Effects of sex of sender and emotional parameters of message on physiological responding in marital communication


Material Information

Effects of sex of sender and emotional parameters of message on physiological responding in marital communication
Physical Description:
vii, 83 leaves : ; 29 cm.
Hermecz, David A., 1957-
Publication Date:


Subjects / Keywords:
Marital Therapy   ( mesh )
Interpersonal Relations   ( mesh )
Problem Solving -- Adult   ( mesh )
Communication   ( mesh )
Clinical and Health Psychology thesis Ph.D   ( mesh )
Dissertations, Academic -- Clinical and Health Psychology -- UF   ( mesh )
bibliography   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph.D.)--University of Florida, 1985.
Bibliography: leaves 75-82.
Statement of Responsibility:
by David A. Hermecz.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000535613
oclc - 16878104
notis - ACV8612
sobekcm - AA00004876_00001
System ID:

Full Text








It is with much gratitude that I acknowledge my committee members

for their part in transporting this project from the realm of the

grandiose to the realm of the merely ambitious. The consistent

support of my chair, Dr. Hugh Davis, was critical in orienting me

through myriad trees in search of a forest. I thank the other members

of the committee--Dr. Lawrence Siegel, Dr. Keith Berg, Dr. Wilse

Webb, and Dr. Mickie Edwardson--for their helpful suggestions on

design, procedures, and data analysis. Without the input of Dr.

David G. Gilbert, with whom I spent countless hours discussing this

paradigm and who scored all psychophysiological records, this project

would have been impossible. Ed Haskins and Sue Williams tirelessly

assisted in running subjects, and Kati Mandoki, Kathy Toohey, and Bill

Winkel served as raters--to them I extend my thanks. Mina Berman

assisted in both the scoring of questionnaires and key punching the

data, for which I am most grateful. I appreciate the willingness of

the participating couples to openly share their conflicts for the sake

of science and twelve dollars. Finally, I must thank my friends, my

family, and my wife, Kim, for their confidence that there is light at

the end of the tunnel, and for their support and love which made it

possible for me to get there.


ACKNOWLEDGMENTS ............................................ ii
LIST OF FIGURES .................................... ................ v
ABSTRACT ....................................................... vi


I LITERATURE REVIEW ........ ........... .................... 1

Communication, Conflict, and Emotion ........ .............. 1
Marital Interaction-A Special Case .................... 2
Verbal Content ....................................... 5
Nonverbal Affect ........................................ 9
Sex Differences ........................................ 14
Statement of the Problem ............................... 23

II METHOD ................................................... 27

Subjects ................ ........... ... ................... 27
Apparatus .................. ...................... 27
Design and Procedure ................................... 29
Notes .......................... ........................ 34

III RESULTS ................................................. 35

Manipulation Check .................................... 35
Subjects ................................................ 36
Data Analyses ........................................... 37
Physiological Scoring .................................. 39
Silent Waiting Period Data .......................... 40
Time of Assessment Effects .............................. 41
Content Effects ......................................... 41
Affect Effects ....................................... 46
Affect by Content Interaction ........................... 46
Sex Effects .... .... ............................... 51
Sex by Content Interaction .............................. 51
Sex by Content by Affect Interaction .................... 58
Nonsignificant Interactions ............................. 62

IV DISCUSSION .............................................. 63


1 EXAMPLES OF PROBLEM TRANSFORMS .......................... 72
2 SENDER'S STANDARD MESSAGE ............................... 74
REFERENCES ..................................................... 75
BIOGRAPHICAL SKETCH ........................................... 83



1 Skin resistance responses as a function of time of
assessment ........................................... 42

2 Skin conductance responses as a function of time of
assessment ........................................... 43

3 Skin conductance responses as a function of content ..... 44

4 Skin conductance responses as a function of time of
assessment and content ............... ........ ........ 45

5 Skin conductance responses as a function of affect ...... 47

6 Skin conductance responses as a function of time of
assessment and affect ............................... 48

7 Skin resistance responses as a function of time of
assessment and affect ............................... 49

8 Skin resistance responses as a function of content and
affect ................................................ 50

9 Heart rate as a function of sex ........................ 52

10 Skin resistance responses as a function of sex .......... 53

11 Skin conductance responses as a function of sex ......... 54

12 Magnitude of change in skin resistance level as a
function of sex ..................................... 55

13 Skin resistance responses as a function of sex and
content ............................................. 56

14 Skin conductance responses as a function of sex and
content .............................................. 57

15 Heart rate as a function of sex, content, and affect .... 59

16 Skin resistance responses as a function of sex, content,
and affect ........................................... 60

17 Skin conductance responses as a function of sex, content,
and affect ......................................... 61

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy




August 1985

Chairman: Hugh C. Davis, Ph.D.
Major Department: Clinical Psychology

Sixty-four couples presented information about specific problem

areas in their relationships. After identifying the area in which the

greatest degree of disagreement existed, one spouse from each couple

was chosen randomly to present a written description of his/her

perception of the target problem to his/her spouse. Half the

presenting spouses were coached to present the problem in an

effectively neutral (calm, matter-of-fact) manner; the others were

coached to present messages in an effectively negative (angry,

impatient) manner. A transformation of the content of the problem

descriptions was also made, with half being modified to express the

complaint in a "mindreading" manner, attributing feelings, attitudes,

opinions, and/or behaviors to the other spouse, and half being

modified to be "problem feeling," expressing the speaker's own

feelings, attitudes, or opinions about the problem area. Half of the

senders of each of these groups were wives and half were husbands.


Heart rate, skin resistance responses, skin conductance responses, and

change in skin resistance level of the listeners were the dependent

measures, monitored over a period of silence, the reading of a

standard message, and the reading of the problem description. Spouses

in the negative affect conditions showed greater increases in the

number of electrodermal responses from standard message to problem

description than subjects in the neutral condition. There was a

tendency for mindreading messages to elicit a greater number of skin

conductance responses than problem feeling messages. Negative affect

combined with mindreading content elicited the greatest number of skin

resistance responses in listeners. Wives showed greater physiological

activity overall than husbands, regardless of the content or affect of

the message. Wives showed more electrodermal activity to problem

feeling than mindreading messages, and husbands were more responsive

to mindreading than problem feeling messages. Content and affect also

interacted with sex, with wives hearing neutral problem feeling

messages showing the highest heart rate response and husbands hearing

negative mindreading messages showing the lowest level of responding.

Results were discussed with respect to complementarity of roles of

husbands and wives in conflict resolution and the potential for

application of the information provided by psychophysiological





Communication, Conflict, and Emotion

The process of communication involves "the passing of information

or the formulation of thought by one person for the sharing or

understanding of one or more other persons, or for the control of the

behavior of one or more observers" (Eisenson, Auer, and Irwin, 1963,

p. 20). Watzlawick, Beavin, and Jackson (1967) have described two

modes whereby communication takes place. One is referred to as the

"digital" mode of communication, corresponding to verbal content. The

primary function of digital communication is to convey information

concerning objects. The second mode of communication--"analogic"--

helps define the nature of the relationship of the communicants.

Watzlawick et al. equate nonverbal communication with analogic

communication. Every communication has both di ital and analogic

components which complement each other; in fact neither mode could

adequately fulfill the function of the other. Because these aspects

of communication are shaped by cultural biases, role conflicts,

conflicting social systems and other factors (Fisher, 1978), and

because the communication process by definition involves more than one

party there is a high likelihood that there will be some mismatch of

goals among interactants. When this occurs, communication has led to,

or become, conflict. Keltner (1970) conceives of interpersonal

conflict as a system of relationships between individuals who are

seeking goals that usually cannot be attained simultaneously under

prevailing conditions. Having been made manifest through the vehicle

of communication, conflict requires communication for its resolution.


However, conflict is more easily initiated than resolved due to

organismic variables which, with the occurrence of conflict, introduce

noise into the transmission/reception of messages. That is to say,

communication does not transpire solely as a systematic exchange of

verbal and nonverbal messages; it is largely influenced by emotional

processes which, once activated, may have an ongoing effect. Sillars

and Parry (1982) have suggested that as stress increases in intensity,

the complexity of human information processing decreases, marked by

decreased perspective-taking and increased other-directed attributions

of blame for conflicts. Hovland, Janis, and Kelley (1953) have also

noted that when emotional tension is strongly aroused distractibility

tendencies are increased and cognitive functions are temporarily

impaired. Thus, as conflicts intensify and conflict management

becomes more important the possibility of engaging in constructive

communication declines. This postulated relationship between arousal

level and ability to utilize cues in the organization of behavior is

not a simple negative linear one however. Several investigators have

suggested an "inverted U" function, with moderate levels of arousal

implicated as best suited for processing of information (Janis, 1967;

Easterbrook, 1959). From this perspective, inasmuch as it is capable

of keeping the arousal level of participants within this optimal level

of arousal, the communication process will be facilitated.

Marital Interaction--A Special Case

There is reason to believe that the communicative behavior of

marital dyads and families is different from that of communication

taking place in most work situations and in other instrumental

activities in daily living. Thomas (1977) discusses several factors

contributing to this discrepancy: 1. Conventional stimulus control

over the topics to be discussed by given family members is relatively

lacking. Usually, practically any topic is able to be discussed at

any time. While this flexibility is at times advantageous, the

relative lack of structure provides an opportunity for communication

difficulties to grow into patterns. 2. Decision-making issues in the

family tend to spill over into general family communication. Thus, if

decision making is not accomplished at a specific time, decision

matters may infuse virtually all talking of family members. 3.

Family communication involves a small, limited audience such that any

given member's verbal behavior is vulnerable to the deficiencies of

the partner's verbal behavior. Thus, the impact of deviant family

members on communication can be greatly exaggerated. 4. The family,

being nuclear and relatively isolated, tends to be detached from most

external corrective influences. Because there is little external

control placed upon the internal workings of most families, highly

unusual patterns of verbal behavior may develop and be sustained

without being detected or altered externally. 5. Even though it is

relatively isolated from external corrective influences, the family is

in large measure an open system inasmuch as virtually any disruptive

input may be brought back into the family by any member. For example

the husband, having had a bad day at work, may bring his troubles

home, insult his wife, and initiate an argument that, given less

intense external pressures, might not have taken place.

There are other ways in which couples are unique in their

interactions. Birchler, Weiss, and Vincent (1975) studied

social-reinforcement exchange in maritally distressed and

nondistressed spouse-and-stranger dyads. When comparing distressed

with nondistressed couples, distressed dyads were found to produce

less positive and more negative social reinforcement. However, when

they had discussions with strangers the distressed couples interacted

considerably more favorably, emitting approximately half as many

negative and about three times more positive statements to strangers

than to spouses. In his attempt to discover what, if anything, is

distinctive in marital interaction Ryder (1968) used a decision-making

task (the Color Matching Task) and paired husbands with their wives or

with female married strangers. He found that husbands were more

likely to take the lead in conversations with their wives than with

strangers, and wives laughed less with spouses than with strangers and

used more disapproval of spouse, as did husbands. Winter, Ferreira,

and Bowers (1973) employed their decision-making task in a study of

married and unrelated couples. Married couples intruded upon and

interrupted each other more frequently than unrelated couples, and

unrelated couples listened respectfully to one another while married

couples were rude. Taken together, these findings suggest that

negatives exchanged indicate a pattern representative of the specific

marital relationship and social-learning history of the partners and

that such partners have the capability to interact more positively

given other interpersonal circumstances.

The present study is an investigation of two primary components

of communication--verbal content and nonverbal affect--and the impact

that various manifestations of each has upon the physiological

responses of individuals listening. The marital dyad is the

interpersonal unit in which the communication process is considered,


representing a persistent relationship with much invested in the

success of the resolution of conflicts around salient issues. Because

both sexes are represented in the marital dyad, the possible effects

of gender on verbal and nonverbal behavior and emotion need also be


Verbal Content

Thomas (1977) recognizes verbal behavior as a direct indicator of

what is commonly meant by communication. He describes human speech as

"probably the largest, most complex, and most finely honed operant

repertoire of all" (p. 3). According to Thomas, verbal stimuli

operate in the same manner as other stimuli in affecting behavior,

controlling behavior by means of reinforcement, punishment,

extinction, discriminative control, and the operation of eliciting

stimuli. Originally neutral verbal stimuli take on the ability of

producing conditioned responses after being paired with unconditioned

stimuli. Assuming that verbal behaviors are capable of eliciting

responses, investigators have studied characteristics of speech and

their effects on listeners. Matarazzo and Wiens (1972) found evidence

for a mutual verbal pattern matching during dyadic interaction. In an

interview setting, they found that the duration of the interviewee's

speech was strongly influenced by the duration of the interviewer's

speech. When the interviewer spoke for longer periods of time, the

interviewee did also, and when the interviewer spoke for brief periods

of time the interviewee likewise reciprocated. Reaction time is

another dimension of verbal behavior which appears to be

interdependent in dyadic interactions. Lauver, Kelley, and Froehle

(1971) found a direct relationship between client and therapist speech

latencies, with longer client latencies leading directly to longer

therapist latencies.

