Effects of interpersonal orientation and language similarity on verbal communication in dyadic interpersonal relationships

PDF ( 4 MBs ) ( .pdf )

effectsofinterpe00hart_Page_03.txt

effectsofinterpe00hart_Page_58.txt

effectsofinterpe00hart_Page_36.txt

effectsofinterpe00hart_Page_06.txt

effectsofinterpe00hart_Page_40.txt

effectsofinterpe00hart_Page_44.txt

effectsofinterpe00hart_Page_64.txt

effectsofinterpe00hart_Page_17.txt

effectsofinterpe00hart_Page_66.txt

effectsofinterpe00hart_Page_71.txt

effectsofinterpe00hart_Page_13.txt

effectsofinterpe00hart_Page_62.txt

effectsofinterpe00hart_Page_30.txt

effectsofinterpe00hart_Page_24.txt

effectsofinterpe00hart_Page_60.txt

effectsofinterpe00hart_Page_39.txt

effectsofinterpe00hart_Page_52.txt

effectsofinterpe00hart_Page_74.txt

effectsofinterpe00hart_Page_21.txt

effectsofinterpe00hart_Page_02.txt

effectsofinterpe00hart_Page_23.txt

effectsofinterpe00hart_Page_65.txt

effectsofinterpe00hart_pdf.txt

effectsofinterpe00hart_Page_32.txt

effectsofinterpe00hart_Page_18.txt

effectsofinterpe00hart_Page_29.txt

effectsofinterpe00hart_Page_49.txt

effectsofinterpe00hart_Page_15.txt

effectsofinterpe00hart_Page_05.txt

effectsofinterpe00hart_Page_12.txt

effectsofinterpe00hart_Page_16.txt

effectsofinterpe00hart_Page_38.txt

effectsofinterpe00hart_Page_54.txt

effectsofinterpe00hart_Page_48.txt

effectsofinterpe00hart_Page_59.txt

effectsofinterpe00hart_Page_45.txt

effectsofinterpe00hart_Page_55.txt

effectsofinterpe00hart_Page_07.txt

effectsofinterpe00hart_Page_26.txt

effectsofinterpe00hart_Page_28.txt

effectsofinterpe00hart_Page_14.txt

effectsofinterpe00hart_Page_56.txt

effectsofinterpe00hart_Page_53.txt

effectsofinterpe00hart_Page_50.txt

effectsofinterpe00hart_Page_51.txt

effectsofinterpe00hart_Page_47.txt

effectsofinterpe00hart_Page_20.txt

effectsofinterpe00hart_Page_09.txt

effectsofinterpe00hart_Page_34.txt

effectsofinterpe00hart_Page_68.txt

effectsofinterpe00hart_Page_31.txt

effectsofinterpe00hart_Page_10.txt

effectsofinterpe00hart_Page_61.txt

effectsofinterpe00hart_Page_70.txt

effectsofinterpe00hart_Page_72.txt

effectsofinterpe00hart_Page_63.txt

effectsofinterpe00hart_Page_35.txt

effectsofinterpe00hart_Page_42.txt

effectsofinterpe00hart_Page_37.txt

effectsofinterpe00hart_Page_69.txt

effectsofinterpe00hart_Page_08.txt

effectsofinterpe00hart_Page_11.txt

effectsofinterpe00hart_Page_33.txt

effectsofinterpe00hart_Page_46.txt

effectsofinterpe00hart_Page_57.txt

effectsofinterpe00hart_Page_43.txt

effectsofinterpe00hart_Page_22.txt

effectsofinterpe00hart_Page_01.txt

effectsofinterpe00hart_Page_25.txt

effectsofinterpe00hart_Page_27.txt

effectsofinterpe00hart_Page_04.txt

effectsofinterpe00hart_Page_67.txt

effectsofinterpe00hart_Page_41.txt

effectsofinterpe00hart_Page_73.txt

effectsofinterpe00hart_Page_19.txt

EFFECTS OF INTERPERSONAL ORIENTATION

AND LANGUAGE SIMILARITY ON VERBAL

COMMUNICATION IN DYADIC

INTERPERSONAL RELATIONSHIPS









By
DONALD M. HARTSOUGH


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
April, 1964


































TO MY WIFE












ACKIOLEDGIENTS


Special gratitude is extended to Dr. Marvin E. Shaw, chairman

of the author's doctoral conrittee, for the encouragement, forebear-

ance, and keen insight with which he directed the study. Acknowledged

also are the helpful inquiries and contributions of the other committee

members, Dr. Albert M. Barrett, Dr. Sidney M. Jourard, Dr. Audrey S.

Schumacher, Dr. Rolland H. Waters, and Dr. Wilse B. Webb. The study

represents not only the research project as supervised by Dr. Shaw,

but a result of hours of valuable and stimulating discussion of inter-

personal behavior in both academic and clinical settings, particularly

with Drs. Jourard and Schumacher. Recognition should be given the

timely assistance of Dr. Clifford C. Courson, Dr. Herbert D. Kimlel, and

Mr. Richard A. Melloh. Finally, the author acknowledges his debt to the

staff of the P. K. Yonge Laboratory School for its support and coopera-

tion, and to the students iho served willingly as subjects for the study.


iii













TABLE OF CONTENTS


ACKNO LEDGME TS . . . . . . . . . . . iii

LIST OF TABLES. . . . . . . . . . . . . v

CHAPTER

I INTRODUCTION. . . . . . . . ... ... 1

Review of the Literature . . . . . . . 5
Statement of the Hypotheses. . . . . . ... 12

II METHOD AND DESIGN . . . . . . . . ... 15

Subjects . . . . . . . . ... . . 15
Experimental Groups. . . . . . . . ... 16
Experimental Dyads . . . . . . . .. 17
Measurement of Language Similarity . . . . 18
Experimental Task. . . . . . . . ... 19
Communication Statements . . . . . .... 21
Communication Responses. . . . . . . ... 24
Summary of Method and Design . . . . ... 27

III RESULTS . . . . . . . . ... .... 30

Characteristics of Subjects. . . . . . ... 30
Orientation, Compatibility, and Communication . . 32
Additional Findings. . . . . . . . ... 35

IV DISCUSSION AND CONCLUSIONS. . . . . . . . 3S

Roles and Communication. . . . . . . ... 39
Compatibility and Communication. . . . . 40
Conformity Behavior and Communication. . . .. 42
Verbal Conditioning and Communication. . . .. 43
Interpersonal Influences on Dyadic Communication . 44
Summary. . . . . . . . . . . . 48

BIBLIOGRAPHY. . . . . . . . . ... ....... 52

APPENDICES. . . . . . . . . ... . . .... 54

BIOGRAPHICAL SKETCH . . . . . . . . ... .. . 68












LIST OF TABLES


TABLE

1. Analysis of Variance of the Communication Ratios ... 33

2. Mean FIRO-B Inclusion Plus Affection (I + A) Scores and
Expressed Control (ec) Scores for Personal Oriented (PO)
and Control Oriented (CO) Groups, by Sex, and Total for
Both Groups . . . . . . . . ... ..... 62

3. Mean FIRO-B Interchange, Originator, Beciprocal, and
Total Compatibility Scores for Compatible and Incom-
patible Dyads . . . . . . . . .. . . 63

4. Frequency of Communication Responses by Orientation-
Compatibility Group and Communication Role. . . .. .64

5. Mean Communication Ratios by Orientation-Compatibility
Group and Communication Role. . . . . . . ... 65

6. Communication Responses by Photograph, Orientation-
Compatibility Group, and Communication Role. . . .. .66

7. Group Mean Communication Ratios by Photograph, Orienta-
tion-Compatibility Group, and Communication Role . . 67












CHAPTER I

INTRODUCTION


"Communication," "interpersonal behavior" and "interpersonal

relationship" are frequently attributed to events as if the terms were

synonymous and referred to the same phenomena. That they are distinct

from one another can be seen by reference to a systems model. An indi-

vidual may be regarded as an open system, i.e., a relatively stable en-

tity which has the capacity to transmit and receive information across

its boundaries. Communication refers to transmission--reception exchanges

with another system. Interpersonal behavior refers to the same interac-

tion process, but also includes the content of messages. Thus, the ex-

change of messages between two people may be effective while the content

of messages is disturbing to one or both. For example, a person may com-

municate rejection or anger, and while his message may be effectively

communicated, the content may be deleterious to another person. On the

other hand, interpersonal behavior which is satisfying to both parti-

cipants may take place in spite of temporary communication disturbances.

An interpersonal relationship refers to another, broader system which

includes both individual systems, as subsystems, the roles attributed to

the systems, and the communication process between the individuals. It

usually denotes a system which is maintained over time.

Students of communication have come to regard it as an essential

variable in changing interpersonal relationships. For example, a psycho-

therapist who wishes to help his client modify unsatisfactory relation-

ships with others must himself establish a therapeutic relationship with












the client. To do this, his interpersonal behavior must be in keeping

with the needs of the client, and he must communicate with precision,

appropriateness, efficiency and flexibility (Ruosch, 1961; Greenhill,

1958).

Just as communication may provide the means of modifying inter-

personal relationships, there is evidence that the pattern of communica-

tion which characterizes a person may be related to his present and past

relationships with significant others. Reiss (1957) found that in psycho-

therapy, hysteric patients tended to use past progressive and active

tenses, whereas obsessive-eompulsives more frequently used evaluative

words, comparative adverbs and polar-opposites. (It is assumed here that

personality type is determined in part by relationships with significant

others.) Lorenz and Cobb (1954) found the communication of neurotics, as

contrasted to normals, characterized by more verbs and pronouns, and

fewer adjectives and prepositions. They interpreted this as reflective

of the neurotic's greater concern for action and affiliation, and greater

avoidance of description and abstract relationships. Weakland (1962)

analyzed letters written by mothers o hospitalized schizophrenics; he

found subtle contradictions which tended to confuse overt messages and

which in some cases visibly upset the patients. Finally, Berg (1958)

found patient improvement in therapy associated with a change in pronouns

used by the patient. The ego words "I," "me," and "my" decreased and the

empathic words "you," "we," and "us" increased with improvement.

In accord with ideas and findings such as these, it seemed worth-

while to devise a study which investigated the general proposition that












communication between two people is related to dimensions of their inter-

personal behavior and the relationship which they create. The present

study investigated effectiveness of dyadic communication in a face-to-

face situation as is influenced by subjects' interpersonal orientations,

the compatibility of their relationship, and the similarity of their

language systems. Effectiveness of communication was evaluated by analy-

sis of feedback communication patterns between dyadic partners. Follow-

ing Ruesch (1957, 1961), communication was regarded as an exchange of

messages in a social context. Since the publication of Wiener's book

on cybernetics in 1948, and the extensive use ow communication theory

which followed, the term "communication" has been used in such a diver-

sity of ways that it is perhaps in danger of being rendered meaningless

as a concept. Communication was therefore amended by the term "interper-

sonal" to limit its application to the exchange of messages between

people who interact in a face-to-face relationship.

The dimension of interpersonal behavior selected for study was

interpersonal orientation. Interpersonal orientation refers to the fact

that people develop expectations about the dimensions along which inter-

personal relationships are formed, and thus tend to approach interactions

with others in certain ways. Schutz (1958) has posited three dimensions

(inclusion, control and affection) which he regards as the significant

orientations available for most relationships. A dimension of interper-

sonal relationships was aloo selected for study; this was the relative

compatibility or incompatibility of relationships. This refers to the

different combinations oi needs and orientations found in relationships,

and the degree of satisfaction offered by the combinations.












The general proposition that interpersonal behavior and relation-

ships are related to communicati n was refined to two specific problems:

(1) the assertion that an interpersonal orientation toward controlling

others is associated with more effective communication than an orienta-

tion toward inclusive and affective relationships, and (2) the assumption

that compatibility of relationship results in more effective communica-

tion than incompatibility. It was supposed that in a task situation the

control-oriented individual would communicate more effectively in order

to maintain control and in keeping with his greater concerns about roles

and tasks. The second problem was based on the assumption of a positive

relationship between compatible psychological needs and the ability to ex-

change messages effectively.

Communication effectiveness in a face-to-face situation was con-

sidered highly associated with use of feedback circuits, or the feedback

contingency, in the communication process. Wiener describes feedback in

the following terms: "In its simplest form, the feedback principle means

that behavior is scanned for its result, and that the success or failure

of this result modifies future behavior." (Wiener, 1950, p. 69.) Thus,

feedback provides a means by which a process is regulated or controlled.

This control results when a system which is producing a process inter-

acts with something outside of the system.

Ruesch treats feedback as a structural property of human cormuni-

cation. He contends that the structural properties determine successful

or unsuccessful comaunicative interactions, and states that utilization

of feedback is perhaps the foremost criterion of successful communication











(Ruesch, 1957). It provides a means of relaying back to the original

sender the effects that his message has had on other participants. It

thus enables him to modify constantly the communication process as the

process itself takes place. By reference to the principle of individual

differences in human behavior, it seemed reasonable to assume that indi-

viduals vary in the extent to which they utilize feedback circuits in

communicating with others. Part of this variation was expected to be

accounted for by interpersonal orientation and interpersonal compati-

bility.

A third variable was included in the study to control for non-

interpersonal factors in feedback communication. Some measure of language

similarity seemed necessary in light of the fact that communication in a

face-to-face situation is a cognitive as well as an interpersonal event.

If two individuals literally do not speak the same language, i.e., have

different language systems, there can be little or no feedback comimunica-

tion. Hence a procedure was employed to assess the similarity of language

systems of dyadic communication partners.


Review of the Literature

The research data of five verbal conditioning experiments supported

the possibility that interpersonal orientation and dyadic (two-person)

compatibility influence a subject's response to a communication inter-

change. These studies investigated the relationship between personality

variables and the response to verbal conditioning.

Sapolsky (1960) found that subjects who were given a set to find

the experimenter an attractive person had a different response rate from











other subjects who were given a set to find the experimenter an unat-

tractive person. During acquisition trials the High Attractive (HA)

subjects established a significant level of conditioning, while the Low

Attractive (LA) subjects did not. During extinction trials, in which the

experimenter was absent from the room, the HA subjects decreased their

rate only slightly, but the response rate of the LA subjects increased

significantly.

These results led Sapolsky to a second experiment which tested the

hypothesis that interpersonal compatibility between subject and experi-

menter would influence verbal conditioning. Five experiments were selected

from the same population as the subjects on the basis of scores on the

FIRO-B (Fundamental Interpersonal Relations Orientation--Behavior, Schutz,

1958). All were high in the control dimension. Each experimenter verb-

ally conditioned six subjects, three of whom were compatible and three of

whom were incompatible with him, according to the FIRO-B. Subjects made

up sentences using one of six pronouns, two of which were reinforced by

the experimenter's "'4m-hm". Extinction trials began when the experimenter

left the room and the subjects were asked to continue making up sentences

into a tape recorder. Results showed that during acquisition trials the

Compatible subjects increased and the Incompatible subjects decreased in

use of the reinforced words. During the extinction phase, the Incompat-

ible subjects increased significantly in response rate to a level com-

parable to that of the Compatible subjects.

A communication theory interpretation of these results would indi-

cate that the dyads comprised of Compatible subjects established and












maintained feedback circuits during the acquisition phase but that dyads

comprised of Incompatible subjects either did not establish feedback cir-

cuits or actively avoided the feedback contingency. Because the output

of one participant (E) was fixed a priori by the output of the other (S),

an extrapolation of the verbal conditioning paradigm to the communication

paradigm was a rather obvious step. The subject's emitted response was

considered as a sender's message or statement and the experimenter's re-

inforcing stimulus was considered as the receiver's acknowledgement. In

this context the subject's readiness to participate in feedback communica-

tion is roughly comparable to his conditioned response rate in that it was

evidence that the receiver's acknowledgements had an effect on his subse-

quent messages to the receiver. While the extrapolation from verbal con-

ditioning to the communication paradigm appeared obvious, it was necessary

to recall that in the conditioning situation it was the receiver (E) who

maintained control over the communication process by virtue of his knowl-

edge of the reinforcement contingency. In other communication dyads, the

sender maintains another kind of control by virtue of possessing more

content information than the receiver (e.g., teacher and student).

Although Sapolsky's study gave no indication of the role of per-

sonality variables as such on communication, it indicated that dyadic

interpersonal compatibility influenced the extent to which feedback cir-

cuits were used during the communication process. That the crucial vari-

able was an interpersonal one was further supported by Sapolsky's finding

that on a sociometric scale completed by both subject and e.perimenter,

19 of 30 pairs rated each other as predicted from their FIRO-B compati-

bility scores. This result reached statistical significance.











Cairns and Lewis (1962) investigated the relationship between

dependency and response to verbal conditioning. The Edwards Personal

Preference Scale was used to select High Dependent (HD) and Low Dependent

(LD) groups. The task consisted of making up sentences using one of three

verbs supplied by the experimenter. The HD and LD groups were further

divided; one-half of each was reinforced for using verbs with strong ag-

gressive connotations, and the other for using verbs with high dependency

connotations. In line with the author's expectations, HD subjects con-

ditioned to a higher level than LD subjects, although conditioning for the

HD subjects was strongly influenced by the type of verb reinforced. The

RD dependency-reinforced subjects initiated a high rate of response before

reinforcement began, and maintained this level during reinforcement. HD

aggressive-reinforced subjects showed initial avoidance of the aggressive

terms, but during reinforcement responded positively. For all LD subjects

the type of word reinforced had no differential effect; in addition, sub-

jects in both LD groups became more resistant to the conditioned stimulus

during reinforcement than they had during the pre-reinforcement series.

This result approximated the "negative" response of Sapolsky's Incompat-

ible dyads to the reinforcing stimulus. In communication terms, the find-

ings of Cairns and Lewis were interpreted to mean that highly dependent

subjects were not only more likely to maintain feedback communication

than low dependent subjects, but that they were also more sensitive to

the content of messages.

Timmons and Noblin (1963) found a differential verbal conditioning

performance between subjects identified as Oral and those identified as

Anal. Orality and anality were determined from a group administration











of the Blacky Test. Scoring was done according to the method developed

by Blun (1950). Orality referred to incorporating perceptions, and anal-

ity to withholding perceptions. Subjects used one of six pronouns to

complete a sentence supplied by the e:perimenter, and were reinforced

for two of then on a 75 per cent variable ratio schedule. The reinforc-

ing stimulus was a "mild, affirmatory utterance." The Orals showed a

typical learning curve, i.e., one which ascended during reinforcement and

descended during extinction. The Anals (true to character, perhaps) pro-

duced the opposite sort of learning curve, one that demonstrated a less

frequent response during reinforcement than during: either pre-reinforce-

ment or extinction. Changes in the Orals' conditioning curve reached

statistical significance, whereas the Anals' curve approached, but did

not reach, the accepted significance level.

Crowne and Strickland (1961) used the Marlowe-Crowne Social De-

sirability Scale to select subjects who were hi,,h in need for social ap-

proval and others who were low on the same scale. Half of each was posi-

tively reinforced and the other half negatively reinforced during a 25-

minute session in which subjects were instructed to produce words ad lib.

Positive reinforcement was "MIl-hm" plus the experimenter nodding his head

affirmatively; negative reinforcement was "Uh-uh" plus the experimenter's

noddin, negatively. Only plural nouns were reinforced; the response rate

was the ratio of plural nouns to total output for ':he session. Results

supported the hypothesis that subjects with a hi h need for social ap-

proval would show significantly greater increases in response rate under

conditions of positive reinforcement and significantly greater decreases

under conditions of negative reinforcement than subjects to whom social












approval was of less consequence. In fact, the low need-for-approval

subjects could not be distinguished from control subjects who received

no reinforcement.

Marlowe (1962) extended the work of Crowns and Strickland to an

interview situation but used only positive reinforcement. High and low

need-for-approval subjects were reinforced by the experimenter's "Zn-hm"

for statements of positive self-regard (PSR) made during a 15-minute in-

terview in which the subject described his "personality characteristics

and traits." Uihen compared with control subjects who did not receive re-

inforcement, hiLh-need subjects emitted more PSR responses and signifi-

cantly increased their rate, while low-nead subjects emitted fewer PSR

responses and significantly decreased their rate in rauch the same way as

had control subjects. A statistically significant correlation of +.42

between need for approval and PSR was obtained for experimental subjects,

but for control subjects the corresponding correlation of +.23 was non-

significant. Marlowe interpreted his results as reflecting a greater

sensitivity and responsiveness to social reinforcers on the part of sub-

jects with high needs for social approval.

