Citation
Self-awareness and cigarette smoking

Material Information

Title:
Self-awareness and cigarette smoking
Creator:
Walker, Gordon Rexal, 1954-
Copyright Date:
1981
Language:
English

Subjects

Subjects / Keywords:
Audiences ( jstor )
Behavior modeling ( jstor )
Carving ( jstor )
Cigarettes ( jstor )
Diaries ( jstor )
Modeling ( jstor )
Personality psychology ( jstor )
Philosophical psychology ( jstor )
Psychological attitudes ( jstor )
Self consciousness ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
07784819 ( OCLC )
ABS1017 ( LTUF )
0028116815 ( ALEPH )

Downloads

This item has the following downloads:


Full Text

SELF-AWAJRENESS AND CIGARETTE SMOKING:
THE INTERACTION OF PUBLIC AND PRIVATE DETERMINANTS






BY


GORDON REXAL WALKER, JR.


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



UNIVERSITY OF FLORIDA


1981





ACKNOWLEDGEMENTS


The author would like to thank the following

experimenters for their invaluable assistance in

collecting data for this research: Ruth Honig, Jennifer

Jeffers, Hindi Kliein, Roberta Ross, Anthony Campos, Eric

Corley, Minou Hirnampour, Paul Wiltse, George Piemonte,

Bill Allen, Shishir Kurup, Jeff Burns, and Sharon Stern.

Additional thanks are due to those who agreed to smoke in

this experiment for the advancement of science.


A final debt of gratitude is due to Dr. William J.

Froming. His patience with me over these years in

graduate school and our trips to the pool hall have made

the entire experience much more enjoyable.





TABLE OF CONTENTS

PAGE


ACKNOWJLEDGEMENTS...........**********************

ABSTRA~CT ..... ..... **** ***** **** ***** **** ***** ****

CHAPTER

I INTRODUCTION............................

Theories of Self-Focused Attention.....2
The Objective Self-Awareness
Perspective............ ......
The Self-Consciousness Perspective....4
Initial Formulations.............4
Modified Theory.....................
The Encoding Perspective.............7

Comparison of the Models...............8

An Alternative Model. .................11
Developing the Model.................11
Example..............................12
Implications.....................14

Evaluation of the Model...............16
A Test of the Model...................1
Alternative Predictions..............21
Objective Self-Awareness
Perspective..... ...........2
Self-Consciousness Perspective......22
Encoding Perspective..............222


II METHOD.................................24

Overview............................2

Subjects..............................25

Procedure..........................2

Design................................29

III HYPOTHESES.......................30


iii





IV RESULTS................................33

V DISCUSSION.............................48

Experimental Procedures...............48

Implications for Self-Awareness
Models..............................50

Implications for the Interactive
Model..............................51

Implications for Other Models........52
Objective Self-Awareness
Perspective......................52
Encoding Perspective................53
Self-Consciousness Perspective......54

Conclusion............................55

REFERENCES. .....................................5


APPENDICES

A QUESTIONS USED IN SELECTING SUBJECTS...60

B SPECIFIC EXPERIMENTAL PROCEDURES.......61

C DEBRIEFING PROCEDURES..................63



BIOGRAPHICAL SKETCH.. .................................. 4





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


SELF-AWARENESS AND CIGARETTE SMOKING:
THE INTERACTION OF PUBLIC AND PRIVATE DETERMINANTS

By

Gordon Rexal Walker, Jr.

March, 1981



Chairman: William J. Froming, Ph.D.

Major Department: Psychology

Several broad conceptualizations of self-attention

processes have emerged from the literature in recent

years. The first of these, the Objective Self-Awareness

Perspective, suggests that objective self-awa pess is

primarily an aversive motivational state. This position

maintains that all environmental inducers of self-

awareness have similar effects, and that self-awareness

itself will motivate an individual to reduce discrepancies

between himself and others.


The Self-Consciousness Perspective describes self-

awareness as a primarily attentional process. This

perspective postulates that there are two types of





self-awareness: private self-awareness and public self-

awareness. It further suggests that certain environmental

stimuli will lead to private self-awareness (e.g. the

presence of a mirror) whereas other stimuli (e.g. the

presence of an audience) will engender a state of public

s el f-awar ene s s. A modification of this position, advanced

by William J. Froming and G. Rex Walker, suggests that

private self-awareness will lead to greater adherence to

privately held attitudes toward a given behavior, whereas

public self-awareness will lead to greater adherence to

public attitudes or social standards.


A final model, the Encoding Perspective, contends that

self-awareness stimuli cause individuals to encode

environmental information in terms of its relevance to the

self. The theory suggests that self-awareness stimuli

will be effective only when an individual's behavior is

publicly observable by others.


The present paper proposes a model of self-attentional

processes based upon a general model of attitude-behavior

relationships. It is suggested that the presence of a

mirror will lead to greater adherence to the particular

attitude salient at a given time. When responses of a

subject are publicly observable by others, self-awareness

(generated by the presence of a mirror) will lead to





greater adherence to public attitudes or social norms.

When responses are not publicly observed, self-awareness

will cause greater adherence to privately held attitudes.


The experiment described herein tests competing

predictions by the different models of self-awareness.

Persons who smoke cigarettes and who hold negative public

attitudes toward smoking (i.e. believe that others

disapprove of smoking) took part in an experiment designed

to test their willingness to smoke under different

circumstances. The model of self-awareness espoused in

this paper predicted that the presence of a mirror would

lead to greater smoking when the subject was alone, but to

less smoking when the subject was with another person.


Data from the experiment showed no significant overall

differences~ due to the experimental manipulations. Follow-

up analyses, however, suggested that mirror presence

increased smoking only when ph~ another individual was

present. Results are discussed in terms of the different

models of self-awareness. It is concluded that the model

of self-awareness espoused in this paper was not

supported, and that the pattern of results generally

supports the Self-Consciousness Perspective.


vii





CHAPTER I

INTRODUCTION


A revival of interest in the self in personality and

social psychology has been evidenced in recent years by a

growing literature dealing with self-focused attention.

Research of this type has attempted to specify the

behavioral and cognitive consequences of attention

directed at the self rather than the environment.


Several broad conceptualizations of self-attention

processes have emerged from the growing literature in this

area. The first of these, the Objective Self-Awareness

perspective, was initially set forth by Duval and Wicklund

(1972). The second point of view, espoused by Arnold Buss

and his associates, may be described as the Self-

Consciousness perspective (Buss, 1980). In addition,

several information processing models of self-focused

attention have been introduced in recent years (Carver,

1979; Hull and Levy, 1979).


The review which follows will consider each of the two

broad theories of self-attention processes, and one of the

information processing models. In addition, an alternative





model will be proposed which addresses some of the

controversial issues among the various perspectives and

attempts to resolve some of the differences.


Theories of Self-Focused Attention


The Objective Self-Awareness Perspective


The concept of self-awareness was first introduced by

Duval and Wicklund (1972). The authors distinguish two

types of self-awareness, objective and subjective, in the

following manner:


It is assumed that the objectively
self-aware individual will focus
attention upon himself--his
consciousness, personal history, or
body--but, in contrast, the same
person in subjective self-awareness
will be aware of himself only as a
source of forces that are exerted on
the environment (page 25).


Objective self-awareness is thus an awareness of

oneself as an object in the environment. By contrast,

subjective self-awareness is attention to the environment.

In this state the individual is the subject rather than the

object of his/her own awareness.


The theory of objective self-awareness contends that

at any point in time attention is wholly directed at the

self or the environment (Wicklund, 1978). Attention may

not be divided between the self and the environment, but





may oscillate rapidly back and forth between the two. The

extent to which one is objectively self-aware is thus the

proportion of time spent focusing on the self.


Changes in objective self-awareness are brought about

by events in the environment. Those events which increase

objective self-awareness are those which "remind the

individual of his object status (Wicklund, 1978; p. 466)."

Experimental manipulations designed to bring about

objective self-awareness have included mirrors, television

cameras, audiences, and tape recordings of the person's

voice. Stimuli which distract the person's attention from

the self to the environment are believed to decrease the

amount of time spent in objective self-awareness. It is

believed that many nervous habits are actually attempts to

reduce objective self-awareness through self-distraction

(Duval and Wicklund, 1972; see also Wicklund, 1975).


The Duval and Wicklund theory of self-awareness is

primarily a motivational theory, suggesting that objective

self-awareness causes the individual to focus on

discrepancies between his/her own behavior and some optimal

standard of behavior. In many cases this discrepancy is

described as one between the "real self" and the "ideal

self" (cf. Liebling, Seiler, and Shaver, 1974). Since

one's behavior is customarily less than optimal, the

objectively self-aware individual is motivated to alter





his/her behavior to bring it more closely into line with

the standard.


The above mentioned discrepancy between one's actual

behavior and an ideal standard of behavior led Duval and

Wicklund (1972) to postulate that objective self-awareness

was an aversive motivational state. Accumulation of

evidence to the contrary (Ickes, Wicklund and Ferris, 1973;

Steenbarger and Aderman, 1979), however, prompted Wicklund

(1978) to modify his position in this regard. It is

therefore believed that objective self-awareness is almost

always, but not necessarily, aversive.


In summary, then, the objective self-awareness

perspective emphasizes the self-evaluative effects of self-

awareness. Resulting changes in behavior are attributable

to the aversive nature of the self-aware state. For this

reason, the objective self-awareness perspective has been

described as a variant of consistency theory (Buss, 1980).


The Self-Consciousness Perspective


Initial Formulations. An alternate conceptualization

of self-attention processes is provided by Buss (1980).

The primary distinction between this approach and that of

Duval and Wicklund (1972) is that the self is divided into

two components: the private self and the public self. The

private self is believed to include those aspects of the





self which cannot be observed by others, such as thoughts,

attitudes, and self-reflections. Conversely, the public

self is believed to include those aspects of the self which

are publicly observable. Examples of these attributes

include physical appearance and voice quality.


The theory of self-consciousness is built upon a

distinction found in the development of the Self-

Consciousness Inventory (Fenigstein, Scheier and Buss,

1974). Factor analysis of the scale, which is believed to

measure chronic self-attentiveness, revealed two distinct

types of self-consciousness. The dimension of private self-

consciousness reflected one's tendency to be self-

reflective or introspective. The second factor, public

self-consciousness, measured one's concern with appearances

or the opinions of others.


Buss (1980) extended the distinction between public

and private self-consciousness to postulate the existence

of states of public and private self-awareness. As the

names suggest, private self-awareness is a state of self-

reflection whereas public self-awareness is a state of

concern with one's outward appearance.


In contrast to the Duval and Wicklund theory, the

theory of self-consciousness does not assume that self-

awareness leads invariably to self-evaluation. For this

reason it is a cognitive or attentional theory rather than





a motivational theory. Moreover, it is assumed that states

of self-awareness are effectively neutral rather than

aversive.


Changes in behavior resulting from states of self-

awareness are believed to result from attentional shifts

rather than from a desire to reduce self-awareness.

Persons who are made self-aware alter their behavior

because the state of self-awareness enhances the salience

of their own attitudes. This increased attention to

attitudes engenders greater adherence to the appropriate

attitude in a given situation, producing a large or small

shift in behavior from that in the nonself-aware state.


Modified theory. A modification of the Buss theory of

self-consciousness is suggested by Froming and Walker

(1981). This modification proposes that stimuli which

produce private self-awareness (such as a mirror) and those

which produce public self-awareness (such as an audience)

can lead to different patterns of behavior under some

circumstances. Specifically, the authors suggest that

persons can hold differing public and private attitudes

toward a given behavior. When this condition exists, a

state of public self-awareness will lead to increased

adherence to public attitudes (or standards) whereas a

state of private self-awareness will increase adherence to

private attitudes.





Froming and Walker demonstrated the theory in an

experiment involving the delivery of punishment to an

experimental confederate. Subjects were pretested using a

modification of the "Attitudes Toward Punishment"

questionnaire of Carver (1975). The questionnaire was

modified to create pairs of questions. The first question

of each pair assessed the subject's own attitude toward

punishment, whereas the second asked for an estimate of the

attitude of "most people." Subjects selected for the

experiment were those who were personally against the use

of physical punishment but who felt that most other people

favored its use. Results indicated that these persons

would decrease their use of punishment (compared to a

control group) in the presence of a mirror, but would

increase their use of punishment in the presence of an

evaluative audience. The authors concluded that

investigations of public and private self-awareness should

take into account the public and private attitudes

appropriate to a given situation.



The Encoding Perspective

A final model of self-attention processes, described

by Hull and Levy (1979), suggests an abandonment of the

notion that mirrors increase self-focused attention in

favor of an information processing model. This model

proposes that self-awareness stimuli generate increased





attention to those aspects of the environment which are

potentially relevant to the self. Instead of focusing on

himself, the individual who is "self-aware" focuses on the

environment in an effort to determine "self-relevant

contingencies." The state of self-awareness is actually an

increased tendency of persons to encode information in

terms of its relevance to the self. This enables the

individual to, in most instances, adopt an appropriate

strategy of self-presentation which will be maximally

effective.


One implication of this approach is that the effects

of self-awareness will depend upon the self-presentational

character of the experimental situation. Specifically,

Hull and Levy predict that the effects of self-awareness

will be shown only when the responses of the subject are

available for examination by the experimenter. When the

subject is allowed to respond anonymously, however, the

self-awareness stimuli will have no effect. This was shown

in two experiments, the first involving self-deprecation

(Hull and Levy, 1979; Experiment II) and a second

involving self-attribution (Experiment III).


Comparison of the Models


In order to compare the usefulness of the various

models which have been described, it is necessary to review

their ability to account for the existing research in self-





awareness. The following paragraphs attempt such a review.

This review, which is not exhaustive, addresses selected

issues wherein different predictions are expected by the

various models of self-awareness.


The first issue concerns situations in which an

individual finds his own judgmei-nt or attitude at odds with

that of other persons. In one such experiment (.Wicklund

and Duval, 1971), it was found that persons made self-aware

by the presence of a mirror were more responsive to

persuasion attempts from a positive reference group. This

outcome is predicted by the objective self-awareness model,

as well as that of Hull and Levy (1979). However, the self-

consciousness perspective would predict that persons made

self-aware by a mirror would focus more strongly on their

own (private) attitudes and therefore be more resistant to

persuasion.


A second general issue is that of attitude-behavior

relationships. Several studies (e.g. Carver, 1975;

Gibbons, 1978) have shown that self-awareness can increase

the consistency between attitudes and behavior. The

objective self-awareness model would suggest that self-

awareness leads the individual to focus on the attitude-

behavior discrepancy, and thereby motivates compliance with

personal attitudes. The self-consciousness model would

attribute the enhanced consistency to a greater awareness





of the attitudes themselves. Both models are therefore

consistent with the data on this issue. The encoding

perspective, however, attributes behavioral change under

conditions of self-awareness to increased focus on

environmental contingencies related to the self. There is

no reason to believe that these environmental contingencies

would be correlated with private attitudes. Thus the

increased correlation between attitudes and behavior

demonstrated in self-aware subjects would not be predicted

by the encoding model.


A final issue concerns focus of attention on

nonattitudinal information. Research has shown that

persons made self-aware are less susceptible to placebo

effects (Gibbons, Carver, Scheier, and Hormuth, 1979) and

more veridical in their assessment of their own bodily

states (Scheier, Carver and Gibbons, 1979). These

effects are predicted by the self-consciousness model

because of its emphasis on attentional focus. However,

because it is primarily a consistency theory, the objective

self-awareness perspective would not predict such an

effect. The encoding model also cannot directly speak to

these data, but one might expect that the model would

predict, if anything, increased responsiveness to placebo

effects under conditions of self-awareness.





