence relating differentiation to verbal

behavior is sparce and inconclusive. The few studies

reported involve TAT protocols and WAIS and WISC responses.

Witkin (1962) reports that these studies and the results

provide tentative generalizations without any direct

findings. Using the unstructured interviews taken from








a group of 10- year-old boys, Witkin (1962) found that

the subjects whose interviews showed general immaturity,

poorly developed self-concept, diminished sense of

awareness, under-developed expressive powers, passivity,

relative lack of orientation to past events, and little

comprehension of relationships between past events, had

perceptual index scores which correlated significantly

with their interview ratings. Those Ss scored at the

non-differentiated levels on the perceptual tasks.

The opposite relationship was found with those Ss whose

interviews gave evidence of developed self-esteem, positive

abilities for interpersonal relationships, active inter-

ests, vitality, greater insight, and clear general

orientation.



Verbal Behavior

The importance of studying verbal behavior becomes

extremely clear when it is recognized that it represents

the primary data observed and collected by clinical

psychologists. It is considered a major part of the raw

material from which personalities are constructed. It

has been shown that "patterns of language variables

themselves may be used to reflect and indicate quanti-

tative aspects of intrapsychic processes" (Gottschalk and

Hambidge, 1955, P. 338).

One of the commonest forms of behavior that








typifies man is verbal behavior. While to a great extent

man's patterns of speech are determined by the grammar and

rhetoric of the language he speaks, it is reasonable to

assume that within this framework the individual's choice

of words, themes, and style of speech may reveal something

of his personality dynamics and his current emotional style"

(Gottschalk et al., 1957, P. 300).

"Language provides a natural and spontaneous

form through which man represents his perception of exter-

nal reality (Lorenz, 1959).

Sanford (1942) states that "Language tradition-

ally has been regarded as the vehicle of thought..." (p. 611).

He points out that as early as 1916, it ,was suggested that

verbal behavior is indicative of psychopathology. The

patient's relationship with his environment is reflected

in grammatical mood, voice, etc. He further states that

the study of "speech as abnormal behavior demonstrates that

abnormality of adjustment is revealed in linguistic

behavior and that more or less specific syndromes are

accompanied by more or less characteristic speech usages"

(p. 386). If speech is looked on as behavior and if

speech can be objectively and quantitatively treated, then

observation can be more precise and inference more in-

cisive" (p. 836).

McClelland and Atkinson (1953), state that "the

frequency with which a certain person employs a certain








grammatical category such as nouns, verbs, adjectives,

and the like, has been shown to be related to various

personality characteristics" (p. 248-249).

Wendel Johnson (1944) views the study of language

as being primarily useful in ascertaining individual

differences. He suggests the following type of words as

being effective as means of studying language behavior:

A. Self-reference words.

B. Quantifying terms precise numerical terms.

C. Pseudo-quantifying terms words less indicative of

amount, size, etc.

D. "Allness" terms superlative or extreme words, such

as "never," "always," "no one," etc.

E. Qualification terms words that serve to qualify or

limit statements, such as "except," "but," "however,"

"if," etc.

F. Terms indicative of consciousness of abstracting -

such words as "apparently," "seems," "appears," "as if."

Johnson also suggested the following measurements:

A. Type-frequency changes that characterize language

development.

B. Type-frequency characteristics of special groups that

may differentiate one group from another.

C. Type-frequency characteristics that correlate with

other variables, e.g., intelligence, emotional

stability, etc.







Jaffe (1961) and Johnson (1944) advocate the

use of a Type Token Ratio (TTR) as a means of analyzing

verbal data. The TTR is a measure of "verbal diversi-

fication" and is represented by the formula TYPE where the

"type" is the total number of each different word used,

and the "token" represents the total number of words

spoken. Jaffe states that "Previous experience with TTR

analysis indicates that low TTR (repetitive verbal trans-

actions) is characteristic of affect, misunderstanding,

confusion, and anxiety, in short, the various agitations

which interrupt the flow of referential verbal communi-

cation" (p. 83). Fairbanks (1944) and Mann (1944) found

lower TTR's in the speech and writing of schizophrenic

patients than in normal controls.

Boder (1940) reports an earlier study by Busemann

who used a variation of the Verb Adjective ratio (V/A).

Busemann viewed the V/A as a reflection of emotional

stability with higher ratios indicating greater instability.

Boder also reports that Rorschach found high V/A to be

related to movement responses, which were produced by

introverted people, while low V/A was related to color

responses, produced by extratensive individuals. Balken

and Masserman (1940) theorized that "high [V/A] values

connote restless, forceful, dramatic action in the phan-

tesies, expressing libidinal tensions and anxiety in the

subject" (p. 79).








Speech disturbances, which are usually not under

control exercised in speaking and are not readily subject

to social or linguistic control,, can serve as useful

indicators of anxiety.

Mahl's (1966) studies showed that speech distur-

bances ratios (Speech disturbances/Total words in sample)

were sensitive discriminators of anxiety which produce

statistically significant individual differences between

interviews for the same individual, and statistically

significant variations from moment to moment within in-

dividual interviews.

The following speech disturbance categories were

used by Mahl (1966) as measures of anxiety:

1. Sentence Change. A correction in form or content of

the expression while the word to word progression

occurs, e.g., Well, she's...'already she's lonesome.

2. Repetition. The serial superfluous representation

of one or more'of the words--usually of one or two

words, e.g.', Cause... cause they get along.

3. Stuttering.

4. Omission. Parts of words or entire words may be

omitted, e.g., She mour... was in mourning.

5. Sentence Incompletion. An expression is interrupted,

clearly left incomplete and the communication

proceeds without correction.

6. Tongue Slips. includes neologisms, the trans-

position of entire words from their "correct serial








position in the sentence, and the substitution

of an 'unintended' for an intended word", e.g.,

We speat the bitches (for "split the beeches").

7. Intruding incoherent sound. A sound which is ab-

solutely incoherent to the listener, e.g., If I

see her now, I just dh... ask her.

Short samples of speech (5-15 minutes) have been

used effectively as a means of studying verbal behavior

(Weintraub and Aronson, 1964; Gottschalk et al., 1961;

1957; 1955; Lorenz and Cobb, 1952). TAT cards have also
been used (Witkin et al., 1961; Gottschalk et al., 1957,

1955). However, Gottschalk and Hambidge (1955) found
significant definable differences in the pattern of speech

elicited by the use of TAT cards and the short samples of

speech. Other investigators have used psychotherapeutic

interviews to obtain their speech samples (Lennard, 1961;

Lennard and Bernstein, 1960; Gottschalk et al., 1961;

Saslow and Matarazzo, 1955).

Gottschalk and Hambidge (1955) studied content

variables, i.e., "the meaning, relationships, objects,

concepts, and processes which the verbal elements in

speech are intended to symbolize" (1955, p. 389). Their

units of analysis consisted of self-references, non-self-

references (other humans, animals, flora, and fauna),

verbs (words expressing action, motivation, perception,

and thought, and which represent processes occurring between








objects and/or concepts), and references to measure,

i.e., time, space, and quantity. "The number and kind

of verbal symbols which express a relationship of measure,

size, quantity, time, space may indicate the degree of

the speaker's exploratory alertness and attention to

orientation. Such orienting activities occur in relation

to sequences of intrapersonal and interpersonal events.

Therefore, they may be of significance in elucidating

certain qualitative and possible quantitative aspects of

psychodynamic processes" (p. 390). In their later study

(1957) Gottschalk and his associates added other units of

analysis, e.g., total qualifying words (which included

adjectives, prepositions, articles, adverbs, conjunctions,

interjections, and negations).

Weintraub and Aronson (1964), using a ten-minute

sample of speech on anything the subject wished, used 12

scoring categories which included:

A. Quantity of speech number of spoken words.

B. Shift to past tense.

C. "Negators" e.g., "no," "not," "nothing."

D. Qualifiers words or phrases indicating uncer-

tainty, e.g., "suppose," "guess," etc., use of

modifiers which detract from the forcefulness of the

statement, e.g., "more or less," etc., and phrases

or words which introduce an element of vagueness,

e.g., "what one might call" or "something like."








E. Explaining or "justifying" use of phrases or words

which indicate (1) relationships, e.g., "because," "due

to," "as if," "as a," (2) a reason for an action,

thought, or attitude, e.g., "My purpose is to...," and

(3) a participal phrase containing a justification

for an action, or thought, e.g., "Having attended

many times, I can speak with authority about...."

F. Evaluators, e.g., goodness and badness, usefulness

and uselessness, right and wrong.

The investigation of perceptual-cognitive styles

has contributed extensively to the field of clinical

psychology. Witkin and his associates have done the

most extensive research in this area in respect to the

concept of differentiation. This cognitive-perceptual

dimension has been related to intelligence, sex differ-

ences, self-awareness, problem solving, personality, and

psychopathology. Yet, one of the most crucial tools of

the clinical psychologist--verbal behavior--has been neg-

lected in respect to its relationship to perceptual-cogni-

tive style.

Witkin's concept of differentiation can be summa-

rized by the following basic personality concepts which

have been found to be associated with high levels of

differentiations

A. A general activity in dealing with the environment;

B. A general tendency to see the world as structured and








analyzed;

C. A sense of separate identity;

D. Effective control of impulses with low anxiety;

The purpose of this study is to test hypothesized

verbal correlates for these four basic personality com-

ponents, in an attempt to demonstrate the relationship

between differentiation and the structure of verbal behavior.