The results of these studies may be summarized as follows:

1. Sensitivity to the verbal feedback contingency (or more pre-

cisely, the tendency to maintain feedback circuits in a dyad) depended

in part on at least one dimension of personality. This was variously

described as high vs. low dependency, orality vs. anality, and high vs.

low need for social approval. From the authors' descriptions and

their means of selecting subjects, this seemed to represent the extreme












positions of a single personality dimension. In terms of interpersonal

theory, this dimension approx:inates a dichotomy between an orientation

toward social and affective needs and an orientation towards needs to

express control over others.

2. While the positive responsivity of the dependent, oral and

high need-for-approval subject was uniformly demonstrated, the response

pattern of the low-dependent, anal and low need-for-approval subject was

less clear. Two studies reported statistically significant or almost

significant decrements in response during reinforcement of the latter

type of subject from the pre-reinforcement level (Cairns and Lewis, 1962;

Timmons and Noblin, 1963). In two other studies (Crowne and Strichland,

1961; Marlowe, 1962), these subjects responded in nmuh the same manner as

the non-reinforced control groups, and would be described as nonresponsive

rather than negatively responsive. Both nonresponsivity and negative

responsivity may be interpreted as this subject's attempt to maintain

control of the communication and to resist becoming part of an interper-

sonal process over which he feels he has no control.

3. Sapolsky's findings demonstrated that differential responsi-

vity can be produced in dyads of naive subjects as well as with experi-

menter and subject. However, since Sapolsky did not provide the FIRO-B

scores for his Compatible and Incompatible groups, it was not possible to

determine whether or not his dyads varied according to the personality

dimension described above.

As stated above, the findings surmmarized here seemed to have im-

plications for the study of interpersonal communication as well as for












verbal conditioning. The communication pattern as applied to the condi-

tioning procedure (i.e., the pattern of statement-acknowledgement-feed-

back-statement) represents simply a translation of terns from verbal

conditioning (i.e., from the procedure of stimulus-response-reinforcement-

stimulus) to communication theory. This same communication pattern, how-

ever, may be extended to rore couple:;, natural dyadic communication, in

which both participants are free to select the content of statements and

acknowledgements, and in which the pattern of statements and acknowledg-

ments becomes modified by the introduction of communication blocks and

other structural properties. Further, it appeared significant that in

the conditioning studies the subject was forced to enter into a rigidly

controlled feedback circuit with the e:perimenter. The fact that it

was a rigidly controlled system--and controlled by the experimenter--nay

have influenced the control- and independence-conscious subject to with-

draw from the system. It could be argued, in fact, that given the oppor-

tunity to be in a more controlling role in the system, for example as a

sender with more information than the receiver, the low-dependent, high-

control subject would produce more feedback communication than his

opposite.


Statement of the Hypotheses

The present study investigated the effects of interpersonal ori-

entation, compatibility, and language similarity on interpersonal commu-

nication in a face-to-face situation. Following Ruesch (1957), the pri-

mary measure of communication was regarded as the extent of use of the

feedback contingency.











The hypotheses were as follows:

1. Individuals highly oriented toward controlling interpersonal

relationships attempt to do so by producing more feedback communication

than individuals highly oriented toward social and affective aspects of

interpersonal relationships.

The first hypothesis was derived from a reinterpretation of the

results of the verbal conditioning studies reviewed above. While these

results indicated that subjects who were oriented primarily toward social

and affective relationships more readily participated in feedback communi-

cation than those who were control-oriented, these results were suspected

to be situationally determined by the inability of the subject to control

the feedback contingency. When control of the communication process is

not so rigid, the latter type of subject was expected to make greater

use of feedback circuits.

2. Dyads which are interpersonally compatible make greater use of

the feedback contingency than dyads which are interpersonally incompatible.

The second hypothesis was derived from Sapolsky's (1960) finding

regarding interpersonal compatibility and conditioning.

3. Individuals who are highly control-oriented make greater use

of the feedback contingency as senders than as receivers.

The third hypothesis was an elaboration of the first in that the

control-oriented subject was expected to be especially sensitive to the

feedback contingency when in the sender role. The sender role may be as-

signed to the communicator with the most information about the communica-

tion process itself or about the content of communication. This hypothe-

sis, supported by results of a pilot study, was based on the assumption











that control orientationincluded greater sensitivity to role and task

factors, hence would produce greater communication differences accord-

ing to tole.

4. Similarity of the language systems of dyadic partners is

positively related to the dyad's use of feedback communication in a

face-to-face situation.

The fourth hypothesis concerned a non-interpersonal variable in

relation to interpersonal communication. Because it was reasonable to

assume that a dyad's capacity for use of feedback circuits has cogni-

tive determinants, a measure of the similarity of written language was

included in the present study.











CHAPTER II

METHOD AND DESIGN


To test the hypotheses relating communication to interpersonal

orientation, compatibility, and communication role, pairs of subjects

were selected according to FIRO-B orientation scores and assigned to

dyads according to compatibility criteria; a second measure assessed

the similarity of their language systems, and finally a sample of their

:orimnication in both sender and receiver roles was obtained. A 2 X 2

X 2 factorial design assigned each subject to one of four treatment

conditions and each of two communication roles. Similarity of language

system was evaluated by separate correlational procedures.


Subjects

The subjects were juniors and seniors (N=170) from the P. K. Yonge

Laboratory School of the University of Florida. Both sexes were used,

but dyads were formed on a like-se: basis. Since Marlowe (1962) and

Crowne and Strickland (1961) found no sex differences in their condition-

ing data, this variable was not expected to demand differential treatment

e.:cepting the assignment of equal numbers of both sexes to each experi-

mental group. Additional information regarding age and intelligence

was obtained from school records. The extent of acquaintance of dyadic

partners was obtained from a brief questionnaire administered subsequent

to all other procedures. Characteristics of subjects by reperimental

groups are provided in Chapter III.


ubsequnt analysis of the data sustained this e: action.
Subsequent analysis of the data sustained this expectation.












Experimental Groups

Sixty-four subjects were selected for placement into one of two

experimental groups on the basis of FIRO-B scores. The FIRO-B is a

questionnaire of 54 statements developed by Schutz (1958) to measure

interpersonal orientation according to three interpersonal needs--inclu-

sion, control, and affection. These needs are further subdivided into

an "expressed" and a "wanted" need, thereby recognizing the distinction

between need to give and need to receive each of the three types of be-
9
havior. Subjects rate themselves on a six point rating scale according

to degree of agreement or disagreement with the statements. Typical

statements are: "I join social groups," "I try to take charge of things

when I'm with people," and "My personal relations with people are cool

and distant." Each subject is described by six scores, a wanted and ex-

pressed score for each need area. The FIRO-B was administered to both

junior and senior classes during a regular class period.

Half of the 64 subjects were selected for high scores in inclu-

sion plus affection (I + A), and the other half for high scores in ex-

pressed control (ec). Generally, males (N = 32) were selected for the

former group if I + A was 21 or more and ec was less than five, and for

the latter group if I + A was less than 21 and ec exceeded five. For

females (N = 32), the corresponding values were I + A score of 23 and

ec score of four. These criteria provided one group significantly ori-

ented toward inclusive and affective needs and a second group signifi-

cantly oriented toward control needs. For convenience, they were

called the Personal Oriented (PO) and Control Oriented (CO) groups.












Experimental Dyads

Thirty-two like-sexed dyads were formed from subjects selected

for the PO and CO groups. To test the hypothesis that compatibility in

interpersonal relationships produces more communication than incompati-

bility, formation of the dyads was related to compatibility procedures

of the FIRO-B. Compatibility is defined by Schutz (1958) as the prop-

erty of interpersonal relationships which leads to mutual satisfaction

of interpersonal needs and harmonious coexistence. Measures have been

developed by Schutz (1958) for three types of compatibility: originator

compatibility, reciprocal compatibility, and interchange compatibility.

Interchange compatibility describes the relative similarity of two indi-

viduals according to the need areas, whereas the other two types describe

their relative similarity according to come combination of expressed and

wanted scores. Interchange compatibility seemed the most appropriate

measure for use with a short-term communication dyad. It is expressed

in the formula: xK = (ei + wi) (ej + wr) where e and w refer to

expressed and wanted scores for a need area.

Sixteen dyads were compatible and sixteen were incompatible. The

compatible dyads were made up of eight pairs of subjects from the PO

group and eight pairs of subjects from tne CO group. The sixteen in-

compatible dyads consisted of one subject from each of the PO and CO

groups. Assignment was made according to a ranking of I + A and ec

scores within each group. Dyad assignment by this procedure accomplished

counterbalancing for partner rank and initial communication role. This

method also permitted the use of multi-variate analysis of variance pro-

cedures to test the first three hypotheses.












Measurement of Language Similarity

A measure of language similarity was obtained for each dyed for

correlation with dyad communication measures. This measure was used to

test the fourth hypothesis, but was not used for establishing the ex-

perimental groups. Use of a written sample of language from each partner

eliminated the possibility that face-to-face effects contaminated the

measure. At the group administration of the FIRO-B, all 170 subjects

were instructed to write a 150-200 word story or essay in response to

card 17BM of the Thematic Apperception Test (TAT). To be included in the

experiment, a subject's passage had to total at least 100 words. The

passages were prepared by the Cloze Procedure and administered prior to

the experimental test for communication to groups of subjects who were

not partners.

The Cloze Procedure was developed by Taylor (1953) as a means of

measuring readability. He describes the Cloze score as appearing to

measure the aggregate influences of all factors which interact to affect

the degree of correspondence between the written language patterns of

transmitter and receiver. This technique deletes every nth word in a

passage and substitutes a blank in its place; subjects then fill in the

missing word. Each of the 64 subjects completed his partner's typewritten

passage after every fifth word had been deleted. Each passage contained

a total of 20 blanks and the complete sentences in which the deletions ap-

peared. Responses were scored as either correct or incorrect by comparing

them with the words deleted from the original passage. The completed pass-

age of each subject was compared to the original passage written by his












partner. The scores of dyadic partners were then summed to provide the

measure of language similarity for the dyad.


Experimental Task

To determine the readiness of a subject to utilize feedback, it

was necessary to create a communication situation in which measurement

could be readily obtained without subjects becoming aware of the critical

variable being measured. The situation also demanded procedural control

to the extent that the communication measures did not become confounded

with one another. An experimental communication task meeting these

criteria was devised and successfully tested in a pilot study. This task

consisted of a pseudo-problem which required each subject in the dyed to

make oral responses to the experimenter and permitted (but did not re-

quire) him to enter into a communication relationship with the partner,

i.e., to influence the partner's messages or be influenced by them. Thus,

the degree to which partners communicated with one another or simply made

oral responses to the experimenter was the dependent variable.

Dyadic partners were seated opposite one another at an oblong

table longitudinally divided by a board one foot high. The divider en-

abled them to see one another but screened material on the partner's half

of the table. The experimenter, seated at one end of the table, directed

the subjects' attention to a photograph mounted on the end of the divider

at the opposite end of the table.

The psuedo-problem consisted of a "personality description" of a

person shown in a photograph. (Two anonymous photographs were selected

for this purpose; both were of young adult males.) The subjects were











told that the experimenter was gathering information about the impres-

sions people get from photographs of people whom they do not know, and

from personal history information about these same people. Instructions

were given which gave the psuedo-problem a rationale and which made it

seem a worthy investigation. (The complete instructions are found in

Appendix A.)

Before the personality description began, one of the subjects was

given a "sheet of information" describing the person in the photograph.

(The information sheets for both photographs are included in Appendix B.)

A different sheet was devised by the experimenter for each photograph.

The personal histories were roughly parallel in that both men were born

in the South, went to college after experiencing difficulties, served in

the armed forces, received vocational counseling, married college women,

and made tentative career choices following graduation.

The provision of the history information to one subject per photo-

graph made possible the establishment of communication roles in the dyad.

That is, the subject who had the information was "sender" and the other

subject was "receiver"--although subjects were given no instructions to

establish the roles. This also made the experiment more believable, as

it demonstrated the necessity of testing two subjects at a time.

Description of the person in the photograph was accomplished by

means of a set of cards containing phrases appropriate for such a task.

Each subject had a different set of 16 cards. Each card contained four

phrases, making a total of 128 for both sets. The description task con-

sisted of each subject choosing, in turn, one of the four phrases which

he thought best fit the person in the photograph.












The instructions gave each partner a different task. The subject

receiving the information sheet was required to select one phrase for

every card. The other subject, i.e., "receiver," had the option of

either selecting a phrase or of making no selection at all. This was in-

cluded in the procedure to resemble the alternative of noncommunication

characteristic of a natural dyad, and to provide a measure of receiver

nonacknowledgement of messages. The latter was thought a sensitive

measure of tendency to withhold communication, and as such would be re-

ciprocal to a measure of "need to communicate."

After the subjects had used, alternatively, each of their 16 cards

to describe one photograph, a second one was substituted by the experi-

menter. For the 16 trials of the second photograph, the subjects ex-

changed card sets and also exchanged roles. The subject who was sender

for the first photograph became receiver for the second, and vice versa.

A second sheet of information was supplied the sender before the descrip-

tion trials began for the second photograph.

As subjects announced their selections from the cards, the ex-

perimenter recorded each choice. He thus made certain that both subjects

heard all choices, as words mispronounced or omitted were repeated by the

experimenter as if for clarity.


Communication Statements

The 128 phrases supplied on the 32 cards (16 per set) constituted

the population of communication statements from which partners selected

messages. To determine whether messages influenced subsequent messages,

i.e., were part of a feedback circuit, the statements had to be systemati-












cally related to one another. This requirement was met in 128 statements

appropriate for personality description selected from the Interpersonal

Check List (ICL) developed by Leary (1957). Leary's interpersonal theory

is a behavior-oriented, multi-level system which classifies data accord-

ing to eight dimensions of interpersonal behavior, each dimension bearing

a constant relationship with every other dimension. The basic model em-

ployed is a circle with each dimension represented by an octant of the

circle. Octants opposed to one another represent opposing behaviors and

adjacent octants represent similar behaviors. Leary's principle measur-

ing device, the ICL, is a self-report technique containing 128 statements

describing interpersonal behavior, 16 statements for each of the eight

dimensions. The eight dimensions as described by Leary, and a representa-

tive adjective phrase for each, are listed below:

1. Managerial Autocratic "Able to give orders"

2. Competitive Exploitive "Able to take care of self"

3. Blunt Aggressive "Can be frank and honest"

4. Skeptical Distrustful "Able to doubt others"

5. Modest Self-effacing "Can be obedient"

6. Docile Dependent "Appreciative"

7. Cooperative Over-conventional "Friendly"

8. Responsible Iypernormal "Helpful"

Statements were selected so that for each item appearing on a card

in Set A, the next adjacent item from the ICL appeared on the same card

in Set B. That is, the 128 ICL items were taken in the order 1-128,

every other item belonging to the same set. This procedure produced two











cards for each of the 16 trials, each card containing descriptions which

resembled one another but were not identical. For example, the items

selected for card No. 3 for each set appeared (in randomized order) as:

A B

friendly affectionate and understanding

can be obedient usually gives in

can be frank and honest forceful

able to give orders critical of others

Statements were selected for the cards from every other octant

of Leary's circular model. The even numbered cards contained statements

from octants 2, 4, 6, and 8 (as numbered above), and the odd numbered

cards from octants 1, 3, 5, and 7. The opposing octants around the

circle are: 1-5, 2-6, 3-7, and 4-8. Thus, statements were selected from

two pairs of diagonally opposed octants. Because diagonally opposed oc-

tants represent contradictory forms of interpersonal behavior in the Leary

system, each pair of phrases appearing on a card described contradictory

interpersonal behavior. A sample from card No. 3, Set B, is the pair "af-

fectionate and understanding" and "critical of others." In addition, pairs

themselves were perpendicular to one another so that for each phrase on a

card there was one contradictory phrase and two other phrases neutral to

it. The four phrases thus selected were randomly placed on the card.

Items as they appear on the ICL vary in an intensity dimension

within each octant, from low to high intensity of the behavior. As items

were selected for the cards in order from 1-128, the cards varied in the

intensity dimension from 1-16. (A sample scoring sheet containing the popu-

lation of 128 communication statements appears in Appendix C.)












Communication Responses

Subjects selected messages from cards containing parallel sets

of statements. Thus, messages could be compared according to their

similarity or dissimilarity with one another by reference to the ICL

octants from which they were drawn. Experimentation with various methods

of scoring and categorizing messages indicated the most suitable scoring

procedure to be a similar--dissimilar dichotomy. When subjects' messages

were compared with one another according to this criterion, four types of

coraunication responses were defined. Definition of the responses was in

terms of the similarity or dissimilarity of the message to the two previ-

ous messages, i.e., one by the respondent and one by his partner. The

responses were labeled congruent, redundant, confirming, and novel. From

the standpoint of receiver, the four responses could be operationally

described in the following manner:

1. Congruent response: Receiver's message was similar

to both sender's message and his own previous message.

2. Redundant response: Receiver's message was similar to

his own previous message, but dissimilar to sender's

message, i.e., he repeated himself without response to

sender.

3. Confirming response: Receiver's message was similar to

sender's message, but dissimilar to his own previous

message, i.e., he changed his message to confirm that

of sender.

4. Novel response: Receiver's message was dissimilar to both












his own and sender's messages, i.e., he introduced a

new message.

Since receiver had the option on each trial of not selecting a

phrase, a problem arose as to the most appropriate way of categorizing

this message. This was done as follows: Non-selections by receiver

were categorized as novel responses if, on the previous trial, receiver

had made a selection, i.e., his non-selection was a "new" message of non-

communication; however, if his non-selection followed a non-selection on

the previous trial, it was categorized as a redundant response, i.e., it

was a repetition of the message of non-coamunication.

Responses which sender made to receiver were also categorized by

comparing his message to his own previous message and to the intervening

message of receiver. This seemed justifiable as, after the first trial,

he received as well as sent messages. Categorizing of communication

responses of both sender and receiver coimenced with the second trial.

The congruent response, although indicative of agreement, was not

considered a sufficient criterion for detecting communication. Although

this undoubtedly resulted in omission of some comrmnication which took

place, it also eliminated instances in which congruence of response sig-

nified nothing more than agreement in judging the photograph. Redundant

responses were accepted as lack of communication, as they reflected

repetition of a message following sender-receiver disagreement. Conmu-

nication, that is, utilization of the feedback circuit, was considered

demonstrated by a confirming response. In this case a subject confirmed

his partner's message by selecting a statement of greater similarity to

the partner's than to his Mwn previous message. In one sense this repre-












sents conformity, but since conformity usually connotes pressure and

there was little in this situation which created pressure to confirm,

the term "conforming" seemed more appropriate. In this sense the

response meant "to support" or "to verify."

To make comparisons between trials, it was again necessary to

refer to Leary's octant system. Trials alternated in the four octants

from which they drew items from the ICL, the odd-numbered trials drawing

from odd-numbered octants and even-numbered trials drawing from even-

numbered octants. It was therefore necessary to decide which odd-

numbered octants represented a dissimilar statement for a statement

made from an even-numbered octant, and vice versa. The most conservative

decision was to use only the octants that were adjacent to the opposite

octant, and to consider all three octants "dissimilar." The following

is a list of each octant and the octants considered dissimilar to it

for scoring purposes; all other octants were considered "similar."

Octant Dissimilar octants

1 4, 5, 6

2 5, 6, 7

3 6, 7, 8

4 7, 3, 1

5 8, 1, 2

6 1, 2, 3

7 2, 3, 4

3 3, 4, 5

In practice, subjects' messages were first translated into octant

numbers, and by reference to the above list, categorized into one of the












four communication responses. As responses were scored starting with

the second trial, each subject had 15 responses per communication role

and 30 for the complete test of communication.