An Alternative Model


The inconsistencies in research in self-focused

attention suggest that none of the above theories is

adequate to handle all of the findings. The model

described herein addresses some of these inconsistencies

and proposes a modified theory of self-attention

processes. The self-awareness theory being proposed

incorporates the attitude-behavior theory of Fishbein

(1967) into the self-consciousness model of Buss (1980)

and Froming and Walker (1981).


Developing the Model. Fishbein (1967) proposed

a theory of attitude-behavior relationships summarized

by the following equation:

B ~ BI = [Aact~w0 + [NB(Mc)]wl



In the equation, B refers to the overt behavior, BI is

the behavioral intention, Aact is the individual's attitude

toward the act, NB is the normative social belief, and

Mc is the motivation to comply with the norm. The weights

(WO and wl ) are empirically derived and are specific to a
given individual in given situation. The model thus

proposes that a person's behavioral intention is a function

of his/her own attitude, the normative social belief, and

the person's motivation to comply with the social norm.





Translating this model into the language of self-

consciousness theory, it can be seen that the attitude

toward the act (Aact) is what Fromina and ~a~lker (1981)

describe as the private attitude, and that the normative

belief (NB) is the public attitude. Both models are thus

suggesting that behavior is a function of both public and

private attitudes.


A model of self-awareness processes can be formulated

by adding another term to the existing model:

B ~ BI = ( [Aactlw0 + [NBs(Mc)]Iwl } SA

In this new equation, SA refers to the effects of mirror-

induced self awareness.


Before assessing the predictions of the model, the

implications of the encoding model of Hull and Levy (1979)

should be considered. Hull and Levy demonstrated that the

effect of self-awareness will in many cases depend upon the

public or private nature of the situation. In terms of the

above model, Hull and Levy are saying that the effect of

self-awareness on behavior will depend on the relative

values of the weights assigned to Aact or NB.


Example. To understand the workings of the above

model, several examples of its operation will be

considered. First, consider a situation in which self-

awareness is manipulated by the presence or absence of a

mirror. Suppose that the weight assigned to Aact is one





(1) and the weight assigned to NB is four (4). Further,

suppose that the motivation to comply with the normative

belief is greater than zero and constant across all

conditions.


The presence or absence of a mirror in the model is

denoted by the values of SA. Suppose that the value of

SA is four (4) when the mirror is present and one (1) when

the mirror is absent. Comparing the two conditions, we

find the two following outcomes:


Mirror Present: BI = 4(Aact) + 16(NB)

Mirror Absent : BI = 1(Aact) + 4(NB)


The difference between the two self-awareness

conditions [3(Aact) + 12(NB)] is largely reflected in

differences in adherence to normative beliefs. Moreover,

as the value of the weight assigned to Aact approaches

zero, the differences between the groups become entirely

attributable to adherence to the normative belief. In a

situation where the weights assigned to the normative

belief and the attitude toward the act are reversed, it

can be easily demonstrated that differences in self-

awareness will be reflected in differential adherence to

attitudes toward the act.


It is thus generally true that the effect of self-

awareness will be to amplify existing differences between





weights assigned to public and private attitudes. Stated

another way, differences in behavior between groups which

are high and low in self-awareness will depend upon the

relative assignment of weights assigned to personal and

social standards.


Implications. The primary implication of the above

model is a modification of an argument set forth by Hull

and Levy (1979). The argument contends that differences in

behavior resulting from self-awareness will depend upon the

public or private character of the situation. Hull and

Levy implied that self-awareness would be primarily

operative in situations where the person's behavior would

be publicly observable by others.


In terms of the present model, the effects of self-

awareness can be demonstrated in situations which are both

public and private. The effects of self-awareness will be

reflected in differential adherence to attitudes, and the

salient attitude in a given situation will depend upon the

public/private character of that situation.


A major implication of the present model is that two

traditionally accepted self-awareness manipulations--mirror

presence and audience presence--are quite different in

their effects. The effect of a mirror is represented by

higher values of SA. Behaviorally, this is reflected in

increased adherence to the salient attitude in a given





situation. The effect of an audience is represented by an

increase in the weight assigned to NB and possibly a

decrease in that assigned to Aact. This is reflected in

increased attention to social norms rather than personal

attitudes.


In a review of self-awareness literature, Diener and

Srull (1979) suggested that most experiments involving

self-awareness have demonstrated increased adherence to

social norms by persons made self-aware. If this is

indeed the case, then in terms of the model being proposed

the finding is not surprising. Virtually all self-

awareness experiments were conducted in a manner such that

the responses of subjects are publicly available. Self-

awareness thus augments socionormative influence in these

situations because the greatest weight is given to social

norms.


A final corollary of the above model is that the

effect of mirror presence can interact with the effect of

audience presence. Such a situation would require that

individuals hold differing public and private attitudes

toward a given behavior. An experiment of this type is

described below.





Evaluation of the Model


In order to assess the utility of the above model, its

ability to account for research in self-awareness will be

considered. Each of the three issues used in comparison of

models of self-awareness will be considered from the

perspective of the new model, and the correspondence of the

model's predictions with the existing data will be

assessed.


The first issue involves situations in which people

find their own attitude at odds with that of others. The

model espoused herein would predict that the presence of a

mirror will enhance responsiveness to the attitudes of

others when the subject's responses are publicly available.

Thus the findings of Wicklund and Duval (1971) are

predicted by the model.


The second issue involves attitude-behavior

relationships. The model would predict adherence to the

attitude which is salient in a given situation. Data from

Froming and Walker (1981) suggest that this is the

case. Data from Carver (1975), which show that the

correlation between private attitudes and behavior is

enhanced by the presence of a mirror, might seem to be at

odds with the prediction, since the experiment was one in

which responses were not anonymous. Supplemental data

analyses performed by Froming and Walker, however, reveal





that public and private attitudes on the Carver "Attitudes

Toward Punishment" scale are highly correlated. Hence the

data from the Carver (1975) experiment do not speak to the

issue of public and private attitudes.


The issue of focus of attention on nonattitudinal

information is not dealt with by the present model. Since

the model is a modification of that of Fishbein (1967),

it retains his emphasis on attitude-behavior

relationships. Nevertheless, it is not unreasonable to

extrapolate private attitudes to self-perception. Thus a

subject participating in the Scheier, Carver and Gibbons

(1979) experiment concerning taste perception would

hold the taste itself as the "private attitude" and the

suggestion of the experimenter as the "public attitude."

In this experiment, subjects were asked to taste various

flavors and rate the intensity of the taste. Subjects

were incorrectly told that others had found one flavor

stronger than another. Results indicated that subjects

made self-aware were more resistant to this suggestion

than nonself-aware subjects.


The experiment of Scheier, Carver and Gibbons,

however, is ambiguous with respect to the nature of the

public and private attitudes. The experimenter

introduced the suggestion by saying that "most people" find

that one taste is stronger than the other. From the





standpoint of the subject, "most people" (that is, those

who had supposedly participated in the experiment

previously) were not audience members. For this reason,

the change in the weight associated with the public

attitude is small when the experimental suggestion is

added. The model would therefore predict increased

salience of the "private standard," the taste.


The model being proposed is therefore consistent with

much of the existing literature in self-awareness.

Nevertheless, no single experiment contains all of the

conditions for a rigorous test of the model. The following

section attempts such a test.


A Test of the Model


The research described in this dissertation will test

the hypothesis that the effects of mirror presence can

interact with those of audience presence. The scenario for

this test is one in which an individual's public and

private standards toward the behavior of interest are

expected to differ. Moreover, a behavior is selected which

can be measured under conditions of anonymity and

nonanonymity of response.


Liebling, Seiler and Shaver (1974) described an

experiment in which the smoking behavior of persons was

measured when a mirror was present, and when it was absent.





In this experiment, the participants were told that the

experimenters were interested in musical preferences of

college students. They were asked to sit in a room for

thirty minutes and listen to several selections of music,

after which their reactions to the music would be assessed.

During the time that the participants listened to the

music, experimental assistants covertly observed them and

measured several indicants of smoking behavior. Results

indicated that the presence of a mirror effected an

increase in smoking.


The authors suggested that the results contradicted

the predictions of the Duval and Wicklund (1972) theory of

objective self-awareness. Pretest results indicated that

the participants rated their "ideal self" as smoking less

than their "real self" (i.e. their actual behavior).

According to Leibling et al. (1974), objective self-

awareness theory would predict that the presence of a

mirror would cause subjects to focus on the discrepancy

between real and ideal behavior, and therefore they should

smoke less.


Liebling et al. concluded that a drive theory was

more useful in explaining mirror effects than a theory of

self-attention. Wicklund's (1975) response to this paper

suggested that smoking served to reduce objective self-

awareness by serving as a distractor. Liebling et al.





(1975) responded to Wicklund's reply by questioning the

utility of a theory which can make two opposing

predictions in a single situation.

The results of the above experiment can be explained

in terms of the model of self-awareness espoused herein.

From the standpoint of smoking behavior, the experiment of

Liebling et al. was conducted under conditions of complete

anonymity, since no one was in the room with the subject

and he/she was unaware of the observer. Accordingly, it

can be expected that the private standard (Aact) would be

most salient. In this case the private standard would be

the person's desire for a cigarette, or his/her general

enjoyment of smoking. For this reason, the presence of a

mirror is expected to enhance the influence of the private

attitude, leading to greater smoking.


The experiment to be described in this dissertation

will address the opposite situation. If the participant in

the experiment is in the presence of another individual, it

is likely that the presence of a mirror will enhance the

effect of the public standard or normative belief. This

would be due to the increased salience of the public

standard resulting from the nonanonymity of the situation.

Thus the presence of a mirror should bring about increased

adherence to a public standard.





In recent years, public attitudes toward smoking

behavior have become decidedly negative. Increasing

concerns with health and increasing attention to the rights

of nonsmokers have placed the smoker in a position of

having differing public and private attitudes toward

smoking. For this reason, a manipulation which serves to

increase one's adherence to public standards toward smoking

would cause a decrease in smoking behavior.


The test of the theory which will be conducted

therefore examines the interactive effect of mirror

presence and audience presence (or anonymity) on smoking

behavior. The theory predicts that the direction of the

effect produced by the presence of a mirror will depend

upon the public/private nature of the situation.

Specifically, it is predicted that the presence of a mirror

will decrease smoking in the presence of an audience but

will increase smoking when the individual is alone.


Alternative Predictions


Objective Self-Awareness Perspective. As Liebling

et al. (1974) have pointed out, the Duval and Wicklund

(1972) model makes two predictions. Nevertheless, with

respect to the critical issue of this paper, their position

is consistent: the effects of the mirror and the audience

will be equivalent. Accordingly, the direction (though not





necessarily the magnitude) of the effect produced by a

mirror without an audience will be the same as that

produced by an audience without a mirror. Moreover, it is
not unreasonable to infer that the effects will be

additive, that is, the effect of the mirror and the

audience together will be greater than that of either

effect alone.


Self-Consciousness Perspective. The modification of

Buss (1980) theory of self-consciousness suggested by

Froming and Walker (1981) would predict opposite effects

for the mirror and audience. Private self-awareness,

produced by mirror alone, will result in increased

responsiveness to private attitudes and thereby lead to

increased smoking. Public self-awareness, produced by

the audience alone, will lead to decreased smoking as

subjects react to the increased salience of the public

attitude against smoking. When both the mirror and

audience are present, however, the effects will tend to

cancel. The rate of smoking by persons in the presence of

both a mirror and an audience should therefore fall

somewhere between that in the presence of the audience

alone and that in the presence of the mirror alone.


Encoding Perspective. The encoding perspective of

Hull and Levy (1979) would predict that the presence of a

mirror would alter the responsiveness of individuals to





self-relevant contingencies in the environment. For

subjects in the audience conditions, those contingencies

concern the response of the audience to their smoking.

Since smoking is viewed negatively by many people, it is

likely that these contingencies would favor a reduction in

the rate of smoking. For this reason, the Hull and Levy

theory would predict reduced smoking in the presence of a

mirror when an audience is present. Since there are no

environmental contingencies related to smoking when the

audience is absent, Hull and Levy would predict no effect

of mirror presence when the audience is absent.





CHAPTER II

METHOD




Overview


The purpose of the experiment was to

demonstrate differences in the effect of a mirror in

situations in which the response of a subject may or may

not be observed by another individual. An experimental

paradigm was selected in which the subject might be expected

to hold a personal belief which is inconsistent with the

attitudes which he/she believes are held by others.


The experiment was designed to replicate and extend

that conducted by Liebling, Seiler, and Shaver (1974). It

was previously mentioned that this experiment was

originally conducted with the subject alone in the

experimental room. The present experiment seeks to

replicate these findings, and extend them to situations in

which another individual is present.





Subjects


Subjects were 11 male and 21 female undergraduates at

the University of Florida. Subjects were recruited from

introductory psychology classes using two methods of

recruitment. The first method utilized sign-up sheets.

These sheets requested that only subjects who smoke more

than 8-10 cigarettes per day volunteer for the

experiment. The second method selected subjects on the

basis of protesting conducted at the beginning of the

academic quarter. Subjects responded to a series of

questions regarding their smoking habits. Those subjects

who indicated that they smoked more than one-half pack of

cigarettes per day were selected for the experiment.


Procedure


Eight subjects signed up for the experiment on the

sign-up sheets. They were contacted by telephone and told

that they would be participating in an experiment dealing

with the effects of cigarette smoking on perception and

memory. A time was arranged for an initial session.


During this initial session, subjects completed a

questionnaire dealing with attitudes toward smoking. The

first question asked the subjects to estimate the number

of cigarettes smoked during an average day, and to





indicate their favorite brand of cigarettes. Several

other questions assessed the subjects' public and private

attitudes toward smoking. The remaining questions were

included to help maintain the deception. In addition to

the smoking attitudes questionnaire, subjects completed

the Self-Consciousness Inventory (Fenigstein, Scheier,

and Buss, 1974) and the Fear of Negative Evaluation Scale

(Watson and Friend, 1969). After completing the

questionnaires, the subject returned them to the

experimenter who, if the subject evidenced the required

discrepancy between public and private attitudes, invited

the subject to take part in the two additional sessions of

the experiment.


Subjects selected from the protesting session (N = 24)

were selected on the basis of the number of cigarettes

smoked and the presence of a discrepancy between public

and private attitudes toward smoking. They were contacted

by telephone and invited to participate in the experiment.

The critical questions used in the selection of subjects

are found in Appendix A.


A protocol of specific experimental procedures is

found in Appendix B. Participants were greeted in a

waiting room by an experimental assistant. The assistant

explained that the experiment dealt with the effects of

smoking on several types of perceptual and motor skills.





The subject was reminded that the experiment was to be

conducted in two sessions, the second of which would take

place in about a week.


Informed consent was obtained from the subject

verbally. If the subject agreed to participate, he/she

was told that there would be a twenty-five minute waiting

period prior to the start of the experiment, since the

experiment dealt with heart rate. The subject was told

that during the waiting period he/she should light a

cigarette and take at least one or two puffs. Any amount

of smoking beyond that was said to have no effect on the

experiment. The subject was asked to wait quietly in the

experiment room, which contained several packs of

cigarettes (in case subjects forgot to bring their own), a

pack of matches and an ashtray. The room also contained a

radio. The subject was told that he/she could not read or

write during the waiting period, but could listen to the

radio if he/she so desired.


On "Mirror Present" trials, the subject was seated at

a table facing a small (24" x 24") mirror placed at eye

level. For trials in which the mirror was absent, the

mirror was covered with a sheet of white paper.


During trials "Audience Present" trials, an

experimental confederate was seated in the room with the

subject. The confederate had been in the room prior to





the subject's arrival, and the experimenter explained that

the confederate was scoring some questionnaires for a

professor. The subject was asked not to talk to the

confederate. If a subject later asked the confederate if

he/she may smoke, the confederate was instructed to

respond with "I don't know" or some other ambiguous

response. This, however, did not occur.