Responses which may be elicited by verbal stimuli are not limited

to the verbal domain. Goldband (1981) examined physiological as well

as self-reported responses to the impositions of latencies and

interruptions during dyadic interactions. Subjects receiving delayed

responses from their partners reported the partners as less empathic

and genuine, and showed greater spontaneous electrodermal activity

than subjects in the no-latency condition. This was not the case for

interruptions, which were actually rated as a somewhat positive


In order to adequately describe the impact of verbal messages it

would be necessary to develop a coding system more complex than

durations, latencies, and frequencies of vocalizations which

incorporates content. Such a system must have categories fine enough

for data reduction but broad enough to prevent "data overload"

(Gottman, 1979). Some studies (Winter and Ferreira, 1967, Waxler and

Mishler, 1965) attempted to adapt Bales' (1950) Interaction Process

Analysis (IPA) for use with couples and families. Due to the

multidimensional meanings of the categories requiring higher order

inferences, reliabilities were unacceptably low, and this system has

been largely abandoned.

One very simple content coding system frequently used is the

ratio of verbally coded agreements to disagreements. This index can

be reliably determined and has been successful at discriminating

distressed from nondistressed populations (Cheek, 1964; Lennard and

Bernstein, 1969; Mishler and Waxier, 1968; Riskin and Faunce, 1970).

However, it is not possible to reconstruct the gist of an interaction

which is described by two codes, a serious shortcoming for a coding

system (Gottman 1979).

Weiss, Hops, and Patterson (1973) developed the Marital

Interaction Coding System (MICS), consisting of 29 categories of

verbal and nonverbal behavior. Using this system, Vincent, Weiss, and

Birchler (1975) found that distressed couples displayed a

significantly greater ratio of negative and smaller ratio of positive

problem-solving behaviors than nondistressed couples. Klier and

Rothberg (1977), using a modified version of the MICS, could

differentiate distressed from nondistressed couples based on negative

but not positive behaviors. Birchler, Weiss, and Vincent (1975) could

discriminate between distressed and nondistressed couples based on

negative and positive exchanges during problem-solving discussions and

negative exchanges during casual discussions. Finally, the MICS has

been shown to be sensitive to changes in interactions resulting from

treatment programs with a problem-solving orientation (Jacobson, 1977,


There have been objections raised to the MICS scoring system,

however (Snyder, 1982). Primary among these are questionable

interrater reliability, an overabundance of categories, and a lack of

functional equivalence among categories; i.e. some are defined by

content, some by grammatical properties, some by intended emotional

impact, etc. In an attempt to develop a reliable, reasonably sized

system which could be used to code distinct messages of a couple's

verbal exchange and then assess the sequence of those messages Gottman

(1979) designed the Couples Interaction Scoring System (CISS). The

number of content categories was collapsed to 8, which Gottman reports

facilitates the assessment of verbal content. These categories are

Agreement (AG), Disagreement (DG), Communication Talk (CT),

Mindreading (MR), Problem Solving (PS), Summarizing Other (SO),

Summarizing Self (SS), and Problem Feeling (PF). Interrater

reliabilities for individual content codes ranged from .84 to .99 with

a mean of .95. Consistent with previous studies Gottman found lower

agreement-to-disagreement ratios for distressed than nondistressed

couples. However, by the application of sequential and nonsequential

analyses Gottman was also able to identify when specific content codes

were most likely to occur over the course of conflict resolution and

what verbal and nonverbal responses were precipitated by the various

codes. Three stages of conflict resolution were identified. The

beginning third, called "agenda building," consisted of MR and PF

codes. Mindreading statements are those which make attributions about

the thoughts, feelings, or behavior of the person being addressed

(e.g. "You don't like to visit my mother"). Such statements have

been presumed to interfere with successful communication (Thomas,

1977), as they increase defensiveness in the listener. Problem

feeling statements present the feelings and attitudes of the speaker

about a problem area. These statements have been thought to be

facilitative in conflict resolution, as they provide information about

the speaker's reactions to a problem while making no inferences about

the thoughts or feelings of the listener (e.g. "I feel angry when I

visit your mother"). During this phase, the task was getting problems

out for discussion. In the middle third of discussion, described as

"arguing," DG and SS codes were most frequent; disagreements and

summaries of opinions were aired. During the end third,

"negotiating," PS, AG, CT, and SO codes were prevalent. Coming to a

mutually satisfying resolution of the problem was the primary task of

this phase. MR and PF codes coupled with negative affect were found

to best discriminate distressed from nondistressed couples. Clinic

couples were more likely to enter cross-complaining sequences

(PF-PF-PF) than nonclinic couples, who were more inclined to engage in

validation sequences (PF-AG-PF). Contract sequences (PS-AG) were more

likely for nonclinic than clinic couples. In all, by comparing the

verbal response patterns of distressed and nondistressed couples

Gottman was able to identify sequences of transactions critical to the

success of conflict resolution.

Nonverbal Affect

The description of the role of nonverbal behaviors in regulating

the communication process offered by Izard, Wehmer, Livsey, and

Jennings (1969) sounds strikingly similar to statements made by

individuals explaining the impact of verbal content on communication:

"Positive affect generally enhances harmonious functioning of the

personality subsystems, while negative affect tends to create

dissonance among the various subsystems. Positive affect

leads to integrative behavior and effective functioning .

negative affect leads to discordance among personality subsystems,

non-integrated behavior, less effective functioning" (p. 278).

However, nonverbal behavior has received considerably more attention

in the marital interaction literature than verbal content, no doubt

because of the pervasive part it plays in the communication of

emotion. Mehrabian (1972) derived the following formula concerning

the delivery of affective messages which dramatically depicts the

importance of nonverbal cues: Total Affect = 7% Verbal + 38% Vocal +

55% Facial. Gottman, Markman, and Notarius (1977) found nonverbal

behavioral codes to be much more powerful discriminators of distressed

and nondistressed couples than any of the verbal codes. In the event

of inconsistency between verbal and nonverbal messages, the valence of

the nonverbal message will be the primary determinant of the

listener's response (Mehrabian and Wiener, 1967).

Accuracy and valence in nonverbal behavior seem to account for

much of the difference between satisfied and dissatisfied couples.

Kahn (1970) concludes that in the absence of accurate non-verbal

communication of intentions, attitudes, and wishes a marriage would be

particularly vulnerable to disharmony. In his comparison of the

nonverbal behavior of 21 dissatisfied and satisfied couples, the

latter group was found to be significantly more accurate in nonverbal

communications. Using the Primary Communication Inventory which

contains seven nonverbal items, Navran (1967) found better nonverbal

communication reported by happily married couples than unhappily

married couples. Regarding valence, positive messages occur more

frequently in well-adjusted couples than in poorly adjusted couples

(Noller, 1982), and negative messages are more common in poorly

adjusted than well-adjusted couples (Noller, 1982; Gottman, 1979).

While it is evident that nonverbal messages play a crucial part

in the communication process, the mechanisms by which they affect the

sequence of communicative events are less clear. The explanatory

framework most commonly invoked to assist in defining the relationship

of nonverbal behavior to emotion, communication and, consequently, to

marital satisfaction is one suggesting some sort of balance between

communicants. Argyle and Dean (1965) proposed an equilibrium theory

which suggests that interpersonal involvement is a cumulative product

of primarily nonverbal behaviors. This approach posits a kind of

pressure between individuals for the maintenance of a comfortable

level of interpersonal intimacy which, upon being reached, represents

a balance between approach and avoidance tendencies in the situation

(Heslin and Patterson, 1982). Any subsequent alteration of behaviors

will require compensatory adjustments in some other behavior to

maintain equilibrium. Research on the equilibrium model has been

supportive overall (Patterson, 1973). For example, in situations

where participants are approached closely or are the recipients of

high levels of gaze, compensatory adjustments (e.g. turning away,

decreasing eye contact, leaving the setting) frequently occur.

However, some studies have shown adjustments contrary to those

predicted by equilibrium theory (Jourard and Freedman, 1970, Breed,

1972, Chapman, 1975). Increased nonverbal intimacy produced

reciprocal rather than compensatory responses in these studies. That

is, when one member of a dyad increased intimacy it was followed by an

increase in nonverbal intimacy in the other member. Patterson (1976)

posits that such discrepancies can be explained by considering arousal

change as a mediating factor. Increased intimacy has been shown to

result in increases in physiological responding. McBride, King, and

James (1965) found increased electrodermal activity with closer

approaches, as did Nichols and Champness (1971) with increased eye

contact. Increases in gaze have been shown to increase heart rate

(Kleinke and Pohlen, 1971) and to change EEG reactions (Gale, Lucas,

Nissim, and Harpham, 1972). Given that changes in nonverbal intimacy

behaviors of Person A precipitate arousal change in Person B, it is

assumed that cognitions in Person B about himself, Person A, or the

situation direct that arousal change into a specific feeling state

(Schacter and Singer, 1962). The type of feeling state experienced by

Person B will direct the course of adjustment made to A's change in

intimacy (Heslin and Patterson, 1982). Negative feeling states (e.g.

anxiety, discomfort, embarrassment) would facilitate compensatory

responses, consistent with equilibrium theory. Positive states (e.g.

liking, love, relief) lead to a reciprocation of intimacy change in

the other. The principle of balance remains integral; both

reciprocity and compensation serve to maximize one's comfort or

satisfaction in an interaction (Heslin and Patterson, 1982).

One difficulty with Patterson's model is research demonstrating

reciprocity of both negative nonverbal cues and physiological arousal

and the absence of reciprocal responding given positive nonverbal

displays. In a study examining exchanges of pleasurable and

displeasurable behaviors Wills, Weiss, and Patterson (1974) concluded

that there was evidence for the reciprocation of displeasurable but

not pleasurable behaviors.' Gottman et al. (1977) found negative

affect reciprocity in distressed couples but failed to find positive

affect reciprocity in either distressed or nondistressed couples.

Gottman (1980) attributes the finding of negative reciprocity and the

failure to find positive affect reciprocity to a temporal

physiological linkage between interacting people created by negative

affect. Support for this notion comes from Kaplan, Burch, and Bloom

(1964) who correlated the electrodermal responses of interacting dyads

paired according to mutual like, mutual dislike, or mutual neutrality

using a sociometric peer rating. Electrodermal responding could be

predicted from one member to another only in the dyads paired on the

basis of mutual dislike. Levenson and Gottman (1982) collected

physiological data (skin conductance, heart rate, heart rate

variability, pulse transit time, pulse transit time variability) from

couples during two discussion periods, one in which events of the day

were discussed and a second during which the couple attempted to

resolve the major issue in their marriage. Multiple correlation

showed a moderate (R = .55) relationship between predictability

measures during the discussion of events of the day (a low conflict

task involving little negative affect) and marital satisfaction. For

the high conflict task in which negative affect and negative affect

reciprocity increase, a considerably stronger relationship was found

(R = .78). Gottman (1982) concluded from these findings that as the

degree of conflict increased so did the physiological coupling which

subsequently better discriminated satisfied and dissatisfied

marriages. He suggested that closeness in marriage is maintained by

symmetry in emotional responsiveness particularly in low-intensity

affective interactions. However, an absence of responsiveness in such

situations would lead to high levels of negative affect which would in

turn produce emotional withdrawal and negative affect reciprocity.

While there is disagreement on the mechanisms by which nonverbal

behaviors affect the communication process, common elements can be

seen among the various models. Nonverbal behaviors are powerful

elicitors of responses, requiring an adjustment in the behavior of the

recipient which may be reciprocal or compensatory. The nature of the

response is largely dependent upon the nature of the behavior, and

both of these are linked with the nature of the relationship which can

be conceptualized on a continuum of satisfaction. Finally, arousal is

implicated as integral in the communication process, being both cause

and effect of changes in nonverbal behavior.