Units of Analysis

The following verbal measures which would seem

to be related to the concept of differentiation were se-

lected for investigation in this study:

1. Adjectives, Adverbs, and Prepositions--usually

express an expansion or addition of character-

ization. They serve as particularization fac-

tors (Lorenz, 1959). They also represent a

means of controlling self and the environment.

2. Verbs--usually express motivation, action,

perception, and thought. They represent pro-

cesses, occurring between objects and/or concepts

(Gottschalk and Hambidge, 1955).

3. Verb Adjective Ratio (V/A)--used to indicate

level of anxiety and amount of fantasy life.

4. Type Token Ratio (TTR)--used as a measure of

verbal diversification.

5. Allness Terms--(e.g., everyone, no one, always,








never, etc.) suggests diffuseness, i.e.,

difficulty in being discrete.

6. Measure Terms--(space, time, quantity, size)

indicate the individual's degree of exploratory

alertness and attention to orientation (Gott-

schalk and Hambidge, 1955).

7. Attenuating Terms--(e.g., "more or less," "a

little bit," "something like," etc.)--words or

phrases which tend to weaken the meaning of the

words they modify, and therefore would indicate

a lesser degree of discreteness.

8. Speech Disturbances--(e.g., stutter, omissions,

tongue slips, etc.) indicate degree of anxiety

in speech,

9. Terms Indicating Causal Relationships--(e.g.,

"because," "due to," "resulting from," etc.)

indicating comprehension of the relationship

between events.


Hypotheses

The findings related to the concept of differ-

eAtiation are used to develop the following hypotheses

which were tested in the context of the four basic person-

ality components.

A. General activity in dealing with the environment.

1. The high-differentiated person will use a








significantly greater number of verbs in

his speech than the low-differentiated per-

son. Rationales The high-differentiated

person has a more "active attitude," is

more goal directed, has a greater aware-

ness of his goals, and is more capable of

relating events in his environment.

B. Tendency for the world to be experienced as

structured and analyzed.

2. The high-differentiated person will use

significantly more total qualifying words

(adjectives, adverbs, and prepositions) in

his speech than will the low-differentiated

person. Rationales High-differentiated

persons are more precise and perceive their

environment in a more discrete manner.

3. The high-differentiated person will use

significantly fewer "allness" terms than

the low-differentiated person. Rationales

The high-differentiated person perceives

his environment in a discrete manner, is

more specific, and less diffuse than the

non-differentiated person.

4. The high-differentiated person will use

significantly fewer attenuating terms than

will the low-differentiated person. Rationales








High-differentiated individuals are more

specific and expressive in their speech,

and attenuating terms tend to weaken the

discreteness of speech.

5. The high-differentiated person will use

significantly more terms indicating causal

relationships. Rationale: High-differen-

tiated persons have greater comprehension

of relationships between events.

C. A sense of separate identity.

6. The high-differentiated person will use

a significantly greater number of measure

terms (time, space, quantity, and size)

than the low-differentiated person.

Rationale: The high-differentiated

person more discretely articulates his

environment and has a greater sense of

separate identity than the low-differ-

entiated person.

7. The high-differentiated person will make

fewer references to self. Rationales

The high-differentiated person has a more

adaptive self-concept and therefore, less

of a need for self-references.

D. Effective control of impulses with low anxiety.

8. The high-differentiated person will have a








lower Verb Adjective Ratio than the low-

differentiated person. Rationales The

high-differentiated person is less ex-

troverted, and has greater control over

his impulses and anxiety than the low-

differentiated person.

9. The high-differentiated person will have

a significantly higher Type Token Ratio

than the low-differentiated person. Rationales

The high-differentiated person has greater

control of his anxiety and is more verbally

expressive.

10. The high-differentiated person will

have significantly fewer speech distur-

bances than the low-differentiated person.

Rationale: The high-differentiated

person has greater control of his impulses

and less anxiety.

A structured interview (Appendix A) was developed

in order to elicit verbal behavior in the context of

structure. As the interview represents a structured

situation for the individual, it is further hypothesized

that it will represent less of a threat to the low-differ-

entiated person than elicited spontaneous speech. Witkin's

findings have shown that the low-differentiated person

has a greater anxiety of a non-specified source, is more








dependent, and requires greater environmental structure

and support than the high-differentiated person (Witkin,

personal communication). Therefore, on the basis of

these findings, the following hypotheses will be tested:

11. The low-differentiated person will have a

significantly lower Verb Adjective Ratio

in the structured interview than in spon-

taneous speech.

12. The low-differentiated person will have a

significantly higher Type Token Ratio

in the structured interview than in spon-

taneous speech.

13. The low-differentiated person will have

significantly fewer speech disturbances

in the structured interview than in spon-.

taneous speech,

14. The high-differentiated person will show

no significant differences in the Verb

Adjective Ratio, Type Token Ratio, or

Speech Disturbances when the structured

interview results are compared with spon-

taneous speech results.













CHAPTER II


METHOD



Subjects

A total of 56 male S's who were enrolled in an

undergraduate, introductory psychology course at the

University of Florida were administered the EFT (Appendix B).

Using Witkins's norms (Witkin, 1950), 15 S's who scored in the

lowest quartile and 15 S's who scored in the highest

quartile were selected and used in this study.



Apparatus

The EFT short form, devised by Witkin (1962), con-

sisting of 12 complex figures and eight simple figures, was

administered to 56 S's in order to determine his degree of

differentiation (field-dependence--field-independence).

Each S selected for further study was given the

structured interview and then asked to give five min-

utes of uninterrupted, spontaneous speech. The interview

and the spontaneous speech were both recorded electronically

on tape.' Typewritten transcripts were made from the tapes to

facilitate the data analysis. Two judges (the experi-

menter and his wife who has a B.A. in psychology) rated








independently each transcript (presented in random order) for

the purpose of determining the reliability of tabulating the

verbal measures.



Procedure

Each S was met by an experimenter and administered

the EFT. At the beginning of the test, the following in-

structions, which were taken from Witkin (1950), were given

to each S:

I am going to show you a series of colored
designs. Each time I show you one of these
designs, I want you to describe the over-all
pattern that you see in it. After you examine
each design I will show you a simpler figure,
which is contained in that larger design. You
will then be given the larger design again,
and your job will be to locate the smaller
figure in it. Let us go through one to show
you how it is done.


The S was then shown the practice complex figure

(P-l) for 15 seconds, after which it was removed and the

practice simple figure (P) was shown for 10 seconds.

When it was removed, the complex figure was presented once

more, with instructions to locate the simple figure in it.

As this was done, a stop watch was started. When S reported

that he had found the figure, the time was recorded and he

was required to trace the figure, so that the E could.be

sure it was the correct one.

After the practice trial, the S was given the

following additional instructions:








This is how we will proceed on all
trials. I would like to add that in
every case the smaller figure will be
present in the larger design. It will
always be in the upright position.
There may be several of the smaller
figures in the same large design, but
you are to look only for the one in
the upright position. Work as quickly
as you possible can, since I will be
timing you; but be sure that the figure
you find is exactly the same as the
original figure both in size and in
proportions. As soon as you have
found the figure, tell me at once.
If you ever forget what the simple
figure looks like, you may ask to
see it again. Are there any questions,?


The same presentation procedure was used on 12

test trials. The S's score on each trial was the time tak-

en to find the simple figure in the complex one after

the latter was presented for the second time. A maximum

of five minutes was allowed for each trial. If the subject

failed to locate the figure in that time, his score was

recorded as 5'. While he searched for the simple figure,

he was permitted to re-examine the copy of it as often as

he wished. This was deemed necessary because the task would

cease to be the one intended if the S no longer remembered

the structure of the figure for which he was searching.

The complex figure was of course removed if he looked back

at the simple one,.so that both figures were never seen

simultaneously, and the S was discouraged from taking

more than 10 seconds for each re-examination of the simpler

figure. The stop watch was stopped during the period of







re-examination, so that this time was not included in the

final score, When the S reported discovery of the simple

figure within the complex one, the time was noted, but the

stop watch was permitted to run while he traced the figure.

If the tracing was done correctly, the score recorded for

the trial was the time of discovery, but if the correct

figure was not traced, the S continued his search and the

time consumed in tracing the incorrect figure was included

in the final score. Witkin's method (1962) for scoring

the short form was applied. For this method, the S's

score for the whole test was the sum of the times taken to

locate the simple figures in all 12 complex figures.

This sum was multiplied by two in order to use the norms

of the long form, which consist of 24 complex figures.

In order to control for experimenter bias, a different

Examiner, who was not aware of the EFT scores, obtained the

speech measures. Each of the 30 S's selected for their

extreme scores on the EFT was administered the speech

tasks individually. Each S was seen in the same room,

with the same meeting arrangements, and the same placement

of the tape recorder in full view. He was informed of the

confidentiality of his verbal productions. He was first

given the structured interview (Appendix A). Following

this, each S was asked to speak for 5 minutes. The in-

structions were as follows:

This is a study of speaking and conver-








national habits. Upon a signal from me,
I would like you to start telling me a-
bout any interesting or dramatic life ex-
periences you have had. Once you have
started, I shall be here listening to you
but would prefer not to reply to any
questions you may feel like asking me
until the five minutes are up. Do you
have any questions you would like to
ask before we start? Well then, you may
start (Gottschalk et al., 1957, P. 301).

After completion-of the spontaneous speech task, the

study was explained to the subject and he was thanked for

his participation and excused.



Method of Analysis

To insure that the high- and low-differentiated

subject groups did not differ significantly in verbal

skills, they were compared by the t test using the verbal

scores of The School and College Ability Test, which were

available from university records.