As the subject waited, he/she was observed by an

experimental assistant through a two-way mirror. The

mirror was almost completely covered with cardboard on the

subject's side, but sufficient space remained uncovered to

permit observation. The measures collected by the

observer were those reported by Liebling, Seiler, and

Shaver (1974). These were: 1) Lighting, 2) Flicking,

3) Puffing, and 4) Holding Time.


Upon completion of first session of the the

experiment, the subject was given the perceptual and

memory tasks. The first task was an abbreviated version

of the "Embedded Figures Test" (Witkin, 1969). The

second was a reverse digit span exercise.


At the conclusion of both parts of the experiment, the

subject was carefully debrief ed by the experimenter using

a predetermined set of questions (Appendix C). Subjects

who were aware that the experiment was concerned with

cigarette smoking during the waiting period, or who were





certain that they were being observed were to be

eliminated from all analyses. Those expressing vague

suspicions about the nature of the experiment were not.


Design


The experimental design was a 2 X 2 factorial with two

Mirror conditions (Mirror and No-mirror) and two Audience

conditions (Audience or No-audience). The design is of

the "mixed" type with repeated measures on the Mirror

factor only.





CHAPTER III

HYPOTHESES


The model of self-awareness proposed in Chapter I

can be used to make predictions concerning the effects of

public and private self-awareness on adherence to public

and private attitudes toward smoking. The model is

described by the following equation:

B ~ BI = ( [Aactlw0 + [NB(Mc)]wl }SA



Predictions about public and private determinants

of cigarette smoking require some assumptions about the

values entered into the model. First, the value of SA

will be I when the mirror is absent, and 2 when it is

present. The value of the weight assigned to Aact (wO

will be 2 when the subject is alone and I when an audience

is present. The value of the weight assigned to NB (wl
will be 1 in the No Audience condition and 2 in the

Audience condition. In all cases the value of NB is 1.


The values assigned to Aact and NB present a more

difficult problem. Subjects selected for the present

experiment are those who personally believe that smoking

is enjoyable, but who believe that others object to

smoking. The selection criteria, however, are based upon





the relative values of public and private attitudes

(requiring only that the private attitude be more

favorable toward smoking than the public attitude), rather

than the absolute values of these attitudes. In making

predictions from the equation, it will be assumed that the

value of Aact is always +1 and the value of NB is

always -1. The reader should note, however, that the

model will make a different set of predictions if both

Aact and NB are positive, and still another set of

predictions if both are negative. Nevertheless, since it

is being assumed that subjects selected for the experiment

are personally in favor of smoking and feel that others

are against smoking, it is believed that the choices of

+1 and -1 are the most appropriate.


Substituting into the equation, the following values

of BI are predicted for the four experimental conditions.

No Mir/No Aud: BI={+]2 -]1()l= 1

Mir/No Aud: BI = {[+1](2) + [-1](1)(1)}2 = 2

No Mir/Aud: BI={+]1 -]1()l= -1

Mir/Aud: BI = {[+1](1) + [-1](1)(2)}2 = -2

These predictions suggested the relative values of the

mean levels of smoking by subjects in each experimental

condition. Based upon these computations, the following

formal experimental hypotheses are advanced.





1. It is predicted that the effect of Mirror Presence

will depend upon the presence or absence of an audience.

An interaction of these effects is thus the first

hypothesis.


2. It is predicted the direction of the Mirror effect

will be different across Audience conditions.

Specifically, it is predicted that the presence of a

mirror will lead to decreased smoking when an audience is

present and will cause increased smoking when the

audience is absent.





CHAPTER IV

RESULTS


A total of 32 subjects participated in the experiment.

Of these, two were eliminated for failure to follow

instructions. Results of this experiment are based upon

the analyses of data from 30 subjects, 15 in each audience

condition.


Table 4-1 summarizes the means of subjects in each

treatment combination on the dependent measures of

smoking. The measures of smoking considered in this

experiment included the number of cigarettes smoked

(Lights), the number of times the subject flicked ashes

off of the cigarette (Flicks), the number of puffs taken

(Puffs), and the total amount of time, in seconds, spent

holding the cigarettes) (Time).


Although order of trials was randomly counterbalanced

within treatment conditions, it was believed to





substantially affect the dependent measures. For this

reason, Tables 4-2 and 4-3 present the means of the

treatment combinations for subjects having the same order

of trials. Means of the fourteen subjects who were not

exposed to the mirror on the first trial are presented in

Table 4-2. Means from the sixteen subjects exposed to the

mirror on the first trial are presented in Table 4-3. A

visual inspection of these tables suggests that subjects

tended to smoke less on the second trial, regardless of

the treatment condition. Accordingly, the principal

analyses include order of presentation as a factor.


The means presented in Table 4-1 suggest that the

presence of a mirror causes an increase in smoking. With

one exception (the Puffs measure for the No Audience

group), all indicants of smoking were greater when

subjects were exposed to a mirror. This was true for

subjects in both the Audience and No Audience

conditions. The means also suggest that the presence of

an audience leads to decreased smoking. With one

exception again (the Flicks measure on Mirror trials),

subjects exposed to an audience smoked less than those not

exposed to an audience on both Mirror Present and Mirror

Absent Trials.


Statistical analyses of the mean differences were

performed using univariate analyses of variance on the





four dependent measures. Each analysis included two

"between group" factors (Condition and Order) and one

"within subject" factor (Trial). The two levels of

Condition were Audience and No Audience; the two levels of

Order were Mirror First and No Mirror First; the two

levels of Trial were Mirror and No Mirror. Analysis of

variance tables are presented in Table 4-4 (Lights), Table

4-5 (Flicks), Table 4-6 (Puffs) and Table 4-7 (Time).


The general pattern of the statistical analyses

suggests no significant effects due to either the presence

of an audience or the presence of a mirror on any of the

dependent measures. The lone exception to this pattern is

a significant main effect of Mirror Presence on the Time

measure (F (1,26) = 4.62, p < .05). However, due to the

large number of comparisons being made in the experiment,

this result can be discounted.


Correlations of the individual difference measures

with the dependent measures of smoking on Mirror Present

and Mirror Absent trials are presented in Tables 4-8

and 4-9 None of the correlations were statistically

significant. Since the magnitude of the correlations

suggested a possible relationship between private

attitudes toward smoking and the measures of smoking, a

partial correlation analysis was performed. Results of

the analysis indicated no significant relationship between





private attitudes and smoking eliminating public

attitudes, or between public attitudes and smoking

eliminating private attitudes.


A comparison of the results of the present study with

the results reported by Liebling, Seiler and Shaver (1974)

is presented in Table 4-10. The table reports percent

changes in the four indicants of smoking and t-tests on

the mean differences for both experiments. The results of

the present study are reported by levels of the audience

factor. Referring to the table, it can be seen that in

the Audience condition, three of the four measures of

smoking showed significant increases as a result of the

mirror manipulation. The data also suggest a great deal

of similarity between the Audience condition of the

present study and the results of the Liebling et al.

study. It is worth noting, however, that the results of

multiple t-tests should be viewed with caution.


Answers to the debriefing questions were tabulated as

a check on manipulations and deceptions and are

summarized in Table 4-11. It should be remembered that

the answers to first three questions in the debriefing

were recorded before the subject was told of the

deception. The answers to the final five questions were

given after the experimental procedures had been


disclosed.












Table 4-1: Means and Standard Deviations on the Dependent
Variables for Each Experimental Treatment
Combination. (N = 15 in each group.)

Trial


Group Variable No Mirror Mirror



No Lights 1.53 (0.51) 1.60 (0.51)
Audience Flicks 8.80 (8.19) 11.33 (8.56)
Puffs 13.47 (7.68) 13.47 (5.51)
Time 495.00 (218.80) 577.57 (227.40)



Audience Lights 1.20 (0.52) 1.47 (0.52)
Flicks 7.47 (3.81) 11.33 (9.70)
Puffs 8.60 (4.60) 12.53 (6.75)
Time 359.60 (118.70) 481.13 (216.49)











Table 4-2: Means and Standard Deviations on the Dependent
Variables for Each Experimental Treatment
Combination, for Subjects WIho Were Exposed to
the Mirror on the First Session. (N = 7 for each
group.)

Trial


Group Variable No Mirror Mirror



No Lights 1.57 (0.53) 1.71 (0.49)
Audience Flicks 9.29 (10.82) 12.00 (9.20)
Puffs 11.71 (7.41) 13.71 (5.93)
Time 486.71 (179.00) 566.14 (243.64)



Audience Lights 1.14 (0.38) 1.57 (0.53)
Flicks 7.28 (4.46) 14.85 (13.06)
Puffs 8.85 (5.67) 15.43 (8.12)
Time 343.71 (121.35) 572.29 (263.86)











Table 4-3: Means and Standard Deviations on the Dependent
Variables for Each Experimental Treatment
Combination, for Subjects Who Were Exposed to
the Mirror on the Second Session. (N = 8 for each
group.)

Trial


Group Variable No Mirror Mirror



No Lights 1.50 (0.53) 1.50 (0.53)
Audience Flicks 8.37 (5.78) 10.75 (8.56)
Puffs 15.00 (8.07) 13.25 (5.52)
Time 502.25 (261.65) 515.37 (297.08)



Audience Lights 1.25 (0.46) 1.37 (0.52)
Flicks 7.62 (5.78) 8.25 (4.33)
Puffs 8.37 (3.81) 10.00 (4.34)
Time 373.50 (122.82) 401.37 (135.94)

















Source df Sum of Squares Mean Square F p


Order 1 0.13 0.13 0.36

Audience 1 0.83 0.83 2.32 .14

Aud x Ord 1 0.04 0.04 0.10

Error (B) 26 9.36 0.36


Mirror 1 0.45 0.45 3.16 .08

Mir x Ord I 0.18 0.18 1.30 .26

Mir x Aud 1 0.16 0.16 1.10 .30

M x O xA 1 0.02 0.02 0.17

Error (W) 26 3.72 0.14


Table 4-4:


Analysis of Variance Table for the Dependent
Variable "Lights."
















Source df Sum of Squares Mean Square F p


Order 1 66.30 66.30 0.91

Audience 1 5.34 5.34 0.07

Aud x Ord 1 15.74 15.74 0.22

Error (B) 26 1891.01 72.73


Mirror 1 164.74 164.74 3.00 .09

Mir x Ord 1 49.54 49.54 0.90

Mir x Aud I 9.01 9.01 0.16

M x O xA 1 40.74 40.74 0.74

Error (W) 26 1427.44 54.90


Analysis of Variance Table for the Dependent
Variable "Flicks."


Table 4-5:

















Source df Sum of Squares Mean Square F p


Order 1 8.90 8.90 0.18

Audience I 113.30 113.30 2.23 .14

Aud x Ord I 71.17 71.17 1.40 .25

Error (B) 26 1320.26 50.78


Mirror 1 66.59 66.59 2.43 .13

Mir x Ord 1 70.58 70.58 2.58 .12

Mir x Aud 1 58.93 58.93 2.15 .15

M x O xA 1 1.33 1.33 0.05

Error (W) 26 712.54 27.40


Analysis of Variance Table for the Dependent
Variable "Puffs."


Table 4-6:

















Source df Sum of Squares Mean Square F p


Order 1 29 028.39 29 028.39 0.43

Audience 1 134 494.81 134 494.81 2.00 .16

Aud x Ord 1 10 465.74 10 465.74 0.16

Error (B) 26 1 744 154.80 67 082.87


Mirror 1 113 680.93 113 680.93 4.62 .04

Mir x Ord 1 66 535.40 66 536.40 2.70 .11

Mir x Aud 1 25 070.14 25 070.14 1.02 .32

M x O xA 1 16 857.34 16 857.34 0.68

Error (W) 26 640 279.59 24 626.14


Table 4-7:


Analysis of Variance Table for the Dependent
Variable "Time."





44

Table 4-8: Correlations of Individual Difference Variables
With Smoking Behavior of Subjects in Mirror
and No Mirror Conditions When an Audience
Was Absent.



Individual Difference Variables


Dependent
Variables


NumSmk PrAtt PuAtt PrSC


PuSc SAnx FNE


NO MIRROR CONDITION


-.01


.00


.09 -.13

.04 -.16


.04

.05

.06

.16


Lights

Flicks

Puffs


.00


.22

.36

.11

.22


-.05 -.27


-.21


.11

.03


.21

.04


.08

.20


-.11

.08


Time


.07


MIRROR CONDITION


Lights

Flicks

Puffs


.04


.07

.19


-03 -.24 -.12 -.17


.02


-16


.07 -.16


.01 .12


-.07 -.17

.44 -.04


-.17


.05


.12


.27


Time


.05


.30 -.01 -.05 -.15


Average number of cigarettes / day
Private attitude toward smoking
Public attitude toward smoking
Private Self-Consciousness
Public Self-Consciousness
Social Anxiety
Fear of Negative Evaluation


Abbreviations : NumSmk
PrAtt
PuAtt
PrSC
PuSC
SAnx
FNE





_ __ __ __


Table 4-9: Correlations of Individual Difference Variables
With Smoking Behavior of Subjects in Mirror
and No Mirror Conditions When an Audience
Was Present.



Individual Difference Variables


Dependent
Variables


NumSmk PrAtt PuAtt PrSC


PuSc SAnx FNE


NO MIRROR CONDITION


.00 -.02 .20


.24


Lights

Flicks

Puffs


-.21 -.37

-.33 -.35

-.07 -.47


-.06


-29 -.36 -.33 .07 -.15


.17


-41 -.06 -.06 .06

-.15 -.03 -.32 -.18


Time


-.17


.02


MIRROR CONDITION


.09 -.27 .13 -.19


Lights

Flicks

Puffs


-.30 -.03


.43


.11 -.22 -.29 .05 -.29


-.19


.04


.14 -.40 .08 -.29


-.15 -.04


.32


Time


-.15


.26 -.11 -.32


.27 -.14


- Average number of cigarettes / day
- Private attitude toward smoking
- Public attitude toward smoking
-Private Self-Consciousness
-Public Self-Consciousness
- Social Anxiety
- Fear of Negative Evaluation


Abbreviations: NumSmk
PrAtt
PuAtt
PrSC
PuSC
SAnx
FNE





LIEBLING, SELLER, AND SHAVER--

No Lights 30 % 1.58 n.s.
Audience Flicks 123 % 3.25 < .01
Puffs 46 % 2.35 < .05
Time 45 % 2.35 < .05


Table 4-10: Comparison of Experimental Results With the
Results Reported by Liebling, Seiler, and
Shaver (1974).


Percent Increase in Smoking
In Mirror Condition and T-tests


Group


Measure


Change


WALKER--

No
Audience




Audience


Lights
Flicks
Puffs
Time


4 %
29 %
O %
17 %


0.43
0.79
0.00
1.11


.67
.45
1.00
.29


Lights
Flicks
Puffs
Time


2.26
1.92
2.37
2.14


.04
.08
.03
.05



























































___


Table 4-11: Tabulation of Responses to the Debriefing
Questions.


RESPONSES
(Grp A = Audience Condition
QUESTION Grp N = No Audience Condition)


Grp A: No 93%

Grp N: No 67%


Grp A: No 80%

Grp N: No 93%


Yes 7%

Yes 33%


Yes 20%

Yes 7%


Were tests
affected by
smok ing?


Other purposes
to experiment?


Guess other
purposes?


Suspect that
being observed?



If yes to #4,
was smoking
affected?


Affected by the
mirror?



Did/would an
audience affect
you?


No correct guesses were given.