One technical adjustment was necessary in order to equate the

number of receiver confirming responses per photograph to those for

sender. Because receiver had a 3/4 by-chance choice of responding to

sender with a similar message, whereas all other comparisons between

messages operated on a 2/4 chance basis, it was necessary to subtract

from receiver's confirming response total, one-third of that total. The

formula was expressed: SCR = RCR (2/3), where SCR is sender confirming

response and RCR is receiver confirming response.


Summary of Method and Design

The following is a summary in chronological order of steps taken

to select subjects for the experiment, form experimental groups, and test

for the presence of cormunication between subjects. Rationale for the

procedures is provided in the foregoing discussion.

1. The FIRO-B was administered to the junior and senior classes

at the P. K. Yonge Laboratory School. At the same time, each student was

instructed to write a story or essay in response to card 17 BM of the TAT.

2. Sixty-four students were selected as subjects for study. Six-

teen males and 16 females were selected for each of the PO and CO groups

on the basis of their FIRO-B score; I + A and ec scores were the criteria

for this selection.

3. PO and CO groups were equally divideC to form dyads for test-

ing by the experimental procedures. Half the subjects were in compatible











dyads and the other half in incompatible dyads. Compatible dyads were

comprised of eight PO-PO and eight CO-CO combinations, and incompatible

dyads were comprised of 16 PO-CO combinations. All dyads were like-sexed.

4. Subsequent to selection of dyads each subject was given his

partner's typewritten TAT passage which had been prepared by the Cloze

Procedure. He was unaware of the authorship of the passage. He was

instructed to fill in the missing words. Cloze scores for each dyad

were obtained by summing the scores of the subjects in the dyad. This

measure was used to test the fourth hypothesis.

5. Dyadic partners were tested for communication together, one

pair of subjects at a time. They were instructed to describe a photo-

graph by means of selecting one of four statements on a card. Sixteen

cards weie used for each of two photographs. One subject was given ad-

ditional information about the person in the photograph; he was required

to select a statement from each of the sixteen cards. The other subject

was given the option of selecting a statement from each card or of not

responding. Partners exchanged roles for the second photograph.

6. Statements used to describe the photographs were from the ICL.

Since the statements were related to one another in a systematic manner

it was possible to record the pattern of a subject's agreements and dis-

agreements with his partner's and with his own previous statements.

From these patterns, four communication responses were defined. The con-

firming response indicated use of the feedback contingency, whereas the

redundant response represented lack of sensitivity to the feedback con-

tingency. These responses were used to test the first three hypotheses.












7. Subsequent to all other procedures, all subjects filled in a

questionnaire on which they indicated the extent of their acquaintance

with the person uho had been their partner in the communication test.

This was used to control for this variable in consideration of the other

results.












CHAPTER III

RESULTS


Experimental procedures were designed to test hypotheses relating

communication to interpersonal orientation, compatibility, and language

similarity. Subjects were assigned to one of four independent treatment

groups, hereafter called orientation-compatibility groups. This term re-

fers to the PO-compatible, PO-incompatible, CO-compatible, and CO-incom-

patible groups, where PO means personal-oriented and CO means control-

oriented approaches to interpersonal relationships. This chapter presents

characteristics of the subjects, gives the communication results which

tested the hypotheses, and presents additional findings pertaining to the

general problem under investigation.


Characteristics of Subjects

PO and CO groups were each comprised of 16 males and 16 females

according to the double criteria of I + A and ec scores, as described in

Chapter II. Mean score differences between PO and CO groups according to

these measures were significant (p < .001) in the required direction.

(See Table 2 in Appendix D.)

From the PO and CO groups, dyads were formed of which half had a

compatible relationship and half had an incompatible relationship. Com-

patibility scores for interchange, originator, reciprocal, and total com-

patibility were calculated for each dyad. (See Table 3 in Appendix E.)

Interchange compatibility was regarded the most appropriate criterion

measure for differentiating dyadic groups for a short-term communication












relationship; it referred to tha relative emphasis placed on the three need

areas by the partners. Originator compatibility referred to each subject's

tendency to initiate or receive behavior within a need area, while recipro-

cal compatibility referred to his tendency to initiate compared to his

partner's tendency to receive. Total compatibility was the sum of the

three separate compatibility measures (Schutz, 1958).

The mean difference between compatible and incompatible dyads on

interchange compatibility scores was significant (p < .001) in the required

direction. Compatible dyads were also more compatible according to recipro-

cal compatibility scores (p < .001) and total compatibility scores (p < .01)

but were less compatible according to originator compatibility scores

(p < .01).

The age range of the 64 subjects was 15-18. There were no signifi-

cant differences in mean age among the four orientation-compatibility

groups. A non-significant t of 1.34 (p. > .05) was obtained for the

greatest mean difference between two groups.

Intelligence test scores on the California Test of Mental Maturity

ranged 77-133. The CO-incompatible group was higher in mean intelligence

score than the PO-compatible group (t = 2.34; p < .05). Differences for

all other orientation-compatibility group comparisons were non-signifi-

cant (p > .05). In addition, a non-significant correlation of +.26

(p > .05) was obtained between intelligence test scores and the communi-

cation ratios for each subject discussed in the next section.

Extent of acquaintance of partners with one another was evaluated

by the median test. There were no significant differences between PO and












CO groups or between compatible and incompatible dyads (p > .05) in

median years of acquaintance.


Orientation, Compatibility,and Communication

Frequencies of the four communication responses (confirming, re-

dundant, congruent, and novel) were obtained for each subject by communi-

cation role. (See Table 4 in Appendix F.) The transformation, X =J7-T.5,

was applied to the response frequencies to eliminate zero totals and to

insure adequate variability (Edwards, 1950, p. 200). A communication

ratio was obtained for each subject in each role by dividing his sum of

confirming responses by his sum of redundant responses. This ratio was

considered the most accurate index of use of feedback circuits. Under

conditions of a discrepancy in subject-partner messages, this ratio com-

pared the subject's tendency to change messages in the direction of his

partner, to his tendency to ignore the partner by repeating his own

previous message.

Table 1 shows the analysis of variance of the communication ratios

by orientation, compatibility, and communication role. Since orientation

and compatibility were independent observations, residual variance between

subjects was used as the error term for their evaluation. Communication

role was a repeated observation, thus the error term for its evaluation

was residual variance within subjects. The criterion accepted for

statistical significance was p < .05.

The first hypothesis predicted communication differences by inter-

personal orientation. A non-significant F for the orientation variable

indicated a lack of support for this hypothesis.












TABLE 1

ANALYSIS OF VALIANCE OF THE


COMMUNICATION RATIOS


Sum of Mean
Source Squares df Square F


Independent observations:

Orientation (0) .24 1 .24 .73

Compatibility (C) 1.02 1 1.02 3.09

0 X C .22 1 .22 .67

Residual between
subjects (error) 19.90 60 .33

Total, independent 21.38 63


Correlated observations:

Role (R) 14.16 1 14.16 42.91*

RX C .05 1 .05 .15

R X O .11 1 .11 .33

RXCXO 0 1 0 0

Residual within
subjects (error) 19.75 60 .33

Total, within subjects 34.07 64

Total for experiment 55.45 127


*p < .001












The second hypothesis predicted greater use of feedback circuits

by compatible dyads than by incompatible dyads. The F for communication

ratios by the compatibility variable approached but did not reach signi-

ficance at the accepted level. Examination of the mean difference showed

incompatible dyads to have a higher average ratio than compatible dyads.

(See Table 5 in Appendix G.) To the extent that this difference was re-

liable, the second hypothesis was contradicted. Further, a correlation

of+.34 was obtained between communication ratio for each dyad and its

interchange compatibility score; this fell just short of significance

(p > .05). Since the magnitude of compatibility scores is inverse to

compatibility, this was further evidence toward contradiction of the

second hypothesis.

The third hypothesis predicted greater use of feedback circuits

for CO subjects as sender than as receiver. A significant F was obtained

for the role variable, but in a direction opposite that predicted for CO

subjects. When CO subjects were considered separately, mean sender ratio

was exceeded by mean receiver ratio to a significant degree (e = 2. 82;

p<.01).This was also true of PO subjects (t = 2.77; p < .01). Thus, the

third hypothesis was contradicted as a result of a highly significant

tendency for all subjects to use feedback circuits more as receiver than

as sender.

The fourth hypothesis predicted a positive relationship between

similarity of partners' language systems and their use of feedback com-

munication. Cloze scores for the partners of each dyad were summed to

give a dyad Cloze score. The possible range of the dyad measure was 0-40,













and the obtained range was 11-33. Since the correlation of -.25 between

dyad Cloze scores and dyad communication ratios was non-significant

(p > .05), the fourth hypothesis was not supported.


Additional Findings

Results are given below in addition to those which tested the

hypotheses and which concerned the general problem under study. They

pertain to: (1) additional variables which were significantly related

to the communication ratio; and (2) variables related to the congruent

response.

Although the second hypothesis (which predicted greater use of

feedback by compatible dyads) was contradicted in terms of interchange

compatibility, it was supported in terms of originator compatibility.

A correlation of -.41 (p < .05) was obtained between originator compati-

bility scores and coarmunication ratios. Since compatibility scores are

inverse to compatibility, this finding signified a positive relationship

between the two variables. This was consistent with the foregoing find-

ing that dyads labeled "incompatible" were actually more compatible than

dyads labeled "compatible" according to originator compatibility scores.

Subjects when in the role of receiver had the option on each trial

of not selecting a statement. Non-selection was regarded as a message of

"non-comiunication" and thus, perhaps, reciprocal to a "need to coramuni-

cate." This possibility was supported by a correlation of -.30 (p < .05)

between the number of non-selections per subject and subject communica-

tion ratios. Thus, subjects who were more sensitive to the feedback












contingency also tended to transmit more messages of content to the

partner.

The congruent response was regarded as indicating agreement be-

tween partners in describing the photograph. Since agreement in the

experimental situation was possible without use of feedback circuits,

congruent responses were not used to test the hypotheses. However,

the finding of a correlation of +.66 (p <.001) between frequency of

congruent responses per subject and subjecL communication ratios lends

support to the possibility that congruent responses, in general, were

indicative of communication. The reverse is also true, i.e., that use

of feedback circuits ray have been strongly influenced by partners'

tendencies to agree. To an extent, this correction may have been

inflated because the communication responses were correlated measures.

The use of a ratio, rather than raw scores, for the index of communica-

tion was considered a sufficient correlation to permit computation of

the correlation.

Three additional findings related the compatibility and orienta-

tion variables to the tendency of partners to agree or disagree. Just

as congruent responses indicated agreement, redundant responses were

interpreted as indicating disagreement. Incompatible dyads produced more

congruent responses than compatible dyads, as sender (p < .001) and as

receiver (p < .01). Conversely, compatible dyads produced more redundant

responses than incompatible dyads, as sender (p < .01) and as receiver

(p < .02). Since the communication responses were correlated measures,

it was not unexpected that high frequency of congruent responses was











associated with low frequency of redundant responses, and vice versa.

However, the finding remains that compatibility of relationship was

associated with disagreement in describing the photographs.

The tendency toward lack of agreement found in compatible dyads

was largely due to PO subjects within the compatible group. The 16 PO-

compatible subjects produced fewer congruent responses than the 16 CO-

compatible subjects; the difference was significant for the receiver

role only (p < .01).

The positive findings may be summarized as follows:

1. Communication role was the only main effect significant in

the variability of communication ratios. All subjects as receiver tended

to use the feedback circuits far more than as sender.

2. Compatibility as related to the communication ratio was com-

plicated by the negative relationship of originator compatibility to the

other compatibility measures. Interchange, reciprocal and total compati-

bility scores were negatively related to the communication ratio, whereas

originator compatibility was positively related to it.

3. A low and positive relationship was found between tendency to

transmit statements (under conditions of choice, i.e., as receiver)and

the communication ratio.

4. A relatively high and positive relationship was found between

subject's tendency to agree with partner and subject communication ratios.

5. Compatible dyads demonstrated significantly less agreement in

describing the photograph than incompatible dyads.

6. PO-compatible subjects were significantly less prone to agree

than CO-compatible subjects.












CHAPTER IV

DISCUSSION AND CONCLUSIONS


The present study c:anined communication effectiveness front the

standpoint of interpersonal behavior and relationships. The prediction

that an interpersonal orientation toward controlling interpersonal re-

lationships results in greater sensitivity to the feedback contingency

was not supported. A related prediction that control-oriented indivi-

duals use feedback circuits more as senders than as receivers was con-

tradicted by the finding that all subjects tended to use then more in

the role of receiver. A third prediction stated that interpersonal

compatibility is associated with greater use of feedback circuits than

incompatibility. For originator compatibility, this was supported; the

finding that reciprocal, interchange, and total compatibility resulted

in less use of feedback circuits was contradictory to the prediction.

Finally, the prediction that similarity of the language systems of

communicative participants is associated with use of the feedback con-

tingency was not supported. Thus, none of the four hypotheses was

directly supported and two were at least partially contradicted.

Despite these negative findings, there was support for the general

proposition that communication is influenced by interpersonal variables.

Additional findings provided evidence that sensitivity to the feedback

contingency was influenced primarily by the presence of communication

roles and the acceptance or non-acceptance of these roles by partici-

pants.












Roles and Communication

Communication role was the only independent variable shown,

without qualification, to be a determinant of subjects' communication

patterns. The F ratio for the role variable was significant at p. (.001,

and the pattern of greater mean receiver communication ratio was con-

sistent for all orientation-compatibility groups. (See Table 5 in

Appendix G.)

This finding was interpreted as stressing the importance of situ-

ational factors for dyadic communication, i.e., aspects of the interper-

sonal relationship which are imposed on partners. In the experimental

procedures, roles were not directly assigned to subjects per se, nor

could they use "ready-made" roles. Instead, roles were suggested in-

directly by means of instructions and procedures. Sender had informa-

tion unknown to receiver, was required to transmit a message with con-

tent on every trial, and responded first on every trial. Thus, factors

external to the dyadic relationship defined three role-related aspects

of the task. Apparently, subjects were nore sensitive to the situation-

ally-suggested roles than to any other interpersonal factors present.

This corresponds to Schutz's assertion that groups develop first through

an inclusion phase in which each member finds where he "fits," i.e., he

seeks his role within the group (Schutz, 1958). To the extent that sub-

jects may have perceived the sender role as "privileged" this finding

also corresponds to the finding by Thibaut (1950) that members of a low

status group increased communication toward members of a high status group

as the privileged position of the latter group became more clearly defined.












Compatibility and Communication

The finding that incompatible dyads tended to utilize feedback

circuits more than compatible dyads was unexpected. Both the F ratio

for compatibility and a correlation between interchange compatibility

and communication ratio approached, but did not reach, significance.

Two alternative interpretations may be offered to account for

the less frequent use of feedback circuits by compatible dyads: (1)

that subjects in compatible dyads could not formulate effective role-

rclationships for the sender-receiver dimension because both attempted

to relate via the same role; and (2) that subjects in compatible dyads

circumvented the roles in order to relate on an individual-to-individual

rather than role-to-role basis, i.e., their compatibility encouraged

greater freedom of response to the total situation.

Supporting the first alternative was the finding that dyads

labeled "compatible" were actually less compatible than dyads labeled

"incompatible" according to originator compatibility scores. The dif-

ference in mean originator compatibility scores was significant at

p. < .01. According to Schutz (1958), originator compatibility refers

to the preference of a subject to initiate or to receive the behaviors

related to a given need area. It is expressed by the formula:

oK = (ei wi) + (ej wu), where e and w refer to expressed and wanted

needs for a need area, and i and j to the partners of the dyad. Posi-

tive scores signify preference for initiating, and negative scores

preference for receiving. Compatibility is indicated by complementary

scores, i.e., by one subject's preference for initiating complemented











by the other's preference for receiving. Since the sign is retained,

complementary scores sum to zero, indicating greatest compatibility.

High positive scores indicate competitive incompatibility (e.g., both

wanting to initiate control), and high negative scores indicate apathetic

incompatibility (e.g., both wanting to be controlled). (Schutz, 1553.)

Of the 16 compatible dyads, ten had originator compatibility

scores above the mean for all dyads. Of the 16 incompatible dyads, only

five had originator compatibility scores above the same mean. Thus it

appears likely that in compatible dyads more than in incompatible dyads

partners experienced difficulty in adopting the suggested communication

roles, and that this role confusion interfered with sensitivity to the

feedback contingency.

The second alternative to explain differences between compati-

bility groups in use of feedback circuits offers a different line of

reasoning. It assumes that subjects in compatible diads were, in fact,

more compatible, and that they rejected a role-oriented relationship in

favor of relating to each other and to the situation as individuals.

That is, compatibility engendered security of relationship, which in

turn encouraged personal freedom of response and made the restriction

of roles undesirable to participants. Conversely, in incompatible dyads

participants were less sure of the relationship and restricted themselves

to role-relating.

Since the communication roles were suggested by the experimenter,

i.e., external to the dyadic relationship, it is possible that compatible

partners cooperated in rejecting them. Schutz recognizes this possibility,











". .. if members of the group do not want to do their assigned task,

compatibility will contribute to their efficiency in not doing it (for

example, a compatible group should be more capable of mutiny)." (Schutz,

1958, p. 115.) The second alternative was supported by the finding that

compatible subjects produced significantly fewer congruent responses,

indicating less agreement, and significantly more redundant responses,

indicating greater disagreement. The probability values ranged from .02

to .001 for differences in mean number of responses by communication

role. Further, PO-compatible subjects as receiver produced fever con-

gruent responses than CO-compatible subjects (p < .01). These find-

ings suggested more sensitivity toward personal factors than situational

factors by partners in compatible dyads, especially partners oriented

toward inclusive and affective relationships. The last finding draws

attention to the PO-compatible group. These subjects, as contrasted to

all others, were less prone to agree with one another and tended to dis-

agree more with one another. This suggested that for them interpersonal

orientation may have contributed to responses to the task even though

orientation as a variable was nonsignificant.


Conformity Behavior and Communication

Results of the present study showed a positive relationship

between conformity behavior, in the sense of partner agreement, and use

of the feedback contingency. A correlation of +.66 was found between

subject communication radios and the frequency of congruent responses

per subject. Incompatible dyads were higher than compatible dyads in

both mean communication ratio and mean frequency of the congruent












response. The issue may be raised: Under conditions of sender-receiver

differences in the content of messages, does subsequent similarity in

messages represent conformity behavior or effective use of feedback cir-

cuits? The issue cannot be resolved from results of the present study.

The procedures were designed so that pressure to conform was minimal,

but it can be assumed that subjects had internal pressures to conform.

(Indeed, in a recent study investigating agreement and disagreement in

dyads, responses similar to the communication responses of this study

were considered indicative of various types of conformity behavior)

(Willis, 1964).

One finding did support the assumption that the communication

ratio was reflective of comsrunicative behavior. A correlation of -.30

(p < .05) was obtained between subjects' number of non-selectionsas re-

ceiver and subject communication ratio. Thus, subjects who were more

sensitive to the feedback contingency also tended to communicate more

content messages to the partner.


Verbal Conditioning and Communication

Inferences about how interpersonal factors influence communica-

tion were made from verbal conditioning studies, as reviewed in Chapter

I. These inferences were given little support by results of the present

study, and were contradicted for the compatibility variable. Incompat-

ible dyads in this study tended to use feedback circuits more than com-

patible dyads, whereas in Sapolsky's (1960) study the reverse was true.

These discrepancies may be attributed, in part, to differences

between this and Sapolsky's study in roles suggested for subjects, to-












gather with differences in the locus of control of the communication

process. In the present study subjects shared control over the com-

munication process and role was defined primarily in terms of knowl-

edge of content, i.e., by sender's possession of information about the

person in the photograph. In the verbal conditioning studies, subjects

were in the sender role, i.e., they transmitted me ssages to the experi-

menter. As senders, however, they did not have effective control over

the communication process, as the experimenter controlled it with his

knowledge of the reinforcement contingency. To an extent, the condi-

tioning situation represents a discrepancy in communication roles from

the natural pattern in which a sender shares control of the comunica-

tion process with a receiver.

Assuming that subjects in incompatible dyads were more control-

conscious and role-conscious than subjects in compatible dyads, one may

speculate that they rejected the role of sender-without-control in the

verbal conditioning studies but found the communication roles of the

present study acceptable. Since there are no data to support this ex-

planation, it must remain speculative.