The verbal measures used in this study were first

transcribed from the tapes into typewritten form. Then

each verbal unit was scored by a judge and single page

selections from each of the 30 S's were scored indepen-

dently by an additional judge for a reliability check. The

two judges achieved 98% agreement.

For each hypothesis, the appropriate verbal units

were then totalled for each subject and transformed into

ratios. The ratios consisted of the total number of words

representative of a given measure (excepting TTR and V/A)








divided by the total number of words used by the subject.

The ratio method was employed to accommodate for individual

differences in total verbal production.

Ten one-way analyses of variance were computed in

order to test for the differences between the two groups

(high- and low-differentiated) on each of the first 10

hypotheses.

For the remaining four hypotheses, three 2 x 2

analyses of variance were done. The two dimensions in

these analyses were the two population groups (high- and

low-differentiated) and the two verbal tasks (structured

and unstructured). Analysis of the role of these variables,

as well as their interactions, were tested by the F statistic.

Dependent variables in these analyses were the V/A, the TTR,

and the speech disturbances.

In order to determine whether a combination of the

verbal measures could result in significant discrimination

between the two groups, three discriminant function analyses

were done. This statistical technique was developed by

R.A. Fisher in 1936 in order to best differentiate criterion

groups where multiple measurements were being used.1

Occasionally, decisions can be made on the basis of a single

variable, but more frequently two groups differ across


1
The following readings are suggested for a more
thorough discussion of discriminant function analysis
(Goulden, C.H., 1952; Fisher, R.A., 1936; Mather, K., 1946).








several variables, each of which contributes to the classi-

fication of an individual. The technique of discriminant

function analysis is essentially a multiple regression

method which is expressed by the function:

Z = XlX 1 2x2 3x3+.. +XInXn

in which z is the Composite predictor score derived from

the individual raw scores (x) on each of the variables

and the corresponding weightings (N). It is necessary to

determine the optimal weights for each of the variables,

i.e., to find values for the lambda coefficients which

will maximize the difference between the composite means

of the criterion groups. Discriminant function should

differentiate groups more efficiently than any other multi-

variate technique.

The three discriminant function analyses used in

this study were as follows: 1) Analysis was made of the

10 verbal units used in this study to determine how effi-

ciently these variables would classify the subjects into

the two appropriate criterion groups; 2) the TTR, V/A and

speech disturbances were analyzed for the structured speech

task. It was expected that this particular analysis would

serve as a control test in that accurate classification

of the criterion groups should not occur, as the structured

task should not produce significantly different anxiety

expressions in the two groups; and 3) TTR, V/A and speech

disturbances were analyzed for the unstructured speech





32


task. With this analysis, accurate classification of the two

criterion groups was expected, as the unstructured context

should increase the anxiety level of the low-differentiated

subjects. This increase should be reflected by a signi-

ficant difference between the two criterion groups in the

use of the three verbal expressions of anxiety.













CHAPTER III


RESULTS



The comparison which was made of the SCAT verbal

scores of the two criterion groups yielded a t of .95,

.50 >p<.20. This finding suggests that the two groups do

not differ significantly in verbal skills and therefore,

differences in their patterns of verbal behavior may be

attributed to factors other than verbal skills.

Table 1 presents the results of the one-way analyses

of variance for the first ten hypotheses. In all of the

ten one-way analyses, the results were non-significant and

the hypotheses regarding differences between the two criter-

ion groups on the ten verbal measures were not confirmed.

The results of the three 2 x 2 analyses of variance

for the three verbal expressions of anxiety are presented

in Appendix C. These three analyses tested hypotheses 11,

12, 13, and 14. The F ratios, in all instances, were

non-significant and the three hypotheses, predicting an in-

crease in anxiety reflected in the three verbal measures for

the low-differentiated group in the structured situation,

were not supported. Hypothesis 14, which predicted no differ-

ence between the structured and unstructured tasks in the










Table 1


Analyses of Variance of the 10 Verbal Measures
for High- and Low-differentiated Subjects


VERBS

Source of Sum of
Variation Squares D/F MS F P


Between groups 374.53 1 374.53 <1 n.s.

Within groups 14732.27 28 526.15

Total 15106.81 29



QUALIFYING TERMS

Source of Sum of
Variation Squares D/F MS F P


Between groups 307.20 1 307.20 <1 n.s.

Within groups 18368.81 28 656.03

Total 18676.01 29



ALLNESS TERMS

Source of Sum of
Variation Squares D/F MS F P


Between groups

Within groups


12.03

759.33


1 12.03


<1 n.s.


27.12


771.37 29


Total










Table 1 Continued


ATTENUATING TERMS

Sum of
Squares D/F


Source of
Variation


Between groups

Within groups

Total





Source of
Variation


Between groups

Within groups

Total


258.13

4890.53

5148.67


1

28

.29


258.13

174.66


CAUSAL RELATIONSHIP TERMS

Sum of
Squares D/F MS


32.03

258.13

290.17


32.03

9.22


MEASURING TERMS


Source of
Variation


Between groups

Within groups

Total


1.48










F


3.47


n.s.










P


n.s.


Sum of
Squares


D/F


213.33

3146.13

3359.47


213.33

112.36


1.90


n.s.


_ _


__ __











Table 1 Continued


Source of
Variation


Between groups

Within groups

Total


SELF-REFERENCES

Sum of
Squares D/F


.30

14199.87

14200.17


.30

507.14


V/A


Source of
Variation


Sum of
Squares
(x 10-3)


MS
D/F (x 10-3)


Between groups

Within groups

Total





Source of
Variation


Between groups

Within groups

Total


253.92

2522.27

2776.19


2539.20

900.81


2.82 n.s.


n.s.


51.50

190.93


n.s.


51.50

5346.06

5397.57




Sum of
Squares


TTR


D/F











Table 1 Continued



SPEECH DISTURBANCES

Source of Sum of
Variation Squares D/F MS F P


Between groups 282.13 1 282.13 3.32 n.s.

Within groups 2379.07 28 84.97

Total 2661.20 29


Note.--n.s. = non-significant.



occurrence of verbal expressions of anxiety for the high-

differentiated group, was supported.



The Discriminant Function Analyses2

Discriminant Function Analysis I tested the effect

of using the combination of the 10 verbal measures in

maximizing the accuracy of classifying appropriately each

subject into the two criterion groups. The mean scores

and differences for each verbal measure between the two

groups are presented in Table 2. The differences, with the

exceptions of V/A, TTR, and Allness Terms are in the pre-

dicted directions.

2
The data were analyzed at the University of
Florida Computer Center. The program used was BMD04--
Discriminant Analysis--Two Groups--Version of May 26,
1964. Health Sciences Computing Facility, UCLA.











Table 2


Mean Ratio Scores and the Differences Between the
Means for the Criterion Groups on the 10 Verbal
Measures (Discriminant Function Analysis I)




Low- High-
differentiated diffentiated
Variables (N = 15) (N = 5) Difference
(x 103) (x 103) (x 103)


1. Verbs

2. Qualifying
Terms

3. Allness
Terms

4. Attenuating
Terms

5. Causal
Relationships

6. Measuring
Terms

7. Self-
References


8. V/A

9. TTR


198.27


393.80


11.60


26.60


4.80


22.87


72.93

1838.27

347.27


205.33


400.20


12.87


20.73


6.87


28.20


72.73

1921.13

328.87


- 7.07


- 6.40


- 1.27


5.87


- 2.07


- 5.33

0.20

-82.87

18.40


10. Speech
Disturbances


28.67 22.53


6.13








The following lambda values (X) for each variable

were obtained: Verbs (X1 = 0.00198), Qualifying Terms

(X2 = -0.003196), Allness Terms (X3 = -0.00576), Attenuating

Terms (% = 0.00319), Causal Relationships (X5 = -0.02326),

Measuring Terms (%6 = -0.00583), Self-References (X =

-0.00121), V/A (x8 = -0.0008), TTR (?9 = 0.00023), and

Speech Disturbances (X10 = 0.00404). The lambda values

suggest how much each variable has contributed to the

classification.

The derived mean composite discriminant score for

each criterion group are as follows: Za = -1.42219 and

Zb = -1.58898, in which Za is the score based on the mean

scores of each variable for the low-differentiated subjects

(Xal, ... a10), and the corresponding weights assigned to

each of the respective variables (X1, ...8)* Zb serves

as the respective mean composite score for the high-differ-

entiated group.

The following strategy was used to determine the

optimal cut-off score for the purpose of classifying any

subject on the basis of his composite score:

If Zi > Za Zb,
2

then classify low-differentiated.

If Zi Za + Zb,
2

then classify high-differentiated.

The ratio, (Za + Zb)/2, yielded the composite








value of Z = -1.50559 as the optimal cut-off score.

Classifications were then made on each subject and

the results are presented in Table 3. These results demon-

strate that only two low-differentiated subjects and three

high-differentiated subjects were misclassified. The

discriminant function analysis accurately classified 83%

of the tested sample of 30 subjects.

The results of Discriminant Function Analysis I

showed significant differentiation between the two criterion

groups (F = 2.38 with d/f = 10, 19, P < .05).

Discriminant Function Analysis II tested the effect

of combining three verbal measures of anxiety (V/A, TTR,

and Speech Disturbances) in the structured speech task upon

accurate classification of the two criterion groups. It

was expected that since structure should not produce differ-

ential anxiety between the high- and low-differentiated

groups that accurate classification would not occur. The

results supported this prediction (see Appendix D for tables).