Sex Differences

Communication between the sexes has long been recognized as a

frequently frustrating endeavor (Haas, 1979). One explanation for

this difficulty is that men and women have been said to speak

"different languages" (Jong, 1977; Reik, 1954). This section will

review studies of sex differences in verbal and nonverbal aspects of


Studies of spoken language can be divided into four areas: form,

topic, content and use (Bloom and Lahey, 1978). Form can be described

in terms of the units of sound (phonology), the units of meanings that

are words or inflections (morphology) and the ways in which units of

meanings are combined (syntax). One way in which men and women have

been found to differ in form is that the language of women tends to be

more conservative than men's. Men more readily coin and use new

phrases, pun, utter slang expressions, and use profanity and obscenity

(Jesperson, 1949). Lakoff (1973) observed that men use stronger

expletives than women, who use softer or weaker profanities. It

appears that the observation of these sorts of differences has a long

history. In Aristophanes' play The Ecclesiazuae (393 B.C.) a woman is

going to portray a man in her part of a conspiracy for women to

overthrow the government. She falters by swearing "By the two

goddesses!" Apparently this was the favorite oath of Athenian women;

ancient Greek men swore only by gods, never by goddesses. Women's

language has also been described as more "polite" than men's (Hartman,

1976; Lakoff, 1975), with women more frequently saying "please" and

"thank you" than men. In a study of 11 male-female dyads, Zimmerman

and West (1975) found that 98% of the interruptions and all the

overlaps were made by the male speakers.

Topic refers to the subject matter of a conversation. In a study

of blue-collar families Komarovsky (1967) found women to be more

inclined to discuss the family and social problems while men

reportedly preferred to talk about cars, sports, work, motorcycles,

and local politics. Klein (1971) made similar observations of the

working class in England. Bernard (1972) and Chesler (1972) found

that in mixed-sex conversations men rarely follow through on topics

initiated by women.

Content refers to the "categorization of the topics that are

encoded in messages" (Bloom and Lahey, 1978, p. 11) and differs from

topic in that topic refers to specific ideas, objects, and events

whereas content refers to the way the topic is referenced (Haas,

1979). Women's language has typically been described as more

emotional and evaluative than men's (Jesperson, 1949; Kramer, 1974;

Lakoff, 1975; Pei, 1969; Reik, 1954). Gleser, Gottschalk, and

Watkins (1959) collected samples of speech from 90 adult men and women

who were asked to describe an interesting or dramatic life experience.

Women were found to use more words implying feeling, emotion, or

motivation (whether positive, negative, or neutral). Women used more

negations and auxiliary words and made more self-references. Men made

more references to time, space, quantity, and destructive action.

Wood (1966) performed a content analysis on the speech of 36 college

students (18 male, 18 female) who described photographs of a man's

face. Males referred more directly to the details of the picture,

while females were more interpretive and subjective in their

descriptions. Barron (1971) studied the speech of teachers and

students during classroom activities. An analysis of the grammatical

cases employed revealed that females used more participative and

purposive cases and males more frequently used instrumental and

objective cases. Similar to Wood's study, men's speech described

objects and actions related to those objects and women discussed how

people felt and why they behaved as they did.

Bloom and Lahey (1978) describe "use" as consisting of "the

socially and cognitively determined selection of behaviors according

to the goals of the speaker and the context of the situation" (p.20).

It has been frequently suggested that male talk is instrumentally

oriented, concerning itself with the task of conveying ideas and

information (Bernard, 1972). Women's speech is commonly described as

tentative. Lakoff (1975) wrote that tag questions (e.g., "It's cold,

isn't it?) are more frequently used by women than men, as this form of

question avoids assertion and offers the addressee the option of

agreeing or disagreeing. Hartman (1976) found qualifiers to be

associated with females in her study of 70 year-old Maine natives. In

Swacker's (1975) study of the speech of college students while

describing pictures all subjects but one who used approximation with

numbers (e.g. "about four books") were female. Baumann (1976),

however, found no sex differences in the use of qualifying prefatory

statements or confirmatory tag questions in her analysis of adults in

various settings. Lakoff (1975) described the disparate ways in which

men and women make requests; men make commands and women state

requests. In an analysis of over 1,000 television commercials

Hennessee and Nicholson (1972) found that men gave nearly 90% of the

commands to buy a particular product.

Studies of sex differences in nonverbal behaviors have focused on

differential abilities in the accurate transmission and reception of

messages and on descriptions and frequencies of specific behaviors

used. One paradigm frequently employed in the study of communication

of affect via facial expressions has one person ("encoder" or

"sender") view a variety of types of scenes on slides or film. The

sender's facial expressions are usually unobtrusively videotaped while

viewing the stimuli and presented to another person ("decoder" or

"receiver"), whose task is to classify the sender's affective

expression into one of the categories of the stimuli being sent.

Using this paradigm, Buck, Savin, Miller, and Caul (1972) found female

pairs to be more effective than male pairs in the transmission and

reception of nonverbal emotional cues. Others using this paradigm

(Buck, 1975; Buck, Miller, and Caul, 1974; Fugita, Harper, and

Wiens, 1980; Woolfolk, Abrams, Abrams, and Wilson, 1979) have

likewise found females to be more accurate senders of spontaneous

facial expressions than males. Noller (1980) had each of 48 married

couples send ambiguous messages to each other employing positive,

neutral, or negative affect. Females were found to be better senders

particularly with regard to positive messages. Hall's (1978) review

of 75 investigations showed females to be superior to males in

decoding nonverbal communication in more studies than would be

expected by chance.

The difference in the ability to encode and decode nonverbal

messages has been attributed to noted discrepancies between men and

women in the use of nonverbal behaviors. In same sex dyads, females

interact more closely than males (Aiello and Aiello, 1974; Aiello and

Jones, 1971; Dosey and Meisels, 1969; Pellegrini and Empey, 1970).

Eye contact is also used more between females than between males

(Exline, 1963; Exline, Gray, and Schuette, 1965; Libby, 1970).

Females use touch more than males and respond more positively to it

(Fisher, Rytting, and Heslin, 1976; Jourard, 1966; Whitcher and

Fisher, 1979). Lochman and Allen (1981) looked at the nonverbal

communication of dating couples enacting standardized interpersonal

conflict vignettes from the Inventory of Marital Conflicts task (Olson

and Ryder, 1970). Compared to males, females were found to emit

significantly higher proportions of nonverbal behaviors and were more

consistent in matching self-ratings to their behavior. Noller (1982)

found sex differences in the use of positive, neutral, and negative

messages in her study of 48 marital dyads. Females were more often

coded positive or negative, while males were more frequently coded

neutral. In a study of six nondistressed couples discussing a salient

relationship issue, Notarius and Johnson (1982) found that wives

reciprocated their husbands' positive and negative speaker turns while

husbands followed their wives' positive speech with a neutral speaking

turn and showed no consistent pattern of responses to their wives'

negative speech.

Given the finding of more nonverbal expressivity in women, it is

necessary to discuss how physiological arousal, already identified as

integral to communicative processes, interacts with nonverbal

behavior. There are currently two models which make opposite

predictions about the direction of physiological responses concomitant

with nonverbal (particularly facial) expressivity. Discharge or

suppression models (Buck, 1979) posit that low levels of expressivity

lead to increased physiological reactivity as the emotional "energy"

which is not overtly channeled is discharged somatically. Jones

(1950) measured galvanic skin responses in elementary school children

exposed to eleven emotion-arousing situations. Those children who

showed low level electrodermal responses were described as high in

overt expressive behavior, while children showing large electrodermal

responses were described as low in overt expressive behavior. In a

study of 100 delinquent boys performing difficult motor tasks Landis

(1932) found that the 14 boys who became frightened, angry, cried, or

showed signs of pain exhibited a smaller frequency of electrodermal

responding than the whole group. Learmonth, Ackerly, and Kaplan

(1959) correlated skin potential fluctuations during various

interpersonal and physical stressors with personality variables

assessed by the MMPI and the Rorschach. They found that increases in

palmar potential during stress were negatively correlated with

personality variables related to expressivity and positively

correlated with variables dealing with restraint and curtailment of

unpleasant or prohibited feelings and actions. Lanzetta and Kleck

(1970) found negative relationships between expression of affect (as

measured by judges from facial and body cues) and skin conductance

liability. Buck, Miller, and Caul (1974) found that more accurate

(expressive) senders, who tended to be female, showed smaller skin

conductance and heart rate responses to emotionally oriented slides.

Buck, Savin, Miller, and Caul (1972) who replicated the finding of

females as "externalizers" and males as "internalizers" reported a

negative correlation between the senders' skin conductance responses

to emotive slides and the accuracy of the observers' ratings. Males

showed more frequent skin conductance fluctuations than women, which

has also been found in classical conditioning situations involving

shock (Graham, Cohen, and Shmavonian, 1966; Shmavonian, Yarmat, and

Cohen, 1965) and in a situation in which observers watched another

person receiving shocks (Craig and Lowrey, 1969). Notarius and

Johnson (1982) found that husbands had a greater tendency to show skin

potential responses to their wives' negative speech during conflict

resolution than did the wives to their husbands' negative speech. In

all, there is much support for the discharge model which suggests an

inverse relationship between overt expressivity and physiological


The second group of models describing the relationship among

facial displays and physiological reactions in response to emotional

stimuli are called "arousal" models. They assume a unidimensional

continuum of responses, thus suggesting a positive relationship

between nonverbal and physiological components of emotion. Darwin

(1872) postulated that overt expressive displays regulated the

emotional experience: "The free expression by outward signs of an

emotion intensifies it. On the other hand, the repression, as far as

possible, of all outward signs softens our emotions" (p. 22). Izard

(1971) and Gellhorn (1964) proposed that central neural circuits are

triggered by the face; these circuits in turn elicit autonomic

changes and the emotional experience as well. Some support has been

found for the arousal perspective in terms of finding relationships

between expressive behavior and reported quality of experienced

affect. Laird (1974) manipulated the facial expressions of subjects

into either a smile or a frown under the guise of a study of

electromyographic responding. He found that subjects' mood ratings

were affected by their facial expressions, as subjects rated their

mood as more positive and rated cartoons more humorous when smiling

than when frowning. Pasquarelli and Bull (1951) had five subjects

under hypnosis hold the physical expression of one emotion while

receiving the suggestion to experience another emotion. Subjects

reported an inability to feel the second emotion without altering

their physical posture. Studies implementing physiological indices

also provide support for the arousal hypothesis. Lanzetta,

Cartwright-Smith, and Kleck (1976) instructed subjects to conceal or

exaggerate the facial display associated with anticipating and

receiving painful shocks of varying intensities. Self-reports of

pain, skin conductance responses, and expressive behavior were all

found to covary. Using a subject-controlled pain tolerance paradigm,

Colby, Lanzetta, and Kleck (1977) also found changes in the facial

expression of pain to be accompanied by parallel changes in autonomic

responding. Vaughan and Lanzetta (1981) instructed subjects to either

amplify or inhibit their facial expressions in response to observing a

model being shocked. They found that the "amplify" group showed more

skin conductance responsivity, heart rate acceleration and activity in

response to the model's expression of pain than the "inhibit" group

and a third group given no facial instructions.

Notarius, Wemple, Ingraham, Burns, and Kollar (1982) discussed

why some studies have found direct relationships between facial

expressivity while others have found inverse relationships. One

explanation given is design differences; between-subjects comparisons

generally support an inverse relationship, and within-subjects

analyses support a direct relationship (Buck, 1980). Methodological

differences have been noted, as some studies expose subjects to

slides, some to shock, some to threat of shock, and some to motion

pictures. Notarius et al. suggest that the different emotional

elicitors may each be associated with different reaction patterns.

External constraint of emotional expression is another difference

across studies. Those studies in which the subjects' expressions were

not constrained have supported an inverse relationship between overt

expression and physiological reactivity, while studies in which the

subjects' expressions were manipulated through instructions offered

support for arousal models. The final explanation offered for

inconsistency in the findings relates to the role of overt expression

in emotional situations. Because facial expressions are subject to

voluntary control, they can be modified in response to the "display

rules" of a given context (Ekman and Friesen, 1969). Autonomic

responses, however, are not easily controlled and represent the

effects of previous emotional conditioning. Thus, Notarius et al.

conclude: "We would expect a strong relationship between self-report

of emotion and autonomic reactivity in response to emotional stimuli

but a variable relationship among facial expression, self-report, and

physiological reactivity as a function of the display rules operative

in any given situation" (p.401). In proposing research designs for

examining the typical relationship of nonverbal expressivity and

autonomic reactivity, they recommend unconstrained emotional

expression in interpersonally relevant situations with as few

environmental constraints that might inhibit expression as possible.