Interpersonal Influences on Dyadic Communication

The primary goal of the present study was contribution to a clearer

understanding of influences which interpersonal factors have on cosruni-

cation. As defined by procedures used, interpersonal orientation was not

significantly related to use of feedback circuits. Language similarity

of dyadic partners was also non-significant in relation to use of feed-

back circuits. However, it is possible that a more heterogeneous popula-












tion than one composed of high school juniors and seniors from a rela-

tively small school would provide a greater range of orientation scores

from which to select subjects for experimental study. It is also poss-

ible that procedures used rejected subjects for study whose interper-

sonal orientations might be highly related to communication effective-
c
ness. By selecting subjects who had high scores on either I + A or e,

the study omitted those who made low scores in all interpersonal need

areas. Assuming the interpersonal orientation of these individuals to

be avoidance of any kind of interpersonal relationships, it night be

this orientation which most influences communication effectiveness.

This appears a problem worthy of future research.

The most distinctive finding of the study was that externally

suggested or imposed roles significantly altered communication patterns.

Subjects as receiver used feedback circuits far more than they did as

sender. Apparently this was a function of sender's having more informa-

tion relating to the task of describing the person in the photograph.

The question of whether receiver was conforming or using effective

communication techniques cannot be answered by results of this study. The

general question of whether changes in an individual's messages represent

conformity behavior or effective communication is perhaps best answered

by examining the general nature of the relationship in which these changes

occur. It may be that as relationships become oriented toward dominance-

submission and/or inclusion-exclusion, the changes represent conformity,

and that as relationships become oriented toward affective and/or cogni-

tive understanding, the changes represent effective communication. For












example, a therapist who frequently makes use of feedback circuits to

develop with a client a mutual understanding of the client's messages

may be thought of as communicating effectively. On the other hand, the

junior executive who invariably changes his statements in accord with

the dominocring boss may be thought of as conforming. This general

question is open to future study.

A second significant finding was the tendency toward a negative

relationship between use of feedback circuits and interpersonal compati-

bility. Two interpretations of this finding wore offerrcd above. The

first was that so-called incompatible dyads were actually more compatible

on a compatibility dimension significant for dyadic communication, i.e.,

originator compatibility. If this were true, the apparent negative re-

sults relating compatibility to communication would be reversed, and the

finding would then be in agreement with those of Sapolsky (1960). How-

ever, this appears unlikely in light of Sapoloky's statement that his in-

compatible dyads were incompatible for all need areas.

The second interpretation explained the incompatible dyads' ten-

dency toward greater use of feedback circuits in terms of role acceptance.

If incompatible dyads accepted the communication roles and compatible

dyads did not, differences in use of feedback would occur. It was argued

that incompatibility of relationship increased the need for relating via

roles, and that compatibility encouraged a feeling of freedom from the

external situation and a sensitivity toward the individual.

Role acceptance was also inferred as a critical factor in explain-

ing the discrepancy between verbal conditioning measures and












dyadic communication neasurcs as influenced by interpersonal compati-

bility. It was assumed that incompatibility increased role-sensitivity

and also sensitivity to the control dimension of communication. It was

then suggested that subjects in incompatible relationships, without means

of effective control over the communication process as in verbal condi-

tioning, might have withdraw from a communication role, but that given

control, as in dyadic communication, might have accepted the same role.

Since the argument depends on a series of inferences, it must be con-

sidered highly tentative.

T;o general conclusions may be stated regarding the effects of

interpersonal variables on communication: (1) externally suggested or

imposed communication roles are significant determinants of comunica-

tion patterns, and (2) acceptance of those roles by participants con-

tributes to communication effectiveness.

One major limitation uust be noted regarding generalization of

the above conclusions. In the present study, effective communication

referred simply to exchanges of messages, i.e., it asked the general

question, "Did a subject acknowledge that he heard the other's mescage?"

Thus, the study disregarded what may be called "exchanges of understand-

ing." The latter refers to use of the feedback contingency to acknowl-

edge not only that a message has been received, but that the meaning

which it has for the sender has also been received and understood. For

this reason, the above conclusions that communication is altered by the

presence of roles, and is facilitated by role-acceptance on the part of

participants, do not necessarily apply when communication is defined as

a total understanding of another's messages. In fact, if one considers











only the restrictive characteristics of roles, they nay be considered

to inhibit communication in a broader sense of exchanges of understand-

ing.

An issue similar to the above distinction between types of ex-

changes was explored by Rogers (1951). In attempting to formulate the

counselor's role, Rogers first thought of it as clarifying and objecti-

fying the client's feelings, i.e., to acknowledge that his feelings were

messages which could be received and codified. He later found this

formulation too intellectualistic and prone to promote a declarative

rather than an empathic attitude on the part of the counselor. He

states:

But when the counselor statement is declarative, it
becomes an evaluation, a judgment made by the counselor,
who is now telling the client what his feelings are.
The process is centered in the counselor, and the feel-
ing of the client would tend to be, "I am being diagnosed."
In order to avoid this latter type of handling, we have
tended to give up the description of the counselor's role
as being that of clarifying the client's attitudes.
[Rogers, 1951, p. 28.]

The effects that interpersonal variables have on an "exchange of under-

standing" between individuals would appear a topic worthy of separate

investigation.


Summary

The present study investigated the general proposition that ef-

fective communication between two people is related to dimensions of

their interpersonal relationship and to the similarity of their language

systems. Communication effectiveness was defined as use of feedback











circuits in a face-to-face situation. Hypotheses predicted that: (1)

subjects oriented toward controlling others would comunicate more ef-

fectively than subjects oriented toward inclusive and affective inter-

personal relationships; (2) pairs of subjects compatible to one another

would communicate more effectively than pairs of subjects incompatible

to one another; (3) control-oriented subjects would communicate more ef-

fectively in a sender role than as receivers; and (4) language similar-

ity of communication pairs would have a positive relationship to their

communication effectiveness in a face-to-face situation.

The FIRO-D was used to select 64 subjects from junior and senior

high school classes for placement into either a personal-oriented (PO)

or a control-oriented (CO) group. PO subjects were selected for an

interpersonal orientation toward inclusive and affective relationships

and CO subjects for an interpersonal orientation toward controlling re-

lationships. Subjects were paired to form 16 compatible and 16 incom-

patible dyads. Compatible dyads consisted of eight PO-PO combinations

and eight CO-CO combinations; incompatible dyads consisted of 16 PO-CO

combinations. Each subject completed a passage written by his partner

which the experimenter had prepared by the Cloce Procedure. Dyad Cloze

scores provided a measure of dyadic language similarity. Dyads were

tested separately for communication effectiveness by means of a tack

which permitted, but did not require them to respond to each other's

messages as they described a photograph. Four communication responses

were defined by comparing a subject's message to the previous mscoage

of his partner and his own previous message. Analysis of variance was












performed on a communication ratio calculated for each subject. This

was determined by dividing the confirming response by the redundant

response.

Interpersonal orientation and language similarity were not sig-

nificantly related to communication effectiveness. There was a tendency

for subjects in incompatible dyads to be more effective communicators

than those in compatible dyads; the F ratio for compatibility approached

but did not reach statistical significance. All subjects tended to com-

municate more effectively in the role of receiver than that of sender.

Thus, none of the hypotheses woo supported and two were at least par-

tially contradicted.

Analysis of the results revealed several additional findings

beyond those used to test the hypotheses. There was a low and positive

relationship between tendency to transmit statements (under conditions

of choice) and the communication ratio. A relatively high and positive

relationship was found between partners' tendency to agree and their com-

munication ratios. Compatible dyads demonstrated significantly less

agreement in describing the photograph than incompatible dyads. PO

subjects in compatible dyads were less prone to agree than CO subjects

in compatible dyads.

Discussion of results focused on the importance of externally sug-

gested or imposed communication roles for dyadic communication, and on

the significance of acceptance or non-acceptance of the roles by parti-

cipants. It 13w concluded that communication roles contributed signifi-






51






cantly to communication patterns and that role acceptance by 3artnoas

increased communication effectiveness ihfien communication effectiveness

was defined as the use of feedback circuits.











BIBLIOGRAPHY


Berg, I. A. Word choice in the interview and personal adjustment. J.
consult. Psychol., 1958, 5, 130-135.

Blum, G. S. The Blacky pictures: A technique for the exploration of
personality dyanics (manual). New York: The Psychological
Corporation, 1950.

Cairns, R. B. & Lewic, M. Dependency and the reinforcement value of
a verbal stimulus. J. consult. Psychol., 1962, 26, 1-8.

Crowne, D. P. & Strickland, B. R. The conditioning of verbal behavior
as a function of need for social approval. J. abnorn. soc.
Psychol., 1961, 63, 395-401.

Edwards, A. L. Experimental design in psychological research. New
York: Rinehart, 1950.

Feldstein, S. & Jaffe, J. Language predictability as a function of
psychotherapeutic interaction. J. consult. Psychol., 1963,
27, 123-126.

Greenhill, M. The focal communication concept. Aner. J. Psychother.,
1958, 12, 30-41.

Leary, T. Interpersonal diagnosis of personality: A functional
theory and methodology for personality evaluation. Neu York:
Ronald Press, 1957.

Lorenz, 1M. & Cobb, S. Language patterns in psychotic and psycho-
neurotic subjects. A. M. A. Arch. Neurol Psychiat., 1954, 72,
665-673.

Marlowe, D. Need for social approval and the operant conditioning of
meaningful verbal behavior. J. consult. Psychol., 1962, 26,
79-83.

Reisc, B. F. Communication in psychotherapy. Amer. J. Psychother.,
1957, 11, 774-789.

Rogers, C. R. Client-centered therapy. Boston: Houghton Mlifflin, 1951.

Puesch, J. Disturbed communication: The clinical assessment of normal
and pathological coamunicative behavior. New York: W. W. Norton,
1957.

Therapeutic communication. New York: W. W. Norton, 1961.











Sapeolky, A. Effect of interpersonal relationships upon verbal condi-
tioning. J. abnorn. soc. Psychol., 1960, 60, 241-246.

Schutz, W. C. FIRO: A three-dinensional theory of interpersonal be-
havior. New York: Rinehart, 1953.

Taylor, U. L. "Cloze procedure": A new tool for measuring readability.
Journalism Quart., 1953, 30, 415-433.

Thibaut, J. An experimental study of the cohesiveness of underprivileged
groups. HEz.. Relat., 1950, 3, 251-278.

Timosnc, E. 0. & Iloblin, C. D. The differential performance of orals
and anals in a verbal conditioning paradigm. J. consult. Psychol.,
1963, 27, 333-385.

Ueakland, J. F. Letters of cother-s of schizophrenics. Ance. J. Ortho-
psychiat., 1962, 32, 604-623.

Uicner, N. Cybernetics or control and cocrunication in the animal and
the machine. Net~ York: Wiley & Sons, 1948.

.__ The human use of human beings. Botton: Houghton Hifflin,
1950.

Willis, R. H. The phanomenology of agreement and disagreement in dyads.
Tech. Rept., Nonr 816(12). St. Louis: ITashington University,
1964.




































APPENDICES










APPENDIX A


INSTRUCTIONS FOR THE EXPERIMENTAL TASK


"First, I will tell you the purpose of the research and then what
your specific job will be during the experiment.

"In this experiment we are getting information on how well people
can judge a person whom they do not know by looking at his photograph
and by reading some factual information about him. We are interested in
the practice of businesses, government agencies, and colleges of request-
ing photographs and personal history data from applicants along with
their application papers. We would like to find out what kind of im-
pressions people get from photographs and from the personal histories
which are written. We will look at two photographs during the experi-
ment.

"I will put up a photograph for you to see. Look at it closely,
especially for clues as to what kind of a person this person might be.
For the purposes of the experiment, one of you will be Subject A and the
other will be Subject B. The cards on the table tell you which one you
are.

"For the first photograph, I will give Subject A a sheet of paper
which tells something about the person shown in the photograph. Only
Subject A may see this sheet or know the information for the first photo-
graph. Therefore, I am asking that subjects refrain from making random
comments or from asking questions once the experiment has begun. When
we look at the second photograph, Subject B will have the information
sheet for that picture.

"Now I will tell you the method for judging or describing the
person in the photograph. To do this, use the stack of 16 cards which
is on the table in front of you. Subject A has List A and Subject B
has List B. Both lists have cards numbered 1-16, but otherwise they are
different.

"Each of these cards has four short statements which could de-
scribe someone. Only one statement can be chosen from each card. You
are to look at these cards, one at a time, and tell me which of these
statements you think best fits the person in the photograph. We will
alternate, Subject A choosing first, then Subject B.

"Now listen carefully, because your tasks are somewhat different.
Subject A will look at each card in turn and decide which of the state-
ments best fits the person in the photograph. Subject B will also look
at each of his cards in turn to see which statement best fits the per-
son in the photograph, but should he decide that none of the statements
fits at all, he should say nothing. In other words, each time A looks












at a particular card, he will select one statement from his card, and
after he has chosen, Subject B will be able to turn to his card of the
came number and either choose one statement from his card or will re-
main silent. In case B does not choose one, I will direct A to the
next card after a few seconds, indicating the beginning of another
trial. Remember, B is the only one that has a chance of not saying
anything on a particular card, as A selects one every time.

"Let me summarize the procedure for you. First I will put up
the photograph. Then Subject A will get the information sheet and will
have a minute to read it. He may refer to this at any time, but cannot
reveal its contents to Subject B. Then I will ask A to turn to card fl,
and he can begin right away by choosing a statement from his card.
After he has done this, B can turn to his card, and should either select
a statement or not say anything, indicating that he thinks that none of
the statements fits at all. We will use the same procedure for all 16
cards.

"Are there any questions about the procedure?

"You will have sufficient time to look at all four statements and
to make your selection, but it is best to decide as quickly as you can,
on a first-impression basis. Do not be concerned with whether you think
your answers are right or wrong, as it is your impressions that we want
you to give. Don't be concerned with duplications or contradictions.

"I will call out the cards for each of you one at a time. Please
do not turn to any card until I have told you to. For example, when I
say 'A 1' only the person who has List A should turn to the first card."

(Another query about questions.)

(Place photograph, give A time to read information sheet.)

(16 trials for first photograph)

"We are still in the middle of the experiment, so I will ask you
not to make any comments while I put up the second photograph. We will
use the same procedure as for the first one, except that B will now have
the sheet of information for the person in the photograph, and will
select a description from every card. And A will be able to select one
for each card or remain silent if he thinks that none of the descriptions
fits at all. In other words, you will switch tasks for the second photo-
graph. I want you also to change lists, so that Subject B has List A,
and vice versa. You may do this now, but please do not turn to any cards
until I tell you to. [Place photograph, give B time to read information
sheet] I am going to call out the card numbers as I did last time. As
before, we will start with the person who has List A. Card A-I."

[16 trials for second photograph. Thank Ss for participation and caution
them not to discuss experiment.]












APPENDIX B

INFORMATION SHEET FOR PHOTOGRAPH #I


Name: Joe M

Age: 26


Mr. M was born in a large southern city and went to high school

there. He was active in sports, especially baseball. He worked part-

time during high school for his uncle who owned a small manufacturing

company. After high school he joined the Army and served in the Quarter-

master Corps for three years. Mr. M planned to enroll in the

University of Georgia immediately after his discharge from the Army,

but his father died suddenly and he had to help support the family for

a year. He had several jobs, most of them working on home construction

crews.


Using the G.I. Bill, Mr. M enrolled at the University of Georgia

and obtained his degree in 1950. During college, he went out for the

baseball team in his freshman year but gave up baseball altogether after

he married one of his classmates. One of his main hobbies now is fish-

ing, and he and his wife are active in a square dancing club. Mr. M

was undecided about what job to take after college and asked for help

from a Veteran's Administration vocational counselor. After several

interviews,he decided to return to his home city to work for the same

uncle he worked for in high school, as the company was now large enough

for him to have a supervisory position.










APPENDIX B, Continued

INFOP IATION SHEET FOR PHOTOGRAPH #2


Name: Bryan W

Age: 25


Mr. W was born in a large southeastern city, but his family moved

to the West when he was six years old. His father was in the Navy, and

the family moved three times while Mr. W was in high school. He

was active in the school band in each school. His main hobbies were

reading, music, and water sports. After high school graduation, HIr.W

worked for almost a year as clerk in a local bookstore to save money for

college. At this time the family moved back to the Southeast again, and

Mr. enrolled in the University of Georgia. He attended college

for only one year because of financial difficulties. He joined the Amy

to fulfill his military obligations, and served in the Medical Corps.


After his discharge from the Army, Ilr. sought help from a Veteran's

Administration vocational counselor. When he found that he had good poten-

tial for finishing college, he returned to the University of Georgia im-

mediately. Two and a half years later, in 1950, he obtained his degree.

Soon after graduation he married a girl whom he had met during his fresh-

man year at the university. He and his wife water ski a great deal, and

he enjoys a nea-found hobby of coin collecting.


lHe was undecided about what to do after college and went back to the vo-

cational counselor. There he made tentative plans to join the training

program of a nearby industrial firm.












APPENDIX C

SAMPLE SCORING FOR EXPERIMENTAL TASK


List A
1.
well thought of
can be strict if necessary
able to criticize self
cooperative

2.
self respecting
can complain if necessary
grateful
considerate

3.
able to give orders
can be frank and honest
can be obedient
friendly

4.
able to take care of self
able to doubt others
appreciative
helpful

5.
often admired
hard-boiled when necessary
easily embarrassed
always pleasant and agreeable

6.
self-confident
resents being bossed
often helped by others
kind and reassuring

7.
good leader
irritable
easily led
sociable and neighborly


1 2 12
AB BA


I I I


List B
1.
makes a good impression
firm but just
apologetic
eager to get along with others

2.
independent
often gloomy
admires and imitates others
encourages others

3.
forceful
critical of others
usually gives in
affectionate and understanding

4.
can be indifferent to others
frequently disappointed
very anxious to be approved of
big-hearted and unselfish

5.
respected by others
stern but fair
lacks self-confidence
wants everyone to like him

6.
self-reliant and assertive
skeptical
very respectful to authority
tender and soft-hearted

7.
likes responsibility
straightforward and direct
modest
uarm











APPENDIX C, Continued

SAMPLE SCORING FOR EXPERIMENTAL TASK


8.
businesslike
hard to impress
accepts advice readily
enjoys taking care of others

9.
always giving advice
impatient with other's mistake
self-punishing
too easily influenced by friend

10.
boastful
bitter
dependent
forgives anything

11.
bossy
outspoken
passive and unaggressive
fond of everyone

12.
thinks only of himself
jealous
lets others make decisions
generous to a fault

13.
tries to be too successful
sarcastic
timid
wants everyone's love


14.
somewhat snobbish
resentful
hardly ever talks back
too lenient with others


1 2 12
AB BA
S.
likes to compete with others
touchy and easily hurt
trusting and eager to please
gives freely of self

9.
acts important
self-seeking
shy
will confide in anyone

10.
proud and self-satisfied
complaining
wants to be led
oversympathetic

11.
dominating
often unfriendly
meek
likes everybody

12.
shrewd and calculating
slow to forgive a wrong
easily fooled
overprotective of others

13.
expects everyone to like him
cruel and unkind
always ashamed of self
agrees with everyone

14.
rebels against everything
egotistical and conceited
clinging vine
tries to comfort everyone













APPENDIX C, Continued

SAMPLE SCORING FOR EPERIMENTAL TASK


1 2 12
AB BA


15.
manages others
frequently angry
obeys too willingly
friendly all the time

16.
selfish
stubborn
likes to be taken care of
too willing to give in to
others


15.
dictatorial
hard-hearted
spineless
loves everyone

16.
cold and unfeeling
distrusts everybody
will believe anyone
spoils people with kindness












APPENDIX D

TABLE 2

IIEAN FIRO-B INCLUSION PLUS AFFECTION (I + A) SCORES AND EXPRESSED

CONTROL (ec) SCORES FOR PERSONAL ORIENTED (PO) AND CONTROL ORIENTED

(CO) GROUPS, BY SEX, AND TOTAL FOR BOTH GROUPS


I + A*


ec**


M

25.88

12.81


27.50

18.06


26.69

15.44


0-

4.77

6.29


3.54

4.50


4.11

6.06


CO differences

PO differences


2.44

5.69


1.00

5.25


1.72

5.47


significant at p. < .001

significant at p. < .001


PO, Males

CO, Males


PO, Females

CO, Females


PO, Total

CO, Total



* All mean

** All mean


O"

1.62

.90


1.17

1.48


1.58

1.24


PO minus

CO ninus












APPENDIX E

TABLE 3


MEAN FIRO-B INTERCHANGE, ORIGINATOR, RECIPROCAL AND TOTAL COMPATI-

BILITY SCORES FOR COMPATIBLE AND INCOMPATIBLE DYADS*




Compatible Dyads Incompatible Dvads

H C3 M O t

Interchange
Compatibility 8.88 4.57 18.56 7.72 5.98g**

Originator
Compatibility 9.96 6.42 5.31 4.20 3.13****

Reciprocal
Compatibility 14.75 5.42 19.94 5.19 4.12***

Total
Compatibility** 33.31 10.88 43.81 13.66 3.29****


*Greater FIRO-B compatibility score = less compatibility
**Total compatibility score = sum of interchange, originator,
and reciprocal scores
***Mean difference significant at p < .001
****lean difference significant at p < .01












APPENDIX F

TABLE 4

FREQUENCY OF COMMUNICATION RESPONSES BY ORIENTATION-COMPATIBILITY

GROUP AND COMMUNICATION ROLE*




Communication Role


Sender


Receiver**


CF R C N CF R C N


PO-
compatible 26 97 70 70 55 42 73 68

PO-
incompatible 28 72 93 47 43 22 99 55

CO-
compatible 28 84 74 54 38 39 100 46

CO-
incompatible 27 69 97 47 32 31 113 49

Total 109 322 334 203 154 134 385 218



*CF = confirming response; R = redundant response; C = congruent
response; N = novel response; PO = personal oriented group;
CO = control oriented group.