The obtained lambda values were as follows: V/A

(X1 = -0.00002), TTR (X2 = 0.0000), and Speech Disturbances

(3 = 0.00072). The derived mean composite discriminant

scores were Za = -0.01581 and Zb = -0.02542 with a cut-off

score (Za + Zb/2), Z = -0.02062. Using this cut-off score,

Discriminant Function Analysis II accurately classified only

57% of the total sample (six low-differentiated and seven
high-differentiated subjects were misclassified). This











Table 3


Classification of Subjects by Use of
Discriminant Function Ia


Low-differentiated High-differentiated
Subject Composite Subject Composite
Rank Scores (N = 15) Scores (N = 15)


-1.320
-1.321
-1.337
-1.377
-1.391
-1.394
-1.404

-1.419
-1.420
-1.450
-1.479

-1.485
-1.487


-1.410



-1.480


-1.488


-1.510

-1.540


-1.538

-1.544
-1.567
-1.596
-1.598
-1.624
-1.627
-1.639
-1.657
-1.661
-1.682
-1.723


z = 1.50558


a
Composite cut-off score: Z = 1.50558; Correct classi-
fication--83%, false classification--17% (Two low-defferen-
tiated subjects and three high-differentiated subjects).








analysis yielded an F ratio = 0.625.

In order to test whether combining the V/A, TTR,

and Speech Disturbances would accurately classify the sub-

jects into the two criterion groups in the unstructured

speech task, Discriminant Function Analysis III was done.

It was expected that the lack of structure would produce

a sufficient increase in the occurrence of verbal evidence

of anxiety for the low-differentiated subjects that the

combination of these measures would significantly differ-

entiate between the two groups. Although the results did

not significantly confirm the prediction, the accuracy of

classification was substantially higher than that of the

classification of the criterion groups in the structured

speech task (see Appendix D for tables). The obtained

lambdas were as follows: V/A ( = 0.00002), TTR (2 =

0.00080), and Speech Disturbances (C3 = 0.00316). The

respective mean composite discriminant scores were Za =

0.44057 and Zb = 0.40404 with a cut-off score of Z = 0.4223.

Discriminant Function Analysis III correctly classified

77% of the total sample (five low-differentiated subjects
and two high-differentiated subjects were misclassified).

The discriminant function analysis yielded an F = 2.37

with d/f = 3, 26. This ratio does not reach significance

at the .05 level.













CHAPTER IV


DISCUSSION


The results failed to confirm the first 10 hypotheses

which predicted significant differences between the two

criterion groups (high- and low-differentiated) on each of

the 10 verbal measures.

Although none of the individual verbal measures sig-

nificantly differentiated the two groups, combining these

same 10 variables into a multivariate analysis did signifi-

cantly differentiate the two criterion groups (Table 3).

Such findings suggest that, although individual variables

by themselves may not describe a population group, a com-

bination of variables may serve as a much more effective

means of predicting or classifying. The results of Discri-

minant Function Analysis I (Tables 3 and 4) lend support to

the general hypothesis that language, as an expressive

behavior, is reflective of personality and perceptual mode.

This seems especially true when various measures are combined.

McClelland and Atkinson's (1953) statement (p. 12 in text),

that "the frequency with which a certain person employs a

certain grammatical category such as nouns, verbs, adjectives,

and the like, has been shown to be related to various person-








ality characteristics," most accurately describes the

results of this study.

It is observed that the high-differentiated group did

use a higher proportion of verbs, qualifying terms, allness

terms, causal relationship terms, measuring terms, and a

higher V/A, whereas the low-differentiated group have a

higher proportion of attenuating terms, self-references,

speech disturbances, as well as a higher TTR.

Allness terms, V/A, and TTR were the three verbal

measures for which the results were not in the predicted

direction. Since the analyses of the differences between

the two criterion groups for the 10 verbal measures demon-

strate that they do not differ significantly from chance,

one might expect the results for two or three variables

to be in the direction opposite of that predicted.

Regarding allness terms, however, it may be that

their use represents a means of being more concise or

speaking with "certainty," characteristics which one might

expect to find more with high-differentiated subjects.

The TTR results were even more unexpected. Johnson (1944)

has found that the TTR represents a measure of verbal diver-

sification. Jaffe (1961) has additionally indicated that

"low TTR (repetitive verbal transactions) is characteristic

of affect, confusion and anxiety." Therefore, one should

expect a higher TTR among high-differentiated subjects

who are reported to be less anxious and more discrete in








their interactions with the environment (Witkin et al.,

1962; 1954). The findings of this study gave evidence of

a higher TTR for the low-differentiated group. This finding

may, in part, be explained on the basis of the limiting

aspects of the structured interview. The interview, al-

though devised to question areas which might reflect

differences in the two criterion groups, tended to restrict

occasionally the diversification of responses. For example,

both high- and low-differentiated subjects frequently respon-

ded to question #1, "How is your health?," with an answer

such as "My health is fine." or "My health is good." This

kind of question, it may be observed, elicits responses

in which both groups tend to utilize the same or similar

words. Use of interview questions which allowed for more

open-ended answers might have provided a greater oppor-

tunity for verbal differentiation. Under such circum-

stances, the high-differentiated subjects may have exhi-

bited the expected greater diversity in their verbal pro-

duction.

The higher V/A for high-differentiated subjects

may be partly explained by this group having a more active

attitude, which is manifested in speech by the use of a

greater number of verbs (Gottschalk and Hambidge, 1955). Bal-

ken and Masserman, also (1940) theorized that high V/A "values

connote restless, forceful, dramatic action in the fanta-

sies ..." The assumption in this study was, however,

that the V/A represents a measure of anxiety as suggested








by Boder (1940), and Balken and Masserman (1940).

Therefore the low-differentiated group was predicted to

show a higher V/A. The lack of support for this predic-

tion may be accounted for also by a fault in the design

of this study. For all 30 subjects, the structured inter-

view task preceded the spontaneous speech task. Such

a procedure may have masked differences in anxiety level

between the two groups as hypothesized. For example, the

kind and amount of structure provided by the interview

situation might not have permitted the full expression

of existing anxiety to be fully reflected in the speech

of the low-differentiated subjects. Thus, the V/A for

the low-differentiated group was not as high as it might

have been in a less structured situation.

In addition, by the time these subjects reached the

spontaneous speech task, which was calculated to give

existing anxiety full expression in the subjects' speech

patterns, the level of their anxiety may have been suffi-

ciently reduced such that measurable differences in an-

xiety level between the two groups no longer existed.

Such reduction in anxiety might have resulted from the

climate of reassurance provided in the structured task,

as well as from rapport established with the examiner.

The most effective means of controlling for this possible

source of experimental bias would be to alternate the or-

der of presentation of the two tasks. In this way, the









effect of the structured task on the unstructured one

could be evaluated.

The next three hypotheses, which predicted increased

anxiety for the low-differentiated subjects in the spon-

taneous speech task as measured by V/A, TTR, and Speech

Disturbances, were not supported. No significant differen-

ces occurred between the two groups on any of the verbal

measures of anxiety for either the structured or the

spontaneous speech task. The absence of significant differ-

ences may, again, be accounted for by the influences of the

structure in the interview situation and the anxiety

reducing impact of the structured situation on the unstructur-

ed. It is also possible that a spontaneous speech task

may not be a more anxiety arousing situation for one group

than for the other, although this would seem unlikely if

one views this task as an unstructured situation. The

results of Witkin's research (1962; 1954) provide consis-

tent evidence for the low-differentiated (field-dependent)

individual's need for structure.

The last hypothesis predicted no difference in the

occurrence of anxiety as measured by the V/A, TTR, and

Speech Disturbances for the high-differentiated group,

between the structured and spontaneous speech task. This

hypothesis was confirmed by the results, however it is of

little relevence in light of the non-significant findings

for the low-differentiated group.








To determine whether, as a composite, the three verbal

measures of anxiety can accurately classify the two criterion

groups on both the structured and spontaneous speech tasks,

Discriminant Function Analyses II and III were done. The

results of Discriminant Function Analysis II tested the

classifying effectiveness of the composite for the struc-

tured speech task and the results were not significant.

This was as predicted, since the presence of structure was

not expected to arouse anxiety for either group. Discrimi-

nant Function Analysis III, which tested the composite of

the same verbal measures of anxiety for the spontaneous

speech task, correctly classified 77% of the total sample,

again providing support for the increased effectiveness of

a composite of variables over single variables in describ-

ing a population. Despite this high rate of classification,

the F ratio was not significant. With this analysis too,

varying the order of presentation of the speech tasks

might have resulted in more significant differences for the

condition in which the unstructured task was presented

first.

The classes of words which followed each individual

speech disturbance were scored and the results of the

frequency count are presented in Tables 7 and 8 (Appen-

dix F). It is interesting to observe that the scores for

each individual category (e.g., articles, prepositions,

nouns, etc.) are markedly similar for both groups, for








each speech task. By inspection, a higher frequency of

speech disturbances preceding active verbs does seem to

exist for the low-differentiated group in the spontaneous

speech task. This suggests that greater anxiety may have

been aroused in the low-differentiated subjects than in

the high-differentiated when action or an action attitude

was to be expressed verbally in the unstructured speech

situation. This observation, although not verified

statistically, is congruent with Witkin's (1962) findings

that high-differentiated (field-independent) individuals

demonstrate evidence of a more active attitude in their

content of their verbalizations.

The highest frequency of speech disturbances pre-

ceded pronouns for both groups in both speech tasks.