Statement of the Problem

From the preceding review it is evident that the course of

marital communication largely depends upon verbal and nonverbal

parameters of the messages being exchanged and which spouse is

sending/receiving a given message. It is also apparent that while

much research has implicated the role of physiological arousal in the

mediation of communicative efforts, surprisingly few investigators

have implemented the recording of autonomic responses during dyadic

interaction. In recognizing the empirical nature of the question of

which types of content actually do elicit emotional responses,

Hovland, Janis, and Kelley (1953) called for a twofold program of

research. The first step would be the development of a classification

scheme for separating emotional appeals from others by using a content

analysis for classifying assertions according to their semantic

meanings. The second step would be conducting controlled

experimentation to differentiate those contents which evoke emotional

reactions from those which evoke other reactions. In his discussion

of nonverbal dimensions of interactions and their effects on emotional

responsivity Patterson (1976) echoed the sentiments of Hovland et al.,

noting the absence of studies taking continuous data on arousal levels

before, during, and after manipulations of the interaction and stating

that data from physiological recording, behavioral reactions and

verbal reports should ideally be collected. In response to the

mandate by Hovland and his colleagues for systems of classification of

interactive behaviors, a number of verbal and nonverbal coding schemes

have been developed and have been discussed previously (Bales, 1950,

Weiss et al., 1973, Gottman, 1979). Unfortunately, investigations

have faltered in taking the second critical step in the Hovland et al.

and Patterson plans for identifying verbal and nonverbal precipitants

of emotional responding. In order to ascertain the impact on a

listener of a message sent with affect X and content Y, one must first

ensure that the particular affect-content configuration will occur.

Using preplanned vignettes Gottman (1979) has demonstrated that affect

and content can be manipulated through instructional control.

However, no one has attempted to modify the affect and content of

personal, salient relationship issues in couples and observe the

subsequent physiological impact of such messages on the listener. The

present study proposed to do that; content was initially presented to

the experimenter in written form by a spouse around a significant

problem area in the marriage and transformed so that its content was

either mind reading or problem feeling (the two categories found by

Gottman (1979) most likely to occur at the beginning of an interaction

and which best discriminate distressed from nondistressed couples).

Affect was manipulated by videotaped modeling and coaching by which

the subject was directed to present the modified message in an

effectively neutral or negative fashion.

Based on the preceding reviews, the following hypotheses were

proposed on the effects of content, affect, and sex of the sender on

physiological responses of the listener during the reading of the

problem description:

Content Effect

Assuming that mind reading statements invoke a defensive set on

the part of the listener, and that a defensive set elicits

physiological responding, problems presented with mind reading content

would elicit in the listener greater physiological reactivity (more

skin resistance responses, more skin conductance responses, larger

skin resistance level changes, and greater heart rate increases) than

problem feeling messages.

Affect Effect

Given the findings on nonverbal affect reciprocity and symmetry

in emotional responsiveness, problems presented with negative affect

would elicit in the listener greater physiological reactivity than

effectively neutral messages.

Content X Affect Interaction

Problems presented with problem feeling content and neutral

affect would elicit least physiological reactivity, followed by, in

increasing order, mind reading content/neutral affect, problem feeling

content/negative affect, mind reading content/negative affect.

Sex effect

Because of the lack of external constraint of emotional

expression on the listener and the concomitant inverse relationship

suggested between affective display and physiological responsivity in

a paradigm of this type, husbands, who are less overtly expressive,

would show greater physiological reactivity to messages than wives.

Greatest discrepancies between husbands and wives would be found on

mindreading messages with negative affect, followed by, in decreasing

order, problem feeling messages with negative affect, mindreading

messages with neutral affect, and problem feeling messages with

neutral affect.




Subjects were sixty-four married couples recruited by means of

advertisements in the classified sections of local newspapers in

Gainesville, Florida. Prior to their participation in the experiment,

couples were informed of the general nature of the experiment and the

time involved (approximately 2 hours) by means of a phone contact.

Couples were paid twelve dollars for their participation.


Two adjacent rooms were used in the study: a subject room 12' 8"

long and 11' 9" wide, in which participants completed questionnaires

and discussed a problem area, and an observation room, 12' 8" long and

9'6" wide, which contained physiological and videotape recording


Physiological recordings of heart rate and skin resistance were

taken from both members of each dyad using a Narco Physiograph Model

DMP-4B. Heart rate was obtained from a pair of stainless steel

electrodes attached to the left and right midforearms, using a Narco

Hi-Gain coupler Model 7171 and a Narco Channel amplifier Model 7070.

On the Hi-Gain coupler, the Time Constant and Gain was set at .03 X

100, and on the Channel amplifier sensitivity was set at 20 mv/cm, and

the low pass filter was set at 10 Hz. Redux Creme was the contact

medium, and the electrodes were secured with adhesive collars. A

Narco GSR coupler Model 7175 supplied a constant 10 microamp current

to obtain skin resistance. Beckman Standard Biopotential electrodes

were attached to the distal phalanxes of the first and second fingers

of the non-dominant hand using 3M Micropore 1/2 inch surgical tape.

Johnson and Johnson K-Y Jelly was the contact medium. Paper speed was

2.5 cm/sec.

Videotape recordings were made using a Panasonic WV-342 video

camera mounted on a tripod and placed across the room from the

subjects, getting a full body view of both participants. Sound was

monitored using two Realistic Highball 2 dual-impedance microphones.

Audio and video signals were recorded on a Sony VO-1600 3/4 inch

videocassette recorder located in the observation room.

A Condor Digital Clock showing hours, minutes, and seconds was

placed between and behind the dyad. A Sony video monitor, driven by

the VTR in the observation room, was used for showing an affect

coaching tape to one spouse. Red and green signal lights, located

between the dyad, were controlled by a switch in the observation room.

An 8 ohm speaker, equidistant from both spouses, was driven by a

Realistic MPA-20 P.A. amplifier, amplifying two two-second

presentations of wideband white noise (rise time = 100 msec.)

pre-recorded and played through the system on a Sony TC-520CS cassette


Design and Procedure

A 2 X 2 X 2 factorial design was used, with affect (neutral or

negative), sex of spouse sending (male or female), and content (mind

reading or problem feeling) as the factors. Each couple was

consecutively assigned in a randomly predetermined order to its

specific condition.

Two experimenters were used in data collection. Experimenter A

(EA) was responsible primarily for interacting directly with the

couple, i.e. reading instructions to them, giving them

questionnaires, debriefing them. Experimenter B (EB) operated the

physiograph and video recording equipment. Upon the arrival of the

couple, the experimenters introduced themselves, described the study,

and obtained consent. EA read this passage:

Because of the necessity of keeping experimental procedures
standardized, it will be necessary for me at times to read you
specific instructions. We will start with each of you completing
several questionnaires. Your responses on these questionnaires will
be kept confidential among the research team. Throughout the
experiment it will be important that you not discuss, comment on, or
look at each other's questionnaire items. Please read and follow all
directions carefully. I will gladly answer any questions you might
have. Please be sure to write your name in the indicated place on
each questionnaire. Here is your first questionnaire.

Four questionnaires were administered including the Couple's

Problem Inventory, a short checklist of problem areas with duration

and intensity ratings for each problem category, a general information

sheet containing demographic information, and the Locke-Williamson

Marital Relationship Inventory (Locke and Williamson, 1958), a measure
of marital adjustment. Spouses were seated in such a way so as to

prevent them from viewing each other's responses. After these

questionnaires were completed, EA read both Couple's Problem

Inventories to himself and tried to identify the problem rated as most

troublesome. If the same problem was ranked highest by both spouses,

the couple was informed of this; if there was disagreement regarding

which problem was most distressing, the top three rated problems were

read aloud and EA helped the couple reach agreement on a mutually

troublesome problem. Once the problem was agreed upon, EA read this:

Please complete this Problem Description sheet in the manner
described in the instructions at the top of the sheet. Be sure to
read the directions carefully and follow them exactly. If you have
any questions let me know. Please write clearly or print. Write
about the problem you just indicated was most troublesome.

Each spouse then completed a Problem Description form which

required a written description of the selected problem. The problem

was to be written directly addressing the other spouse "as though you

were trying to communicate to him/her your feelings about the

problem." The problem description was to be between 100 and 130 words,

and numbers on the problem description form corresponding to each word

placement were provided to facilitate staying within these boundaries.

The spouse assigned to the "listener" condition was then escorted
to a waiting room and given a questionnaire to complete. The spouse

assigned to the "sender" condition remained in the subject room and

filled out the same questionnaire. Both experimenters then worked

together in another room on transforming the content of the problem

description to being either "mind reading" or "problem feeling,"

depending upon the assigned condition. Transform guidelines were as

follows: 1) The number of words in the transformed problem

description should be between 80% and 120% of the number used in the

sender's original problem description. 2) No fewer than three

mindreading or problem feeling thought units should be incorporated

into the text. 3) Mindreading transforms may contain some problem

feeling statements but must contain more mind reading than problem

feeling phrases. 4) Problem feeling transforms should contain no

mindreading thought units. 5) Similarity of the transformed problem

description should be maximized by using original words, sentences,

and paragraph syntax where possible. (See Appendix 1 for samples of

problem descriptions and their transforms.) After the sender finished

filling out the questionnaire he or she was instructed to watch a

videotape describing the affective manner (i.e. either emotionally or

neutrally, depending upon the assigned condition) in which he or she

was to present the problem written about to the spouse. At this

point, EA turned on the video monitor, while EB started the affect

coaching videotape from the observation room. After the tape was

finished, EA presented the transformed problem description to the

sender and read the following:

In order to assure scientific standardization, I have made a few
slight modifications of the problem description you wrote earlier. I
would like you to read this standardized problem description to
yourself twice and then to read it to me in an
(emotional/non-emotional) manner similar to that you just viewed on
the television.

Subjects in the neutral affect condition were coached to speak in

a non-emotional, businesslike, matter-of-fact manner with moderate

loudness, pitch, rate, inflection, and enunciation. Subjects in the

negative affect condition were coached to speak in a manner conveying

anger and/or impatience, presenting the message loudly with high

pitch, blaring timbre, fast rate, irregular inflection, and clipped

enunciation. The experimenter praised the subject when he or she was

affect-appropriate during the practice readings and had the sender

practice more affect-appropriate cues when necessary (i.e. if a

neutral-assigned sender was being negative or if a negative-assigned

sender was not emotional enough).

Following coaching, the sender was told that the affective set

practiced was to be maintained only during the actual reading of the

problem description to the spouse and that during the discussion

subsequent to the reading of the problem, both spouses were to express

themselves in whatever manner seemed natural. Both the sender (in the

subject room) and the listener (in the waiting room) were then

dismissed to wash their hands with soap and water in preparation for

the attachment of the sensors (Martin and Venables, 1980). While both

subjects were out of the room, EB turned on the video camera. When

the subjects returned they were seated in separate chairs at right

angles to and approximately three feet from each other with the

listener on the sender's right. Skin resistance and heart rate

electrodes were attached to both spouses as described previously.

Both subjects completed the Eysenck Personality Questionnaire (Eysenck
and Eysenck, 1964) and another questionnaire. At this time, EB

activated the physiograph, identifying zero baselines and calibrating

both skin resistance channels with an applied 10 kohm signal

deflecting each pen 10 mm. After the verification of all

physiological signals by EB and the completion of the questionnaires

by the couple, EA read the following passage:

In a few minutes (sender) will read the problem description which
he/she wrote earlier in the experiment. Immediately following his/her
reading this problem the two of you are asked to come to a mutually

satisfactory solution of the problem. Spend at least 10, but no more
than 15 minutes mutually discussing the problem. Be natural, and try
to come to a real solution that satisfies both of your needs. The
only thing that we ask is that you keep your hands, arms, and body
relatively quiet so that the sensors will not be disturbed. You can,
however, use your face and voice in whatever manner seems most
natural. What and how you say and do is up to you after (sender)
reads his/her problem description. (Listener), you must not talk
until (sender) has read all of his/her problem. (Sender) will
indicate that he/she is done reading by saying, 'OK, this is the way I
view the problem.' Then the two of you are free to talk when and how
you want for from 10 to 15 minutes as you try to reach a satisfying
solution to the problem. Any questions? (Sender), you are to begin
reading in several minutes when the light (point to it) turns green.
Please be silent until then. Prior to the green signal light and your
conversation two tones, a soft and a loud one, will be presented, once
each. Relax now until the green light.

EA then set the digital clock to synchronize with all recording

systems and joined EB in the observation room. The video recorder was

activated and the pre-recorded white noise segments were presented,

with a 75 dB sound at two minutes and a 95 dB sound at three minutes.