**Receiver confirming responses (CF) adjusted by the formula:
Sender confirming response (SCR) = Receiver confirming re-
sponse (RCR) times two-thirds (2/3).













APPENDIX G

TABLE 5

M1EA C(GtMUNICATION RATIOS BY ORIENTATION-COMPATIBILITY

GROUP AND Cc0tBIUNICATION ROLE*



Sender Recciver

it M ___

PO-
compatible .58 .22 1.28 .66

PO-
incompatible .83 .61 1.56 .75

CO-
compatible .66 .26 1.20 .50

CO-
incompatible .68 .26 1.37 .80


*Cormunication ratio = confirming response/oedundant response; PO =
personal oriented group; CO = control oriented group.











APPENDIX H

TABLE 6

COXmIUNICATION RESPONSES BY PHOTOGRAPH, ORIENTATION-COMPATIBILITY GROUP,

AND CMMUNICATION ROLE*




Photograph Number 1

Sender Receiver**
CF R C N CF R C N

PO-compatible 11 46 32 31 19 20 36 34

PO-incompatible 13 44 45 18 17 15 44 35

CO-compatible 18 40 40 22 24 13 55 15

CO-incompatible 18 37 41 24 10 20 63 22

Total 60 167 158 95 70 68 198 106


Photograph Number 2

CF R C N

PO-compatible 12 51 33 24

PO-incompatible 15 28 48 29

GO-compatible 10 45 33 32

CO-incompatible 8 29 59 24

Total 45 153 173 109


CF R C N

21 22 33 33

25 7 55 20

14 24 45 30

21 11 50 27

81 64 183 110


*CF = confirming response; R = redundant response; C = congruent
response; N = novel response; PO = personal oriented group;
CO = control oriented group.
**Receiver confirming responses (CF) adjusted by the formula; Sender
confirming response (SCR) = Receiver confirming response (RCR) times
two-thirds (2/3).











APPENDIX I

TABLE 7

GROUP MEAtI CCMBUNICATION RATIOS BY PHOTOGRAPH, ORIENTATION-COMIPATIBILITY

GROUP, AND CCMUNICATION ROLE*




Photograph Number 1 Photograph Number 2

Sender Receiver Sender Receiver


PO-compatible .24 .95 .24 .95

PO-incompatible .30 1.33 .54 3.57

CO-conpatible .45 1.85 .22 .58

CO-incompatible .49 .50 .28 1.91


*Communicition ratio = confirming response/redundant response; PO = personal
oriented group; CO = control oriented group.












DIOGRAIICAL SKrTCI'


Donald M. Hartsough was born November 27, 1934, at Seville, Ohio.

He was graduated from Cuyahoga Palls High School in June, 1951. He at-

tended the College of Wooster, whore he served as Student Body President,

and in June, 1955, received the degree of Bachelor of Arts with a major

in economics. The following year from February until June he attended

the University of California. In 1956 he enrolled in the Graduate School

of the University of Florida. lHe worked as a graduate assistant in the

Department of Psychology from February, 1957, until August, 1958, when

he received the degree of Master of Arts. During the academic years

1958-1959 and 1960-1961, he received a training grant from the Bureau

of Vocational Rehabilitation in order to pursue work toward the degree

of Doctor of Philosophy in clinical psychology. From September, 1959,

until August, 1960, he was a research assistant in the College of Health

Related Services of the University of Florida. His clinical internship

was served at the J. Hillis Miller Health Center the following year. In

August, 1961, he joined the Alachua County Health Department at Gaines-

ville, Florida, as staff psychologist. From August, 1962, until the

present, he has held the combined appointment of psychologist at the

P. K. Yonge Laboratory School and instructor in the College of Education

of the University of Florida.

Donald M. Hartsough is married to the former Dalyte Rose Ellis and

is the father of three children. lie is a member of the Florida Psycho-

logical Association and the American Association of University Professors.

lie is an associate member of the American Psychological Association.











This dissertation uas 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.

April 18, 1964
Dean, College of Arts an/ SCiences


Dean, Graduate School


SUPERVISORY COMMITTEE:


Chaiiman


(~ (2~ 6~(2

PAGE 1

EFFECTS OF INTERPERSONAL ORIENTATION AND LANGUAGE SIMILARITY ON VERBAL COMMUNICATION IN DYADIC INTERPERSONAL RELATIONSHIPS By DONALD M. HARTSOUGH 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 April, 1964

PAGE 2

TO MY WIFE

PAGE 3

ACKNOWLEDGMENTS Special gratitude is extended to Dr. Marvin E. Shaw, chairman of the author's doctoral committee, for the encouragement, forebearance, and keen insight with which he directed the study. Acknowledged also are the helpful inquiries and contributions of the other committee members, Dr. Albert M. Barrett, Dr. Sidney M. Jourard, Dr. Audrey S. Schumacher, Dr. Rolland H. Waters, and Dr. Wilse B. Webb. The study represents not only the research project as supervised by Dr. Shaw, but a result of hours of valuable and stimulating discussion of interpersonal behavior in both academic and clinical settings, particularly with Drs. Jourard and Schumacher. Recognition should be given the timely assistance of Dr. Clifford C. Courson, Dr. Herbert D. Kimmel, and Mr. Richard A. Mellob. Finally, the author acknowledges his debt to the staff of the P. K. Yonge Laboratory School for its support and cooperation, and to the students who served willingly as subjects for the study. iii

PAGE 4

TABLE OF CONTENTS ACKNOWLEDGMENTS ii:L LIST OF TABLES v CHAPTER I INTRODUCTION Reviev; of the Literature Statement of the Hypotheses 12 II METHOD AND DESIGN 15 Subjects " Experimental Groups Experimental Dyads 17 Measurement of Language Similarity Experimental Task Communication Statements 21 Communication Responses Summary of Method and Design Ill RESULTS 30 Characteristics of Subjects Orientation, Compatibility, and Communication .... Additional Findings IV DISCUSSION AND CONCLUSIONS 33 Roles and Communication Compatibility and Communication 40 Conformity Behavior and Communication Verbal Conditioning and Communication Interpersonal Influences on Dyadic Communication . . 44 Summary BIBLIOGRAPHY APPENDICES 5A BIOGRAPHICAL SKETCH 68 iv

PAGE 5

LIST OF TABLES TABLE 1. Analysis of Variance of the Communication Ratios 33 2. Mean FIRO-B Inclusion Plus Affection (I + A) Scores and Expressed Control (e c ) Scores for Personal Oriented (PO) and Control Oriented (CO) Groups, by Sex, and Total for Both Groups 62 3. Mean FIRO-B Interchange, Originator, Reciprocal, and Total Compatibility Scores for Compatible and Incompatible Dyads 63 4. Frequency of Communication Responses by OrientationCompatibility Group and Communication Role • 64 5. Mean Communication Ratios by Orientation-Compatibility Group and Communication Role 65 6. Communication Responses by Photograph, OrientationCompatibility Group, and Communication Role 66 7. Group Mean Communication Ratios by Photograph, Orientation-Compatibility Group, and Communication Role . . . . 67

PAGE 6

CHAPTER I INTRODUCTION "Communication," "interpersonal behavior" and "interpersonal relationship" are frequently attributed to events as If the terms were synonymous and referred to the same phenomena. That they are distinct from one another can be seen by reference to a systems model. An individual may be regarded as an open system, i.e., a relatively stable entity which has the capacity to transmit and receive information across its boundaries. Communication refers to transmission— reception exchanges with another system. Interpersonal behavior refers to the same interaction process, but also includes the content of messages. Thus, the exchange of messages between two people may be effective while the content of messages is disturbing to one or both. For example, a person may communicate rejection or anger, and while his message may be effectively communicated, the content may be deleterious to another person. On the other hand, interpersonal behavior which is satisfying to both participants may take place in spite oi temporary communication disturbances. An interpersonal relationship refers to another, broader system which includes both individual systems, as subsystems, the roles attributed to the systems, and the communication process between the individuals. It usually denotes a system which is maintained over time. Students of communication have come to regard it as an essential variable in changing interpersonal relationships. For example, a psychotherapist who wishes to help his client modify unsatisfactory relationships with others must himself establish a therapeutic relationship with

PAGE 7

the client. To do this, his interpersonal behavior must be in keeping with the needs of the client, and he must communicate -with precision, appropriateness, efficiency and flexibility (Rucsch, 1961; Greenhill, 1958). Just as communication may provide the means of modifying interpersonal relationships, there is evidence that the pattern of communication which characterizes a person may be related to his present and past relationships with significant others. Reiss (1957) found that in psychotherapy, hysteric patients tended to use past progressive and active tenses, whereas obsesslve-eompulsives more frequently used evaluative words, comparative adverbs and polar-opposites. (It is assumed here that personality type is determined in part by relationships with significant others.) Lorenz and Cobb (1954) found the communication of neurotics, as contrasted to normals, characterized by more verbs and pronouns, and fewer adjectives and prepositions. They interpreted this as reflective of the neurotic's greater concern for action and affiliation, and greater avoidance odescription and abstract relationships. Weakland (1962) analyzed letters written by mothers o. hospitalized schizophrenics; he found subtle contradictions which tended to confuse overt messages and which in some cases visibly upset the patients. Finally, Berg (1958) found patient improvement in therapy associated with a change in pronouns used by the patient. The ego words "I," "me," and "my" decreased and the empathic words "you," "we," and "us" increased with improvement. In accord with ideas and findings such as these, it seemed worthwhile to devise a study which investigated the general proposition that

PAGE 8

communication between two people is related to dimensions of their interpersonal behavior and the relationship which they create. The present study investigated effectiveness of dyadic communication in a face-toface situation as is influenced by subjects' interpersonal orientations, the compatibility of their relationship, and the similarity of their language systems. Effectiveness of communication was evaluated by analysis of feedback communication patterns between dyadic partners. Following Ruesch (1957, 1961), communication was regarded as an exchange of messages in a social context. Since the publication of Wiener's book on cybernetics in 1948, and the extensive use o: communication theory which followed, the term "communication" has been used in such a diversity of ways that it is perhaps in danger of being rendered meaningless as a concept. Communication was therefore amended by the term "interpersonal" to limit its application to the exchange of messages between people who interact in a iace-to-iace relationship. The dimension of interpersonal behavior selected for study was interpersonal orientation. Interpersonal orientation reiers to the fact that people develop expectations about the dimensions along which interpersonal relationships are formed, and thus tend to approach interactions with others in certain ways. Schutz (1953) has posited three dimensions (inclusion, control and affection) which he regards as the significant orientations available for most relationships. A dimension of interpersonal relationships was also selected for study; this was the relative compatibility or incompatibility of relationships. This refers to the different combinations oi needs and orientations found in relationships, and the degree of satisfaction offered by the combinations.

PAGE 9

The general proposition that interpersonal behavior and relationships are related to communicati n was refined to two specific problems: (1) the assertion that an interpersonal orientation toward controlling others is associated with more effective communication than an orientation toward inclusive and affective relationships, and (2) the assumption that compatibility of relationship results in more effective communication than incompatibility. It was supposed that in a task situation the control-oriented individual would communicate more effectively in order to maintain control and in keeping with his greater concerns about roles and tasks. The second problem was based on the assumption of a positive relationship between compatible psychological needs and the ability to exchange messages effectively. Communication effectiveness in a face-to-face situation was considered highly associated with use or feedback circuits, or the feedback contingency, in the communication process. Wiener describes feedback in the following terms: "In its simplest form, the feedback principle means that behavior is scanned for its result, and that the success or failure of this result modifies future behavior. " (Wiener, 1950, p. 69.) Thus, feedback provides a means by which a process is regulated or controlled. This control results when a system which is producing a process interacts with something outside of the system. Ruesch treats feedback as a structural property of human communication. He contends that the structural properties determine successful or unsuccessful communicative interactions, and states that utilization of feedback is perhaps the foremost criterion of successful communication

PAGE 10

(Ruesch, 1957). It provides a means of relaying back to the original sender the effects that his message has had on other participants. It thus enables him to modify constantly the communication process as the process itself takes place. By reference to the principle of individual differences in human behavior, it seemed reasonable to assume that individuals vary in the extent to which they utilize feedback circuits in communicating with others. Part of this variation was expected to be accounted for by interpersonal orientation and interpersonal compatibility. A third variable was included in the study to control for noninterpersonal factors in feedback communication. Some measure of language similarity seemed necessary in light of the fact that communication in a face-to-face situation is a cognitive as well as an interpersonal event. If two individuals literally do not speak the same language, i.e., have different language systems, there can be little or no feedback communication. Hence a procedure was employed to assess the similarity of language systems of dyadic communication partners. Review of the Literature The research data of five verbal conditioning experiments supported the possibility that interpersonal orientation and dyadic (two-person) compatibility influence a subject's response to a communication interchange. These studies investigated the relationship between personality variables and the response to verbal conditioning. Sapolsky (I960) found that subjects who were given a set to find the experimenter an attractive person had a different response rate from

PAGE 11

other subjects who were given a set to find the experimenter an unattractive person. During acquisition trials the High Attractive (IIA) subjects established a significant level of conditioning, while the Low Attractive (LA) subjects did not. During extinction trials, in which the experimenter was absent from the room, the IIA subjects decreased their rate only slightly, but the response rate of the LA. subjects increased significantly. These results led Sapolsky to a second experiment which tested the hypothesis that interpersonal compatibility between subject and experimenter would influence verbal conditioning. Five experiments were selected from the same population as the subjects on the basis of scores on the FIRO-B (Fundamental Interpersonal Relations Orientation— Behavior, Schutz, 1958). All were high in the control dimension. Each experimenter verbally conditioned six subjects, three of whom were compatible and three of whom were incompatible with him, according to the FIRO-B. Subjects made up sentences using one of six pronouns, two of which were reinforced by the experimenter's "Mm-hm". Extinction trials began when the experimenter left the room and the subjects were asked to continue making up sentences into a tape recorder. Results showed that during acquisition trials the Compatible subjects increased and the Incompatible subjects decreased in use of the reinforced xrords. During the extinction phase, the Incompatible subjects increased significantly in response rate to a level comparable to that of the Compatible subjects. A communication theory interpretation of these results would indicate that the dyads comprised of Compatible subjects established and

PAGE 12

maintained feedback circuits during the acquisition phase but that dyads comprised of Incompatible subjects either did not establish feedback circuits or actively avoided the feedback contingency. Because the output of one participant (E) was fixed a priori by the output of the other (S), an extrapolation of the verbal conditioning paradigm to the communication paradigm was a rather obvious step. The subject's emitted response was considered as a sender's message or statement and the experimenter's reinforcing stimulus was considered as the receiver's acknowledgement. In this context the subject's readiness to participate in feedback communication is roughly comparable to his conditioned response rate in that it was evidence that the receiver's acknowledgements had an effect on his subsequent messages to the receiver . While the extrapolation from verbal conditioning to the communication paradigm appeared obvious, it was necessary to recall that in the conditioning situation it was the receiver (E) who maintained control over the communication process by virtue of his knowledge of the reinforcement contingency. In other communication dyads, the sender maintains another kind of control by virtue of possessing more content information than the receiver (e.g., teacher and student). Although Sapolsky's study gave no indication of the role of personality variables as such on communication, it indicated that dyadic interpersonal compatibility influenced the extent to which feedback circuits were used during the communication process. That the crucial variable was an interpersonal one was further supported by Sapolsky's finding that on a sociometric scale completed by both subject and experimenter, 19 of 30 pairs rated each other as predicted from their FIRO-B compatibility scores. This result reached statistical significance.