However, this was also the only class of words in which

the number of speech disturbances for the combined speech

tasks was greater for the high-differentiated group than

for the low-differentiated group. This aspect of speech

disturbances, although not a formal part of this study,

appears to justify a more thorough investigation in future

related research.

In summary, the results of this study point to the

likelihood that relationships do exist between the cog-

nitive-perceptual style of differentiation and the struc-

ture of verbal behavior. The combining of all 10 verbal

measures in a discriminant function analysis did sig-








nificantly classify (or dichotomize) the two criterion

groups. )Such a finding also offers support for the hy-

pothesis that the structure of verbal behavior lends itself

as a potentially effective representation of personality,

perception, and cognition.

The importance of verbal behavior in typifying and

understanding man is paramount (Lorenz, 1959; Gottschalk

and Hambidge, 1955). Verbal behavior is also one of the

basic diagnostic and prognostic tools of the clinician.

Therefore, positive findings in this study would have

strengthened the usefulness of the concept of differen-

tiation in the understanding of the human organism. Addi-

tional, more extensive and intensive research of the re-

lationship between the concept of differentiation and

verbal structure needs to be undertaken. In order to

correct for the problems in the present design, further

experimental controls, as discussed above, should be in-

cluded. The use of a structured interview which allowed for

more open-ended answers may have resulted in significantly

higher TTR's for the high-differentiated subjects.

The two speech tasks should have been randomly presented

in order to control for the effect of one task on the other.

However, the minimal significant relationships found

in this study between differentiation, as measured by the

EFT, and the structure of verbal behavior, suggest the

possibility that differentiation, as a personality con-









struct, may not generalize to the extent of including

the structure of verbal behavior. A further question is

concerning the validity of the construct itself. It may

well be that differentiation, as described and studied

by Witkin and his associates, may represent no more than

an overt expression of more basic processes, such as

biochemical functions, anxiety levels, etc. As such,

the construct of differentiation would serve only as a

convenient and descriptive label which has little or no

relevance in the nature of the human organism.

An additional, possible explanation of the non-

significant findings in this study is that the popu-

lation group studied, although scoring in the extremes

(upper and lower 25 percentiles) on the EFT, may be very

similar in the structure of their verbal behavior.

They were all college students who scored high in the

verbal section of the SCAT. Such a population, whose

scholastic training emphasizes verbal skills and who

are also similar in their verbal skills, are most likely

very unrepresentative of the general population. More

extreme differences might be found among subjects more

representative of the general population.

Additional relevant factors to study might be as

follows:

1. Content variables, e.g., expressions of affect,

self-concept, somatic complaints, etc. Low-differ-








entiated subjects might be expected to use a great-

er frequency of affect expression and somatic com-

plaints in their speech, and should have more diffi-

culty describing themselves verbally.

2. Additional populations, e.g., nosological cate-

gories, vocations, etc. It would seem relevant to

have a better understanding of the verbal behavior,

both structure and content, of different nosolo-

gical categories and how such information relates

to their specific problems in respect to dynamics,

onset, deterioration, and/or growth. This may be

especially important since Witkin (1965) has al-

ready demonstrated a relationship between various

types of pathology and the cognitive-perceptual style

of differentiation. One might expect, for example,

asthmatic, alcoholic, obese, and ulcerative pa-

tients, whom Witkin (1965) found to be related

to non-differentiation, to use a smaller proportion

of verbs, qualifying terms, etc., than would

patients with.paranoid difficulties, whom Witkin

found to be more differentiated.

Since Witkin (1962) has found developmental

trends to be operating in the cognitive-perceptual

style of differentiation, it would be useful to

examine the existence of parallel developmental

trends in speech measures for various age groups








(children, geriatric, etc.).

3. Other verbal structure measures, e.g., the use of

negators, evaluators, shifts to past tense, etc.

Such additional speech measures may be equally or

more related to differentiation than some of those

measures used in this study. One might expect,

for example, low-differentiated individuals to

make more references to past tense, as such re-

ferences may provide a form of structure.

4. Further study of the occurrence of speech distur-

bances and the words (structure and content) which

they precede. Such studies would provide infor-

mation relating the verbal expression of anxiety

to other speech measures.



In conclusion, it should be strongly emphasized

that, although the use of the multivariate instrument of

discriminant function analysis appeared to be an effective

means of classifying the two criterion groups, the pre-

dictive validity of the instrument can not be demonstrated

without additional cross-validation studies. This same

caution is even more strongly stressed when one observes

that only seven of the 10 verbal measures which made up

the composite in Discriminant Function Analysis I were

in the predicted direction. Therefore, three of the

variables used in the significant differentiation of the








two groups may have contributed to the discrimination for

reasons other than those hypothesized. The lambda weights of

these variables suggest only Allness Terms may have signi-

ficantly contributed to the composite, whereas it is possi-

ble that the removal of V/A and TTR might have added to the

accuracy of classification of the two criterion groups.

Relationships of the concept of differentiation to

the structure of verbal behavior have been only suggested

by this study. Further research utilizing more stringent

controls is needed before these relationships may be more

fully understood.













CHAPTER V


SUMMARY



The cognitive-perceptual style of differentiation,

as developed by H. A. Witkin, has been found to be re-

lated to many dimensions of the human being, e.g., per-

ception, intelligence, personality, sex differences,

body concept, and pathology. The area of verbal behavior

and its relationship to differentiation has not been

investigated beyond anecdotal observations. The purpose

of this study was, therefore, to study the relationship

between the concept of differentiation and the structure

of verbal behavior.

An initial total of 56 male S's were administered

the Embedded Figures Test. Using Witkin's norms, 30 S's,

15 who scored in the highest quartile (designated low-

differentiated) and 15 who scored in the lowest quartile

(designated high-differentiated), were selected and used

in this study. Each S selected for further study was

engaged in a structured interview and then asked to give

five minutes of spontaneous speech.

Ten verbal measures which were seen as related to

the personality correlates of differentiation, as reported








by Witkin end his associates, were used to predict sig-

nificant differences between the two criterion groups.

It was predicted that for the total speech production,

the high-differentiated group would use a significantly

higher proportion of verbs, qualifying terms, causal

relationship terms, measuring terms, and that they would

produce a higher type Token Ratio and a lower Verb Ad-

jective Ratio than the low-differentiated group. For the

low-differentiated group it was predicted that they would

use a higher proportion of Allness Terms, attenuating

terms, self-references, and speech disturbances. For the

two speech tasks (structured and unstructured), it

was predicted that the low-differentiated group would

show a significant increase in the use of verbal measures

reflecting anxiety in the spontaneous speech task over the

use of these measures in the structured interview.

The results did not support the hypotheses. A

discriminant function analysis was used to determine the

effectiveness of the composite of the 10 verbal measures

in accurately classifying the two criterion groups. The

results demonstrated that the composite significantly

differentiated the two groups, accurately classifying

83% of the total sample.

Individually, the three verbal measures of anxiety

failed to differentiate the two groups for the two speech

tasks. The multivariate test of the composite of the three








variables for the structured task also yielded non-signi-

ficant results, accurately classifying only 57% of the total

sample. Using the same composite for the spontaneous

speech task, the results, although non-significant, did

accurately classify 77% of the total sample. This

finding suggests that the unstructured task may have

aroused greater anxiety in the low-differentiated subjects,

as was predicted.

The results of this study point to the likelihood

that relationships do exist between the cognitive-per-

ceptual style of differentiation and the structure of

verbal behavior. Further research utilizing more strin-

gent controls is needed before these relationships may

be more fully understood.

The use of the multivariate instrument of discrimi-

nant function analysis appears to be an effective means of

classifying the two groups; however the predictive validity

of this instrument cannot be demonstrated without addi-

tional cross-validation studies.
































APPENDICES








APPENDIX A


Structured Interview



The structured interview was devised to elicit

approximately four to five minutes of speech from each sub-

ject. The specific questions asked are those which were, in

some way, related to personality differences found between

high- and low-differentiated persons (Witkin, 1954; 1962).

These personality differences, corresponding to each number-

ed question in the structured interview, are as follows:

1. Somatic complaints.

2. Dependency factors.

3. Need for social acceptance.

4. Self-concept.

5. Articulated future goals.


The specific questions asked were as follows:

1. How is your health?

a. What illnesses are you prone to?

b. (when applicable) When does this occur?

2. What is your relationship with your family like?

Describe it.

3. What do you think of campus social life?

a. What are your social activities?

b. How about friends?

4. How would you describe yourself?

5. What do you plan to do after graduation?

59







APPENDIX B


Representative Cards from the
Embedded Figures Test


Simple Figure
(achromatic)


Complex Figure
(chromatic)


Simple Figure
(achromatic)


Complex Figure'
(chromatic)


Simple Figure
(achromatic)


Complex Figure
(chromatic)


Z-








APPENDIX C


Table 4


Two-Way Analyses of Variance for the Three
Verbal Expressions of Anxiety


V/A

Source of Sum of
Variation Squares D/F MS F P


Between subjects 29

High- and low-
differentiated(A) 0.11 1 0.11 <1 n.s.

Subjects with-
in groups 9.11 28 0.30

Within subjects 30

Structured and
unstructured(B) 0.65 1 0.65 <1 n.s.

A x B 0.95 1 0.95 2.52 n.s.

B x subjects
within groups 10.95 28 0.38



TTR

Source of Sum of
Variation Squares D/F MS F P


Between subjects

High- and low-
differentiated

Subjects with-
in groups


0.00


0.11


1 0.00


0.00


n.s.