At five minutes, EA switched on the green light and the sender read

the Sender's Standard Message (Appendix 2) with neutral affect.

Following the reading of the Standard Message the sender read the

problem description with neutral or negative affect, depending on the

condition being implemented. A ten minute conflict resolution

followed the reading of the problem description, after which both

spouses completed the Conflict Recall Report (a set of questions

concerning each spouse's reactions to their discussion) and two other
questionnaires. Couples were informed of the research questions of

interest and thanked for their participation.


1 The other questionnaire, to be used in another investigation, was
the first half of the Loevinger Sentence Completion Test (Loevinger
and Wessler, 1970) a measure of ego development. Hereafter, all
procedures administered and questionnaires given without direct
relevance to this study are noted.

2 The Problem Descriptions of both senders and listeners will be
content analyzed in a subsequent study.

3 The Trait form (i.e. "How you usually feel") of the Profile of Mood
States (Lorr, Daston, and Smith, 1967).

4 The State form of the Profile of Mood States (i.e. "How you feel
right now").

5 Heart rate and electrodermal responses to the tones will be examined
in a subsequent investigation.

6 The Conflict-Recall form of the Profile of Mood States (i.e. "How
you felt during your just-completed attempt to solve the problem with
your spouse"), and the second half of the Loevinger scale.



Manipulation Check

In order to ensure that content and affect were effectively

manipulated, undergraduate students blind to experimental conditions

were used as raters. For content, one rater attempted to determine

whether each transformed problem description was mindreading or

problem feeling, if it contained at least three content-appropriate

thought units, and if it was between 80 and 120% of the length of the

original. Of all the transforms 95.3% were judged to be mindreading

or problem feeling as assigned. Untransformed problem descriptions

were an average of 118.0 words in length, 102.4% of the mean length of

transformed problem descriptions (115.2 words). For affect, two

raters viewed the reading of the problem description on videotape and

rated the intensity of affect on a scale of 0 to 3 with higher scores

signifying more intense affect. Scores of 0 and 1 were called

neutral; scores of 2 and 3 were called negative. At least one of the

two raters judged the sender to be neutral or negative as assigned in

96.9% of the cases. Rater l's ratings agreed with the assigned

condition in 85.9% of the cases; Rater 2's ratings agreed in 88.9% of

the cases. Rater 1 agreed with Rater 2 in 77.8% of the cases.

Perceived similarity of the sender's transformed problem

description to the original problem description was determined by

examining the response to the following item on the Conflict Recall

Report: "Please recall your original complaint as you wrote it. If

you had only spoken your complaint, how similar would it have been to

what you wrote down as your complaint?" Senders reported a similarity

rating of 6.83 on a scale of 1 (not at all) to 9 (just like),

suggesting a fairly high degree of similarity. An analysis of

variance revealed no significant differences among the various

conditions in response to this item.

To determine how typical the discussion period was, responses to

the question "How much like his/her typical problem-solving behavior

with you was your partner's behavior/style?" were examined. On a

scale of 1 (not at all like) to 9 (just like), the mean rating was

7.22, suggesting that the discussion was more than less typical. An

analysis of variance grouped by sex, content, and affect showed no

significant differences among respondents in any condition on

responses to this item.


In obtaining data for 64 couples, it was necessary to run a total

of 78 couples. Data were not usable from 14 couples for the following

reasons: failure to read the standard message (2), failure to employ

negative affect (2), laughter during the standard message or problem

description (3), failure to interact (2), inappropriate verbalizations

not germane to the selected conflict area (1), health problems

directly affecting physiological recording (1), equipment failure (1),

and improper recording of skin resistance levels or event

onsets/offsets (2).

Mean age of participants was 30.96 years (SD=11.27). Spouses had

been married an average of 6.78 years (SD=9.29) and had 15.51 years of

education (SD=2.42). Mean marital satisfaction as measured by the

Locke- Williamson Marital Research Inventory was 98.61 (SD=12.83).

The average intensity of the specific problem area discussed on a

scale of 0 to 100 was 50.96 (SD=30.59).

Data Analyses

Analyses of variance were performed on heart rate, skin

resistance responses, skin conductance responses, and the magnitude of

change in skin resistance level during the silent waiting period to

determine if subjects in all conditions were at comparable levels

physiologically prior to the standard message and problem description.

Repeated measures analyses of variance were performed for each of the

four response variables (heart rate, skin resistance responses, skin

conductance responses, magnitude of change in skin resistance level)

for two sequential events: presentation of the standard message, and

presentation of the problem description. Content (mindreading vs.

problem feeling), affect (neutral vs. negative), and sex of listener

(wives vs. husbands) were the grouping factors. Because of

correlations found between physiological measures during the problem

description and both marital satisfaction on the Locke-Williamson and

Neuroticism on the Eysenck scale (see Table 1), analyses of covariance

were also run with scores from each of these self-report measures as

covariates. The intensity of the problem discussed on a scale of 0 to

100 from the Couple's Problem Inventory was also used as a covariate

to control for variations in severity of problems discussed among


Pearson Correlations of Marital Satisfaction and Neuroticism
with Heart Rate, Skin Resistance, and Skin Conductance Responses
During the Problem Description

Heart Rate Skin Resistance Skin Conductance
Responses Responses






Eysenck r=.242 r=.130
Neuroticism p<.054 p<.305
N=64 N=64

subjects. Responding during the standard message was used as a

covariate to control for possible differences at that point in the

procedure and thus allow inferences to be made regarding the

contribution of the problem description alone to subjects' responding.

Statistical results reported are analyses of variance unless a

covariate is noted.

Physiological Scoring

Heart rate, frequency of skin resistance responses, frequency of

skin conductance responses, and magnitude of skin resistance responses

were scored for listeners during the last thirty seconds of the five

minute silent waiting period, the last thirty seconds of the standard

message, and the entire reading of the problem description. For heart

rate, the number of R-waves between epoch lines marked on the

physiograph record was counted to the nearest one-tenth of a beat.

For the thirty second silent waiting and standard message epochs this

total was multiplied by two to yield beats per minute. Beats per

minute were calculated for the problem description epoch by taking the

number of beats during the problem description, dividing by the number

of seconds duration, and multiplying this quotient by 60. Frequencies

of skin resistance responses were determined by counting the number of

decreases in resistance of at least 2 kilohms (2 mm) from a slope of

0. This response was expressed as the number of deflections per

thirty seconds. For the thirty second silent waiting and standard

message epochs this was simply a total of the number of deflections.

For the problem description, the number of deflections was divided by

the duration in seconds of the problem description and multiplied by

thirty. Skin conductance responses of .2 micromhos or more were

scored for the same intervals as skin resistance responses. Unlike

skin resistance responses, their occurrence was dependent upon skin

resistance level; thus a 2 mm deflection was a .2 micromho skin

resistance response if the skin resistance level was equal to 100

kilohms. At greater resistance levels, a larger deflection was

required to be counted as a response, and at levels less than 100

kilohms a smaller deflection was needed to be counted as a .2 micromho

response. Skin resistance response magnitudes were defined as the

difference between the peak upward excursion (minimum resistance) and

the lowest excursion (maximum resistance) preceding that particular

peak for a given epoch.

Silent Waiting Period Data

Heart rate was marginally different during the silent waiting

period for husbands and wives, F (1,56)=3.58, p<.064, with wives

having slightly higher heart rates (Wives: Mean HR = 76.20 bpm, SD =

8.09; Husbands: Mean HR = 71.61 bpm, SD = 10.66). A borderline

effect was also found for content, F (1,56)= 3.09, p<.084, with

subjects in the mindreading conditions having lower heart rates during

the waiting period than those in the problem feeling conditions

(Mindreading: Mean HR = 71.77 bpm, SD = 9.10; Problem feeling: Mean

HR = 76.04 bpm, SD = 9.65). A borderline sex by content by affect

interaction was found, F (1,56)=3.75, p<.058, with wives in the

neutral problem feeling condition having the highest heart rate (Mean

= 80.65 bpm, SD = 7.99) and husbands in the negative mindreading

condition having the lowest heart rates (Mean = 65.05 bpm, SD = 6.78).

No significant differences for any combination of conditions were

found for the frequency of skin resistance or skin conductance

responses or for the magnitude of skin resistance level change during

the waiting period.

Time of Assessment Effects

Skin resistance responses and skin conductance responses varied

as a function of time of assessment (standard message, problem

description) regardless of sex, content, and affect conditions [Skin

resistance: F (1,56) =6.11, p<.017; skin conductance: F

(1,56)=5.61, p<.021]. Both skin resistance and skin conductance

responses increased significantly from standard message to problem

description [Skin resistance: t(63)=2.48, p<.016; skin conductance:

t(63)=2.28, p<.026] (Figures 1 and 2).

Content Effects

There was a tendency for a greater number of skin conductance

responses in the mindreading than problem feeling conditions

[F(1,55)=2.87, p<.096, covariate = standard message] (Figure 3). Time

and content also interacted [F(1,56)=2.98, p<.090], with the change

from standard message to problem description being significant in the

mindreading condition but not in the problem feeling condition,

t(63)=2.71, p<.011 (Figure 4). No main effects for content were found

with analysis of variance.

.3.0 -


2.0 -

1.0 -

Standard Problem
Message Description

Figure 1. Skin resistance responses as a function of time
of assessment.


S3.0 -


Standard Problem
Message Description

Figure 2. Skin conductance responses as a function of time
of assessment.


3.0 -


o 2.0 -


z 1.0 -

Mindreading Problem

Figure 3. Skin conductance responses as a function of content.


3.0 -

o 2.0 -


z 1.0 -

Standard Problem
Message Description
Figure 4. Skin conductance responses as a function of time
of assessment and content.


A Problem Feeling

Affect Effects

A main effect was found for affect on the number of skin

conductance responses at a borderline level of significance

[F(1,55)=3.79, p<.057, covariate = standard message] with more skin

conductance responses occurring in the negative than in the neutral

condition (Figure 5). Time by affect interactions were found for both

skin conductance and skin resistance responses [F(1,56)=5.66, p<.021

and F(1,56)=3.45, p<.068 respectively] (Figures 6 and 7). The number

of responses was not different between neutral and negative conditions

at either standard message or problem description; however, the

change from standard message to problem description was a significant

increase for mindreading statements [Skin conductance: t=3.41,

p<.002; skin resistance: t=3.07, p<.004]. There was no significant

change in frequency of responses in the problem feeling condition. No

main effects for affect were found with analysis of variance.

Affect by Content Interaction

Frequency of skin resistance responses varied as a function of

the interaction of affect and content [F (1,56)=4.16, p<.046] (Figure

8). Subjects in the negative mindreading condition had the most skin

resistance responses, more than those in the negative problem feeling

condition [t (30)=2.09, p<.046]. There were no differences between

subjects in the mindreading neutral and negative, mindreading neutral

and problem feeling neutral, or problem feeling neutral and problem

feeling negative conditions. This interaction was not significant

using responses during the standard message as a covariate.


3.0 -

2.0 -


z 1.0-

Neutral Negative

Figure 5. Skin conductance responses as a function of affect.


S3.0 -



Z 1.0 -

Figure 6. Skin conductance responses as a function of time
of assessment and affect.


A Negative




Figure 7. Skin resistance responses as a function of time
of assessment and affect.

A Negative

3.0 -

2.0 -



S3.0 -

o 1

0 2.0 -



Neutral Negative

Figure 8. Skin resistance responses as a function of content
and affect.

MR = Mindreading

PF = Problem Feeling

Sex Effects

Wives showed greater physiological reactivity to the experimental

situation than husbands as reflected by significant differences in the

number of electrodermal responses and a tendency toward higher heart

rate across procedures [Skin resistance responses: F (1,49)=4.38,

p<.042, covariate = problem intensity; skin conductance responses: F

(1,56)=5.93, p<.018; skin resistance level change: F (1,56)=5.10,

p<.028; heart rate: F (1,56)=3.29, p<.075] (Figures 9, 10, 11, and

12.) Analysis of covariance using responding during the standard

message as a covariate showed borderline sex differences for heart

rate and skin conductance responses [Heart rate: F (1,55)=2.85,

p<.097; skin conductance: F (1,55)=3.82, p<.056] and no differences

on skin resistance responses and magnitude of skin resistance level


Sex by Content Interaction

Husbands and wives show differential sensitivity to content for

frequency of both skin resistance and skin conductance responses [Skin

resistance: F(1,56)=5.70, 2<.020; skin conductance: F (1,49)=5.33,

p<.025, covariate=problem intensity] (Figures 13 and 14). Wives in

the problem feeling condition showed the most electrodermal activity,

having a significantly greater number of skin resistance and skin

conductance responses than husbands in the same condition, t

(30)=2.95, p<.006 and t (30) =2.87, p<.007 respectively. There were

no significant differences between wives in the problem feeling and

mindreading conditions. Husbands showed significantly more skin

80 -

79 -

S78 -

S77 -


75 -
4- I
74 -

73 -

72 -

71 -

Wives Husbands

Figure 9. Heart rate as a function of sex.