PAGE 13

Cairns and Lewis (1962) investigated the relationship between dependency and response to verbal conditioning. The Edwards Personal Preference Scale was used to select High Dependent (HD) and Low Dependent (LD) groups. The task consisted of making up sentences using one of three verbs supplied by the experimenter. The HD and LD groups were further divided; one-half of each was reinforced for using verbs with strong aggressive connotations, and the other for using verbs with high dependency connotations. In line with the author's expectations, HD subjects conditioned to a higher level than LD subjects, although conditioning for the HD subjects was strongly influenced by the type of verb reinforced. The HD dependency-reinforced subjects initiated a high rate of response before reinforcement began, and maintained this level during reinforcement. HD aggressive-reinforced subjects showed initial avoidance of the aggressive terms, but during reinforcement responded positively. For all LD subjects the type of word reinforced had no differential effect; in addition, subjects in both LD groups became more resistant to the conditioned stimulus during reinforcement than they had during the pre-reinforcement series. This result approximated the "negative" response of Sapolsky's Incompatible dyads to the reinforcing stimulus. In communication terms, the findings of Cairns and Lewis were interpreted to mean that highly dependent subjects were not only more likely to maintain feedback coimnunication than low dependent subjects, but that they were also more sensitive to the content of messages. Timmons and Noblin (1963) found a differential verbal conditioning performance between subjects identified as Oral and those identified as Anal. Orality and anality vexe determined from a group administration

PAGE 14

of the Blacky Test. Scoring was done according to the method developed by Blum (1950). Orality referred to incorporating perceptions, and anality to withholding perceptions. Subjects used one of si;: pronouns to complete a sentence supplied by the ercperimenter, and were reinforced for two of them on a 75 per cent variable ratio schedule. The reinforcing stimulus was a "mild, affirmatory utterance." The Orals showed a typical learning curve, i.e., one which ascended during reinforcement and descended during extinction. The Anals (true to character, perhaps) produced the opposite sort of learning curve, one that demonstrated a less frequent response during reinforcement than durin , either pre-reinforcement or extinction. Changes in the Orals' conditioning curve reached statistical significance, whereas the Anals* curve approached, but did not reach, the accepted significance level. Crowne and Strickland (1961) used the llarlowe-Crowne Social Desirability Scale to select subjects vjho were hi ; h in need for social approval and others who were low on the sane scale. Half of each was positively reinforced and the other half negatively reinforced during a 25ninute session in which subjects were instructed to produce words ad lib . Positive reinforcement was "Mm-hm" plus the experimenter nodding his head affirmatively; negative reinforcement was "Uh-uh" plus the experimenter's noddin^ negatively. Only plural noune were reinforced; the response rate was the ratio of plural nouns to total output for '.he session. Results supported the hypothesis that subjects with a hi h need for social approval would show significantly greater increases in response rate under conditions of positive reinforcement and significantly greater decreases under conditions of negative reinforcement than subjects to whom social

PAGE 15

i • approval was of less consequence. In fact, the low need-for-approval subjects could not be distinguished from control subjects who received no reinforcement. Marlowe (1962) extended the work of Crowne and Strickland to an interview situation but used only positive reinforcement. High and low need-for-approval subjects were reinforced by the experimenter's "mm-hm" for statements of positive self-regard (PSR) made during a 15-oinute interview in which the subject described his "personality characteristics and traits." lilien compared with control subjects who did not receive reinforcement, high-need subjects emitted more PSR responses and significantly increased their rate, while low-need subjects emitted fewer PSR responses and significantly decreased their rate in much the same way as had control subjects. A statistically significant correlation of +.42 between need for approval and PSR was obtained for experimental subjects, but for control subjects the corresponding correlation of -I-.23 was nonsignificant. Marlowe interpreted his results as reflecting a greater sensitivity and responsiveness to social reinforcers on the part of subjects with high needs for social approval. The results of these studies may be summarized as follows: 1. Sensitivity to the verbal feedback contingency (or more precisely, the tendency to maintain feedback circuits in a dyad) depended in part on at least one dimension of personality. This was variously described as high vs. low dependency, orality vs. anality, and high vs. low need for social approval. From the authors' descriptions and their means of selecting subjects, this seemed to represent the extreme

PAGE 16

II positions of a single personality dimension. In terras of interpersonal theory, this dimension approximates a dichotomy between an orientation toward social and affective needs and an orientation towards needs to express control over others. 2. While the positive responsivity of the dependent, oral and high need-for-approval subject was uniformly demonstrated, the response pattern of the low-dependent, anal and low need-for-approval subject was less clear. Two studies reported statistically significant or almost significant decrements in response during reinforcement of the latter type of subject from the pre-reinf orceent level (Cairns and Lewis, 1962; Timmons and Noblin, 1963). In two other studies (Crowne and Strickland, 1961; Marlowe, 1962), these subjects responded in mu h the same manner as the non-reinforced control groups, and would be described as nonresponsive rather than negatively responsive. Both nonresponsivity and negative responsivity nay be interpreted as this subject's attempt to maintain control of the communication and to resist becoming part of an interpersonal process over which he feels he has no control. 3. Sapolsky's findings demonstrated that differential responsivity can be produced in dyads of naive subjects as well as with experimenter and subject. However, since Sapolsky did not provide the FIRQ-B scores for his Compatible and Incompatible groups, it was not possible to determine whether or not his dyads varied according to the personality dimension described above. As stated above, the findings summarized here seemed to have implications for the study of interpersonal communication as well as for

PAGE 17

r; verbal conditioning. The communication pattern as applied to the conditioning, procedure (i.e., the pattern of statement-acknowledgeraent-feedback-statement) represents simply a translation of terns from verbal conditioning (i.e., from the procedure of stimulus-response-reinforcecientstimulus) to communication theory. This same communication pattern, however, ray be extended to more complex, natural dyadic communication, in which both participants are free to select the content of statements and acknowledgements, and in which the pattern of statements and acknowledgments becomes modified by the introduction of communication blocks and other structural properties. Further, it appeared significant that in the conditioning studies the subject was forced to enter into a rigidly controlled feedback circuit with the experimenter. The fact that it was a rigidly controlled system— and controlled by the experimenter— may have influenced the controland independence-conscious subject to withdraw from the system. It could be argued, in fact, that given the opportunity to be in a more controlling role in the system, for example as a sender with more information than the receiver, the low-dependent, highcontrol subject would produce more feedback communication than his opposite. Statement of the Hypotheses The present study investigated the effects of interpersonal orientation, compatibility, and language similarity on interpersonal communication in a face-to-face situation. Following Ruesch (1957), the primary measure of communication was regarded as the extent of use of the feedback contingency.

PAGE 18

The hypotheses were as follows: 1. Individuals highly oriented toward controlling interpersonal relationships attempt to do so by producing more feedback communication than individuals highly oriented toward social and affective aspects of interpersonal relationships. The first hypothesis was derived from a reinterpretation of the results of the verbal conditioning studies reviewed above. While these results indicated that subjects who were oriented primarily toward social and affective relationships more readily participated in feedback communication than those who were control-oriented, these results were suspected to be situationally determined by the inability of the subject to control the feedback contingency. When control of the communication process is not so rigid, the latter type of subject was expected to make greater use of feedback circuits. 2. Dyads which are interpersonally compatible make greater use of the feedback contingency than dyads which are interpersonally incompatible The second hypothesis was derived from Sapolsky's (1960) finding regarding interpersonal compatibility and conditioning. 3. Individuals who are highly control-oriented make greater use of the feedback contingency as senders than as receivers. The third hypothesis was an elaboration of the first in that the control-oriented subject was expected to be especially sensitive to the feedback contingency when in the sender role. The sender role may be assigned to the communicator with the most information about the communication process itself or about the content of communication. This hypothesis, supported by results of a pilot study, was based on the assumption

PAGE 19

r that control orientation included greater sensitivity to role and task factors, hence would produce greater communication differences according to role. ,. Similarity of the language systems of dyadic partners is positively related to the dyad's use of feedback communication in a face-to-face situation. The fourth hypothesis concerned a noninterpersonal variable in relation to interpersonal communication. Because it was reasonable to assume that a dyad's capacity for use of feedback circuits has cognitive determinants, a measure of the similarity of written language was included in the present study.

PAGE 20

CHAPTER II METHOD AND DESIGN To test the hypotheses relating communication to interpersonal orientation, compatibility, and comraunication role, pairs of subjects were selected according to FIRO-B orientation scores and assigned to dyads according to compatibility criteria; a second measure assessed the similarity of their language systems, and finally a sample of their communication in both sender and receiver roles was obtained. A 2 X 2 X 2 factorial design assigned each subject to one of four treatment conditions and each of two communication roles. Similarity of language system was evaluated by separate correlational procedures. Subjects The subjects were juniors and seniors (N=17G) from the P. K. Yonge Laboratory School of the University of Florida. Both sexes were used, but dyads were formed on a like-sex basis. Since Marloxre (1962) and Crowne and Strickland (1961) found no sex differences in their conditioning data, this variable was not expected to demand differential treatment excepting the assignment of equal numbers of both sexes to each experimental group. Additional information regarding age and intelligence was obtained from school records. The extent of acquaintance of dyadic partners was obtained from a brief questionnaire administered subsequent to all other procedures. Characteristics of subjects by experimental groups are provided in Chapter III. Subsequent analysis of the data sustained this expectation. 1..

PAGE 21

I Experimental Groups Sixty-four subjects were selected for placement into one of txm experimental groups on the basis of FIRO-B scores. The FIRO-B is a questionnaire of 54 statements developed by Schutz (1953) to measure interpersonal orientation according to three interpersonal needs— inclusion, control, and affection. These needs are further subdivided into an "expressed" and a "wanted" need, thereby recognizing the distinction between need to give and need to receive each of the three types of behavior. Subjects rate themselves on a six point rating scale according to degree of agreement or disagreement with the statements. Typical statements are: "I join social groups," "I try to take charge of things when I'm with people," and "My personal relations with people are cool and distant." Each subject is described by six scores, a wanted and expressed score for each need area. The FIRO-B was administered to both junior and senior classes during a regular class period. Half of the 64 subjects were selected for high scores in inclusion plus affection (I + A), and the other half for high scores in expressed control (ec). Generally, males (N 32) were selected for the former group if I + A was 21 or more and e c was less than five, and for the latter group if I -'. A was less than 21 and e c exceeded five. For females (N « 32), the corresponding values were I + A score of 23 and e c score of four. These criteria provided one group significantly oriented toward inclusive and affective needs and a second group significantly oriented toward control needs. For convenience, they were called the Personal Oriented (PO) and Control Oriented (CO) groups.

PAGE 22

17 Experimental Dyads Thirty-two like-sexed dyads were formed from subjects selected for the PO and CO groups. To test the hypothesis that compatibility in interpersonal relationships produces more communication than incompatibility, formation of the dyads was related to compatibility procedures of the FIRO-B. Compatibility is defined by Schutz (1958) as the property of interpersonal relationships which leads to mutual satisfaction of interpersonal needs and harmonious coexistence. Measures have been developed by Schutz (1958) for three types of compatibility: originator compatibility, reciprocal compatibility, and interchange compatibility. Interchange compatibility describes the relative similarity of two individuals according to the need areas, whereas the other two types describe their relative similarity according to eome combination of expressed and wanted scores. Interchange compatibility seemed the most appropriate measure for use with a short-term communication dyad. It is expressed in the formula: xK I (e ± + w ± ) (e.. + w j)|» where e and w refer to expressed and wanted scores for a need area. Sixteen dyads were compatible and sixteen were incompatible. The compatible dyads were made up of eight pairs of subjects from the PO group and eight pairs of subjects from the CO group. The sixteen incompatible dyads consisted of one subject from each of the PO and CO groups. Assignment was made according to a ranking of I + A and e scores within each group. Dyad assignment by this procedure accomplished counterbalancing for partner rank and initial communication role. This method also permitted the use of multi-variate analysis of variance procedures to test the first three hypotheses.

PAGE 23

:u; Measurement of Language Similarity A measure of language similarity was obtained for each dyad for correlation with dyad communication measures. This measure was used to test the fourth hypothesis, but was not used for establishing the experimental groups. Use of a written sample of language from each partner eliminated the possibility that face-to-face effects contaminated the measure. At the group administration of the FIRO-B, all 170 subjects were instructed to write a 150-200 word story or essay in response to card 17BM of the Thematic Apperception Test (TAT). To be included in the experiment, a subject's passage had to total at least 100 words. The passages were prepared by the Cloze Procedure and administered prior to the experimental test for conmunication to groups of subjects who were not partners. The Cloze Procedure was developed by Taylor (1953) as a means of measuring readability. He describes the Cloze score as appearing to measure the aggregate influences of all factors which interact to affect the degree of correspondence between the written language patterns of transmitter and receiver. This technique deletes every n th word in a passage and substitutes a blank in its place; subjects then fill in the missing word. Each of the 64 subjects completed his partner's typewritten passage after every fifth word had been deleted. Each passage contained a total of 20 blanks and the complete sentences in which the deletions appeared. Responses were scored as either correct or incorrect by comparing them with the words deleted from the original passage. The completed passage of each subject was compared to the original passage written by his

PAGE 24

!partner. The scores of dyadic partners were then summed to provide the measure of language similarity for the dyad. Experimental Task To determine the readiness of a subject to utilize feedback, it was necessary to create a communication situation in which measurement could be readily obtained without subjects becoming aware of the critical variable being measured. The situation also demanded procedural control to the extent that the communication measures did not become confounded with one another. An experimental communication task meeting these criteria was devised and successfully tested in a pilot study. This task consisted of a pseudo-problem which required each subject in the dyad to make oral responses to the experimenter and permitted (but did not require) him to enter into a communication relationship with the partner, i.e., to influence the partner's messages or be influenced by them. Thus, the degree to which partners communicated with one another or simply made oral responses to the experimenter was the dependent variable. Dyadic partners were seated opposite one another at an oblong table longitudinally divided by a board one foot high. The divider enabled them to see one another but screened material on the partner's half of the table. The experimenter, seated at one end of the table, directed the subjects' attention to a photograph mounted on the end of the divider at the opposite end of the table. The psuedo-problem consisted of a "personality description" of a person shown in a photograph. (Two anonymous photographs were selected for this purpose; both were of young adult males.) The subjects were

PAGE 25

?. told that the experimenter was gathering information about the impressions people get from photographs of people whom they do not know, and from personal history information about these same people. Instructions were given which gave the psuedo-problem a rationale and which made it seem a worthy investigation. (The complete instructions are found in Appendix A.) Before the personality description began, one of the subjects was given a "sheet of information" describing the person in the photograph. (The information sheets for both photographs are included in Appendix B.) A different sheet was devised by the experimenter for each photograph. The personal histories were roughly parallel in that both men were born in the South, went to college after experiencing difficulties, served in the armed forces, received vocational counseling, married college women, and made tentative career choices following graduation. The provision of the history information to one subject per photograph made possible the establishment of communication roles in the dyad. That is, the subject %)ho had the information was "sender" and the other subject was "receiver"— although subjects were given no instructions to establish the roles. This also made the experiment more believable, as it demonstrated the necessity of testing two subjects at a time. Description of the person in the photograph was accomplished by means of a set of cards containing phrases appropriate for such a task. Each subject had a different set of 16 cards. Each card contained four phrases, making a total of 128 for both sets. The description task consisted of each subject choosing, in turn, one of the four phrases which he thought best fit the person in the photograph.

PAGE 26

:l The instructions gave each partner a different task. The subject receiving the information sheet was required to select one phrase for every card. The other subject, i.e., "receiver," had the option of either selecting a phrase or of making no selection at all. This was included in the procedure to resemble the alternative of noncommunication characteristic of a natural dyad, and to provide a measure of receiver nonacknowledgement of messages. The latter was thought a sensitive measure of tendency to withhold communication, and as such would be reciprocal to a measure of "need to communicate.'' After the subjects had used, alternatively, each of their 16 cards to describe one photograph, a second one was substituted by the experimenter. For the 16 trials of the second photograph, the subjects exchanged card sets and also exchanged roles. The subject tdio was sender for the first photograph became receiver for the second, and vice versa. A second sheet of information was supplied the sender before the description trials began for the second photograph. As subjects announced their selections from the cards, the experimenter recorded each choice. He thus made certain that both subjects heard all choices, as words mispronounced or omitted were repeated by the experimenter as if for clarity. Communication Statements The 123 phrases supplied on the 32 cards (16 per set) constituted the population of communication statements from which partners selected messages. To determine whether messages influenced subsequent messages, i.e., were part of a feedback circuit, the statements had to be systeraati-

PAGE 27

?.?. cally related to one another. This requirement was met in 128 statements appropriate for personality description selected from the Interpersonal Check List (ICL) developed by Leary (1957). Leary's interpersonal theory is a behavior-oriented, multi-level system which classifies data according to eight dimensions of interpersonal behavior, each dimension bearing a constant relationship with every other dimension. The basic model employed is a circle with each dimension represented by an octant of the circle. Octants opposed to one another represent opposing behaviors and adjacent octants represent similar behaviors. Leary's principle measuring device, the ICL, is a self-report technique containing 128 statements describing interpersonal behavior, 16 statements for each of the eight dimensions. The eight dimensions as described by Leary, and a representative adjective phrase for each, are listed below: 1. Managerial Autocratic "Able to give orders" 2. Competitive Exploitive "Able to take care of self" 3. Blunt Aggressive "Can be frank and honest" 4. Skeptical Distrustful "Able to doubt others" 5. Modest Self-effacing "Can be obedient" G. Docile Dependent "Appreciative" 7. Cooperative Over-conventional "Friendly" 8. Responsible Hypernormal "Helpful" Statements were selected so that for each item appearing on a card in Set A, the next adjacent item from the ICL appeared on the same card in Set B. That is, the 128 ICL items were taken in the order 1-128, every other item belonging to the same set. This procedure produced two

PAGE 28

::.; cards for each of the 16 trials, each card containing descriptions which resembled one another but were not identical. For example, the items selected for card No. 3 for each set appeared (in randomized order) as: A B friendly affectionate and understanding can be obedient usually gives in can be frank and honest forceful able to give orders critical of others Statements were selected for the cards from every other octant of Leary's circular model. The even numbered cards contained statements from octants 2, 4, 6, and 8 (as numbered above), and the odd numbered cards from octants 1, 3, 5, and 7. The opposing octants around the circle are: 1-5, 2-6, 3-7, and 4-8. Thus, statements were selected from two pairs of diagonally opposed octants. Because diagonally opposed octants represent contradictory forms of interpersonal behavior in the Leary system, each pair of phrases appearing on a card described contradictory interpersonal behavior. A sample from card No. 3, Set B, is the pair "affectionate and understanding" and "critical of others." In addition, pairs themselves were perpendicular to one another so that for each phrase on a card there was one contradictory phrase and two other phrases neutral to it. The four phrases thus selected were randomly placed on the card. Items as they appear on the ICL vary in an intensity dimension within each octant, from low to high intensity of the behavior. As items were selected for the cards in order from 1-123, the cards varied in the intensity dimension from 1-16. (A sample scoring sheet containing the population of 123 communication statements appears in Appendix C.)

PAGE 29

U Communication Responses Subjects selected messages from cards containing parallel sets of statements . Thus, messages could be compared according to their similarity or dissimilarity with one another by reference to the ICL octants from which they were drawn. Experimentation with various methods of scoring and categorizing messages indicated the most suitable scoring procedure to be a sinilar--dissimilar dichotomy. When subjects' messages were compared with one another according to this criterion, four types of communication responses were defined. Definition of the responses was in terms of the similarity or dissimilarity of the message to the two previous messages, i.e., one by the respondent and one by his partner. The rcsponr.es were labeled congruent, redundant , confirming , and novel . From the standpoint of receiver, the four responses could be operationally described in the following manner: 1. Congruent response: Receiver's message was similar to both sender's message and his own previous message. 2. Redundant response: Receiver's message was similar to his own previous message, but dissimilar to sender's message, i.e., he repeated himself without response to sender . 3. Confirming response: Receiver's message was similar to sender's message, but dissimilar to his own previous message, i.e., he changed his message to confirm that of sender. 4. Novel response: Receiver's message was dissiroilar to both

PAGE 30

:: his own and sender's messages, i.e., he introduced a new message. Since receiver had the option on each trial of not selecting a phrase, a problem arose as to the most appropriate way of categorizing this message. This was done as follows: Non-selections by receiver were categorized as novel responses if, on the previous trial, receiver had made a selection, i.e., his non-selection was a "new" message of noncommunication; however, if his non-selection followed a non-selection on the previous trial, it was categorized as a redundant response, i.e., it was a repetition of the message of non-communication. Responses which sender made to receiver were also categorized by comparing his message to his own. previous message and to the intervening message of receiver. This seemed justifiable as, after the first trial, he received as well as sent messages. Categorizing of communication responses of both sender and receiver commenced with the second trial. The congruent response, although indicative of agreement, was not considered a sufficient criterion for detecting communication. Although this undoubtedly resulted in omission of some communication which took place, it also eliminated instances in which congruence of response signified nothing more than agreement in judging the photograph. Redundant responses were accepted as lack of communication, as they reflected repetition of a message following sender-receiver disagreement. Communication, that is, utilization of the feedback circuit, was considered demonstrated by a confirming response. In this case a subject confirmed his partner's message by selecting a statement of greater similarity to the partner's than to his XWtt previous message. In one sense this repre-

PAGE 31

26 sents conformity, but since conformity usually connotes pressure and there was little in this situation which created pressure to confirm, the term "conforming" seemed more appropriate. In this sense the response meant "to support" or "to verify." To make comparisons between trials, it was again necessary to refer to Leary's octant system. Trials alternated in the four octants from which they drew items from the ICL, the odd-numbered trials drawing from odd-numbered octants and even-numbered trials drawing from evennumbered octants. It was therefore necessary to decide which oddnumbered octants represented a dissimilar statement for a statement made from an even-numbered octant, and vice versa. The most conservative decision was to use only the octants that were adjacent to the opposite octant, and to consider all three octants "dissimilar." The following is a list of each octant and the octants considered dissimilar to it for scoring purposes; all other octants were considered "similar." Octant Dissimilar octants 1 4, 5, 6 2 5, 6, 7 3 6, 7, 3 4 7, 3, 1 5 S, 1, 2 6 1, 2, 3 7 2, 3, 4 3 3, 4, 5 In practice, subjects' messages were first translated into octant numbers, and by reference to the above list, categorized into one of the

PAGE 32

21 four communication responses. As responses were scored starting with the second trial, each subject had 15 responses per communication role and 30 for the complete test of communication. One technical adjustment was necessary in order to equate the number of receiver confirming responses per photograph to those for sender. Because receiver had a 3/4 by-chance choice of responding to sender with a similar message, whereas all other comparisons between messages operated on a 2/4 chance basis, it was necessary to subtract from receiver's confirming response total, one-third of that total. The formula was caressed: SCR RCR (2/3), where SCR is sender confirming response and RCR is receiver confirming response. Summary of Method and Design The following is a summary in chronological order of steps taken to select subjects for the experiment, form experimental groups, and test for the presence of communication between subjects. Rationale for the procedures is provided in the foregoing discussion. 1. The FIRO-B was administered to the junior and senior classes at the P. K. Yonge Laboratory School. At the same time, each student was instructed to write a story or essay in response to card 17 BM of the TAT. 2. Sixty-four students were selected as subjects for study. Sixteen males and 16 females were selected for each of the PO and CO groups on the basis of their FIRO-B score; I + A and e c scores were the criteria for this selection. 3. PO and CO groups were equally divided to form dyads for testing by the experimental procedures. Half the subjects were in compatible

PAGE 33

28 dyads and the other half in incompatible dyads. Compatible dyads were comprised of eight PO-PO and eight CO-CO combinations, and incompatible dyads were comprised of 16 PO-CO combinations. All dyads were like-sexed. 4. Subsequent to selection of dyads each subject was given his partner's typewritten TAT passage which had been prepared by the Cloze Procedure. He was unaware of the authorship of the passage. He was instructed to fill in the missing words. Cloze scores for each dyad were obtained by summing the scores of the subjects in the dyad. This measure was used to test the fourth hypothesis. 5. Dyadic partners were tested for communication together, one pair of subjects at a time. They were instructed to describe a photograph by means of selecting one of four statements on a card. Sixteen cards we;e used for each of two photographs. One subject was given additional information about the person in the photograph; he was required to select a statement from each of the sixteen cards. The other subject was given the option of selecting a statement from each card or of not responding. Partners exchanged roles for the second photograph. 6. Statements used to describe the photographs were from the ICL. Since the statements were related to one another in a systematic manner it was possible to record the pattern of a subject's agreements and disagreements with his partner's and with his own previous statements. From these patterns, four communication responses were defined. The confirming response indicated use of the feedback contingency, whereas the redundant response represented lack of sensitivity to the feedback contingency. These responses were used to test the first three hypotheses.