All: Definitely yes 0%
Maybe -63%
No -37%


All: No 100%
Yes -0%


Yes 33%

Yes 20%


Yes 27%

Yes 33%


Grp A: No 67%

Grp N: No 80%


Grp A: No 73%

Grp N: No 67%





CHAPTER V

DISCUSSION


The statistical analyses presented in the previous

chapter suggest that none of the experimental

manipulations significantly affected the smoking behavior

of the subjects. The discussion which follows offers

possible reasons why this may be the case, and uses the

data for a limited comparison of the models of self-

awareness presented herein.


Experimental Procedures


The procedure of this experiment attempted to parallel

that of Liebling, Seiler and Shaver (1974). The primary

difference between the two experiments concerns the method

used to generate smoking behavior. Liebling et al. told

subjects that the experiment concerned reactions to

selections of music, and that they could smoke while

listening if they wished. The present experiment informed





subjects that purpose of the research was to study

smoking, but misinformed subjects as to the aspects of

smoking which were under investigation. More important,

the present study required that subjects smoke a minimal

amount during the waiting period.


The results of the experiment suggest that this

procedural modification substantially altered the

subjects' responses to the situation. One result of the

of the change was that subjects smoked more in the present

experiment than in the Liebling et al. experiment. The

mean number of cigarettes smoked in the Liebling et al.

experiment was 1.06 for the No Mirror condition and 1.38

for the Mirror condition. Referring to Table 1, it can be

seen that subjects in the present experiment smoked

considerably more than their counterparts in the Liebling

et al. experiment. This difference is especially striking

when one takes into account the fact that the subjects in

the Liebling et al. experiment were observed for 30

minutes, as compared to 25 minutes in the present

experiment.


It would appear that the requirement that the subjects

smoke greatly altered the situation. This is evidenced

by the results of the debriefing. Only 27% of those in

the Audience condition reported being affected by the

audience. Most reported that since the experiment was





concerned with smoking, the confederate had no right to

object. Furthermore, the effect of the audience was often

manifested in other behaviors. A large number of

subjects reported being concerned that the noise from the

radio in the experimental room would bother the

confederate, and as a result they tried to keep the

volume as low as possible. Thus it is possible that there

was a substantial audience effect, but that the nature of

the experiment prevented it from affecting the subjects'

smoking behavior.


One can conclude from the present study that the

method used in this experiment to study attitude-behavior

relationships related to smoking is of limited utility.

One might suppose that the decision to smoke or not smoke

in a given situation is a separate issue from the decision

of how much to smoke. By eliminating the first issue, the

present experiment reduced the probability of showing

significant effects.


Implications for Self-Awareness Models


The primary and follow-up analyses of this

experiment allow only a very limited number of conclusions

to be drawn from the present research. Those conclusions





which are drawn herein should therefore be considered very

tentative at the present time.


Implications for the Interactive Model


The initial purpose of the experiment was to test the

notion that the effect of a mirror could, under certain

circumstances, depend upon the presence or absence of an

audience. Specifically, it was predicted that mirror

presence would increase smoking in the absence of an

audience and decrease smoking when an audience was

present. The data from this experiment are in

contradiction to these predictions, and suggest that the

effect of a mirror will be the same in the presence or

absence of an audience. Moreover, the results of the

multiple t-test analyses (Table 4-10) suggest that the

increase in smoking generated by the presence of a mirror

may actually be enhanced by an audience.


The data from this experiment thus provide no support

for the model of self-awareness presented in this paper.

Although these results are damaging to the theory, they do

not discount this model of self-awareness entirely.


A major problem with operationalizing the implications

of this model of self-awareness concerns the nature of the

public standards. Every attempt was made in this

experiment to make it appear that the confederate





(audience) was not connected with the experiment itself.

This was designed to create a situation in which the

subject would have to infer the attitude of the subject

based on his/her beliefs about the attitudes of "most

other people."


Evidence from the debriefing suggests that the attempt

to separate the confederate from the experiment was not

successful. The rationalizations used by the subjects

suggest that the public standard appropriate to this

situation was more complex than was previously imagined.


Since there is no clear indication of the nature of

the public standard in this situation, and no evidence

that the standard was the same for all subjects, the

present experiment cannot be regarded as an adequate test

of the theory of self-attention espoused herein. Future

research in this area should attempt to create situations

in which public and private standards can be more

rigorously controlled and more accurately documented.


Implications for Other Models


Objective Self-Awareness Perspective. The Objective

Self-Awareness model of Duval and Wicklund (1792) receives

no support in this experiment. This model would predict

that the effect of the mirror and the effect of the

audience should differ only in magnitude. The model would





further suggest that the combination of two manipulations

of self-awareness should have the greatest effect. In the

present experiment, this model would predict that smoking

would be greatest in the Audience/Mirror condition and

least in the No Audience/No Mirror condition. Neither

prediction is supported by this experiment. These data,

taken together with those of Froming and Walker (1981),

suggest that the notion that all self-awareness

manipulations are in all cases unidirectional may be

incorrect.


Encoding Perspective. The Encoding Perspective of

Hull and Levy (1979) suggests that the effect of mirror

presence will only be shown when the responses of an

individual are publicly observable by others. Referring

to Tables 4-1 and 4-10, it would appear that this

prediction is supported by the experimental results. The

t-tests of Table 4-10 are only significant in the Audience

condition.


The Encoding Perspective, however, does not explain

the (non-significant) differences between the Audience and

No Audience conditions. From the standpoint of this

model, the increase in smoking resulting from the presence

of a mirror implies that the "self-relevant

contingencies" are related to increased smoking. If this

is the case, then within a given Mirror condition, the





effect of an audience should be to increase smoking

behavior. Referring to Table 4-1, it can be seen that

the effect of an audience probably was to decrease

smoking in both the Mirror and No Mirror conditions. The

Encoding Perspective is thus not supported by the

experimental results.


Self-Consciousness Perspective. The modified Self-

Consciousness Perspective of Froming and Walker (1981)

suggested that the effect of a mirror would be to increase

smoking behavior, whereas the effect of an audience would

be to decrease smoking. It was further predicted that the

combination of stimuli (Mirror/Audience condition) would

result in a cancelling out of the effects. The prediction

of the Self-Consciousness model is thus a main effects

model.


The analyses of variance presented in Chapter IV offer

some limited support for thnis perspective. The pattern of

the univariate analyses suggests that the main effects

model is a more likely model than an interactive one.

With one exception, however, these differences are

nonsignificant


The results of the multiple t-test analyses presented

in Table 4-10, however, suggest an interactive model. This

model appears to show an effect of mirror presence only

in the Audience condition. This would appear to





contradict the predictions of the Self-Consciousness

Perspective as well as the results of Liebling et al.

(1974).


In defense of the Self-Consciousness position, it

appears that the absence of a significant difference in

the No Audience condition was due to a ceiling effect.

Subjects in the No Audience/Mirror condition smoked an

average of 1.60 cigarettes in a twenty-five minute period.

Clearly there are limits to how much a person will smoke,

especially in a small room without windows. If this is

indeed the case, then the discrepancy between the present

experiment and that of Liebling et al. (1974) can be

explained. Subjects in that experiment smoked

substantially less than those in the present experiment.

Hence there was room for an increase due to the mirror.


Conclusion


The model of self-awareness proposed herein suggested

that the influence of self-focused attention on attitude-

behavior relationships could be summarized by a

modification of an equation developed by Fishbein (1967).

Specifically, it was proposed that a self-awareness term

be added to the model, producing the following equation.

B ~ BI = ( [Aactlw0 + [NB(Mc)]wl } SA

The equation suggested that the effect of self-awareness

multiplicative with the situational cues, and thus that it





served to enhance the salience of the standards

appropriate to a given situation.


The data seem to suggest that the self-consciousness

model more accurately accounts for the mirror and audience

effects. If this model is indeed the most appropriate,

then the above equation is incorrect. The Self-

Consciousness Perspective would suggest that the SA term

in the model is unnecessary, and that self-awareness

effects are shown through changes in the weights

associated with Aact and NB. Specifically, the model

would suggest that the weight assigned to Aact is to some

degree a function of private self-awareness and the weight

assigned to NB is a function of public self-awareness.

It would thus appear that models of self-awareness will

in the future have to take account of public and private

self-awareness along with public and private attitudes.





REFERENCES


Buss, A. H. Self-Consciousness and Social Anxiety.
San Francisco: Freeman, 1980.

Carver, C. S. Physical aggression as a function of
objective self-awareness and attitudes toward
punishment. Journal of Experimental Social
Psychology. 1975, 11, 510-519.

Carver, C. S. A cybernetic model of self-attention
processes. Journal of Personality and Social
Psychology. 1979, 37, 1251-1281.

Carver, C. S., & Scheier, M. F. Self-focusing effects of
dispositional self-consciousness, mirror presence, and
audience presence. Journal of Personality and Social
Psychology, 1978, 36, 324-332.

Diener, E., & Srull, T. K. Self awareness, psychological
perspective and self-reinforcement in relation to
personal and social standards. Journal of Personality
and Social Psychology, 1979, 37, 413-423.

Diener, E., & Wallbom, M. Effects of self-awareness on
antinormative behavior. Journal of Research in
Personality 1976, 10, 107-111.

Duval, S. & Wicklund, R. A. A Theory of Objective Self-
Awareness. New York: Academic Press, 1972.

Exner, J. E. The self-focus sentence completion: A study
of egocentricity. Journal of Personality Assessment,
1973, 37, 437-455.

Fenigstein, A., Scheier, M. F., & Buss, A. H. Public and
Private Self-Consciousness: Assessment and Theory.
Journal of Consulting and Clinical Psychology,
1975, 43, 522-527.

Fishbein, M. Attitude and the prediction of behavior.
In M. Fishein (Ed.) Readings In Attitude Theory
and Measurement. New York: Wiley, 1967.

Froming, W. J. The relationship of moral judgement, self-
awareness, and sex to compliance behavior. Journal of
Research in_ Personality, 1978, 12, 396-409.





Froming, W. J., & Walker, G. R. Self-awareness and public
vs. private standards for behavior. Under editorial
review, 1981.

Gibbons, F. X. Sexual standards and reactions to
pornography: Enhancing behavioral consistency
through self-focused attention. Journal of Personalit~y
and Social Ps chology, 1978, 36, 976-987.

Gibbons, F. X., Carver, C. S., Scheier, M. F., &
Hormnuth, S. E. Self-focused attention and the placebo
effect: Fooling some of the people some of the time.
Journal of Experimental Social Psychology, 1979,
15, 263-274.

Hull, J. G., & Levy, A. S. The organizational functions
of the self: An alternative to the Duval and Wicklund
model of self-awareness. Journal of Personality and
Social Ps~ychology, 1979, 37, 756-768.

Ickes, W. J., Wicklund, R. A., & Ferris, C. B. Objective
self-awareness and self-esteem. Journal of
Ex erimental Social Psychology, 1973, 9, 202-219.

Liebling, B. A., Seiler, M., & Shaver, P. Self-awareness
and cigarette smoking behavior. Journal of
Ex erimental Social Ps chology, 1974, 11, 325-332.

Pryor, J. B., Gibbons, F. X., Wicklund, R. A., Fazio, R.
H., & Hood, R. Self-focused attention and self-
report validity. Journal of Personality, 1977,
45, 513-527.

Scheier, M. F., Self-awareness, self-consciousness, and
angry aggression. Journal of Personality, 1976,
4 627-644.

Scheier, M. F., & Carver, C. S. Self-focused attention
and the experience of emotion: Attraction, repulsion,
elation, and depression. Journal of Personality and
Social Psychology, 1977, 35, 625-636.

Scheier, M. F., Carver, C. S., & Gibbons, F. X. Self-
directed attention, awareness of bodily states, and
suggestibility. Journal of Personality and Social
Psychology, 1979, 37, 1576-1588.

Scheier, M. F., Fenigstein, A., & Buss, A. H. Self-
awareness and physical aggression. Journal of
Ex erimental Social Ps chology, 1974, 1, 264-273.





Steenbarger, B. N., & Aderman, D. Objective self-
awareness as a nonaversive state: Effects of
anticipating discrepancy reduction. Journal of
Personality, 1979, 47, 330-339.

Watson, D., & Friend, R. Measurement of Social-Evaluative
Anxiety. Journal of Consulting and Clinical
Psychology, 1969, 33, 448-457.

Wicklund, R. A. Objective self-awareness. In L.
Berkowitz (Ed.), Advances in Experimental Social
Psychology (Vol. 8).New York: Academic Press, 1975.

Wicklund, R. A. Three years later. In L. Berkowitz (Ed.),
Cognitive Theories in Social Ps~ycholog~y. New York:
Academic Press, 1978.

Wicklund, R. A., & Duval, S. Opinion change and performance
facilitation as a result of objective self-awareness.
Journal of Ex~perimental Social Ps~ychology, 1971, 7,
319-342.

Witkin, H. A. The Embedded Figures Test. Palo Alto:
Consulting Psychologists Press, 1969.

Wuebben, P. L., Straits, B. C., &( Schulman, G. I. The
Experiment as a Social Occasion. Berkeley: The
Glendessary Press, 1974.





APPENDIX A


QUESTIONS USED IN SELECTING SUBJECTS

1. How much do you smoke?

1) Not at all
2) Less than 5 cigarettes a day
3) About one-half pack a day
4) Around a pack a day
5) More than a pack a day

For each of the remaining items, indicate how much you
agree or disagree with each statement by using the
following scale:


1) Strongly
2) Disagree
3) Neutral
4) Agree
5) Strongly


disagree



agree


2. I find smoking to be an enjoyable habit.


3. Smoking is a disgusting
have.

4. Most people (other than
annoying and offensive.

5. Most other people have a
the habit of smoking.


habit that no one should


myself) find smoking to be


favorable attitude toward





APPENDIX B


SPECIFIC EXPERIMENTAL PROCEDURES AND INSTRUCTIONS


1. Instructions read to subject:

By showing up you have earned one experimental credit.
Before we go any farther, I need to tell you something
about the experimental procedure so that you can decide
whether or not you want to continue.

This study involves smoking, and is specifically
interested in differences in memory functions and heart
rate between smokers and non-smokers. I'll tell you more
about the specific predictions of the experiment later,
after the experiment is over, since we want the tests to
be unbiased.

Today we will have you rest for about twenty-five
minutes, and then we will measure your heart rate and let
you try the memory tests. When you come back for the
second part of the experiment, you will again rest for
about twenty-five minutes, and we will do a different set
of tests. If you complete both parts of the experiment
you will receive two hours credit. You must, however,
complete both parts of the experiment to receive any
credit.

During the waiting period, I would like for you to
light up a cigarette and take at least one or two puffs.
If possible, it would be better if you did this at the
beginning of the waiting period, since it takes a little
while for the effects of nicotine to be felt.

Some people wonder if it is all right to smoke more
than this. The answer is yes, you can smoke as much as
you like, but you have to take at least one or two puffs.
Smoking any more than this will not affect our experiment,
but not smoking at all could change the results.

If you do not have any objections to the experiment, I
will take you to the waiting room and we can begin.





2. Proceed to experiment room and continue:

You will be waiting here for about twenty-five minutes
while your heart rate stabilizes. Remember to light up a
cigarette and take at least one or two puffs during this
time. I will come by when the waiting period is over and
we can start the experiment.

3. If the confederate is present:

We are going to be conducting an experiment in this
room in about twenty-five minutes. You can stay here
until then, but at that time you will have to leave. I
only ask that you two (addressing subject and confederate)
do not talk to each other.

4. At end of waiting period, present deception tasks.


SECOND SESSION

1. Instructions: Experimenter reminds subject that the
instructions for the second session are identical
to those in the first.