3.0 -


1 .0 -h

Wives Husbands

Figure 10. Skin resistance responses as a function of sex.


u 3.0 -


i 2.0 -


Wives Husbands

Figure 11. Skin conductance responses as a function of sex.

14.0 -

^ 13.0 -

12.0 -

) I __

Wives Husbands

Figure 12. Magnitude of change in skin resistance level
as a function of sex.



2.0 -



z 1.0 -

Mindreading Problem

Figure 13. Skin resistance responses as a function of sex
and content.

W = Wives

H = Husbands

3.0 -




1.0 -

Mindreading Problem
Figure 14. Skin conductance responses as a function of sex
and content.

W = Wives

H = Husbands

resistance responses and marginally more skin conductance responses in

the mindreading condition than in the problem feeling condition, t

(30)=2.54, p<.017 and t (30)=1.80, p<.083. No sex by content

interactions were found using analysis of covariance with responses

during the standard message as a covariate.

Sex by Content by Affect Interaction

Heart rates of husbands and wives were differentially responsive

to the affect and the content conditions, F(1,55)=4.63, p<.036,

covariate=neuroticism (Figure 15). Wives in the neutral problem

feeling condition had the highest heart rates, marginally higher than

those of husbands in the same condition (p<.069), wives in the

negative problem feeling condition (p<.090), and significantly greater

than those of husbands in the negative mindreading condition, who had

the lowest heart rates (2<.016). Husbands in the negative mindreading

condition had marginally lower rates than husbands in the negative

problem feeling condition (p<.081). Wives and husbands in the neutral

mindreading condition did not differ in heart rate, nor did wives and

husbands in the negative problem feeling condition. This interaction

was not statistically significant for skin conductance or skin

resistance responses; however, a comparison of skin resistance

responses (Figure 16) with heart rate shows that within the neutral

affect conditions relative positions of husbands' to wives' responses

are the same to both content categories; for negative affect

conditions, the relative positions of husbands' to wives' responses

are reversed. Skin conductance responses followed a similar pattern

except in the negative mindreading condition (Figure 17). No sex by

content by affect interactions were found using analysis of covariance

85 -
84 -
83 -
82 -
80 -
. 79 -
\ 78 -
, 77 -
S76 -
75 -
73 -
72 -
71 -
70 L
Neutral Neutral Negative Negative
Mind- Problem Mind- Problem
Reading Feeling Reading Feeling
Figure 15. Heart rate as a function of sex, content,
and affect.
W = Wives

H = Husbands

4.0 -


N 23.0

2 .0-

Neutral Neutral Negative Negative
Mind- Problem Mind- Problem
Reading Feeling Reading Feeling

Figure 16. Skin resistance as a function of sex, content,
and affect.

W = Wives

H = Husbands


4.0 -

\ 3.0 -


CN 2.0

o I

Neutral Neutral Negative Negative
Mind- Problem Mind- Problem
Reading Feeling Reading Feeling

Figure 17. Skin conductance as a function of sex, content,
and affect.
W = Wives

H = Husbands


with responses during the standard message as a covariate.

Nonsignificant Interactions

None of the following interactions for response variables were

found to be statistically significant: sex by affect, time by sex,

time by sex by content, time by sex by affect, time by content by

affect, time by sex by content by affect.



The present study attempted to ascertain some of the

psychophysiological consequences of listening to spouses presenting

experimentally controlled complaints about their marital

relationships. From the preceding results, it is apparent that the

research paradigm did provide an opportunity to observe spouses

present and respond to issues of genuine emotional salience. The act

of simply listening to a spouse reading a personal specific marital

complaint elicited responding at a level significantly greater than

when listening to that spouse reading a standard message. Thus, even

though the communicative environment created was somewhat artificial,

the responses of the participants suggest that this situation was

meaningful and potent.

The use of both analyses of variance and covariance made it

possible to examine potential response differences of various

conditions both during and prior to the problem description period.

When the responses occurring during the standard message were used as

a covariate, a main effect for content was found. Coupled with

significant time by affect and time by content interactions with both

statistical procedures, it is evident that these effects are

attributable to the messages themselves. Thus, across sex and affect

conditions, mindreading statements can be said to have elicited


significant increases in skin conductance activity, while problem

feeling statements did not. This finding supports the inclusion of

content in an analysis of message impact and suggests that the

relative importance of verbal content to nonverbal affect as an

interpersonal stimulus has perhaps been underestimated. Using the

same covariance analysis, nonverbal affect emerged as a main effect,

as messages delivered with negative affect were responded to

physiologically with increased electrodermal responding, while neutral

messages failed to elicit increased responding. This finding is

consistent with the work of Levenson and Gottman (1982) who found

evidence for increased physiological activity concomitant with

increased intensity of affect. Sex differences found with the

analysis of variance were attenuated but not eliminated with the

covariance analysis, suggesting that the specific marital complaints

were augmenting the already discrepant responding of the different

sexes. Wives showed greater overall responsivity than husbands,

contrary to hypothesis. One explanation for this pattern of results

is that the listeners, even though given no instructions about

generating emotions (as in the Lanzetta studies) or no controls on

what they could say during the conflict resolution period, may have

been constrained in that they had to listen to their spouses and could

not respond until cued by the speaking spouse. Thus, the arousal

model, which assumes a positive relationship between nonverbal and

physiological components of emotion, would probably have been a more

appropriate theoretical basis on which to have made predictions.

It is possible that some of the effects found may not be

explainable exclusively by the specific marital problem presented by

the senders, and the statistical procedures employed also help to

identify these sources of variance. It may be that group differences

at the points of the waiting period or the standard message may

account in part for effects found. The slight heart rate differences

during the silent waiting period for sex by content by affect could

explain the finding of the sex by content by affect interaction with

an analysis of variance and a failure to find this interaction when

the initial differences, carried over to the standard message, are

eliminated with covariance procedures. However, another possibility

is that group differences at the points of the waiting period and the

standard message were not independent of the problem description, i.e.

that there was a preliminary "leakage" by the coached sender of the

content and affect of the message about to be delivered (Ekman and

Friesen, 1969). One indication of this is the finding that analyses

of variance were sensitive to interactions of factors (e.g. sex by

content, content by affect, sex by content by affect), yet these

interactions were not significant when the responses during the

standard message were used as a covariate. Thus, covarying out

responses during the standard message not only appeared to enhance

finding main effects, but it also reduced the possibility of finding

effects related to sex differences or affect and content preliminarily

leaked during the standard message or perhaps as early as the silent

waiting period. The design of the present study might be improved to

address this question by initially measuring physiological response

levels prior to affect coaching and content transformation, thus

eliminating the possibility of nonverbal leakage of the nature of the

upcoming problem description.

In spite of the aforementioned effects of content and affect

examined separately, it is evident from the data that various

combinations of these emotional parameters yield different

physiological results, and these variables should be considered as

they interact. Subjects in the negative condition with mindreading

content did show the greatest number of skin resistance responses, as

predicted. However, when the content of the message was problem

feeling, negative affect subjects showed the fewest skin resistance

responses. Thus, knowing only the affect of a given message would not

be sufficient for explaining the effect to be expected. Gottman

(1979) found that mindreading statements could be facilitative or

destructive, depending upon the affect used. With negative affect,

mindreading statements were seen as personal attacks, and were more

characteristic of distressed couples. With neutral content, however,

mindreading statements functioned as "feeling probes," engaging the

other spouse in a nonthreatening manner to agree or disagree with the

attribution in the process of working through a conflict, and these

were more characteristic of nondistressed couples.

Sex and content were also found to interact, with husbands and

wives responding differentially to content conditions. Wives showed

the greatest electrodermal responsivity in the problem feeling

conditions, while husbands showed more responses in the mindreading

conditions. This discrepancy could be related to differences in what

Bloom and Lahey (1978) call "use," in a sense paralleling the "display

rules" described by Ekman and Friesen (1969) for nonverbal behavior.

Husbands may be more accustomed to hearing the more personally

expressive problem feeling messages from their wives and respond more

to the novel mindreading messages. Similarly, because of the tendency

of men to use more instrumental language describing object relations

(Barron, 1971), conceivably including the relationships of other

people to behaviors, wives may more frequently hear husbands make

attributions about their wives' thoughts, feelings, or behaviors than


The manner in which sex differences influenced the interaction of

content and affect shows yet another important dimension to be

considered in explaining responsivity. Husbands in neutral affect

conditions responded similarly if content was mindreading or problem

feeling. Wives, however, showed high levels of responsivity in the

neutral problem feeling condition. When wives were assigned to

effectively negative conditions, the content seemed to have little

influence over physiological responding. Conversely, husbands in

effectively negative conditions showed a large discrepancy in response

depending upon content. Heart rate was actually lowest in husbands in

the negative mindreading condition, contrary to what was expected.

One explanation for the disparity in responsivity for husbands and

wives is that physiological responding may be paralleling what Gottman

(1979) refers to as "cognitive editing processes." In a sequential

analysis of verbal and nonverbal behaviors of nondistressed spouses he

discovered complementary roles of wives and husbands in conducting

conflict resolution. Upon hearing negative affect, husbands were

likely to be "negative listeners" (i.e. to show nonverbal disapproval

of a message while listening) and to reciprocate the negative affect;

wives, however, were less likely to be negative listeners and also

less likely to return negative affect. During discussion of low

conflict issues, Gottman found that neutral affect on the part of

husbands was likely to be perceived as negative by wives and responded

to with disagreement or counterproposals with negative affect. The

reverse was not found to be true; wives' neutral behavior was not

followed by negative affect or content by the husbands. Gottman

concluded that husbands and wives had different tasks in the process

of successful problem resolution. Because of their lack of reciprocal

responding to negative affect wives were seen as responsible for

breaking up negative affect cycles in the interest of resolution.

Similarly, when discussing low conflict issues, husbands were deemed

responsible for maintaining a climate of agreement. When these

response patterns are present in both spouses, the probability for

successfully handling either high or low conflict areas is increased.

Findings from the present study of nonclinic couples suggest a similar

complementarity in wives' and husbands' physiological response

patterns. Wives listening to an expressive problem feeling statement

with neutral affect were unable to respond overtly while listening,

but may have been showing physiologically the disagreement and

counterarguments that would typify their hearing this sort of message.

Husbands show a depressed level of heart rate coupled with the highest

number of electrodermal responses in the negative mindreading

condition. This pattern of responding is suggestive of orienting

which is often seen when presented with a novel stimulus. It is

likely that more instrumental mindreading language coupled with

negative affect is unusual for nonclinic wives. When negative affect

is used by wives it is probably more frequently used in conjunction

with problem feeling statements; husbands in this study, in fact,

showed highest heart rates to this combination, suggesting a more

defensogenic response.

The relationship of heart rate and electrodermal activity is an

interesting one, appearing linear in some situations and inverse in

others. In the neutral conditions, heart rate and electrodermal

activity covaried, with both heart rate and electrodermal responses

being higher in problem feeling than mindreading conditions (c.f.

Figures 15 and 16). The opposite was true for the negative affect

condition; heart rates in problem feeling conditions were relatively

higher than in mindreading conditions, while skin resistance responses

were more frequent for mindreading than problem feeling conditions.

This relationship is evident when considering the sex of the listener

as well; relative positions of level of responding for wives to

husbands are similar when considering neutral affect conditions and

inverse in the negative affect conditions (c.f. Figures 15 and 16).

It may be that given increased affect intensity differential demands

are made on cardiovascular and electrodermal systems, further modified

according to the sex of the listener. The correlational data

presented in Table 1 offer another example of how heart rate and

electrodermal measures may be convergent or divergent depending upon

the task. When related to the level of reported marital satisfaction,

couples with higher satisfaction scores had more electrodermal

responses and lower heart rates than less satisfied couples. When

correlating the dimension of neuroticism, conceptualized by Eysenck as

arousability, with heart rate and electrodermal measures, both classes

of responses showed a somewhat positive relationship. One might

cautiously conclude that the pattern of reduced heart rate with

increased electrodermal activity is a crude index of attentive

listening, characteristic of satisfied couples, but that defensogenic

responding (i.e. high heart rate and electrodermal responding) is

more typical of neurotic individuals. Less cautiously, it is apparent

that measures of heart rate and electrodermal activity offer more

information when considered together and probably do not provide

sufficient information when only one or the other is used.