PAGE 34

29 7. Subsequent to all ether procedures, all subjects filled in a questionnaire on which they indicated the extent of their acquaintance with the person who had been their partner in the cocaBunication test. This was used to control for this variable in consideration of the other results.

PAGE 35

CHAPTER III RESULTS Experimental procedures were designed to test hypotheses relating communication to interpersonal orientation, compatibility, and language similarity. Subjects were assigned to one of four independent treatment groups, hereafter called orientation-compatibility groups. This term refers to the PO-compatible, PO-incompatible, CO-compatible, and CO-incompatible groups, where PO means personal-oriented and CO means controloriented approaches to interpersonal relationships. This chapter presents characteristics of the subjects, gives the communication results which tested the hypotheses, and presents additional findings pertaining to the general problem under investigation. Characteristics of Subjects PO and CO groups were each comprised of 16 males and 16 females according to the double criteria of I + A and e c scores, as described in Chapter II. Mean score differences between PO and CO groups according to these measures were significant (p < .001) in the required direction. (See Table 2 in Appendix D.) From the PO and CO groups, dyads were formed of which half had a compatible relationship and half had an incompatible relationship. Compatibility scores for interchange, originator, reciprocal, and total compatibility were calculated for each dyad. (See Table 3 in Appendix E.) Interchange compatibility was regarded the most appropriate criterion measure for differentiating dyadic groups for a short-term communication 30

PAGE 36

31 relationship; it referred to the relative emphasis placed on the three need areas by the partners. Originator compatibility referred to each subject's tendency to initiate or receive behavior within a need area, while reciprocal compatibility referred to his tendency to initiate compared to his partner's tendency to receive. Total compatibility was the sum of the three separate compatibility measures (Schutz, 1958). The mean difference between compatible and incompatible dyads on interchange compatibility scores was significant (p < .0C1) in the required direction. Compatible dyads were also more compatible according to reciprocal compatibility scores (p < .001) and total compatibility scores (p < .01) but were less compatible according to originator compatibility scores (p < .01). The age range of the 64 subjects was 15-13. There were no significant differences in mean age among the four orientation-compatibility groups. A nonsignificant t of 1.34 (p. > .05) was obtained for the greatest mean difference between two groups. Intelligence test scores on the California Test of Mental Maturity ranged 77-133. The CO-incompatible group was higher in mean intelligence score than the PO-compatible group (t 2.34; p < .05). Differences for all other orientation-compatibility group comparisons were non-significant (p > .05). In addition, a non-significant correlation of + .26 (p > .05) was obtained between intelligence test scores and the communication ratios for each subject discussed in the next section. Extent of acquaintance of partners with one another was evaluated by the median test. There were no significant differences between PO and

PAGE 37

:;: CO groups or between compatible and incompatible dyads (p > .05) in median years of acquaintance. Orientation, Compatibility, and Communication Frequencies of the four communication responses (confirming, redundant, congruent, and novel) were obtained for each subject by communication role. (See Table 4 in Appendix F.) The transformation, X =JX + .5, was applied to the response frequencies to eliminate zero totals and to insure adequate variability (Edwards, 1950, p. 200). A communication ratio was obtained for each subject in each role by dividing his sum of confirming responses by his sum of redundant responses. This ratio was considered the most accurate index of use of feedback circuits. Under conditions of a discrepancy in subject-partner messages, this ratio compared the subject's tendency to change messages in the direction of his partner, to his tendency to ignore the partner by repeating his own previous message. Table 1 shows the analysis of variance of the communication ratios by orientation, compatibility, and communication role. Since orientation and compatibility were independent observations, residual variance between subjects was used as the error term for their evaluation. Communication role was a repeated observation, thus the error term for its evaluation was residual variance within subjects. The criterion accepted for statistical significance was p < .05. The first hypothesis predicted communication differences by interpersonal orientation. A non-significant F for the orientation variable indicated a lack of support for this hypothesis.

PAGE 38

TABLE 1 ANALYSIS OF VARIANCE OF THE COMMUNICATION RATIOS 33

PAGE 39

34 The second hypothesis predicted greater use of feedback circuits by compatible dyads than by incompatible dyads. The F for communication ratios by the compatibility variable approached but did not reach significance at the accepted level. Examination of the mean difference showed incompatible dyads to have a higher average ratio than compatible dyads. (See Table 5 in Appendix G.) To the extent that this difference was reliable, the second hypothesis was contradicted. Further, a correlation of +.34 was obtained between communication ratio for each dyad and its interchange compatibility score; this fell just short of significance (p > .05). Since the magnitude of compatibility scores is inverse to compatibility, this was further evidence toward contradiction of the second hypothesis. The third hypothesis predicted greater use of feedback circuits for CO subjects as sender than as receiver. A significant F_ was obtained for the role variable, but in a direction opposite that predicted for CO subjects. When CO subjects were considered separately, mean sender ratio was exceeded by mean receiver ratio to a significant degree (t = 2. 82; p<.01). This was also true of PO subjects (t = 2.77; p < .01). Thus, the third hypothesis was contradicted as a result of a highly significant tendency for all subjects to use feedback circuits more as receiver than as sender. The fourth hypothesis predicted a positive relationship between similarity of partners' language systems and their use of feedback communication. Cloze scores for the partners of each dyad were summed to give a dyad Cloze score. The possible range of the dyad measure was 0-40,

PAGE 40

35 and the obtained range was 11-33. Since the correlation of -.25 between dyad Cloze scores and dyad communication ratios was nonsignificant (p > .05), the fourth hypothesis was not supported. Additional Findings Results are given below in addition to those which tested the hypotheses and which concerned the general problem under study. They pertain to: (1) additional variables which were significantly related to the coinminication ratio; and (2) variables related to the congruent response. Although the second hypothesis (which predicted greater use of feedback by compatible dyads) was contradicted in terms of interchange compatibility, it was supported in terms of originator compatibility. A correlation of -.41 (p < .05) was obtained between originator compatibility scores and communication ratios. Since compatibility scores are inverse to compatibility, this finding signified a positive relationship between the two variables. This was consistent with the foregoing finding that dyads labeled "Incompatible" were actually more compatible than dyads labeled "compatible" according to originator compatibility scores. Subjects when in the role of receiver had the option on each trial of not selecting a statement. Non-selection was regarded as a message of "non-communication" and thus, perhaps, reciprocal to a "need to communicate." This possibility was supported by a correlation of -.30 (p < .05) between the number of non-selections per subject and subject communication ratios. Thus, subjects who were more sensitive to the feedback

PAGE 41

36 contingency also tended to transmit nore aessages of content to the partner. The congruent response was regarded as indicating agreement between partners in describing the photograph. Since agreement in the experimental situation was possible without use of feedback circuits, congruent responses were not used to test the hypotheses. However, the finding of a correlation of + .66 (p <,001) between frequency of congruent responses per subject and subject, communication ratios lends support to the possibility that congruent responses, in general, were indicative of communication. The reverse is also true, i.e., that use of feedback circuits may have been strongly influenced by partners' tendencies to agree. To an extent, this correction may have been inflated because the communication responses were correlated measures. The use of a ratio, rather than raw scores, for the index of communication was considered a sufficient correlation to permit computation of the correlation. Three additional findings related the compatibility and orientation variables to the tendency of partners to agree or disagree. Just as congruent responses indicated agreement, redundant responses were interpreted as indicating disagreement. Incompatible dyads produced more congruent responses than compatible dyads, as sender (p < «001) and as receiver (p < .01). Conversely, compatible dyads produced more redundant responses than incompatible dyads, as sender (p < .01) and as receiver (p < .02). Since the communication responses were correlated measures, it was not unexpected that high frequency of congruent responses was

PAGE 42

37 associated with low frequency of redundant responses, and vice versa. However, the finding remains that compatibility of relationship was associated with disagreement in describing the photographs. The tendency toward lack of agreement found in compatible dyads was largely due to PO subjects within the compatible group. The 16 P0compatible subjects produced fewer congruent responses than the 16 C0compatible subjects; the difference was significant for the receiver role only (p < .01). The positive findings may be summarized as follows: 1. Communication role was the only main effect significant in the variability of communication ratios. All subjects as receiver tended to use the feedback circuits far more than as sender. 2. Compatibility as related to the communication ratio was complicated by the negative relationship of originator compatibility to the other compatibility measures. Interchange, reciprocal and total compatibility scores were negatively related to the communication ratio, whereas originator compatibility was positively related to it. 3. A low and positive relationship was found between tendency to transmit statements (under conditions of choice, i.e., as receiver) and the communication ratio. 4. A relatively high and positive relationship was found between subject's tendency to agree with partner and subject communication ratios. 5. Compatible dyads demonstrated significantly less agreement in describing the photograph than incompatible dyads. 6. PO-compatible subjects were significantly less prone to agree than CO-compatible subjects.

PAGE 43

CHAPTER IV DISCUSSIOII AND CONCLUSIONS The present study examined communication effectiveness from the standpoint of interpersonal behavior and relationships. The prediction that an interpersonal orientation toward controlling; interpersonal relationships results in greater sensitivity to the feedback contingency was not supported. A related prediction that control-oriented individuals use feedback circuits more as senders than as receivers was contradicted by the finding that all subjects tended to use them more in the role of receiver. A third prediction stated that interpersonal compatibility is associated with greater use of feedback circuits than incompatibility. For originator compatibility, this was supported; the finding that reciprocal, interchange, and total compatibility resulted in less use of feedback circuits was contradictory to the prediction. Finally, the prediction that similarity of the language systems of communicative participants is associated with use of the feedback contingency was not supported. Thus, none of the four hypotheses was directly supported and two were at least partially contradicted. Despite these negative findings, there was support for the general proposition that communication is influenced by interpersonal variables. Additional findings provided evidence that sensitivity to the feedback contingency was influenced primarily by the presence of communication roles and the acceptance or non-acceptance of these roles by participants.

PAGE 44

Roles and Communication Communication role was the only independent variable shown, without qualification, to be a determinant of subjects' communication patterns. The F ratio for the role variable was significant at p. <.001, and the pattern of greater mean receiver communication ratio was consistent for all orientation-compatibility groups. (See Table 5 in Appendix G.) This finding was interpreted as stressing the importance of situational factors for dyadic communication, i.e., aspects of the interpersonal relationship which are imposed on partners. In the experimental procedures, roles were not directly assigned to subjects per se , nor could they use "ready-made" roles. Instead, roles were suggested indirectly by means of instructions and procedures. Sender had information unknown to receiver, was required to transmit a message with content on every trial, and responded first on every trial. Thus, factors external to the dyadic relationship defined three role-related aspects of the task. Apparently, subjects were more sensitive to the situationally-cuggested roles than to any other interpersonal factors present. This corresponds to Schutz's assertion that groups develop first through an inclusion phase in which each member finds where he ''fits," i.e., he seeks his role within the group (Schutz, 1958). To the extent that subjects may have perceived the sender role as "privileged" this finding also corresponds to the finding by Thibaut (1950) that members of a low status group increased communication toward members of a high status group as the privileged position of the latter group became more clearly defined.

PAGE 45

40 Compatibility and Communication The finding that incompatible dyads tended to utilize feedback circuits more than compatible dyads was unexpected. Both the F ratio for compatibility and a correlation between interchange compatibility and conmunication ratio approached, but did not reach, significance. Two alternative interpretations may be offered to account for the less frequent use of feedback circuits by compatible dyads : (1) that subjects in compatible dyads could not formulate effective rolerelationships for the sender-receiver dimension because both attempted to relate via the same role; and (2) that subjects in compatible dyads circumvented the roles in order to relate on an individual-to-individual rather than role-to-role basis, i.e., their compatibility encouraged greater freedom of response to the total situation. Supporting the first alternative was the finding that dyads labeled "compatible" were actually less compatible than dyads labeled "incompatible" according to originator compatibility scores. The difference in mean originator compatibility scores was significant at I < .01. According to Schutz (1953), originator compatibility refers to the preference of a subject to initiate or to receive the behaviors related to a given need area. It is expressed by the formula: oK (e^ Wj) + (c.-. xj.) , where e and w refer to expressed and wanted needs for a need area, and i and j to the partners of the dyad. Positive scores signify preference for initiating, and negative scores preference for receiving. Compatibility is indicated by complementary scores, i.e., by one subject's preference for initiating complemented

PAGE 46

41 by the other's preference for receiving. Since the sign is retained, complementary scores sum to zero, indicating greatest compatibility. High positive scores indicate competitive incompatibility (e.g., both wanting to initiate control) , and high negative scores indicate apathetic incompatibility (e.g., both wanting to be controlled). (Schutz, 1550.) Of the 16 compatible dyads, ten had originator compatibility scores above the mean for all dyads. Of the 16 incompatible dyads, only five had originator compatibility scores above the same mean. Thus it appears likely that in compatible dyads more than in incompatible dyads partners experienced difficulty in adopting the suggested connaunication roles, and that this role confusion interfered with sensitivity to the feedback contingency. The second alternative to explain differences between compatibility groups in use of feedback circuits offers a different line of reasoning. It assumes that subjects in compatible diads were, in fact, more compatible, and that they rejected a role-oriented relationship in favor of relating to each other and to the situation as individuals. That is, compatibility engendered security of relationship, which in turn encouraged personal freedom of response and made the restriction of roles undesirable to partici pants. Conversely, in incompatible dyads participants were less sure of the relationship and restricted themselves to role-relating. Since the communication roles x-;ere suggested by the experimenter, i.e., external to the dyadic relationship, it is possible that compatible partners cooperated in rejecting them. Schutz recognizes this possibility,

PAGE 47

-:: ". . .if members of the group do not want to do their assigned task, compatibility will contribute to their efficiency in not doing it (for example, a compatible group should be more capable of mutiny) ." (Schuts, 1958, p. 115.) The second alternative was supported by the finding that compatible subjects produced significantly fewer congruent responses, indicating less agreement, and significantly more redundant responses, indicating greater disagreement. The probability values ranged from .02 to .001 for differences in mean number of responses by conraunication role. Further, PO-compatible subjects as receiver produced fewer congruent responses than COcompatible subjects (p < .01) . These findings suggested more sensitivity toward personal factors than situational factors by partners in compatible dyads, especially partners oriented toward inclusive and affective relationships. The last finding drawc attention to the PO-compatible group. These subjects, as contrasted to all others, were less prone to agree with one another and tended to disagree more with one another. This suggested that for them interpersonal orientation may have contributed to responses to the task even though orientation as a variable was nonsignificant. Conformity Behavior and Communication Results of the present study showed a positive relationship between conformity behavior, in the sense of partner agreement, and use of the feedback contingency. A correlation of +.66 was found between subject coiiaunication rauios and the frequency of congruent responses per subject. Incompatible dyads were higher than compatible dyads in both mean communication ratio and mean frequency of the congruent

PAGE 48

<-.: response. The issue may be raised: Under conditions of sender-receiver differences in the content of messages, does subsequent similarity in messages represent conformity behavior or effective use of feedback circuits? The issue cannot be resolved from results of the present study. The procedures were designed so that pressure to conform was minimal, but it can be assumed that subjects had internal pressures to conform. (Indeed, in a recent study investigating agreement and disagreement in dyads, responses similar to the coraraunication responses of this study were considered indicative of various types of conformity behavior ) (Willis, 1964). One finding did support the assumption that the communication ratio was reflective of communicative behavior. A correlation of -.30 (p < .05) was obtained between subjects' number of non-selectionsas receiver and subject consaunication ratio. Thus, subjects who were more sensitive to the feedback contingency also tended to conmunicate more content messages to the partner. Verbal Conditioning and Communication Inferences about how interpersonal factors influence communication were made from verbal conditioning studies, as reviewed in Chapter I. These inferences were given little support by results of the present study, and were contradicted for the compatibility variable. Incompatible dyads in this study tended to use feedback circuits more than compatible dyads, whereas in Sapolsky's (1960) study the reverse was true. These discrepancies may be attributed, in part, to differences between this and Sapolsky's study in roles suggested for subjects, to-

PAGE 49

•/: gether with differences; in the locus of control of the coomunication process. In the present study subjects shared control over the coomunication process and role was defined primarily in terms of knowledge of content, i.e., by sender's possession of information about the person in the photograph. In the verbal conditioning studies, subjects were in the sender role, i.e., they transmitted ne s sages to the experimenter. As senders, however, they did not have effective control over the communication process, as the experimenter controlled it with his knowledge of the reinforcement contingency. To an extent, the conditioning situation represents a discrepancy in communication roles from the natural pattern in which a sender shares control of the communication process with a receiver. Assuming that subjects in incompatible dyads were more controlconscious and role-conscious than subjects in compatible dyads, one may speculate that they rejected the role of sender-without-control in the verbal conditioning studies but found the communication roles of the present study acceptable. Since there are no data to support this explanation, it must remain speculative. Interpersonal Influences on Dyadic Communication The primary goal of the present study was contribution to a clearer understanding of influences which interpersonal factors have on communication. As defined by procedures used, interpersonal orientation was not significantly related to use of feedback circuits. Language similarity of dyadic partners was also nonsignificant in relation to use of feedback circuits. However, it is possible that a more heterogeneous popula-

PAGE 50

45 tion than one composed of high school juniors and seniors from a relatively small school would provide a greater range of orientation scores from which to select subjects for experimental study. It is also possible that procedures used rejected subjects for study whose interpersonal orientations might be highly related to communication effectiveness. By selecting subjects who had high scores on either I + A or e , the study omitted those who made low scores in all interpersonal need areas. Assuming the interpersonal orientation of these individuals to be avoidance of any kind of interpersonal relationships, it might be this orientation which most influences communication effectiveness. This appears a problem worthy of future research. The most distinctive finding of the study was that externally suggested or imposed roles significantly altered coaraunication patterns. Subjects as receiver used feedback circuits far more than they did as sender. Apparently this was a function of sender's having more information relating to the task of describing the person in the photograph. The question of whether receiver was conforming or using effective communication techniques cannot be answered by results of this study. The general question of whether changes in an individual's messages represent conformity behavior or effective communication is perhaps best ansv;ered by examining the general nature of the relationship in which these changes occur. It may be that as relationships become oriented toward dominancesubmission and/or inclusion-exclusion, the changes represent conformity, and that as relationships become oriented toward affective and/or cognitive understanding, the changes represent effective communication. For