2. Debriefing procedures (Appendix C) follow waiting
period.





APPENDIX C

DEBRIEFING PROCEDURES


1. Did you believe that the tests at the end of the
first session were affected smoking?


2. Did you think there were any other purposes to the
experiment, i.e. anything we were not telling you?


3. If yes, what did you think the experiment was really
about?




At this point, explain the experiment



4. Did you suspect that you were being watched?


5. If yes, did this affect your smoking?


6. Did the presence of the mirror affect you or bother
you at all?

If so, how?


7. Did/Would the presence of another person in the room
affect you or bother you at all?

If so, how?


8. Other comments:





BIOGRAPHICAL SKETCH


Gordon Rexal Walker, Jr., was born in Burlington,

North Carolina, on September 11, 1954, to Mary Speight

Walker and Gordon Rexal Walker. He completed his

secondary education at Walter Williams High School in

Burlington, North Carolina, and entered Wake Forest

University in September, 1972. After graduation from Wake

Forest in 1976, he entered the University of Richmond,

where he received the degree of Master of Arts in 1978.


He entered the Graduate School at the University of

Florida in September, 1978. During that time, he studied

in the Personality Area of the Department of Psychology,

under the direction of Dr. William J. Froming. On July 1,

1979, he was married to Jane Linda Curry of Hingham,

Massachusetts.





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.



William J. Froming, ChairmAd
Assistant P A~fessor of Ps biology



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.



Franz,9 AJ- p
Professor ~/ Psych ~ogy



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.



)arry A. Crater
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.



Thomas K. Srull
Assistant 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.



Jo ~Lynchh
Ass stant Professor of Marketing



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.



J l~es Algina
Aociate Professor of Foundations
of Education



This dissertation was submitted to the Graduate
Faculty of the Department of Psychology in the College of
Liberal Arts and Sciences and to the Graduate Council, and
was accepted as partial fulfillment of the requirements for
the degree of Doctor of Philosophy.

March, 1981




Dean for Graduate Studies and
Research




Full Text

PAGE 1

SELF-AWARENESS AND CIGARETTE SMOKING: THE INTERACTION OF PUBLIC AND PRIVATE DETERMINANTS BY GORDON REXAL WALKER, JR. A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1981

PAGE 2

ACKNOWLEDGEMENTS The author would like to thank the following experimenters for their invaluable assistance in collecting data for this research: Ruth Honig, Jennifer Jeffers, Hindi Kliein, Roberta Ross, Anthony Campos, Eric Corley, Minou Hirnampour, Paul Wiltse, George Piemonte, Bill Allen, Shishir Kurup, Jeff Burns, and Sharon Stern. Additional thanks are due to those who agreed to smoke in this experiment for the advancement of science. A final debt of gratitude is due to Dr. William J. Froming. His patience with me over these years in graduate school and our trips to the pool hall have made the entire experience much more enjoyable. ii

PAGE 3

TABLE OF CONTENTS PAGE ACKNOvlLEDGEMENTS ii ABSTRACT V CHAPTER I II III INTRODUCTION .......................... 1 Theories of Self-Focused Attention ..... 2 The Objective Self-Awareness Perspective ......... ............. 2 The Self-Consciousness Perspective .... 4 Initial Formulatio.ns ................ 4 Modified Theory .................... 6 The Encoding Perspective .............. 7 Comparison of the Models ............. 8 An Alternative Model ................. 11 Developing the Model ............... 11 Example ............................ 12 Implications ........................ 14 Evaluation of the Model 16 A Test of the Model .................. 18 Alternative Predictions .............. 21 Objective Self-Awareness Perspective ...................... 21 Self-Consciousness Perspective ..... 22 Encoding Perspective ................ 22 METHOD 24 0vervie,11 ............................. 24 Subjects ............................. 25 Procedure ............................ 25 Design ................................ 29 HYPOTHESES 30 iii

PAGE 4

IV V RESULTS ........................... 3 3 DISCUSSION .. .......................... 48 Experimental Procedures ............... 48 Implications for Self-Awareness Mode 1 s 5 O Implications for the Interactive Model ............................. 51 Implications for Other Models ....... 52 Objective Self-Awareness Perspective ..................... 52 Encoding Perspective .............. 53 Self-Consciousness Perspective ..... 54 Conclusion ............................ 55 REFERENCES 5 7 APPENDICES A B C QUESTIONS USED IN SELECTING SUBJECTS ... 60 SPECIFIC EXPERIMENTAL PROCEDURES ....... 61 DEBRIEFING PROCEDURES ................. 63 BIOGRAPHICAL SKETCH .................................. 64 iv

PAGE 5

r Abstract of Dissertation Presented to the Graduate Council of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy SELF-AWARENESS AND CIGARETTE SMOKING: THE INTERACTION OF PUBLIC AND PRIVATE DETERMINANTS By Gordon Rexal Walker, Jr. March, 1981 Chairman: William J. Framing, Ph.D. Major Department: Psychology Several broad conceptualizations of self-attention processes have emerged from the literature in recent years. The first of these, the Objective Self-Awareness Perspective, suggests that objective self-awa ?'1 ess is primarily an aversive motivational state. This position maintains that all environmental inducers of selfawareness have similar effects, and that self-awareness itself will motivate an individual to reduce discrepancies between himself and others. The Self-Consciousness Perspective describes self awareness as a primarily attentional process. This perspective postulates that there are two types of V

PAGE 6

self-awareness: private self-awareness and public self awareness. It further suggests that certain environmental stimuli will lead to private self-awareness (e.g. the presence of a mirror) whereas other stimuli (e.g. the presence of an audience) will engender a state of public self-awareness. A modification of this position, advanced by William J. Froming and G. Rex Walker, suggests that private self-awareness will lead to greater adherence to privately held attitudes toward a given behavior, whereas public self-awareness will lead to greater adherence to public attitudes or social standards. A final model, the Encoding Perspective, contends that self-awareness stimuli cause individuals to encode environmental information in terms of its relevance to the self. The theory suggests that self-awareness stimuli will be effective only when an individual's behavior is publicly observable by others. The present paper proposes a model of self-attentional processes based upon a general model of attitude-behavior relationships. It is suggested that the presence of a mirror will lead to greater adherence to the particular attitude salient at a given time. When responses of a subject are publicly observable by others, self-awareness (generated by the presence of a mirror) will lead to vi

PAGE 7

greater adherence to public attitudes or social norms. When responses are not publicly observed, self-awareness will cause greater adherence to privately held attitudes. The experiment described herein tests competing predictions by the different models of self-awareness. Persons who smoke cigarettes and who hold negative public attitudes toward smoking (i.e. believe that others disapprove of smoking) took part in an experiment designed to test their willingness to smoke under different circumstances. The model of self-awareness espoused in this paper predicted that the presence of a mirror would lead to greater smoking when the subject was alone, but to less smoking when the subject was with another person. Data from the experiment showed no significant overall differences due to the experimental manipulations. up analyses, however, suggested that mirror presence c.,/ Followincreased smoking only when ~ another individual was present. Results are discussed in terms of the different models of self-awareness. It is concluded that the model of self-awareness espoused in this paper was not supported, and that the pattern of results generally supports the Self-Consciousness Perspective. vii

PAGE 8

r CHAPTER I INTRODUCTION A revival of interest in the self in personality and social psychology has been evidenced in recent years by a growing literature dealing with self-focused attention. Research of this type has attempted to specify the behavioral and cognitive consequences of attention directed at the self rather than the environment. Several broad conceptualizations of self-attention processes have emerged from the growing literature in this area. The first of these, the Objective Self-Awareness perspective, was initially set forth by Duval and Wicklund (1972). The second point of view, espoused by Arnold Buss and his associates, may be described as the SelfConsciousness perspective (Buss, 1980). In addition, several information processing models of self-focused attention have been introduced in recent years (Carver, 1979; Hull and Levy, 1979). The review which follows will consider each of the two broad theories of self-attention processes, and one of the information processing models. In addition, an alternative l

PAGE 9

( model will be proposed which addresses some of the controversial issues among the various perspectives and attempts to resolve some of the differences. Theories of Self-Focused Attention The Objective Self-Awareness Perspective The concept of self-awareness was first introduced by Duval and Wicklund (1972). The authors distinguish two types of self-awareness, objective and subjective, in the following manner: It is assumed that the objectively self-aware individual will focus attention upon himself--his consciousness, personal history, or body--but, in contrast, the same person in subjective self-awareness will be aware of himself only as a source of forces that are exerted on the environment (page 25). Objective self-awareness is thus an awareness of oneself as an object in the environment. By contrast, subjective self-awareness is attention to the environment. In this state the individual is the subject rather than the object of his/her own awareness. The theory of objective self-awareness contends that at any point in time attention is wholly directed at the self or the environment (Wicklund, 1978), Attention may not be divided between the self and the environment, but 2

PAGE 10

may oscillate rapidly back and forth between the two. The extent to which one is objectively self-aware is thus the proportion of time spent focusing on the self. Changes in objective self-awareness are brought about by events in the environment. Those events which increase objective self-awareness are those which "remind the individual of his object status (Wicklund, 1978; p. 466)." Experimental manipulations designed to bring about objective self-awareness have included mirrors, television cameras, audiences, and tape recordings of the person's voice. Stimuli which distract the person's attention from the self to the environment are believed to decrease the amount of time spent in objective self-awareness. It is believed that many nervous habits are actually attempts to reduce objective self-awareness through self-distraction (Duval and Wicklund, 1972; see also Wicklund, 1975). The Duval and Wicklund theory of self-awareness is primarily a motivational theory, suggesting that objective self-awareness causes the individual to focus on discrepancies between his/her own behavior and some optimal standard of behavior. In many cases this discrepancy is described as one between the "real self" and the "ideal self" (cf. Liebling, Seiler, and Shaver, 1974). Since one's behavior is customarily less than optimal, the objectively self-aware individual is motivated to alter 3

PAGE 11

his/her behavior to bring it more closely into line with the standard. The above mentioned discrepancy between one's actual behavior and an ideal standard of behavior led Duval and Wicklund (1972) to postulate that objective self-awareness was an aversive motivational state. Accumulation of evidence to the contrary (Ickes, Wicklund and Ferris, 1973~ Steenbarger and Aderman, 1979), however, prompted Wicklund (1978) to modify his position in this regard. It is therefore believed that objective self-awareness is almost always, but not necessarily, aversive. In summary, then, the objective self-awareness perspective emphasizes the self-evaluative effects of self awareness. Resulting changes in behavior are attributable to the aversive nature of the self-aware state. For this reason, the objective self-awareness perspective has been described as a variant of consistency theory (Buss, 1980). The Self-Consciousness Perspective Initial Formulations. An alternate conceptualization of self-attention processes is provided by Buss (1980). The primary distinction between this approach and that of Duval and Wicklund (1972) is that the self is divided into two components: the private self and the public self. The private self is believed to include those aspects of the 4

PAGE 12

r self which cannot be observed by others, such as thoughts, attitudes, and self-reflections. Conversely, the public self is believed to include those aspects of the self which are publicly observable. Examples of these attributes include physical appearance and voice quality. The theory of self-consciousness is built upon a distinction found in the development of the Self Consciousness Inventory (Fenigstein, Scheier and Buss, 1974). Factor analysis of the scale, which is believed to measure chronic self-attentiveness, revealed two distinct types of self-consciousness. The dimension of private selfconsciousness reflected one's tendency to be selfreflective or introspective. The second factor, public self-consciousness, measured one's concern with appearances or the opinions of others. Buss (1980) extended the distinction between public and private self-consciousness to postulate the existence of states of public and private self-awareness. As the names suggest, private self-awareness is a state of self reflection whereas public self-awareness is a state of concern with one's outward appearance. In contrast to the Duval and Wicklund theory, the theory of self-consciousness does not assume that self awareness leads invariably to self-evaluation. For this reason it is a cognitive or attentional theory rather than 5

PAGE 13

a motivational theory. Moreover, it is assumed that states of self-awareness are affectively neutral rather than aversive. Changes in behavior resulting from states of self awareness are believed to result from attentional shifts rather than from a desire to reduce self-awareness. Persons who are made self-aware alter their behavior because the state of self-awareness enhances the salience of their own attitudes. This increased attention to attitudes engenders greater adherence to the appropriate attitude in a given situation, producing a large or small shift in behavior from that in the nonself-aware state. Modified theory. A modification of the Buss theory of self-consciousness is suggested by Froming and Walker (1981). This modification proposes that stimuli which produce private self-awareness (such as a mirror) and those which produce public self-awareness (such as an audience) can lead to different patterns of behavior under some circumstances. Specifically, the authors suggest that persons can hold differing public and private attitudes toward a given behavior. When this condition exists, a state of public self-awareness will lead to increased adherence to public attitudes (or standards) whereas a state of private self-awareness will increase adherence to private attitudes. 6

PAGE 14

Froming and Walker demonstrated the theory in an experiment involving the delivery of punishment to an experimental confederate. Subjects were pretested using a modification of the "Attitudes Toward Punishment" questionnaire of Carver (1975). The questionnaire was modified to create pairs of questions. The first question of each pair assessed the subject's own attitude toward punishment, whereas the second asked for an estimate of the attitude of "most people." Subjects selected for the experiment were those who were personally against the use of physical punishment but who felt that most other people favored its use. Results indicated that these persons would decrease their use of punishment (compared to a control group) in the presence of a mirror, but would increase their use of punishment in the presence of an evaluative audience. The authors concluded that investigations of public and private self-awareness should take into account the public and private attitudes appropriate to a given situation. The Encoding Perspective A final model of self-attention processes, described by Hull and Levy (1979), suggests an abandonment of the notion that mirrors increase self-focused attention in favor of an information processing model. This model proposes that self-awareness stimuli generate increased 7

PAGE 15

r attention to those aspects of the environment which are potentially relevant to the self. Instead of focusing on himself, the individual who is "self-aware" focuses on the environment in an effort to determine "self-relevant contingencies." The state of self-awareness is actually an increased tendency of persons to encode information in terms of its relevance to the self. This enables the individual to, in most instances, adopt an appropriate strategy of self-presentation which will be maximally effec t ive. One implication of this approach is that the effects of self-awareness will depend upon the self-presentational character of the experimental situation. Specifically, Hull and Levy predict that the effects of self-awareness will be shown only when the responses of the subject are available for examination by the experimenter. When the subject is allowed to respond anonymously, however, the self-awareness stimuli will have no effect. This was shown in two experiments, the first involving self-deprecation (Hull and Levy, 1979; Experiment II) and a second involving self-attribution (Experiment III). Comparison of the Models In order to compare the usefulness of the various models which have been described, it is necessary to review their ability to account for the existing research in self8

PAGE 16

awareness. The following paragraphs attempt such a review. This review, which is not exhaustive, addresses selected issues wherein differ e nt predictions are expected by the various models of self-awareness. The first issue concerns situations in which an individual finds his own judg ./rne nt or attitude at odds with that of other persons. In one such experiment (Wicklund and Duval, 1971), it was found that persons made self-aware by the presence of a mirror were more respqnsive to persuasion attempts from a positive reference group. This outcome is predicted by the objective self-awareness model, as well as that of Hull and Levy (1979). However, the self consciousness perspective would predict that persons made self-aware by a mirror would focus more strongly on their own (private) attitudes and therefore be more resistant to persuasion. A second general issue is that of attitude-behavior relationships. Several studies (e.g. Carver, 1975; Gibbons, 1978) have shown that self-awareness can increase the consistency between attitudes and behavior. The objective self-awareness model would suggest that self awareness leads the individual to focus on the attitude behavior discrepancy, and thereby motivates compliance with personal attitudes. The self-consciousness model would attribute the enhanced consistency to a greater awareness 9