Table 4 Continued


Source of
Variation


Within subjects


Structured and
unstructured(B)


AxB


B x subjects
within groups


0.00

0.00


0.04


1 0.00

1 0.00


0.00


Speech Disturbances


Source of
Variation


Between subjects


High- and low-
differentiated(A)

Subjects with-
in groups

Within subjects


0.00


0.00


1 0.00


3.84 n.s.


0.00


Structured and
unstructured(B)


AxB


B x subjects
within groups


0.00

0.00


0.00


1 0.00

1 0.00


<1 n.s.


n.s.


0.00


Note.--n.s. = non-significant.


Sum of
Squares


D/F


n.s.


Sum of
Squares


D/F








APPENDIX D


Results of Discriminant Function
Analyses II and III


Table 5

Classification of Subjects in the Structured Speech
Task by Use of Discriminant Function IIa


Low-differentiated High-differentiated
Subject Composite Subject Composite
Rank Scores (N = 15) Scores (N = 15)


1 0.00794
2 0.00652
3 0.00170
4 0.00058
5 -0.00011
6 -0.00181
7 -0.00507
8 -0.00953
9 -0.00987
10 -0.01273
11 -0.01445
12 -0.01515
13 -0.01680
14 -0.01811
15 -0.01941
16 -0.02021 Z = -.02062
17 -0.02224
18 -0.02602
19 -0.02612
20 -0.02652
21 -0.02667
22 -0.02847
23 -0.02924
24 -0.03175
25 -0.03340
26 -0.03582
27 -0.03774











Table 5 continued


Low-differentiated
Subject Composite
Rank Scores (N = 15)


High-differentiated
Subject Composite
Scores (N = 15)


28 -0.03879
29 -0.04095
30 -0.08822


aComposite cut-off score: Z = -.02062; Correct classi-
fication--57%, false classification--43% (six low-differ-
entiated subjects and seven high-differentiated subjects).




Table 6


Classification of Subjects in the Spontaneous Speech
Task by Use of Discriminant Function IIIa




Low-differentiated High-differentiated
Subject Composite Subject Composite
Rank Scores (N = 15) Scores (N = 15)


0.52244
0.50248
0.47835
0.47768


0.45019
0.44935
0.44835
0.44407
0.44138


0.46813
0.45205












Table 6 continued


Low-differentiated
Subject Composite
Rank Scores (N = 15)


High-differentiated
Subject Composite
Scores (N = 15)


0.43115


0.40854


0.40117
0.39846

0.39346




0.36152


0.41396
0.41057
0.40992
0.40917

0.40212
0.40163
0.40122


0.39365

0.39057
0.38573
0.37975
0.37406
0.36805


Z = .84450


aComposite cut-off score: Z = .84450; Correct classi-
fication--77%, false classification--23% (five low-differ-
entiated subjects and two high-differentiated subjects).








APPENDIX E


Representative Protocols for a Low-differentiated
and a High-differentiated Individual



Subject E--Low-differentiated


Structured Interview

(How is your health?) Excellent, good, very good. I have

no problems at all that I know of as far as that goes.



(What illnesses are you prone to?) Just a common cold.



(When does this occur?) Probably, oh, one or two times a

year during winter.



(What is your relationship with your family like?) We get

along all right, We get along great. We have no problems

that I may have had when I was younger. I don't live with

them now, so I just see them maybe several times a year

and any contact I have is usually by phone.



(What do you think of campus social life?) I think there

is plenty to do. There is certainly all year long. I've

been up here--this is my third summer in a row. I don't

find anything a problem. Sometimes I think there is just

too much to do, especially in the fall. There is something

going all the time that you can't take advantage of all of it.

66








(What are your social activities?) Depending on the time

of year--right now, I think it is mostly swimming or I like

to go out at night a lot. Night life--there isn't too much

to offer in Gainesville, especially to a dance or a bar or

something.



(What about friends?) Well, most of them I think right now

are connected around what my interest is over in Journalism.

I'm interested in television, so most of them are--well--

instructors that are directors over there, technicians or

people working in radio and television.



(How would you describe yourself?) I don't know, I don't

know--this--I don't know what you want really.



(What do you plan to do after graduation?) I graduate this

August and I would like to work right for a commercial

television station, like Field's Broadcasting and I've done

a little work with Channel 5 on a part time basis. I think

I eventually would like to get back into educational tele-

vision, but it seems the opportunity, or maybe the money,

right now, is in commercial television. I would like to

go out and work for them a year, a year and a half or so,

and then come back and work on the master's degree in

broadcasting and eventually into the educational end of it.

That's about it. It would be in television production

work. That's what I would like to get into eventually.








Spontaneous Speech

Oh, I think some of the interesting things probably

happened during--well, I spent two years in the army. Of

course it wasn't my choice. I was drafted, but I think the

interesting thing about it was I was sent overseas to

Germany. I spent about 18 months in Germany right outside

of Nuremberg. There is a little college town called Erlang

and I think those two years gave me an opportunity to see

Europe which I probably wouldn't have done if I had stayed

in Florida. I think the people were most interesting. The

Germans were the most friendly of--compared to Italy or

even England. I seem to have had more language problems

in England than anywhere else. They seem to think I was

doing something for the English language and they couldn't

understand it. But I managed to travel through Italy,

France, Germany, Austria, England, and found the German

people most easily to'get along with, most friendly and

found the Italians, even though I am an Italian, a little

bit after the money rather than the individual. They

spotted us as Americans right off and were interested in

how much we were going to spend rather than who we were.

I spent about a week in Berlin--contrast between East and

West was most striking. For instance, West Berlin was

built up to, oh, compared to any modern city we have--

New York, Los Angeles, Chicago, anywhere. East Berlin is

still run down, still using cars over twenty years old and








you could see vast areas that were bombed down that had

just been cleared away, grass planted, and nothing built

in this city of two and one-half million people--see whole

blocks that was just blank which was grass and churches

still standing. You could see large churches on corners,

but they were all blocked and boarded with concrete and

stone, brick with wire fences around them. To get into

East Berlin, you had to wear your uniform and you could

feel the people looking at you and knowing you were an

American and could walk in and out whenever you felt like

it or wanted to, felt people looking at you, but I think

the difference in the cities--West Berlin had, well, the

neon lights, the shops and the stores, the cars and the

people. I really enjoyed my stay in Berlin more than in

any other city. I went back several times just to wander

around. I would like to go back some day and see more of

it. I like to travel. I came back to the United States

and traveled up--I haven't been too far west--ILwent as

far west as Chicago, went up to Canada. In our travels

to Europe we drove most of the way, bought a Volkswagen

over there. It was about 18 years old, but it ran anyway

so we drove around Italy, Austria, France--Two of my bud-

dies and myself .... I can remember going through Germany.








Subject GG--High-differentiated


Structured Interview

(How is your health?) I thought it was pretty good.



(What illnesses are you prone to?) None.



(What is your relationship with your family like?) It is

above normal. I have a very good relationship with my

parents. Most kids up here treat their--talk about their

parents as if they were just individuals, just people, and

I don't think I do that. I think I have a better relation-

ship than most people, especially with my mother. I have

a better relationship with my mother than I do with my

father, but we do have a good relation.



(What do you think of campus social life?) Greek or inde-

pendent? Well, Greek is great and independent is very poor,

and I can say that because I've been both. I mean, I waited

a trimester and a half to pledge a fraternity so I know

what--more or less what's it like for an independent and I

know what it's like for a Greek, and it, the difference,

socially, is tremendous. There's no comparison.



(What are your social activities?) Well, dating and foot-

ball games, dances, parties, things like that, football

games, whatever you do on dates, I mean, you know, horse-

back riding, whatever there is to do.








(How about friends?) Well now, what about friends? I'm

not trying to be beligerant--what--oh, the friends that I

have are very good friends and I can usually count on them

any time I need them, and can count on them even when I don't

need them. I mean, you know, even if it is not a need, I

know they are there and they are pretty good to me, the ones

I have. I don't know. I don't think I have any trouble

making friends, but I don't have too many very good ones,

but the ones I have, I, you know, I'm glad I've got.



(How would you describe yourself?) Average. Well, I'm

5'9", weight, approximately 150 pounds, brownish or blonde--

brownish-blonde hair, brown eyes, with Polish and Italian

origin. I have a tendency to have a hot gut.and I'm pretty

active. I like to have a good time.



(What do you plan to do after graduation?) Well, if I can,

I'd like to go on for my masters. Actually, what I'd like

to do is travel. I would like to go to Europe and just, if

I could, I'd like to spend a year in Europe on a--just have

a little car and just travel all around and see the cities,

of course, you know, the big sights and everything, but get

out into the countryside so I could see the true--the true

Europe, the real Europe, and just try to communicate more,

and see more of the things I'm studying about now, especial-

ly in Humanities.








Spontaneous Speech

In my junior year in high school, a very good friend

of mine, well, three very good friends of mine and myself

decided to go camping on a weekend, and we had a jeep, and

we took the jeep as far in as we could go into this grove,

and it turned into--from the grove on was a pasture, and

the pasture was fenced off, and in the pasture were horses

that were there for a summer camp for the riding--for you

to ride in this summer camp. So we camped out there and

rounded up the horses that night, and it was very inter-

esting, and we finally got four horses, but we didn't have

any saddles or bridles, so we opened the shed there and

took out a bridle--there was only one there. But one of

the boys that was with us was the guy that was taking care

of the horses during the day, so he had the key, and it

wasn't really that--as illegal as it sounds. So we had one

brible, and then we improvised. We used what they call an

Indian bridle, I believe, just ropes, and then we rode bare-

back, and I had a horse called Sundown. I'll never forget

that. It was a white horse and it was very old--at least

he looked old--and I got on him and fell right off. So I

got on him again, and he threw me off this time. So George,

this good friend of mine, decided he was going to try to

calm him down a little because he was probably just, you

know, probably from being ousted out in the night like this.