PAGE 51

46 example, a therapist who frequently makes use of feedback circuits to develop with a client a mutual understanding of the client's messages nay be thought of as communicating effectively. On the other hand, the junior executive who invariably changes his statements in accord with the domineering boss may be thought of as conforming. This general question is open to future study. A second significant finding was the tendency toward a negative relationship between use of feedback circuits and interpersonal compatibility. Two interpretations of this finding were offerred above. The first was that so-called incompatible dyads were actually more compatible on a compatibility dimension significant for dyadic communication, i.e., originator compatibility. If this were true, the apparent negative results relating compatibility to communication would be reversed, and the finding would then be In agreement with those of Sapolsky (1960) . However, this appears unlikely in light of Sapolsky' s statement that his incompatible dyads were incompatible for all need areas. The second interpretation explained the incompatible dyads' tendency toward greater use of feedback circuits in terms of role acceptance. If incompatible dyads accepted the communication roles and compatible dyads did not, differences in use of feedback would occur. It was argued that incompatibility of relationship increased the need for relating via roles, and that compatibility encouraged a feeling of freedom from the external situation and a sensitivity toward the individual. Role acceptance was also inferred as a critical factor in explaining the discrepancy between verbal conditioning measures and

PAGE 52

-.-.. dyadic coomunication measures as influenced by interpersonal compatibility. It was assumed that incompatibility increased role-sensitivity and also sensitivity to the control dimension of communication. It was then suggested that subjects in incompatible relationships, without means: of effective control over the coramunicatioa process as in verbal conditioning, might have withdrawn from a coramunication role, but that given control, as in dyadic communication, might have accepted the same role. Since the argument depends on a series of inferences, it must be considered highly tentative. Two general conclusions may be stated regarding the effects of interpersonal variables on communication: (1) externally suggested or imposed cooounication roles are significant determinants of communication patterns, and (2) acceptance of these roles by participants contributes to communication effectiveness. One major limitation must be noted regarding generalization of the above conclusions. In the present study, effective coramunication referred simply to exchanges of messages, i.e., it asked the general question, "Did a subject acknowledge that he heard the other's message?" Thus, the study disregarded what may be called "exchanges of understanding." The latter refers to use of the feedback contingency to acknowledge not only that a message has been received, but that the meaning which it has for the sender has also been received and understood. For this reason, the above conclusions that communication is altered by the presence of roles, and is facilitated by role-acceptance on the part of participants, do not necessarily apply when cooounication is defined as a total understanding of another's messages. In fact, if one considers

PAGE 53

only the restrictive characteristics of roles, they aay be considered to inhibit cocnunication in a broader sense of exchanges of understanding. An issue similar to the above distinction between types of exchanges was explored by Rogers (1951) . In attempting to formulate the counselor's role, Rogers first thought of it as clarifying and objectifying the client's feelings, i.e., to acknowledge that his feelings were messages which could be received and codified. He later found this formulation too intellectualistic and prone to promote a declarative rather than an empathic attitude on the part of the counselor. He states : But when the counselor statement is declarative, it becomes an evaluation, a judgment made by the counselor, who is now telling the client what his feelings are. The process is centered in the counselor, and the feeling of the client would tend to be, "I am being diagnosed." In order to avoid this latter type of handling, we Ixave tended to give up the description of the counselor's role as being that of clarifying the client's attitudes. [Rogers, 1951, p. 28.] The effects that interpersonal variables have on an "exchange of understanding" between individuals would appear a topic worthy of separate investigation. Summary The present study investigated the general proposition that effective communication between tiro people is related to dimensions of their interpersonal relationship and to the similarity of their language systems. Communication effectiveness was defined as use of feedback

PAGE 54

•':•; circuits in a facc-to-face situation. Hypotheses predicted that: (1) subjects oriented toward controlling otherc would communicate more effectively than subjects oriented toward inclusive and affective interpersonal relationships; (2) pairs of subjects compatible to one another would communicate aore effectively than pairs of subjects incompatible to one another; (3) control-oriented subjects would communicate more effectively in a sender role than as receivers; and (4) language similarity of communication pairs would have a positive relationship to their conmunication effectiveness in a face-to-face situation. The FIRO-B was used to select 64 subjects from junior and senior high school classes for placement into either a personal-oriented (PO) or a control-oriented (CO) group. PO subjects were selected for an interpersonal orientation toward inclusive and affective relationships and CO subjects for an interpersonal orientation toward controlling relationships. Subjects were paired to form 16 compatible and 16 incompatible dyads. Compatible dyads consisted of eight PO-PO combinations and eight CO-CO combinations; incompatible dyads consisted of 16 PO-CO combinations. Each subject completed a passage written by his partner which the experimenter had prepared by the Close Procedure. Dyad Cloze scores provided a measure of dyadic language similarity. Dyads were tested separately for communication effectiveness by means of a task which permitted, but did not require them to respond to each other's messages as they described a photograph. Four communication responses were defined by comparing a subject's message to the previous message of his partner and his own previous message. Analysis of variance was

PAGE 55

5C performed on a communication ratio calculated for each subject. Thic was determined by dividing the confirming response by the redundant response. Interpersonal orientation and language similarity were not significantly related to communication effectiveness. There was a tendency for subjects in incompatible dyads to be more effective communicators than those in compatible dyads; the F ratio for compatibility approached but did not reach statistical significance. All subjects tended to communicate more effectively in the role of receiver than that of sender. Thus, none of the hypotheses was supported and two vera at least partially contradicted. Analysis of the results revealed several additional findings beyond those used to test the hypotheses. There was a low and positive relationship between tendency to transmit statements (under conditions of choice) and the communication ratio. A relatively high and positive relationship was found between partners 1 tendency to agree and their communication ratios. Compatible dyads demonstrated significantly less agreement in describing the photograph than incompatible dyads. PO subjects in compatible dyads were less prone to agree than CO subjects in compatible dyads. Discussion of results focused on the importance of externally suggested or imposed communication roles for dyadic communication, and on the significance of acceptance or non-acceptance of the roles by participants. It was concluded that communication roles contributed signifi-

PAGE 56

cantly to conaunication patterns and that role acceptance by partners increased communication effectiveness when coiarauni cation effectiveness was defined as the use of feedback circuits.

PAGE 57

BIBLIOGRAPHY Berg, I. A. VJord choice in the interview and personal adjustment. J. consult. Psychol ., 1950, 5, 130-135. Blum, G. S. The Blacky pictures: A technique for the exploration or personality dyanlcs (manual) . New York: The Psychological Corporation, 1950. Cairns, R. B. &. Lewie, H. Dependency and the reinforcement value of a verbal stimulus. J. consult. Psychol ., 1962, 26, 1-8. Crowne, D. P. & Strickland, B. R. The conditioning of verbal behavior as a function of need for social approval. J. abnorn. soc . Psychol ., 1961, 63, 395-401. Edwards, A. L. Experimental design in psychological research . New York: Rinehart, 1950. Feldstein, S. & Jaffe, J. Language predictability as a function of psychotherapeutic interaction. J. consult. Psychol ., 1963, 27, 123-126. Greenhill, M. The focal communication concept. Amer. J. Psychother ., 1958, 12, 30-41. Loary, T. Interpersonal diagnosis of personality: A functional theory and methodology for personality evaluation. New York: Ronald Press, 1957. Lorenz, M. & Cobb, S. Language patterns in psychotic and psychoneurotic subjects. A. M. A. Arch. Neurol Psychiat ., 1954, 72 , 665-673. llarlowe, D. Need for social approval and the operant conditioning of meaningful verbal behavior. J. consult. Psychol ., 1962, 26 , 79-83. Relss, 3. F. Communication in psychotherapy. Amer. J. Psychother ., 1957, 11, 774-789. Rogers, C. R. Client-centered therapy . Boston: Houghton Mifflin, 1951. Ruesch, J. Disturbed communication: The clinical assessment of normal and pathological communicative behavior . New York: W. W. Norton, 1957. . Therapeutic communication. Nov; York: H. W. Norton, 1961. 52

PAGE 58

53 Sapolsky, A. Effect of interpersonal relationships upon verbal conditioning. J , abnorm. soc. Psychol ., 1960, 00, 241-246. Schutz, W. C. FIRO: A t hree-diaensional theory of interpersonal be havior . Hew Yorkl Rinehart, 1950. Taylor, U. L. "Close procedure": A new tool for measuring readability. Journalism Quart .. 1953, 30, 415-433. Thibaut, J. An experimental study of the coheciveness of underprivileged groups. Bunt. Rela t., 1950, 3, 251-273. Tiraaons, E. 0. & Noblin, C. D. The differential performance of orals and anals in a verbal conditioning paradigm. J. co nsult. Psychol., 1963, 27, 383-335. Ueakland, J. F. Letters of mothers of schisophrenics. Aner. J. Ortho psychiat. , 1962, 32, 604-623. Uicner, H. Cybernetics or control and comuni cation in the animal and the machine . Star York: Wiley u Sons, 1943. The human use of human beings . Boston: Houghton Mifflin, 1950. Willis, R. 11. The phenomenology of agreement and disagreement in dyads. Tech. Rept., Nonr 316(12) . St. Louis: i/ashington University, 1964.

PAGE 59

APPENDICES

PAGE 60

APPENDIX A INSTRUCTIONS FOR THE EXPERIMENTAL TASK i: First, I will tell you the purpose of the research and then what your specific job will be during the experiment. "In this experiment we are getting information on how well people can judge a person whom they do not know by looking at his photograph and by reading some factual information about him. We are interested in the practice of businesses, government agencies, and colleges of requesting photographs and personal history data from applicants along with their application papers. We would like to find out what kind of impressions people get from photographs and from the personal histories which are written. We will look at two photographs during the experiment. "I will put up a photograph for you to see. Look at it closely, especially for clues as to what kind of a person this person might be. For the purposes of the experiment, one of you will be Subject A and the other will be Subject B . The cards on the table tell you which one you are. "For the first photograph, I will give Subject A a sheet of paper which tells something about the person shown in the photograph. Only Subject A may see this sheet or know the information for the first photograph. Therefore, I am asking that subjects refrain from making random comments or from asking questions once the experiment has begun. When we look at the second photograph, Subject B will have the information sheet for that picture. "Now I will tell you the method for judging or describing the person in the photograph. To do this, use the stack of 16 cards which is on the table in front of you. Subject A has List A and Subject B has List B . Both lists have cards numbered 1-16, but otherwise they are different. "Each of these cards has four short statements which could describe someone. Only one statement can be chosen from each card. You are to look at these cards, one at a time, and tell me which of these statements you think best fits the person in the photograph. We will alternate, Subject A choosing first, then Subject B . "Now listen carefully, because your tasks are somewhat different. Subject A will look at each card in turn and decide which of the statements best fits the person in the photograph. Subject B will also look at each of his cards in turn to see which statement best fits the person in the photograph, but should he decide that none of the statements fits at all, he should say nothing. In other words, each time A looks 55

PAGE 61

56 at a particular card, he will select one statement from his card, and after he has chosen, Subject B will be able to turn to his card of the came number and either choose one statement from his card or will remain silent. In case B does not choose one, I will direct A to the next card after a few seconds, indicating the beginning of another trial. Remember, B is the only one that has a chance of not saying anything on a particular card, as A selects one every time. "Let me summarize the procedure for you. First I will put up the photograph. Then Subject A will get the information sheet and will have a minute to read it. He may refer to this at any time, but cannot reveal its contents to Subject B . Then I will ask A to turn to card #1, and he can begin right away by choosing a statement from his card. After he has done this, B can turn to hie card, and should either select a statement or not say anything, indicating that he thinks that none of the statements fits at all. We will use the same procedure for all 16 cards. "Are there any questions about the procedure? "You will have sufficient time to look at all four statements and to make your selection, but it is best to decide as quickly as you can, on a first-impression basis. Do not be concerned with whether you think your answers are right or wrong, as it is your impressions that we want you to give. Don't be concerned with duplications or contradictions. "I will call out the cards for each of you one at a time. Please do not turn to any card until I have told you to. For example, when I say 'A 1' only the person who lias List A should turn to the first card." (Another query about questions.) (Place photograph, give A time to read information sheet.) (16 trials for first photograph) "We are still in the middle of the experiment, so I will ask you not to make any comments while I put up the second photograph. We will use the same procedure as for the first one, except that B will now have the sheet of information for the person in the photograph, and will select a description from every card. And A will be able to select one for each card £r remain silent if he thinks that none of the descriptions fits at all. In other words, you will switch tasks for the second photograph. I want you also to change lists, so that Subject B has List A, and vice versa . You may do this now, but please do not turn to any cards until I tell you to. [Place photograph, give B time to read information sheet] I am going to call out the card numbers as I did last time. As before, we will start with the person who has List A. Card A-l." [16 trials for second photograph. Thank Ss for participation and caution them not to discuss experiment.]

PAGE 62

57 APPENDIX B INFORMATION SHEET FOR PHOTOGRAPH #1 Name: Joe M Age: 26 Mr. M was born in a large southern city and went to high school there. He was active in sports, especially baseball. He worked parttime during high school for his uncle who owned a small manufacturing company. After high school he joined the Army and served in the Quartermaster Corps for three years. Mr. M planned to enroll in the University of Georgia immediately after his discharge from the Army, but his father died suddenly and he had to help support the family for a year. He had several jobs, most of them working on home construction cr ewe . Using the G.I. Bill, Mr. M enrolled at the University of Georgia and obtained his degree in 1950. During college, he went out for the baseball team in his freshman year but gave up baseball altogether after he married one of his classmates. One of his main hobbies now is fishing, and he and his wife are active in a square dancing club. Mr. M was undecided about what job to take after college and asked for help from a Veteran's Administration vocational counselor. After several interviews, he decided to return to his home city to work for the same uncle he worked for in high school, as the company was now large enough for him to have a supervisory position.

PAGE 63

58 APPENDIX B, Continued INFORMATION SHEET FOR KIOTOGRAPH #2 Name: Bryan W Age: 25 Mr. W was born in a large southeastern city, but his family moved to the West when he was six years old. His father was in the Navy, and the family moved three times while Mr. W was in high school. He was active in the school band in each school. His main hobbies were reading, music, and water sports. After high school graduation, Mr.W worked for almost a year as clerk in a local bookstore to save money for college. At this time the family moved back to the Southeast again, and Mr. enrolled in the University of Georgia. He attended college for only one year because of financial difficulties. He joined the Army to fulfill his military obligations, and served in the Medical Corps. After his discharge from the Army, Mr. sought help from a Veteran's Administration vocational counselor. When he found that he had good potential for finishing college, he returned to the University of Georgia immediately. Two and a half years later, in 1950, he obtained his degree. Soon after graduation he married a girl whom he had met during his freshman year at the university. He and his wife water ski a great deal, and he enjoys a new-found hobby of coin collecting. He was undecided about what to do after college and went back to the vocational counselor. There he made tentative plans to join the training program of a nearby industrial firm.

PAGE 64

L_ APPENDIX C SAMPLE SCORING FOR EXPERIMENTAL TASK List A 1. xjell thought of can be strict if necessary able to criticize self cooperative 2. self respecting can complain if necessary grateful considerate 3. able to give orders can be frank and honest can be obedient friendly 4. able to take care of self able to doubt others appreciative helpful 5. often admired hard-boiled when necessary easily embarrassed always pleasant and agreeable 6. self-confident resents being bossed often helped by others kind and reassuring 7. good leader irritable easily led sociable and neighborly 1 A

PAGE 65

60 APPENDIX C, Continued SAMPLE SCORING FOR EXPERIMENTAL TASK 1 2 A B 1 2 B A 8. businesslike hard to impress accepts advice readily enjoys taking care of other* 9. always giving advice inpatient with other's mistakes self-punishing too easily influenced by frienc 10. boastful bitter dependent forgives anything 11. bossy outspoken passive and unaggressive fond of everyone 12. thinks only of himself jealous lets others make decisions generous to a fault 13. tries to be too successful sarcastic timid wants everyone's love 14. somewhat snobbish resentful hardly ever talks back too lenient with others

PAGE 66

51 APPENDIX C, Continued SAMPLE SCORING FOR EXPERIMENTAL TASK 1 2 A B 1 2 B A 15. raanages others frequently angry obeys too willingly friendly all the tiiae 16. selfish stubborn likes to be taken care of too willing to give in to others 15. dictatorial hard-hearted spineless loves everyone 16. cold and unfeeling distrusts everybody will believe anyone spoils people with kindness

PAGE 67

62 APPENDIX D TABLE 2 MEAN FIRO-B INCLUSION PLUS AFFECTION (I + A) SCOREti AND EXPRESSED CONTROL (e c ) SCORES FOR PERSONAL ORIENTED (PO) AND CONTROL ORIENTED (CO) GROUPS, BY SEX, AND TOTAL FOR BOTH GROUPS

PAGE 68

63 APPENDIX E TABLE 3 MEAN FIRO-B INTERCHANGE, ORIGINATOR, RECIPROCAL AND TOTAL COMPATIBILITY SCORES FOR COMPATIBLE AND INCOMPATIBLE DYADS* Compatible Dyads Incompatible Dyac'.s M C M CT Interchange Compatibility 8.88 4.57 18.56 7.72 Originator Compatibility 9.96 6.42 5.31 4.20 Reciprocal Compatibility 14.75 5.42 19.94 5.19 Total Compatibility** 33.31 10.88 43.81 13.66 3.13**** 4.12*** 3.29**** *Greater FIRO-B compatibility score = less compatibility **Total compatibility score " sum of interchange, originator, and reciprocal scores ***Mean difference significant at p < .001 ****Mean difference significant at p < .01

PAGE 69

APPENDIX F TABLE 4 FREQUENCY OF COMMUNICATION RESPONSES BY ORIENTATION-COMPATIBILITY GROUP AND COMMUNICATION ROLE* J Conmunication Role Sender Receiver** CF R N CF R PO-

PAGE 70

Oij APPENDIX G 2AI3LE 5 MEAN COMMUNICATION RATIOS BY ORIENTATIONCOMPATIBILITY GROUP AND COMMUNICATION ROLE*

PAGE 71

6o APPENDIX H TABLE 6 COMMUNICATION RESPONSES BY KWTOGRAPH, ORIENTATIONCOMPATIBILITY GROUP, AND COMMUNICATION ROLE*

PAGE 72

67 APPENDIX I TABLE 7 GROUP MEAN COMMUNICATION RATIOS BY PHOTOGRAPH, ORIENTATION-COMPATIBILITY GROUP, AND COMMUNICATION ROLE*

PAGE 73

BIOGKAKIICAL SKETCH Donald M. Ilartsough was born November 27, 1934, at Seville, Ohio. He was graduated from Cuyahoga Falls High School in June, 1951. He attended the College of Wooster, where he served as Student Body President, and in June, 1955, received the degree of Bachelor of Arts with a major in economics. The following year from February until June he attended the University of California. In 1956 he enrolled in the Graduate School of the University of Florida. He worked as a graduate assistant in the Department of Psychology from February, 1957, until August, 1950, when he received the degree of Master of Arts. During the academic years 1958-1959 and 1900-1961, he received a training grant from the Bureau of Vocational Rehabilitation in order to pursue work toward the degree of Doctor of Philosophy in clinical psychology. From September, 1959, until August, I960, he was a research assistant in the College of Health Related Services of the University of Florida. His clinical internship was served at the J. Ilillis Miller Health Center the following year. In August, 1961, he joined the Alachua County Health Department at Gainesville, Florida, as staff psychologist. From August, 1962, until the present, he has held the combined appointment of psychologist at the P. K. Yonge Laboratory School and instructor in the College of Education of the University of Florida. Donald M. Ilartsough is married to the former Dalyte Rose Ellis and is the father of three children. He is a member of the Florida Psychological Association and the American Association of University Professors. He is an associate member of the American Psychological Association. 6G

PAGE 74

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. April 18, 1964 Dean, Graduate School SUPERVISORY COMMITTEE: f Chairc U£^t. ftty^cL^t^~-y'/0 fol^Kj^X p> \^\£h+tX^

PAGE 75

UNIVERSITY OF FLORIDA 3 1262 08552 9849