PAGE 17

of the attitudes themselves. Both models are therefore consistent with the data on this issue. The encoding perspective, however, attributes behavioral change under conditions of self-awareness to increased focus on environmental contingencies related to the self. There is no reason to believe that these environmental contingencies would be correlated with private attitudes. Thus the increased correlation between attitudes and behavior demonstrated in self-aware subjects would not be predicted by the encoding model. A final issue concerns focus of attention on nonattitudinal information. Research has shown that persons made self-aware are less susceptible to placebo effects (Gibbons, Carver, Scheier, and Hormuth, 1979) and more veridical in their assessment of their own bodily states (Scheier, Carver and Gibbons, 1979). These effects are predicted by the self-consciousness model because of its emphasis on attentional focus. However, because it is primarily a consistency theory, the objective self-awareness perspective would not predict such an effect. The encoding model also cannot directly speak to these data, but one might expect that the model would predict, if anything, increased responsiveness to placebo effects under conditions of self-awareness. 10

PAGE 18

i An Alternative Model The inconsistencies in research in self-focused attention suggest that none of the above theories is adequate to handle all of the findings. The model described herein addresses some of these inconsistencies and proposes a modified theory of self-attention processes. The self-awareness theory being proposed incorporates the attitude-behavior theory of Fishbein (1967) into the self-consciousness model of Buss (1980) and Froming and Walker (1981). Developing the Model. Fishbein (1967) proposed a theory of attitude-behavior relationships summarized by the following equation: B ~BI= [Aact]wo + [NB(Mc)]wl In the equation, B refers to the overt behavior, BI is the behavioral intention, Aact is the individual's attitude toward the act, NB is the normative social belief, and Mc is the motivation to comply with the norm. The weights (w 0 and w 1 ) are empirically derived and are specific to a given individual in given situation. The model thus pro~oses that a person's behavioral intention is a function of his/her own attitude, the normative social belief, and the person's motivation to comply with the social norm. 11

PAGE 19

Translating this model into the language of selfconsciousness theory, it can be seen that t~e ~ttitud e toward the act ( Aact) is what Fro::nina ann W;:i 1 ker (1981) describe as the private attitude, and that the normative belief (NB) is the public attitude. Both models are thus suggesting that behavior is a function of both public and private attitudes. A model of self-awareness processes can be formulated by adding another term to the existing model: } SA In this new equation, SA refers to the effects of mirror induced self awareness. Before assessing the predictions of the model, the implications of the encoding model of Hull and Levy (1979) should be considered. Hull and Levy demonstrated that the effect of self-awareness will in many cases depend upon the public or private nature of the situation. In terms of the above model, Hull and Levy are saying that the effect of self-awareness on behavior will depend on the relative values of the weights assigned to Aact or NB. Example. To understand the workings of the above model, several examples of its operation will be considered. First, consider a situation in which self awareness is manipulated by the presence or absence of a mirror. Suppose that the weight assigned to Aact is one 12

PAGE 20

(1) and the weight assigned to NB is four (4). Further, suppose that the motivation to comply with the normative belief is greater than zero and constant across all conditions. The presence or absence of a mirror in the model is denoted by the values of SA. Suppose that the value of SA is four (4) when the mirror is present and one (1) when the mirror is absent. Comparing the two conditions, we find the two following outcomes: Mirror Present: BI= 4(Aact) + 16(NB) Mirror Absent: BI= l(Aact) + 4(NB) The difference between the two self-awareness conditions [3(Aact) + 12(NB)] is largely reflected ~n differences in adherence to normative beliefs. Moreover, as the value of the weight assigned to Aact approaches zero, the differences between the groups become entirely attributable to adherence to the normative belief. In a situation where the weights assigned to the normative belief and the attitude toward the act are reversed, it can be easily demonstrated that differences in self awareness will be reflected in differential adherence to attitudes toward the act. It is thus generally true that the effect of self awareness will be to amplify existing differences between 13

PAGE 21

weights assigned to public and private attitudes. Stated another way, differences in behavior between groups which are high and low in self-awareness will depend upon the relative assignment of weights assigned to personal and social standards. Implications. The primary implication of the above model is a modification of an argument set forth by Hull and Levy (1979). The argument contends that differences in behavior resulting from self-awareness will depend upon the public or private character of the situation. Hull and Levy implied that self-awareness would be primarily operative in situations where the person's behavior would be publicly observable by others. In terms of the present model, the effects of self awareness can be demonstrated in situations which are both public and private. The effects of self-awareness will be reflected in differential adherence to attitudes, and the salient attitude in a given situation will depend upon the public/private character of that situation. A major implication of the present model is that two traditionally accepted self-awareness manipulations--mirror presence and audience presence--are quite different in their effects. The effect of a mirror is represented by higher values of SA. Behaviorally, this is reflected in increased adherence to the salient attitude in a given 14

PAGE 22

I situation. The effect of an audience is represented by an increase in the weight assigned to NB and possibly a decrease in that assigned to Aact. This is reflected in increased attention to social norms rather than personal attitudes. In a review of self-awareness literature, Diener and Srull (1979) suggested that most experiments involving self-awareness have demonstrated increased adherence to social norms by persons made self-aware. If this is indeed the case, then in terms of the model being proposed the finding is not surprising. Virtually all self awareness experiments were conducted in a manner such that the responses of subjects are publicly available. Self~ awareness thus augments socionormative influence in these situations because the greatest weight is given to social norms. A final corollary of the above model is that the effect of mirror presence can interact with the effect of audience presence. Such a situation would require that individuals hold differing public and private attitudes toward a given behavior. An experiment of this type is described below. 15

PAGE 23

Evaluation of the Model In order to assess the utility of the above model, its ability to account for research in self-awareness will be considered. Each of the three issues used in comparison of models of self-awareness will be considered from the perspective of the new model, and the correspondence of the mod~l's predictions with the existing data will be assessed. The first issue involves situations in which people find their own attitude at odds with that of others. The model espoused herein would predict that the presence of a mirror will enhance responsiveness to the attitudes of others when the subject's responses are publicly available. Thus the findings of Wicklund and Duval (1971) are predicted by the model. The second issue involves attitude-behavior relationships. The model would predict adherence to the attitude which is salient in a given situation. Data from Froming and Walker (1981) suggest that this is the case. Data from Carver (1975), which show that the correlation between private attitudes and behavior is enhanced by the presence of a mirror, might seem to be at odds with the prediction, since the experiment was one in which responses were not anonymous. Supplemental data analyses performed by Framing and Walker, however, reveal 16

PAGE 24

that public and private attitudes on the Carver "Attitudes Toward Punishment" scale are highly correlated. Hence the data from the Carver (1975) experiment do not speak to the issue of public and private attitudes. The issue of focus of attention on nonattitudinal information is not dealt with by the present model. Since the model is a modification of that of Fishbein (1967), it retains his emphasis on attitude-behavior relationships. Nevertheless, it is not unreasonable to extrapolate private attitudes to self-perception. Thus a subject participating in the Scheier, Carver and Gibbons (1979) experiment concerning taste perception would hold the taste itself as the "private attitude" and the suggestion of the experimenter as the "public attitude." In this experiment, subjects were asked to taste various flavors and rate the intensity of the taste. Subjects were incorrectly told that others had found one flavor stronger than another. Results indicated that subjects made self-aware were more resistant to this suggestion than nonself-aware subjects. The experiment of Scheier, Carver and Gibbons, however, is ambiguous with respect to the nature of the public and private attitudes. The experimenter introduced the suggestion by saying that "most people" find that one taste is stronger than the other. From the 17

PAGE 25

standpoint of the subject, "most people 11 (that is, those who had supposedly participated in the experiment previously) were not audience members. For this reason, the change in the weight associated with the public attitude is small when the experimental suggestion is added. The model would therefore predict increased salience of the "private standard," the taste. The model being proposed is therefore consistent with much of the existing literature in self-awareness. Nevertheless, no single experiment contains all of the conditions for a rigorous test of the model. The following section attempts such a test. A Test of the Model The reseach described in this dissertation will test the hypothesis that the effects of mirror presence can interact with those of audience presence. The scenario for this test is one in which an individual's public and private standards toward the behavior of interest are expected to differ. Moreover, a behavior is selected which can be measured under conditions of anonymity and nonanonymity of response. Liebling, Seiler and Shaver (1974) described an experiment in which the smoking behavior of persons was measured when a mirror was present, and when it was absent. 18

PAGE 26

In this experiment, the participants were told that the experimenters were interested in musical preferences of college students. They were asked to sit in a room for thirty minutes and listen to several selections of music, after which their reactions to the music would be assessed. During the time that the participants listened to the music, experimental assistants covertly observed them and measured several indicants of smoking behavior. Results indicated that the presence of a mirror effected an increase in smoking. The authors suggested that the results contradicted the predictions of the Duval and Wicklund (1972) theory of objective self-awareness. Pretest results indicated that the participants rated their "ideal self" as smoking less than their "real self" (i.e. their actual behavior). According to Leibling et al. (1974), objective self awareness theory would predict that the presence of a mirror would cause subjects to focus on the discrepancy between real and ideal behavior, and therefore they should smoke less. Liebling et al. concluded that a drive theory was more useful in explaining mirror effects than a theory of self-attention. Wicklund's (1975) response to this paper suggested that smoking served to reduce objective self awareness by serving as a distractor. Liebling et al. 19

PAGE 27

(1975) responded to Wicklund's reply by questioning the utility of a theory which can make two opposing predictions in a single situation. The results of the above experiment can be explained in terms of the model of self-awareness espoused herein. From the standpoint of smoking behavior, the experiment of Liebling et al. was conducted under conditions of complete anonymity, since no one was in the room with the subject and he/she was unaware of the observer. Accordingly, it can be expected that the private standard (Aact) would be most salient. In this case the private standard would be the person's desire for a cigarette, or his/her general enjoyment of smoking. For this reason, the presence of a mirror ; is expected to enhance the influence of the private attitude, leading to greater smoking. The experiment to be described in this dissertation will address the opposite situation. If the participant in the experiment is in the presence of another individual, it is likely that the presence of a mirror will enhance the effect of the public standard or normative belief. This would be due to the increased salience of the public standard resulting from the nonanonymity of the situation. Thus the presence of a mirror should bring about increased adherence to a public standard. 20

PAGE 28

In recent years, public attitudes toward smoking behavior have become decidedly negative. Increasing concerns with health and increasing attention to the rights of nonsmokers have placed the smoker in a position of having differing public and private attitudes toward smoking. For this reason, a manipulation which serves to increase one's adherence to public standards toward smoking would cause a decrease in smoking behavior. The test of the theory which will be conducted therefore examines the interactive effect of mirror presence and audience presence (or anonymity) on smoking behavior. The theory predicts that the direction of the effect produced by the presence of a mirror will depend upon the public/private nature of the situation. Specifically, it is predicted that the presence of a mirror will decrease smoking in the presence of an audience but will increase smoking when the individual is alone. Alternative Predictions Objective Self-Awareness Perspective. As Liebling et al. (1974) have pointed out, the Duval and Wicklund (1972) model makes two predictions. Nevertheless, with respect to the critical issue of this paper, their position is consistent: the effects of the mirror and the audience will be equivalent. Accordingly, the direction (though not 21

PAGE 29

necessarily the magnitude) of the effect produced bv a mirror without an audience will be the same as that produced by an audience without a mirror. Moreover, it is not unreasonable to infer that the effects will be additive, that is, the effect of the mirror and the audience together will be greater than that of either effect alone. Self-Consciousness Perspective. The modification of Buss (1980) theory of self-consciousness suggested by Froming and Walker (1981) would predict opposite effects for the mirror and audience. Private self-awareness, produced by mirror alone, will result in increased responsiveness to private attitudes and thereby lead to increased smoking. Public self-awareness, produced by the audience alone, will lead to decreased smoking as subjects react to the increased salience of the public attitude against smoking. When both the mirror and audience are present, however, the effects will tend to cancel. The rate of smoking by persons in the presence of both a mirror and an audience should therefore fall somewhere between that in the presence of the audience alone and that in the presence of the mirror alone. Encoding Perspective. The encoding perspective of Hull and Levy (1979) would predict that the presence of a mirror would alter the responsiveness of individuals to 22

PAGE 30

self-relevant contingencies in the environment. For subjects in the audience conditions, those contingencies concern the response of the audience to their smoking. Since smoking is viewed negatively by many people, it is likely that these contingencies would favor a reduction in the rate of smoking. For this reason, the Hull and Levy theory would predict reduced smoking i? the presence of a mirror when an audience is present. Since there are no environmental contingencies related to smoking when the audience is absent, Hull and Levy would predict no effect of mirror presence when the audience is absent. 23

PAGE 31

CHAPTER II METHOD Overview The purpose of the experiment was to demonstrate differences in the effect of a mirror in situations in which the response of a subject may or may not be observed by another individual. An experimental paradigm was selected in which the subject might be expected to hold a personal belief which is inconsistent with the attitudes which he/she believes are held by others. The experiment was designed to replicate and extend that conducted by Liebling, Seiler, and Shaver (1974). It was previously mentioned that this experiment was originally conducted with the subject alone in the experimental room. The present experiment seeks to replicate these findings, and extend them to situations in which another individual is present. 24

PAGE 32

Subjects Subjects were 11 male and 21 female undergraduates at the University of Florida. Subjects were recruited from introductory psychology classes using two methods of recruitment. The first method utilized sign-up sheets. These sheets requested that only subjects who smoke more than 8-10 cigarettes per day volunteer for the experiment. The second method selected subjects on the basis of pretesting conducted at the beginning of the academic quarter. Subjects responded to a series of questions regarding their smoking habits. Those subjects who indicated that they smoked more than one-half pack of cigarettes per day were selected for the experiment. Procedure Eight subjects signed up for the experiment on the sign-up sheets. They were contacted by telephone and told that they would be participating in an experiment dealing with the effects of cigarette smoking on perception and memory. A time was arranged for an initial session. During this initial session, subjects completed a questionnaire dealing with attitudes toward smoking. The first question asked the subjects to estimate the number of cigarettes smoked during an average day, and to 25

PAGE 33

indicate their favorite brand of cigarettes. Several other questions assessed the subjects' public and private attitudes toward smoking. The remaining questions were included to help maintain the deception. In addition to the smoking attitudes questionnaire, subjects completed the Self-Consciousness Inventory (Fenigstein, Scheier, and Buss, 1974) and the Fear of Negative Evaluation Scale (Watson and Friend, 1969). After completing the questionnaires, the subject returned them to the experimenter who, if the subject evidenced the required discrepancy between public and private attitudes, invited the subject to take part in the two additional sessions of the experiment. Subjects selected from the pretesting session (N = 24) were selected on the basis of the number of cigarettes smoked and the presence of a discrepancy between public and private attitudes toward smoking. They were contacted by telephone and invited to participate in the experiment. The critical questions used in the selection of subjects are found in Appendix A. A protocol of specific experimental procedures is found in Appendix B. Participants were greeted in a waiting room by an experimental assistant. The assistant explained that the experiment dealt with the effects of smoking on several types of perceptual and motor skills. 26

PAGE 34

The subject was reminded that the experiment was to be conducted in two sessions, the second of which would take place in about a week. Informed consent was obtained from the subject verbally If the subject agreed to participate, he/she was told that there would be a twenty-five minute waiting period prior to the start of the experiment, since the experiment dealt with heart rate. The subject was told that during the waiting period he/she should light a cigarette and take at least one or two puffs. Any amount of smoking beyond that was said to have no effect on the experiment. The subject was asked to wait quietly in the experiment room, which contained several packs of cigarettes (in case subjects forgot to bring their own), a pack of matches and an ashtray. The room also contained a radio. The subject was told that he/she could not read or write during the waiting period, but could listen to the radio if he/she so desired. On "Mirror Present" trials, the subject was seated at a table facing a small (24" x 24") mirror placed at eye level. For trials in which the mirror was absent, the mirror was covered with a sheet of white paper. During trials "Audience Present" trials, an experimental confederate was seated in the room with the subject. The confederate had been in the room prior to 27