So he got on him, and the horse tried to throw him, but









George stayed on pretty well, and then the horse took off

and went around this big clump of bushes and trees that we

couldn't-see. Of course, it was dark, and we couldn't see

anyway, and all of a sudden we heard this scream, and the

horse came trotting back by himself. So we went around

there to see what had happened, and George had been bucked

off and had fallen onto an old fence post, and in the fore-

arm are two bones, the ulner and the radius, and I can't

remember which, but the one that runs along with the small

finger had been popped out of joint at the elbow, and it

was sticking up out of the arm. It hadn't broken the skin,

but it was sticking up out of the arm. So we just forgot

the horses and took him to a car, and it was raining--it

started to rain. It was raining like heck, so we couldn't

see the road, and the nearest hospital at this time was 11

miles away. That was after we got out of the pasture.

That took us quite a while, and he was taking it pretty

good. We were more shook than he was, as a matter of fact.

And then, when we finally got to the road, it was raining

fits. It was raining so bad that we couldn't see, so, what

we were doing, we were following the white line on the road,

and that was the only way we could see where we were going,

was to follow this white line, and it seemed like an endura-

ble time, but--I mean, an unendurable time, but we finally

made it to the hospital, and all they did, they brought him

in, didn't give him anything for the pain at all. They








brought him to the X-ray room and called the guy--the

technician, and it took him about half an hour to get there,

and, of course, George was getting worse and worse. His

arm was swelling even more so. When they got him there,

they laid his arm on the table, took an X-ray of it, and

then--his hand was palm down--and then they took and turned

his palm so that the palm was up and, of course, George

was back to the ceiling, and they took another X-ray that

way. And then, when they developed the X-rays, they said

that it was just dislocated, and there was nothing that

they could do there as they didn't have a specialist, and

that we were going to have to take him to Lakeland, which

is 60 miles from there, and we didn't think the car would

make it. It was an old--it was an old junky car anyway.

We had gotten out of the jeep in the meantime. It was an

old junky car. So one of us had to call our parents, and

I was elected. So at 2:30 in the morning, I called my

father, got him out of bed and he came down to the hospital,

picked us up, and drove us to Lakeland, and we arrived in

Lakeland, and I had the X-rays, and they gave him--they

finally gave him something for his pain, but it didn't help

very much. And then, we got there, they wanted--the recep-

tionist and the people there wanted my father to sign a

paper saying it was alright to put him under for the pain,

so they could reset the bone and stuff. My father couldn't

take that responsibility anyway, and his parents, George's






75

parents were in Miami, and we didn't have any idea where, so

the only--the nurse told us to call first was his grandmother,

and he didn't want to call his grandmother because, he kept

saying, she'd have a heart attack if we did.








APPENDIX F


Frequencies of Word Categories Which are
Preceded by a Speech Disturbance



Table 7


Frequencies of Word Categories Which are Preceded
by a Speech Disturbance for a Low-differentiated
Subject in Structured and Unstructured
Speech Conditions





Part of
Speech Structured Unstructured Total


Articles

Prepositions

Pronouns

Conjunctions

Active Verbs

Passive Verbs

Adjectives

Adverbs

Nouns

Miscellaneous


Total


37

52

131

18

44

17

56

31

42


26


257


--


197


454













Table 8


Frequencies of Word Categories Which are Preceded
by a Speech Disturbance for a High-differentiated
Subject in Structured and Unstructured
Speech Conditions


Part of
Speech Structured Unstructured Total


Articles 8 25 33

Prepositions 26 26 52

Pronouns 75 69 144

Conjunctions 4 12 16

Active Verbs 11 13 24

Passive Verbs 5 12 17

Adjectives 13 25 38

Adverbs 19 21 40

Nouns 7 23 30

Miscellaneous 8 13 21


176 239


Total


415








APPENDIX G


Ratio Scores for All 30 Subjects



Table 9


Ratio Scores for the 10 Verbal Measures
for the Combined Speech Tasksa


LOW-DIFFERENTIATED

Quali- Attenu-
Sub- fying Allness eating Causal
jects Verbs Terms Terms Terms Terms


.204
.220
.196
.189
.162
.182
.220
.207
.223
.115
.221
.188
.234
.218
.206


.391
.393
.364
.383
.432
.421
.395
.405
.395
.390
.408
.433
.346
.366
.385


.009
.004
.021
.007
.020
.015
.015
.009
.009
.014
.018
.007
.011
.007
.008


.280
.038
.009
.038
.022
.074
.037
.014
.012
.014
.026
.029
.016
.013
.029


.011
.008
.003
.004
.006
.007
.002
.002
.004
.006
.000
.004
.003
.005
.007


aAll the ratios used in the tables in Appendix except-
ing V/A and TTR, were derived by dividing the total number
of words representative of each measure by the total number
of words used by the subject.















Table 9 extended


Self- Speech
Measure Refer- Distur-
Terms ences V/A TTR bances


.013
.015
.011
.022
.019
.028
.022
.047
.007
.032
.022
.030
.032
.023
.020


.083
.058
.075
.084
.048
.074
.068
.098
.049
.050
.072
.074
.078
.110
.073


1.611
2.008
2.361
1.779
1.396
1.342
2.082
1.652
2.148
1.074
2.156
1.181
2.500
2.111
2.173


.372
.361
.316
.349
.323
.374
.372
.353
.327
.316
.337
.388
.366
.339
.316


.028
.031
.033
.025
.016
.032
.032
.053
.026
.025
.022
.026
.047
.018
.016














Table 9 continued


HIGH-DIFFERENTIATED

Quali- Attenu-
Sub- fying Allness eating Causal
jects Verbs Terms Terms Terms Terms

V .189 .435 .009 .014 .004
0 .197 .395 .010 .015 .012
J .211 .393 .021 .020 .005
GG .228 .349 .007 .006 .012
BB .223 .367 .011 .034 .00-
DD .189 .416 .023 .020 .010
B .198 .420 .009 .018 .007
TT .200 .440 .012 .015 .003
M .211 .377 .020 .019 .005
X .200 .395 .007 .008 .004
F .215 .390 .013 .036 .002
R .195 .436 .009 .031 .009
D .189 .421 .017 .025 .008
P .225 .381 .010 .029 .008
AA .210 .388 .015 .021 .010















Table 9 extended


Self-
Refer-
ences


.099
.086
.057
050
.090
.065
.033
.042
.087
.085
.126
.044
.053
.105
.069


V/A


1.352
2.021
1.941
2.613
2.386
1.799
1.724
1.598
1.871
1.807
2.742
1.309
1.605
2.373
1 .676


TTR


Speech
Distur-
bances


Measure
Terms


.049
.011
.038
.016
.024
.026
.050
.033
.023
.019
.023
.026
.032
.033
.020


.347
.338
.331
.309
.318
.335
.292
.363
.324
.383
.308
.360
.379
.260
.286


.032
.017
.016
.029
.019
.014
.021
.021
.038
.032
.013
.021
.012
.030
.023










Table 10


Ratio Scores for the Three Verbal
Expressions of Anxiety for the
Structured Interview


LOW-DIFFERENTIATED

Speech
Subject V/A TTR Disturbances

Y 1.486 .507 .034
C 1.833 .467 .028
FF 1.977 .374 .035
N 2.567 .434 .019
T 1.860 .448 .008
K 1.344 .450 .038
E 2.324 .431 .035
S 1.878 .444 .064
H 2.250 .358 .027
G 1.879 .371 .028
Q 2.125 .466 .023
L 1.098 .518 .032
I 1.647 .490 .055
A 1.576 .427 .017
LL 2.271 .401 .018



HIGH-DIFFERENTIATED

Speech
Subject V/A TTR Disturbances


2.111
2.194
2.091
2.222
2.465
1.453
1.804
1.656
2.906


.436
.525
.377
.398
.431
.491
.420
.458
.392


.046
.022
.023
.035
.013
.020
.010
.033
.047












Table 10 continued


Subject


V/A


TTR


Speech
Disturbances


X 1.381 .476 .038
F 4.579 .438 .009
R 1.123 .486 .028
D 2.200 .506 .012
P 2.905 .312 .031
AA 1.480 .332 .024


Table 11



Ratio Scores for the Three Verbal
Expressions of Anxiety for the
Spontaneous Speech Task


LOW-DIFFERENTIATED

Speech
Subject V/A TTR Disturbances


1.676
2.529
2.667
1.415
1.092
1.340
1.934
1.527
2.000
1.659


.422
.402
.364
.402
.342
.418
.461
.374
.415
.382


.025
.033
.031
.029
.021
.028
.030
.047
.024
.022











Table 11 continued


Speech
Disturbances


.021
.023
.044
.018
.016


HIGH-DIFFERENTIATED


Speech
Disturbances


.026
.015
.014
.026
.024
.011
.026
.017
.034
.030
.016
.019
.013
.029
.019


Subject


V/A


TTR


2.172
1.227
3.225
2.779
2.114


.363
.418
.396
.349
.381


Subject


V/A


1.094
1.938
1.778
2.851
2.328
2.025
1.675
1.576
1.511
2.039
2.000
1.443
1.371
1.958
1.873


TTR


.382
.359
.388
.336
.363
.365
.337
.396
.393
.415
.378
.391
.428
.337
.352













BIBLIOGRAPHY


Asch, S. E. and Witkin, H. A. Studies in space orienta-
tion: Perception of the upright with displaced
visual fields. Journal of Experimental Psycho-
logy, 1948, 38, 325-337.