The present findings have implications for treatment programs

which are currently being implemented for marital distress. It is

apparent that, at least at a psychophysiological level, the impact of

a given message may vary significantly depending upon the sex of the

speaker and listener. If physiological activity is related to

information processing and conflict resolution, this finding conflicts

with programs which suggest, for example, always speaking in a

neutral, business-like manner for best conflict-resolution results.

It is also apparent that the use of mindreading messages does not

categorically elicit large responses and may, in fact, lead to a

reduction in physiological activity. Some researchers and clinicians,

recognizing the importance of individual and sex differences (e.g.

Noller, 1980) have suggested teaching husbands and wives different

communication skills as necessary, such as teaching husbands to be

more expressive.

Future work in this area should directly investigate the

relationship between physiological responsivity and actual conflict

resolution to see if the data provided by physiological indices are

useful in predicting, on an immediate or delayed basis, the outcome of

attempts at resolution. Real-time investigations of unconstrained

interactions around salient relationship issues would be the ideal

situations to examine, as manipulations such as those used in the

present study are somewhat artificial and may interfere with

interpreting conflict resolution solely as a function of the couples'

communication skills and styles. Given that information derived from

psychophysiological measures is found to be both predictive and

nonredundant, the potential arises for implementing intervention

strategies using otherwise untapped data from the interactants as

tools for resolution.



Untransformed Problem Description

You become defensive when I point out things when they're not

meant at you. I get negative and draw away from you. I know I can be

difficult but I'd like you to express your thoughts anyway. I wish

you were more playful and active than you sometimes are. I feel we

haven't truly given our relationship time to develop because we

haven't yet had the free time or money to work towards our goals.

Sometimes you give in to my wants on little things and it hurts in

other areas. I'd much prefer you to speak up or remind me because I

am forgetful.

Problem Description Transformed to Mindreading Content

You become defensive when I point out things when they're not

meant at you. I get negative and draw away from you. You know I can

be difficult but I'd like you to express your thoughts anyway. I wish

you were more playful and active than you sometimes are. You know we

haven't truly given our relationship time to develop because we

haven't yet had the free time or money to work towards our goals.

Sometimes you give in to my wants on little things and it hurts in

other areas. I'd much prefer you to speak up more than you do or

remind me because I am forgetful.

Untransformed Problem Description

I think we have a problem concerning recreational activities in

that we hardly ever interact with each other indoors or out. You know

how sports minded I am, if you would only spend two or three hours a

weekend watching televised sports or playing a sport outdoors with me

our relationship in this area would be greatly improved. Not only

would we spend more time together, but our health could only improve.

In addition to that, (our son) could sometimes participate which would

keep him occupied in a constructive manner. I know you feel that if

you spent time watching sports with me or playing sports outside

nothing would get done around the house but I think we could set down

and make a schedule to alleviate that kind of problem.

Problem Description Transformed to Problem Feeling Content

I think there is a problem concerning recreational activities in

that there is hardly any interaction either indoors or out. I am very

sports minded and feel our relationship in this area would be greatly

improved if two or three hours a weekend were spent watching televised

sports or playing a sport outdoors together. Not only would more time

be spent together, but I feel our health could only improve. In

addition to that, (our son) could sometimes participate which would

keep him occupied in a constructive manner. I know it seems that

spending time watching sports with me or playing sports outside would

keep things from getting done around the house, but I think a schedule

could alleviate that kind of problem.



I am going to read to you two things. The first is a series of

statements I have been asked to read concerning this marital research

project. The second is the problem description that I wrote earlier,

and which we are to discuss after I read. Here is the first of the


This study's procedures took over a year to design and develop.

The study was designed to learn about aspects of marriage that have

not until now been studied carefully and in a scientific manner.

Therefore our participation in this study will hopefully provide

psychologists with knowledge that will eventually help them to more

effectively help married couples lead more enjoyable relationships.

Our participation in the present research is, thus, greatly

appreciated by the research staff.

The researchers are always concerned with trying to improve their

experiments. Therefore, it is helpful to them if we give them our

impressions about the study and how it might be improved. We will,

therefore, have an opportunity to talk about the experiment at its


Now I am going to read my problem description.


Aiello, J.R., and Aiello, T. The development of personal space:
Proxemic behavior of children 6 through 16. Human Ecology, 1974, 2,

Aiello, J.R., and Jones, S.E. Field study of the proxemic behavior of
young school children in three subcultural groups. Journal of
Personality and Social Psychology, 1971, 19, 351-356.

Argyle, M., and Dean, I. Eye contact, distance, and affiliation.
Sociometry, 1965, 28, 289-304.

Bales, R.F. Interaction process analysis. Cambridge, Mass.:
Addison- Wesley, 1950.

Barron, N. Sex-typed language: The production of grammatical cases.
Acta Sociologica, 1971, 14, 24-72.

Baumann, M. Two features of "women's speech?" In B.L. Dubois and I.
Crouch (Eds.), The sociology of the language of American women. San
Antonio, Tex.: Trinity University Press, 1976.

Bernard, J. The sex game. Englewood Cliffs, N.J.: Prentice-Hall,

Birchler, G., Weiss, R., and Vincent, J. Multimethod analysis of
social reinforcement exchange between maritally distressed and
nondistressed spouse and stranger dyads. Journal of Personality and
Social Psychology, 1975, 31, 349-360.

Bloom, L., and Lahey, M. Language development and language disorders.
New York: Wiley, 1978.

Breed, G. The effect of intimacy: Reciprocity or retreat? British
Journal of Social and Clinical Psychology, 1972, 11, 135-142.

Buck, R. Nonverbal communication of affect in children. Journal of
Personality and Social Psychology, 1975, 31, 646-653.

Buck, R. Individual differences in nonverbal sending accuracy and
electrodermal responding: The externalizing-internalizing dimension.
In R. Rosenthal (Ed.), Skill in nonverbal communication. Cambridge,
Mass.: Oelgeschlager, Gun, and Hain, Publishers, 1979.

Buck, R.W. Nonverbal behavior and the theory of emotion: The facial
feedback hypothesis. Journal of Personality and Social Psychology,
1980, 38, 811-824.

Buck, R.W., Miller, R.E., and Caul, W.F. Sex, personality, and
physiological variables in the communication of affect via facial
expression. Journal of Personality and Social Psychology, 1974, 30,

Buck, R.W., Savin, V.J., Miller, R.E., and Caul, W.F. Communication

of affect through facial expressions in humans. Journal of
Personality and Social Psychology, 1972, 23, 362-371.

Chapman, A.J. Eye contact, physical proximity, and laughter: A
reexamination of the equilibrium model of social intimacy. Social
Behavior and Personality, 1975, 3, 143-155.

Cheek, F.E. The "schizophrenogenic" mother in word and deed. Family
Process, 1964, 3, 155-177.

Chesler, P. Women and madness. Garden City, N.Y.: Doubleday, 1972.

Colby, C.Z., Lanzetta, J.T., and Kleck, R.E. Effects of the
expression of pain on autonomic and pain tolerance responses to
subject-controlled pain. Psychophysiology, 1977, 14, 537-540.

Craig, K., and Lowrey, H.J. Heart rate components of conditioned
vicarious autonomic responses. Journal of Personality and Social
Psychology, 1969, 11, 381-387.

Darwin, C. The expression of the emotions in man and animals.
London: John Murray, 1872.

Dosey, M.A. and Meisels, M. Personal space and self-protection.
Journal of Personality and Social Psychology, 1969, 11, 93-97.

Easterbrook, J.A. The effect of emotion on cue utilization and thee
organization of behavior. Psychological Review, 1959, 66, 183-201.

Eisenson, J., Auer, J.J., and Irwin, J.V. The psychology of
communication. New York: Appleton-Century-Crofts, 1963.

Ekman, P., and Friesen, W.V. Nonverbal clues to deception.
Psychiatry, 1969, 32, 88-106.

Exline, R.V. Explorations in the process of person perception:
Visual interaction in relation to competition, sex, and need for
affiliation. Journal of Personality, 1963, 31, 1-20.

Exline, R., Gray, D., and Schuette, D. Visual behavior in a dyad as
affected by interview content and sex of respondent. Journal of
Personality and Social Psychology, 1965, 1, 201-209.

Eysenck, H.J., and Eysenck, S.B.G. Eysenck Personality Inventory.
San Diego: San Diego Educ. Indust. Testing Service, 1964.

Fisher, B.A. Perspectives on human communication. New York:
Macmillan Publishers, 1978.

Fisher, J.D., Rytting, M., and Heslin, R. Hands touching hands:
Affective and evaluative effects of interpersonal touch. Sociometry,
1976, 39, 416-421.

Fugita, B.N., Harper, R.G., and Wiens, A.N. Encoding-decoding of
noverbal emotional messages: Sex differences in spontaneous and

enacted expressions. Journal of Nonverbal Behavior, 1980, 4, 131-145.

Gale, A., Lucas, B., Nissim, R., and Harpham, B. Some EEG correlates
of face-to-face contact. British Journal of Social and Clinical
Psychology, 1972, 11, 326-332.

Gellhorn, E. Motion and emotion: The role of proprioception in the
physiology and pathology of emotions. Psychological Review, 1964, 71,

Gleser, G.C., Gottschalk, L.A., and Watkins, J. The relationship of
sex and intelligence to choice of words: A normative study of verbal
behavior. Journal of Clinical Psychology, 1959, 15, 182-191.

Goldband, S. Imposed latencies, interruptions, and dyadic
interaction: Physiological response and interpersonal attraction.
Journal of Research in Personality, 1981, 15, 221-232.

Gottman, J.M. Marital interaction--Experimental investigations. New
York: Academic Press, 1979.

Gottman, J.M. Consistence of nonverbal affect and affect reciprocity
in marital interaction. Journal of Consulting and Clinical
Psychology, 1980, 48, 711-717.

Gottman, J.M. Emotional responsiveness in marital conversations.
Journal of Communication, 1982, 32, 108-120.

Gottman, J., Markman, H., and Notarius, C. The topography of marital
conflict: A sequential analysis of verbal and nonverbal behavior.
Journal off Marriage and the Family, 1977, 39, 461-477.

Graham, L.A., Cohen, S.I., and Shmavonian, B.M. Sex differences in
autonomic responses during instrumental conditioning. Psychosomatic
Medicine, 1966, 28, 264-271.

Haas, A. Male and female spoken language differences: Stereotypes
and evidence. Psychological Bulletin, 1979, 86, 616-626.

Hall, J.A. Gender effects in decoding nonverbal cues. Psychological
Bulletin, 1978, 85, 845-857.

Hartman, M.A. A descriptive study of the language of men and women
born in Maine around 1900 as it reflects the Lakoff hypotheses in
"Language and Woman's Place." In B.L. Dubois and I. Crouch (Eds.),
The sociology of the languages of American women. San Antonio, Tex.:
Trinity University Press, 1976.

Hennessee, J., and Nicholson, J. NOW says: TV commercials insult
women. New York Times Magazine, May 28, 1972, pp. 12-13, 48-51.

Heslin, R., and Patterson, M.L. Nonverbal behavior and social
psychology. New York: Plenum Press, 1982.

Hovland, C.I., Janis, I.L., and Kelley, H.H. Communication and

persuasion- Psychological studies of opinion change. New Haven,
Conn.: Yale University Press, 1953.

Izard, C.E. The face of emotion. New York: Appleton-Century-Crofts,

Izard, C.E., Wehmer, G.M., Livsey, W.J., and Jennings, J.R. Affect,
awareness, and performance. In Markel, N. (Ed.),
Psycholinguistics--An introduction to the study of speech and
personality. Homewood, Ill.: Dorsey Press, 1969.

Jacobson, N.S. Problem solving and contingency contracting in the
treatment of marital discord. Journal of Consulting and Clinical
Psychology, 1977, 45, 92-100.

Jacobson, N.S. Specific and nonspecific factors in the effects of a
behavioral approach to the treatment of marital discord. Journal of
Consulting and Clinical Psychology, 1978, 46, 442-452.

Janis, I.L. Effects of fear arousal on attitude change: Recent
developments in theory and experimental research. In L. Berkowitz
(Ed.), Advances in experimental social psychology. New York:
Academic Press, 1967,, pp. 167-225.