PAGE 35

the subject's arrival, and the experimenter explained that the confederate was scoring some questionnaires for a professor. The subject was asked not to talk to the confederate. If a subject later asked the confederate if he/she may smoke, the confederate was instructed to respond with "I don't know" or some other ambiguous response. This, however, did not occur. As the subject waited, he/she was observed by an experimental assistant through a two-way mirror. The mirror was almost completely covered with cardboard on the subject's side, but sufficient space remained uncovered to permit observation. The measures collected by the observer were those reported by Liebling, Seiler, and Shaver (1974). These were: 1) Lighting, 2) Flicking, 3) Puffing, and 4) Holding Time. Upon completion of first session of the the experiment, the subject was given the perceptual and memory tasks. The first task was an abbreviated version of the "Embedded Figures Test" (Wi tkin, 1969). The second was a reverse digit span exercise. At the conclusion of both parts of the experiment, the subject was carefully debriefed by the experimenter using a predetermined set of questions (Appendix C). Subjects who were aware that the experiment was concerned with cigarette smoking during the waiting period, or who were 28

PAGE 36

I certain that they were being observed were to be eliminated from all analyses. Those expressing vague suspicions about the nature of the experiment were not. Design The experimental design was a 2 X 2 factorial with two Mirror conditions (Mirror and No-mirror) and two Audience conditions (Audience or No-audience). The design is of the "mixed" type with repeated measures on the Mirror factor only. 29

PAGE 37

CHAPTER III HYPOTHESES The model of self-awareness proposed in Chapter I can be used to make predictions concerning the effects of public and private self-awareness on adherence to public and private attitudes toward smoking. The model is described by the following equation: B BI = { [Aact]w 0 + [NB(Mc)Jw 1 } SA Predictions about public and private determinants of cigarette smoking require some assumptions about the values entered into the model. First, the value of SA will be 1 when the mirror is absent, and 2 when it is present. The value of the weight assigned to Aact (w 0 ) will be 2 when the subject is alone and 1 when an audience is present. The value of the weight assigned to NB (w 1 ) will be 1 in the No Audience condition and 2 in the Audience condition. In all cases the value of NB is 1. The values assigned to Aact and NB present a more difficult problem. Subjects selected for the present experiment are those who personally believe that smoking is enjoyable, but who believe that others object to smoking. The selection criteria, however, are based upon 30

PAGE 38

the relative values of public and private attitudes (requiring only that the private attitude be more favorable toward smoking than the public attitude), rather than the absolute values of these attitudes. In making predictions from the equation, it will be assumed that the value of Aact is always +land the value of NB is always -1. The reader should note, however, that the model will make a different set of predictions if both Aact and NB are positive, and still another set of predictions if both are negative. Nevertheless, since it is being assumed that subjects selected for the experiment are personally in favor of smoking and feel that others are against smoking, it is believed that the choices of +land -1 are the most appropriate. Substituting into the equation, the following values of BI are predicted for the four experimental conditions. No Mir/No Aud: BI= ([+1](2) + [-1](1)(1)}1 = 1 Mir/No Aud: BI= {[+1](2) + [-1](1)(1)}2 = 2 No Mir/Aud: Mir/Aud: BI= {[+l](l) + [-1](1)(2)}1 BI= {[+l](l) + [-1](1)(2)}2 = -1 = -2 These predictions suggested the relative values of the mean levels of smoking by subjects in each experimental condition. Based upon these computations, the following formal experimental hypotheses are advanced. 31

PAGE 39

1. It is predicted that the effect of Mirror Presence will depend upon the presence or absence of an audience. An interaction of these effects is thus the first hypothesis. 2. It is predicted the direction of the Mirror effect will be different across Audience conditions. Specifically, it is predicted that the presence of a mirror will lead to decreased smoking when an audience is present and will cause increased smoking when the audience is absent. 32

PAGE 40

CHAPTER IV RESULTS A total of 32 subjects participated in the experiment. Of these, two were eliminated for failure to follow instructions. Results of this experiment are based upon the analyses of data from 30 subjects, 15 in each audience condition. Table 4-1 summarizes the means of subjects in each treatment combination on the dependent measures of smoking. The measures of smoking considered in this experiment included the number of cigarettes smoked (Lights), the number of times the subject flicked ashes off of the cigarette (Flicks), the number of puffs taken (Puffs), and the total amount of time, in seconds, spent holding the cigarette(s) (Time). Although order of trials was randomly counterbalanced within treatment conditions, it was believed to 33

PAGE 41

substantially affect the dependent measures. For this reason, Tables 4-2 and 4-3 present the means of the treatment combinations for subjects having the same order of trials. Means of the fourteen subjects who were not exposed to the mirror on the first trial are presented in Table 4-2. Means from the sixteen subjects exposed to the mirror on the first trial are presented in Table 4-3. A visual inspection of these tables suggests that subjects tended to smoke less on the second trial, regardless of the treatment condition. Accordingly, the principal analyses include order of presentation as a factor. The means presented in Table 4-1 suggest that the presence of a mirror causes an increase in smoking. With one exception (the Puffs measure for the No Audience group), all indicants of smoking were greater when subjects were exposed to a mirror. This was true for subjects in both the Audience and No Audience conditions. The means also suggest that the presence of an audience leads to decreased smoking. With one exception again (the Flicks measure on Mirror trials), subjects exposed to an audience smoked less than those not exposed to an audience on both Mirror Present and Mirror Absent Trials. Statistical analyses of the mean differences were performed using univariate analyses of variance on the 34

PAGE 42

four dependent measures. Each analysis included two "between group" factors (Condition and Order) and one "within subject" factor (Trial). The two levels of Condition were Audience and No Audience: the two levels of Order were Mirror First and No Mirror First: the two levels of Trial were Mirror and No Mirror. Analysis of variance tables are presented in Table 4-4 (Lights), Table 4-5 (Flicks), Table 4-6 (Puffs) and Table 4-7 (Time). The general pattern of the statistical analyses suggests no significant effects due to either the presence of an audience or the presence of a mirror on any of the dependent measures. The lone exception to this pattern is a significant main effect of Mirror Presence on the Time measure (F (1,26) = 4.62, p < .OS). However, due to the large number of comparisons being made in the experiment, this result can be discounted. Correlations of the individual difference measures with the dependent measures of smoking on Mirror Present and Mirror Absent trials are presented in Tables 4-8 and 4-9 None of the correlations were statistically significant. Since the magnitude of the correlations suggested a possible relationship between private attitudes toward smoking and the measures of smoking, a partial correlation analysis was performed. Results of the analysis indicated no significant relationship between 35

PAGE 43

------------private attitudes and smoking eliminating public attitudes, or between public attitudes and smoking eliminating private attitudes. A comparison of the results of the present study with the results reported by Liebling, Seiler and Shaver (1974) is presented in Table 4-10. The table reports percent changes in the four indicants of smoking and t-tests on the mean differences for both experiments. The results of the present study are reported by levels of the audience factor. Referring to the table, it can be seen that in the Audience condition, three of the four measures of smoking showed significant increases as a result of the mirror manipulation. The data also suggest a great deal of similarity between the Audience condition of the present study and the results of the Liebling et al. study. It is worth noting, however, that the results of multiplet-tests should be viewed with caution. Answers to the debriefing questions were tabulated as a check on manipulations and deceptions and are summarized in Table 4-11. It should be remembered that the answers to first three questions in the debriefing were recorded before the subject was told of the deception. The answers to the final five questions were given after the experimental procedures had been disclosed. 36

PAGE 44

Table 4-1: Means and Standard Deviations on the Dependent Variables for Each Experimental Treatment Combination. (N = 15 in each group.) Trial Group Variable No Mirror Mirror No Lights 1.53 (0.51) 1.60 (0.51) Audience Flicks 8.80 (8.19) 11.33 (8.56) Puffs 13.47 (7.68) 13.47 (5.51) Time 495.00 (218.80) 577.57 (227.40) ----------------------------------------------Audience Lights 1.20 (0.52) 1.47 (0.52) Flicks 7.47 (3.81) 11.33 (9.70) Puffs 8.60 (4.60) 12.53 (6.75) Time 359.60 (118.70) 481.13 (216.49) 37

PAGE 45

Table 4-2: Means and Standard Deviations on the Dependent Variables for Each Experimental Treatment Combination, for Subjects Who Were Exposed to the Mirror on the First Session. (N = 7 for each group.) Trial Group Variable No Mirror Mirror No Lights 1.57 (0.53) 1.71 (0.49) Audience Flicks 9.29 (10.82) 12.00 (9.20) Puffs 11.71 (7.41) 13.71 (5.93) Time 486.71 (179.00) 566.14 (243.64) ----------------------------------------------Audience Lights 1.14 (0.38) 1.57 (0.53) Flicks 7.28 (4.46) 14.85 (13.06) Puffs 8.85 (5.67) 15.43 (8.12) Time 343.71 (121.35) 572.29 (263.86) 38

PAGE 46

Table 4-3: Means and Standard Deviations on the Dependent Variables for Each Experimental Treatment Combination, for Subjects Who Were Exposed to 39 the Mirror on the Second Session. (N = 8 for each group.) Trial Group Variable No Mirror Mirror No Lights 1.50 (0.53) 1.50 (0.53) Audience Flicks 8.37 (5.78) 10.75 (8.56) Puffs 15.00 (8.07) 13.25 (5.52) Time 502.25 (261.65) 515.37 (297.08) ----------------------------------------------Audience Lights 1.25 (0.46} 1.37 (0.52} Flicks 7.62 (5.78) 8.25 (4.33) Puffs 8.37 (3.81} 10.00 (4.34) Time 373.50 (122.82) 401.37 (135.94)

PAGE 47

Table 4-4: Analysis of Variance Table for the Dependent Variable "Lights." Source df Sum of Squares Mean Square F Order 1 0.13 0.13 0.36 Audience 1 o.83 0.83 2.32 Aud x Ord 1 0 04 0.04 0.10 Error ( B) 26 9.36 0.36 Mirror l 0.45 0.45 3.16 Mir x Ord 1 0.18 0.18 1.30 Mir x Aud l 0.16 0.16 1.10 M X 0 X A l 0.02 0.02 0.17 Error (W) 26 3.72 0 .14 40 p .14 .os .26 .30

PAGE 48

Table 4-5: Analysis of Variance Table for the Dependent Variable "Flicks." Source df Sum of Squares Mean Square F Order 1 66.30 66.30 0.91 Audience 1 5.34 5.34 0.07 Aud x Ord 1 15.74 15.74 0.22 Error ( B) 26 1891.01 72.73 Mirror 1 164.74 164.74 3.00 Mir x Ord 1 49.54 49.54 0.90 Mir x Aud 1 9.01 9.01 0.16 M X 0 X A 1 40.74 40.74 0.74 Error ( W) 26 1427.44 54.90 41 p .09

PAGE 49

Table 4-6: Analysis of Variance Table for the Dependent Variable "Puffs." Source df Sum of Squares Mean Square F Order 1 8.90 8.90 0.18 Audience 1 113.30 113.30 2.23 Aud x Ord 1 71.17 71.17 1.40 Error (B) 26 1320.26 50.78 Mirror 1 66.59 66.59 2.43 Mir X Ord 1 70.58 70.58 2.58 Mir x Aud 1 58.93 58.93 2.15 M X 0 X A l 1.33 1.33 0.05 Error ( W) 26 712.54 27.40 42 p .14 .25 .13 .12 .15

PAGE 50

Table 4-7: Analysis of Variance Table for the Dependent Variable "Time." Source df Sum of Squares Mean Square F Order 1 29 028.39 29 028.39 0.43 Audience 1 134 494.81 134 494. 81 2.00 Aud x Ord 1 10 465.74 10 465.74 0.16 Error ( B) 26 1 744 154.80 67 082.87 Mirror 1 113 680.93 113 680.93 4.62 Mir x Ord l 66 535.40 66 536.40 2.70 Mir x Aud l 25 070.14 25 070.14 1.02 M X 0 x A l 16 857.34 16 857.34 0.68 Error (W) 26 640 279.59 24 626.14 43 p .16 .04 .11 .32 I

PAGE 51

44 Table 4-8: Correlations of Individual Difference Variables With Smoking Behavior of Subjects in Mirror and No Mirror Conditions When an Audience Was Absent. Individual Difference Variables Dependent NumSmk PrAtt PuAtt PrSC PuSc SAnx FNE Variables NO MIRROR CONDITION Lights .oo .22 -.01 .oo .09 -.13 .04 Flicks -.21 .36 -.05 -.27 .04 -.16 .os Puffs -.07 .11 -.11 .11 .21 .08 ,06 Time ,07 .22 .08 .03 .04 .20 ,16 MIRROR CONDITION Lights .04 .07 -.03 -.24 -.12 -.17 .02 Flicks -,16 .19 .07 -.16 .01 .12 .35 Puffs -,07 -.17 -.17 .OS .12 ,27 .28 Time .44 -.04 .05 .30 -.01 -.OS -.15 Abbreviations: NumSmk Average number of cigarettes I day PrAtt Private attitude toward smoking PuAtt Public attitude toward smoking PrSC Private Self-Consciousness PuSC Public Self-Consciousness SAnx Social Anxiety FNE Fear of Negative Evaluation

PAGE 52

45 Table 4-9: Correlations of Individual Difference Variables With Smoking Behavior of Subjects in Mirror and No Mirror Conditions When an Audience Was Present. Individual Difference Variables Dependent Variables NumSmk PrAtt PuAtt PrSC NO MIRROR CONDITION Lights Flicks Puffs Time -.21 -.33 -.07 -.17 MIRROR CONDITION Lights Flicks Puffs Time -.30 -.19 -.15 -.15 -.37 -.35 -.47 .02 -.03 .04 04 .21 06 00 -.29 -.36 -.41 -.06 -.15 -.03 43 09 .11 -.22 .32 .14 .26 -.11 PuSc SAnx -.02 .20 -.33 .07 06 06 -.32 -.18 -.27 -.29 -.40 -.32 .13 .05 .08 .27 FNE .24 -.15 .17 -.26 -.19 -.29 -.29 -.14 Abbreviations: NumSmk PrAtt PuAtt PrSC PuSC SAnx FNE Average number of cigarettes/ day Private attitude toward smoking Public attitude toward smoking Private Self-Consciousness Public Self-Consciousness Social Anxiety Fear of Negative Evaluation

PAGE 53

Table 4-10: Comparison of Experimental Results With the Results Reported by Liebling, Seiler, and Shaver (1974). Group WALKER-No Audience Audience Measure Lights Flicks Puffs Time Lights Flicks Puffs Time Percent Increase in Smoking In Mirror Condition and T-tests Change 4 % 29 % 0 % 17 % 16 % 52 % 46 % 34 % t 0.43 0. 79 o.oo 1.11 2.26 1.92 2.37 2.14 p .67 .45 1.00 .29 .04 .08 .03 .OS LIEBLING, SEILER, AND SHAVER-No Audience Lights Flicks Puffs Time 30 % 123 % 46 % 45 % 1.58 3.25 2.35 2.35 n.s. < 01 < .os < 05 46