Bailey, W., Hustmyer, F., and Kristofferson, A. Alco-
holism. Brain damage and perceptual dependence.
Quarterly Journal of Studies in Alcoholism,
1961, 22, 387-393.

Balken, Eva, R. and Masserman, J. H. The language of
the phantasies of patients with conversion hy-
steria, anxiety state, and obsessive-compulsive
neuroses. Journal of Psychology, 1940, 10,
75-86.

Bennet, D. H. Perception of the upright in relation
to body image. The Journal of Mental Science,
1956, 102, 487-50-.

Boder, D. P. The adjective-verb quotient; a contri-
bution to the psychology of language. Psycho-
logical Record, 1940, 2, 309-344.

Couch, A., and Keniston, K. Yeasayers and Naysayers:
Agreeing response set as a personality variable.
Journal of Abnormal and Social Psychology, 1960,
62, 151-174.

Fairbanks, Helen. Studies in language behavior: II.
The quantitative differentiation of samples
of spoken language. Psychological Monographs,
1944, 56, 19-38.

Fisher, R. A. The use of multiple measurements in taxo-
nomic problems. Annals of Eugenics, 1936, ~,
87-104.

Gardner, R. W., Jackson, D. N., and Messick, S. J.
Personality organization in cognitive controls
and intellectual abilities. Psychological
Issues, 1960, (Whole No. 8), 4, 1-149.








Gardner, R. W., Holzman, P. S., Klein, G., Linton,
Harriet, and Spence, D. P. Cognitive control:
A study of individual consistencies in cognitive
behavior. Psychological Issues, 1959, (Whole
No. 4), 4, 1-186.

Goodenough, D. R. and Karp, S. A. Field dependence
and intellectual functioning. Journal of Ab-
normal and Social Psychology, 1961, 2, 2,71-246.

Gottschalk, L. A., Springer, K. J., and Gleser, G. C.
Experiments with a method of assessing the
variations in intensity of certain psychologi-
cal states occurring during two psychothera-
peutic interviews. In Gottschalk, L. A.
(Ed.) Comparative psycholinguistic analysis
of two psychotherapeutic interviews, New York:
International Universities Press, Inc., 1961,
Pp. 115-138.

Gottshalk, L. A., Gleser, G. C., and Hambidge, G.
Verbal behavior analysis: Some content and form
variables in speech related to personality ad-
justment. A.M.A. Archives of Neurology and
Psychiatry, 1957, 77, 300-311.

Gottshalk, L. A. and Hambidge, G. Verbal behavior
analysis: A systematic approach to the problem
of quantifying psychologic processes. Journal
of Protective Techniques, 1955, 19, 387-409.

Goulden, C. H. Methods of statistical analysis, (2nd
ed.) New York: John Wiley and Sons, Inc.,
1952.

Jackson, D. N., Messnick, S., and Myers, C. T. Evalu-
ation of group and individual forms of embedded-
figures measures of field-independence. Edu-
cational and Psychological Measurement, 1964,
24, 177-191.

Jackson, D. N. Intellectual ability and mode of per-
ception. Journal of Consulting Psychology,
1957, 21, 458.

Jackson, D. N. A short form of Witkin's Embedded-
Figures Test. Journal of Abnormal and Social
Psychology, 1956, j5, 235-255.








Jaffe, J. Dyadic analysis of two psychotherapeutic
interviews. In Gottschalk, L. (Ed.) Comparative
psycholinguistic analysis of two psychothera-
peutic interviews, New York: International
Universities Press, Inc., 1961, Pp. 73-90.

Johnson, W. Studies in language behavior. Psycho-
logical Monographs, 1944, (Whole No. 255)
56, 19-38.
Karp, S. A., Poster, Dorothy, and Goodman, A. Differ-
entiation in Alcoholic Women. Journal of Per-
sonality, 1963, 31, 386-393.

Lennard, H. L. Some aspects of the psycho-therapeutic
system. In Strupp, H. H. and Luborsky, L.
(Eds.) Research in psychotherapy. Proceedings
of a conference, Chapel Hill, North Carolina,
May 17-20, 1961, Pp. 218-236.

Lennard, H. L. and Bernstein, A. The anatomy of psycho-
therapy: systems of communication and expectation.
New York: Columbia University Press, 1960.

Linton, H. B. Dependence on external influence: Correlates
in perception, attitudes, and judgment. Journal
of Abnormal and Social Psychology, 1955, 51,
502-507.

Lorenz, Maria. Language as an index to perceptual
modes. Journal of Projective Techniques,
1959, 23, 441-52.
Lorenz, Maria and Cobb, S. Language behavior in manic
patients. A.M.A. Archive Neurology and Psychi-
atry, 1952, 67, 763-770.

Mahl, G. F. Measuring the patient's anxiety during
interviews from "expressive" aspects of speech.
In A. P. Goldstein and S. J. Dean (Eds.)
The investigation of psychotherapy, New York:
John Wiley and Sons, Inc., 1966, Pp. 133-139.

Mann, Mary, B. Studies in language behavior: III.
The quantitative differentiation of samples
of written language. Psychological Monographs,
1944, 56, 41-74.

Marlowe, D. Some psychological correlates of field
independence. Journal of Consulting Psychology,
1958, 22, 334.








Mather, K. Statistical analysis in ioloy (2nd ed.)
London: Bethuen and Co., Ltd., 1946.

McClelland, D. and Atkinson, J. W. Clark, R. A., and
Lowell, E. L. The achievement motive, New
York: Appleton-Century Crofts, 1953.

Messick, B. and Damarin, F. Cognitive styles and memory
for faces. Journal of Abnormal and Social
Psychology, 196, 19 62, 313-318.

Saslow, G., Matarazzo, J. D., and Guze, S. B. The
stability of interaction chronograph patterns
in psychiatric interviews. Journal of Con-
sulting Psychology, 1955, 19, 417-430.

Sanford, F. H. Speech and personality. Psychological
Bulletin, 1942, 39, 811-841.

Taft, R. and Coventry. J. Neuroticism, Extraversion,
and the Perception of the vertical. Journal
of Abnormal and Social Psychology, 1958, 56,
139-142.

Weintraub, W. and Aronson, H. The application of verbal
behavior analysis to the study of psychological
defense mechanisms. II: Speech patterns associ-
ated with impulsive behavior. The Journal of
Nervous and Mental Disease, 1964, 132, 75-82.

Witkin, H. A. Psychological differentiation and forms of
pathology. Journal of Abnormal and Social Psycho-
logy, 1965, 70, 317-336.

Witkin, H. A. Personal communication.

Witkin, H. A., Dyk, D. B., Faterson, H. F., Goodenough,
D. R., and Karp, S. A. Psychological differen-
tiation, New York: Wiley, 1962.

Witkin, H. A., Karp, S. A. and Goodenough, D. R. Depend-
ence in alcoholics. Quarterly Journal of Studies
in Alcoholism, 1959, 20, 493-504.

Witkin, H. A., Lewis, H. B., Hertzman, M., Machover,
Karen, Meissner, P. and Wapner, S. Personality
through perception, New York: Harper and Bros.,
1954.





89


Witkin, H. A. Individual differences in ease of percep-
tion of embedded figures. Journal of Personality,
1950, 19, 1-15.
Witkin, H. A. The nature and importance of individual
differences in perception. Journal of Personality,
1949, 18, 145-170.
Young, R. H. A test of Witkin's field-dependence hypo-
thesis. Journal of Abnormal and Social Psycho-
logy, 1959, 59, 1i-192.













BIOGRAPHICAL SKETCH


William H. Davidov was born October 22, 1936, at

Philadelphia, Pennsylvania. He was graduated from Camden

High School, Camden, New Jersey. In June, 1958, he receiv-

ed the degree of Bachelor of Arts from Rutgers College of

south Jersey. In August, 1963, he received the degree of

Master of Arts with a major in Psychology from Temple

University. From February until September of 1964, he

served as a Psychology Intern at the Philadelphia State

Hospital. In the Fall of 1964, William H. Davidov enrolled

in the Graduate School of the University of Florida to work

toward the degree of Doctor of Philisophy. Since then he

has worked as a Psychology Trainee at the Veteran's Adminis-

tration Hospital in Coral Gables, Florida, was a Graduate

Fellow for one year, and was a United States Public Health

Fellow for one year. He completed his internship at the

Irving Schwartz Institute for Children and Adolescents,

Philadelphia, Pennsylvania, in July, 1967. William H.

Davidov took a position as a Staff Psychologist with the

West Philadelphia Community Mental Health Consortium from

July, 1967 until December, 1967. From January, 1968 until

September, 1968, he worked as a Clinical Psychologist for

the Alachua County Health Department, Gainesville, Florida.






91


In 1966, William H. Davidov married the former Jane

Rebecca Winnie of Buffalo, New York. He is an associate

member of the American Psychological Association.







This dissertation was prepared under the direction
of the chairman of the candidate's supervisory committee
and has been approved by all members of that committee.
It was submitted to the Dean of the College of Arts and
Sciences and to the Graduate Council, and was approved as
partial fulfillment of the requirements for the degree of
Doctor of Philosophy.

December, 1968


Dean, College of rts and Sciences



Dean, Graduate School



Supervisory Committee:


--Chairman
Chairman


MiV- nL




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - Version 2.9.7 - mvs