Jesperson, 0. Language. New York: Macmillan, 1949.

Jones, H.E. The study of patterns of emotional expression. In M.C.
Reymert (Ed.), Feelings and emotions. New York: McGraw-Hill, 1950.

Jong, E. How to save your own life. New York: Holt, Rinehart, and
Winston, 1977.

Jourard, S.M. An exploratory study of body-accessibility. British
Journal of Social and Clinical Psychology, 1966, 5, 221-231.

Jourard, S.M., and Freedman, R. Experimenter-subject "distance" and
self- disclosure. Journal of Personality and Social Psychology, 1970,
15, 278-282.

Kahn, M. Nonverbal communication and marital satisfaction. Family
Process, 1970, 9, 449-456.

Kaplan, H.B., Burch, N.R., and Bloom, S.W. Psychological covariation
and sociometric relationship in small peer groups. In P.H. Leiderman
and D. Shapiro (Eds.), Psychobiological approaches to social
behavior. Stanford, Calif.: Stanford University Press, 1964.

Keltner, J.W. Interpersonal speech-communication: Elements and
Structures. Belmont, Calif.: Wadsworth Publishing, 1970.

Klein, J. The family in "traditional" working-class England. In M.
Anderson (Ed.), Sociology of the family. Baltimore, Md.: Penguin
Books, 1971.

Kleinke, C.L., and Pohlen, P.D. Affective and emotional responses as

a function of other persons gaze and co-operativeness in a two-person
game. Journal of Personality and Social Psychology, 1971, 17,

Klier, J.L., and Rothberg, M. Characteristics of conflict resolution
in couples. Paper presented at the annual meeting of the Association
for the Advancement of Behavior Therapy, Atlanta, December, 1977.

Komarovsky, M. Blue-collar marriage. New York: Vintage Books, 1967.

Kramer, C. Folklinguistics. Psychology Today, June, 1974, pp 82-85.

Laird, J.P. Self-attributions of emotion: The effects of expresive
behavior on the quality of emotional experience. Journal of
Personality and Social Psychology, 1974, 29, 475-486.

Lakoff, R. Language and woman's place. Language in Society, 1973, 2,

Lakoff, R. Language and woman's place. New York: Colophon/Harper
and Row, 1975.

Landis, C. An attempt to measure emotional traits in juvenile
delinquency. In K.S. Lashley (Ed.), Studies on the dynamics of
behavior, 1932.

Lanzetta, J.T., Cartwright-Smith, J., and Kleck, R.E. Effects of
nonverbal dissimulation on emotional experience and autonomic arousal.
Journal of Personality and Social Psychology, 1976, 33, 354-370.

Lanzetta, J.T., and Kleck, R.E. Encoding and decoding of nonverbal
affect in humans. Journal of Personality and Social Psychology, 1970,
16, 12-19.

Lauver, P.J., Kelley, J.D., and Froehle, T.C. Client reaction time
and counselor verbal behavior in an interview setting. Journal of
Counseling Psychology, 1971, 18, 26-31.

Learmonth, G.J., Ackerly, W., and Kaplan, M. Relationship between
palmar skin potential during stress and personality variables.
Psychosomatic Medicine, 1959, 21, 150-157.

Lennard, H.L., and Bernstein, A. Patterns in human interaction. San
Francisco: Jossey-Bass, 1969.

Levenson, R.W., and Gottman, J.M. Why marriages fail: Affective and
ph physiological patterns in marital interaction. Unpublished
manuscripts, University of Illinois, 1982.

Libby, W.L. Eye contact and direction of looking as stable individual
differences. Journal of Experimental Research in Personality, 1970,
4, 303-312.

Lochman, J.E., and Allen, G. Nonverbal communication of couples in
conflict. Journal of Research in Personality, 1981, 15, 253-269.

Locke, H.J., and Williamson, R.C. Marital adjustment: A factor
analysis study. American Sociological Review, 1958, 23, 562-569.

Loevinger, J., and Wessler, R. Measuring ego development-Volume 1-
Construction and use of a sentence completion test. San Francisco:
Jossey-Bass, 1970.

Lorr, M., Daston, P., and Smith, I.R. An analysis of mood states.
Educational and Psychological Measurement, 1967, 27, 89-96.

Martin, I., and Venables, P.H. Techniques in psychophysiology. New
York: Wiley, 1980.

Matarazzo, J.D., and Wiens, A.N. The interview: Research on its
anatomy and structure. Chicago: Aldine-Atherton, 1972.

McBride, G., King, M.G., and James, J.W. Social proximity effects on
GSR in adult humans. Journal of Psychology, 1965, 61, 153-157.

Mehrabian, A. Nonverbal communication. New York: Alpine-Atherton,

Mehrabian, A., and Wiener, M. Decoding of inconsistent
communications. Journal of Personality and Social Psychology, 1967,
6, 109-114.

Mishler, E.G., and Waxler, N.E. Interaction in families: An
experimental study of family process in schizophrenia. New York:
Wiley, 1968.

Navran, L. Communication and adjustment in marriage. Family Process,
1967, 6, 173-184.

Nichols, K.A., and Champness, B.G. Eye gaze and the GSR. Journal of
Experimental Social Psychology, 1971, 7, 623-626.

Noller, P. Misunderstandings in marital communication: A study of
couples' nonverbal communication. Journal of Personality and Social
Psychology, 1980, 39, 1135-1148.

Noller, P. Channel consistency and inconsistency in the
communications of married couples. Journal of Personality and Social
Psychology, 1982, 43, 732-741.

Notarius, C.I., and Johnson, J.S. Emotional expression in husbands
and wives. Journal of Marriage and the Family, 1982, 4, 483-489.

Notarius, C.I., Wemple, C., Ingraham, L.J., Burns, T.J., and Kollar,
E. Multichannel responses to an interpersonal stressor:
Interrelationships among facial display, heart rate, self-report of
emotion, and threat appraisal. Journal of Personality and Social
Psychology, 1982, 43, 400-408.

Olson, D.H., and Ryder, R.G. Inventory of marital conflicts: An
experimental interaction procedure. Journal of Marriage and the


Family, 1970, 32, 443-448.

Pasquarelli, B., and Bull, N. Experimental investigations of the
mind-body continuum in affective states. Journal of Nervous and
Mental Disease, 1951, 113, 512-521.

Patterson, M.L. Stability of nonverbal immediacy behaviors. Journal
of Experimental Social Psychology, 1973, 235-245.

Patterson, M. An arousal model of interpersonal intimacy.
Psychological Review, 1976, 83, 235-245.

Pei, M. Words in sheep's clothing. New York: Hawthorn Books, 1969.

Pellegrini, R.T., and Empey, J. Interpersonal spatial orientation in
dyads. Journal of Psychology, 1970, 76, 67-70.

Reik, T. Men and women speak different languages. Psychoanalysis,
1954, 2, 3-15.

Riskin, J., and Faunce, E.E. An evaluative review of family
interaction research. Family Process, 1972, 11, 365-455.

Ryder, R.G. Husband-wife dyads versus married strangers. Family
Process, 1968, 7, 233-238.

Schacter, S., and Singer, J.E. Cognitive, social, and physiological
determinants of emotional state. Psychological Review, 1962, 69,

Shmavonian, B., Yarmat, A., and Cohen, S. Relations between the
autonomic nervous system and central nervous system in age differences
and behavior. In Behavior, aging, and the nervous system.
Springfield, Ill: C.C. Thomas, 1965.

Sillars, A., and Parry, D. Stress, cognition, and communication in
interpersonal conflicts. Communication Research, 1982, 9, 201-226.

Snyder, D.K. Advances in marital assessment: Behavioral
communications and psychometric approaches. In C.D. Spielberger and
J.N. Butcher (Eds.), Advances in personality assessment, Vol. 1.
Hillsdale, New Jersey: Lawrence Erlbaum Associates, 1982.

Swacker, M. The sex of the speaker as a sociolinguistic variable. In
B. Thorne and N. Henley (Eds.), Language and sex: Difference and
dominance. Rowley, Mass.: Newbury House, 1975.

Thomas, E.J. Marital communication and decision making: Analysis, a
assessment, and change. New York: The Free Press, 1977.

Vaughan, K.B., and Lanzetta, J.T. The effect of modification of
expressive displays on vicarious emotional arousal. Journal of
Experimental Social Psychology, 1981, 17, 16-30.

Vincent, J.P., Weiss, R.H., and Birchler, G.R. A behavioral analysis

of problem solving in distressed and nondistressed married and
stranger dyads. Behavior Therapy, 1975, 6, 475-487.

Watzlawick, P., Beavin, J.H., and Jackson, D.D. Pragmatics of human
communication. New York: W.W. Norton and Company, 1967.

Waxier, N.E., and Mishler, E.G. Scoring and reliability problems in
Interaction Process Analysis: A methodological note. Sociometry,
1965, 29, 32-49.

Weiss, R.L., Hops, H., and Patterson, G.R. A framework for
conceptualizing marital conflict: A technology for alterint it, some
data for evaluating it. In L.A. Hamerlynch, I.C. Handy, and E.J.
Mash (Eds.), Behavior Change: The fourth Banff conference on behavior
modification. Champaign, Ill: Research Press, 1973

Whitcher, S.J., and Fisher, J.D. Multidimensional reaction to
therapeutic touch in a hospital setting. Journal of Personality and
Social Psychology, 1979, 37, 87-91.

Wills, T.A., Weiss, R.L., and Patterson, G.R. A behavioral analysis
of the determinants of marital satisfaction. Journal of Consulting
and Clinical Psychology, 1974, 42, 802-811.

Winter, W.D., and Ferreira, A.J. Interaction process analysis of
family decision-making. Family Process, 1967, 6, 155-172.

Winter, W.D., Ferreira, A., and Bowers, N. Decision-making in married
and unrelated couples. Family Process, 1973, 12, 83-94.

Wood, M. The influence of sex and knowledge of communication
effectiveness on spontaneous speech. Word, 1966, 22, 112-137.

Woolfolk, A.E., Abrams, L.N., Abrams, D.B., and Wilson, G.T. Effects
of alcohol on the nonverbal communication of anxiety: The impact of
beliefs on nonverbal behavior. Environmental Psychology and Nonverbal
Behavior, 1979, 3, 205-218.

Zimmerman, D.H., and West, C. Sex roles, interruptions and silences
in conversation. In B. Thorne and N. Henley (Eds.), Language and
sex: Difference and dominance. Rowley, Mass.: Newbury House, 1975.


David Arthur Hermecz was born on February 10, 1957, to Julius and

Frances Hermecz in Fairhope, Alabama. He graduated from McGill-Toolen

High School in Mobile, Alabama, in 1975 and attended Auburn University

from 1975 to 1979. Majoring in both psychology and sociology, he was

a member of Psi Chi psychology honorary, Alpha Kappa Delta sociology

honorary, Omicron Delta Kappa leadership honorary, Alpha Phi Omega

service fraternity, and was the trainer of the Auburn University

mascot, War Eagle IV, from 1976 to 1979. Upon receiving traineeships

from the U.S. Public Health Service in 1979-1980 and again in

1980-1981, he was involved in research in behavioral medicine under

the supervision of Barbara G. Melamed. He received the Master of

Arts in psychology from the University of Florida in 1982. He was

awarded a grant from the Division of Sponsored Research in 1983 to

investigate psychosocial interventions for children receiving medical

treatment for cancer and has published work on behavioral medicine

with an emphasis on children's fears, psychophysiology, and imagery in

Behavior Therapy, the Journal of Consulting and Clinical Psychology,

and Psychosomatic Medicine. He is a student member of the Association

for the Advancement of Behavior Therapy, the Society for

Psychophysiological Research, and the Christian Association for

Psychological Studies.

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.

ug avis, Chairman
Professor of Clinical Psychology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.

SLawrence J. Si6#y
Associate Pro ssor of Clinical

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the-degree
of Doctor of Philosophy.

W. Keith Berg j 7/
Associate Professor of

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.

Wilse B. Web6
Graduate Research Professor
of Psychology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.

Mickie N.Edwardson
Professor of Journalism
and Communication

This dissertation was submitted to the Graduate Faculty of the College
of Health Related Professions and to the Graduate School, and was
accepted as partial fulfillment of the requirements for the degree of
Doctor of Philosophy.

August 1985

Dean, College of Health Related

Dean for Graduate Studies and

11II11 IIII 11111111 111111 lill 11111 11 1111
3 1262 08554 3691