PAGE 54

Table 4-11: Tabulation of Responses to the Debriefing Questions. 1 2 3 4 5 6 7 QUESTION Were tests affected by smoking? Other purposes to experiment? Guess other purposes? Suspect that being observed? If yes to #4, was smoking affected? Affected by the mirror? Did/would an audience affect you? RESPONSES (Grp A= Audience Condition Grp N = No Audience Condition) Grp A: No 93% Grp N: No 67% Grp A: No 80% Grp N: No 93% Yes 7% Yes 33% Yes 20% Yes 7% No correct guesses were given. All: Definitely yes Maybe 0% 63% 37% No All: No Yes Grp A: No 67% Grp N: No 80% Grp A: No 73% Grp N: No 67% 100% 0% Yes 33% Yes 20% Yes 27% Yes 33% 47

PAGE 55

CHAPTER V DISCUSSION The statistical analyses presented in the previous chapter suggest that none of the experimental manipulations significantly affected the smoking behavior of the subjects. The discussion which follows offers possible reasons why this may be the case, and uses the data for a limited comparison of the models of self awareness presented herein. Experimental Procedures The procedure of this experiment attempted to parallel that of Liebling, Seiler and Shaver (1974). The primary difference between the two experiments concerns the method used to generate smoking behavior. Liebling et al. told subjects that the experiment concerned reactions to selections of music, and that they could smoke while listening if they wished. The present experiment informed 48

PAGE 56

subjects that purpose of the research was to study smoking, but misinformed subjects as to the aspects of smoking which were under investigation. More important, the present study required that subjects smoke a minimal amount during the waiting period. The results of the experiment suggest that this procedural modification substantially altered the subjects' responses to the situation. One result of the of the change was that subjects smoked more in the present experiment than in the Liebling et al. experiment. The mean number of cigarettes smoked in the Liebling et al. experiment was 1.06 for the No Mirror condition and 1.38 for the Mirror condition. Referring to Table 1, it can be seen that subjects in the present experiment smoked considerably more than their counterparts in the Liebling et al. experiment. This difference is especially striking when one takes into account the fact that the subjects in the Liebling et al. experiment were observed for 30 minutes, as compared to 25 minutes in the present experiment. It would appear that the requirement that the subjects smoke greatly altered the situation. This is evidenced by the results of the debriefing. Only 27% of those in the Audience condition reported being affected by the audience. Most reported that since the experiment was 49

PAGE 57

concerned with smoking, the confederate had no right to object. Furthermore, the effect of the audience was often manifested in other behaviors. A large number of subjects reported being concerned that the noise from the radio in the experimental room would bother the confederate, and as a result they tried to keep the volume as low as possible. Thus it is possible that there was a substantial audience effect, but that the nature of the experiment prevented it from affecting the subjects' smoking behavior. One can conclude from the present study that the method used in this experiment to study attitude-behavior relationships related to smoking is of limited utility. One might suppose that the decision to smoke or not smoke in a given situation is a separate issue from the decision of how much to smoke. By eliminating the first issue, the present experiment reduced the probability of showing significant effects. Implications for Self-Awareness Models The primary and follow-up analyses of this experiment allow only a very limited number of conclusions to be drawn from the present research. Those conclusions 50

PAGE 58

which are drawn herein should therefore be considered very tentative at the present time. Implications for the Interactive Model The initial purpose of the experiment was to test the notion that the effect of a mirror could, under certain circumstances, depend upon the presence or absence of an audience. Specifically, it was predicted that mirror presence would increase smoking in the absence of an audience and decrease smoking when an audience was present. The data from this experiment are in contradiction to these predictions, and suggest that the effect of a mirror will be the same in the presence or absence of an audience. Moreover, the results of the multiplet-test analyses (Table 4-10) suggest that the increase in smoking generated by the presence of a mirror may actually be enhanced by an audience. The data from this experiment thus provide no support for the model of self-awareness presented in this paper. Although these results are damaging to the theory, they do not discount this model of self-awareness entirely. A major problem with operationalizing the implications of this model of self-awareness concerns the nature of the public standards. Every attempt was made in this experiment to make it appear that the confederate 51

PAGE 59

(audience) was not connected with the experiment itself. This was designed to create a situation in which the subject would have to infer the attitude of the subject based on his/her beliefs about the attitudes of "most other people." Evidence from the debriefing suggests that the attempt to separate the confederate from the experiment was not successful. The rationalizations used by the subjects suggest that the public standard appropriate to this situation was more complex than was previously imagined. Since there is no clear indication of the nature of the public standard in this situation, and no evidence that the standard was the same for all subjects, the present experiment cannot be regarded as an adequate test of the theory of self-attention espoused herein. Future research in this area should attempt to create situations in which public and private standards can be more rigorously controlled and more accurately documented. Implications for Other Models Objective Self-Awareness Perspective. The Objective Self-Awareness model of Duval and Wicklund (1792) receives no support in this experiment. This model would predict that the effect of the mirror and the effect of the audience should differ only in magnitude. The model would 52

PAGE 60

further suggest that the combination of two manipulations of self-awareness should have the greatest effect. In the present experiment, this model would predict that smoking would be greatest in the Audience/Mirror condition and least in the No Audience/No Mirror condition. Neither prediction is supported by this experiment. These data, taken together with those of Framing and Walker (1981), suggest that the notion that all self-awareness manipulations are in all cases unidirectional may be incorrect. Encoding Perspective. The Encoding Perspective of Hull and Levy (1979) suggests that the effect of mirror presence will only be shown when the responses of an individual are publicly observable by others. Ref~rring to Tables 4-1 and 4-10, it would appear that this prediction is supported by the experimental results. The t-tests of Table 4-10 are only significant in the Audience condition. The Encoding Perspective, however, does not explain the (non-significant) differences between the Audience and No Audience conditions. From the standpoint of this model, the increase in smoking resulting from the presence of a mirror implies that the ''self-relevant contingencies" are related to increased smoking. If this is the case, then within a given Mirror condition, the 53

PAGE 61

effect of an audience should be to increase smoking behavior. Referring to Table 4-1, it can be seen that the effect of an audience probably was to decrease smoking in both the Mirror and No Mirror conditions. The Encoding Perspective is thus not supported by the experimental results. Self-Consciousness Perspective. The modified Self Consciousness Perspective of Framing and Walker (1981) suggested that the effect of a mirror would be to increase smoking behavior, whereas the effect of an audience would be to decrease smoking. It was further predicted that the combination of stimuli (Mirror/Audience condition) would result in a cancelling out of the effects. The prediction of the Self-Consciousness model is thus a main effects model. The analyses of variance presented in Chapter IV offer some limited support for this perspective. The pattern of the univariate analyses suggests that the main effects model is a more likely model than an interactive one. With one exception, however, these differences are nonsignificant The results of the multiplet-test analyses presented in Table 4-10, however, suggest an interactive model. model appears to show an effect of mirror presence only in the Audience condition. This would appear to This 54

PAGE 62

-----------------contradict the predictions of the Self-Consciousness Perspective as well as the results of Liebling et al. (1974). In defense of the Self-Consciousness position, it appears that the absence of a significant difference in the No Audience condition was due to a ceiling effect. Subjects in the No Audience/Mirror condition smoked an average of 1.60 cigarettes in a twenty-five minute period. Clearly there are limits to how much a person will smoke, especially in a small room without windows. If this is indeed the case, then the discrepancy between the present experiment and that of Liebling et al. (1974) can be explained. Subjects in that experiment smoked substantially less than those in the present experiment. Hence there was room for an increase due to the mirror. Conclusion The model of self-awareness proposed herein suggested that the influence of self-focused attention on attitude behavior relationships could be summarized by a modification of an equation developed by Fishbein (1967). Specifically, it was proposed that a self-awareness term be added to the model, producing the following equation. B ~BI= { [Aact]wo + [NB(Mc)Jwl J SA The equation suggested that the effect of self-awareness multiplicative with the situational cues, and thus that it 55

PAGE 63

served to enhance the salience of the standards appropriate to a given situation. The data seem to suggest that the self-consciousness model more accurately accounts for the mirror and audience effects. If this model is indeed the most appropriate, then the above equation is incorrect. The SelfConsciousness Perspective would suggest that the SA term in the model is unnecessary, and that self-awareness effects are shown through changes in the weights associated with Aact and NB. Specifically, the model would suggest that the weight assigned to Aact is to some degree a function of private self-awareness and the weight assigned to NB is a function of public self-awareness. It would thus appear that models of self-awareness will in the future have to take account of public and private self-awareness along with public and private attitudes. 56

PAGE 64

REFERENCES Buss, A.H. Self-Consciousness and Social Anxiety. San Francisco: Freeman, 1980. Carver, C. s. Physical aggression as a function of objective self-awareness and attitudes toward punishment. Journal of Experimental Social Psychology. 1975, .!..!_, 510-519. Carver, c. s. A cybernetic model of self-attention processes. Journal of Personality and Social Psychology. 1979, ~' 1251-1281. Carver, c. s., & Scheier, M, F. Self-focusing effects of dispositional self-consciousness, mirror presence, and audience presence. Journal of Personality and Social Psychology, 1978, 36, 324-332. Diener, E., & Srull, T, K. Self awareness, psychological perspective and self-reinforcement in relation to personal and social standards. Journal of Personality -and Social Psychology, 1979, l2,, 413-423. Diener, E., & Wallbom, M. Effects of self-awareness on antinormative behavior. Journal of Research in Personality. 1976, 10, 107-111. Duval, S. & Wicklund, R. A. Awareness. New York: A Theory of Objective Self Academic Press, 1972. Exner, J.E. The self-focus sentence completion: A study of egocentricity. Journal of Personality Assessment, 1973, 37, 437-455. Fenigstein, A., Scheier, M. F,, & Buss, A, H. Public and Private Self-Consciousness: Assessment and Theory. Journal of Consulting and Clinical Psychology, 1975, 43, 522-527. Fishbein, M, Attitude and the prediction of behavior. In M. Fishein (Ed.) Readings In Attitude Theory and Measurement. New York: Wiley, 1967. Froming, w. J. The relationship of moral judgement, self awareness, and sex to compliance behavior. Journal of Research in Personality, 1978, g, 396-409. 57

PAGE 65

Froming, W. J., & Walker, G. R. Self-awareness and public vs. private standards for behavior. Under editorial review, 1981. Gibbons, F. x. Sexual standards and reactions to pornography: Enhancing behavioral consistency through self-focused attention. Journal of Personality and Social Psychology, 1978, 36, 976-987. Gibbons, F. x., Carver, C. s., Scheier, M. F., & Hormuth, S. E. Self-focused attention and the placebo effect: Fooling some of the people some of the time. Journal of Experimental social Psychology, 1979, 15, 263-274. Hull, J. G., & Levy, A. s. The organizational functions of the self: An alternative to the Duval and Wicklund model of self-awareness. Journal of Personality and Social Psychology, 1979, 37, 756-768. Ickes, w. J., Wicklund, R. A., & Ferris, C. B. Objective self-awareness and self-esteem. Journal of Experimental Social Psychology, 1973, 9, 202-219. Liebling, B. A., Seiler, M., & Shaver, P. Self-awareness and cigarette smoking behavior. Journal of Experimental Social Psychology, 1974, g, 325-332. Pryor, J. B., Gibbons, F. X., Wicklund, R. A., Fazio, R. H., & Hood, R. Self-focused attention and self report validity. Journal of Personality, 1977, 45, 513-527. Scheier, M. F., Self-awareness, self-consciousness, and angry aggression. Journal of Personality, 1976, 44, 627-644. Scheier, M. F., & Carver, c. s. Self-focused attention and the experience of emotion: Attraction, repulsion, elation, and depression. Journal of Personality and Social Psychology, 1977, l2_, 625-636. Scheier, M. F., Carver, C. s., & Gibbons, F. X. Self directed attention, awareness of bodily states, and suggestibility. Journal of Personality and Social Psychology, 1979, 37, 1576-1588. Scheier, M. F., Fenigstein, A., & Buss, A. H. Self awareness and physical aggression. Journal of Experimental Social Psychology, 1974, 10, 264-273. 58

PAGE 66

Steenbarger, B, N,, & Aderman, D, Objective self awareness as a nonaversive state: Effects of anticipating discrepancy reduction. Journal of Personality, 1979, 47, 330-339. Watson, D., & Friend, R. Measurement of Social-Evaluative Anxiety. Journal of Consulting and Clinical Psycholog~, 1969, 33, 448-457. Wicklund, R, A. Objective self-awareness. In L. Berkowitz (Ed,), Advances in ExperimentalSocial Psychology (Vol. 8). New York: Academic Press, 1975. Wicklund, R. A. Three years later. In L, Berkowitz (Ed.), Cognitive Theories in Social Psychology. New York: Academic Press, 1978. Wicklund, R. A., & Duval, s. Opinion change and performance facilitation as a result of objective self-awareness. Journal of Experimental Social Psychology, 1971, 2, 319-342. Witkin, H. A. The Embedded Figures Test. Palo Alto: Consulting Psychologists Press, 1969. Wuebben, P. L., Straits, B. c., & Schulman, G. I. The Experiment as a Social Occasion. Berkeley: Th~ Glendessary Press, 1974. 59

PAGE 67

APPENDIX A QUESTIONS USED IN SELECTING SUBJECTS 1. How much do you smoke? 1) Not at all 2) Less than 5 cigarettes a day 3) About one-half pack a day 4) Around a pack a day 5) More than a pack a day For each of the remaining items, indicate how much you agree or disagree with each statement by using the following scale: l ) Strongly disagree 2) Disagree 3) Neutral 4) Agree 5) Strongly agree 2. I find smoking to be an enjoyable habit. 3. Smoking is a disgusting habit that no one should have. 4. Most people (other than myself) find smoking to be annoying and offensive. 5. Most other people have a favorable attitude toward the habit of smoking. 60

PAGE 68

APPENDIX B SPECIFIC EXPERIMENTAL PROCEDURES AND INSTRUCTIONS 1. Instructions read to subject: By showing up you have earned one experimental credit. Before we go any farther, I need to tell you something about the experimental procedure so that you can decide whether or not you want to continue. This study involves smoking, and is specifically interested in differences in memory functions and heart rate between smokers and non-smokers. I'll tell you more about the specific predictions of the experiment later, after the experiment is over, since we want the tests to be unbiased. Today we will have you rest for about twenty-five minutes, and then we will measure your heart rate and let you try the memory tests. When you come back for the second part of the experiment, you will again rest for about twenty-five minutes, and we will do a different set of tests. If you complete both parts of the experiment you will receive two hours credit. You must, however, complete both parts of the experiment to receive any credit. During the waiting period, I would like for you to light up a cigarette and take at least one or two puffs. If possible, it would be better if you did this at the beginning of the waiting period, since it takes a little while for the effects of nicotine to be felt. Some people wonder if it is all right to smoke more than this. The answer is yes, you can smoke as much as you like, but you have to take at least one or two puffs. Smoking any more than this will not affect our experiment, but not smoking at all could change the results. If you do not have any objections to the experiment, I will take you to the waiting room and we can begin. 61

PAGE 69

I I 62 2 Proceed to experiment room and continue: You will be waiting here for about twenty-five minutes while your heart rate stabilizes. Remember to light up a cigarette and take at least one or two puffs during this time. I will come by when the waiting period is over and we can start the experiment. 3. If the confederate is present: We are going to be conducting an experiment in this room in about twenty-five minutes. You can stay here until then, but at that time you will have to leave. I only ask that you two (addressing subject and confederate) do not talk to each other. 4. At end of waiting period, present deception tasks. SECOND SESSION 1. Instructions: Experimenter reminds subject that the instructions for the second session are identical to those in the first. 2. Debriefing procedures (Appendix C) follow waiting period.

PAGE 70

APPENDIX C DEBRIEFING PROCEDURES 1. Did you believe that the tests at the end of the first session were affected smoking? 2. Did you think there were any other purposes to the experiment, i.e. anything we were not telling you? 3. If yes, what did you think the experiment was really about? At this point, explain the experiment 4. Did you suspect that you were being watched? 5. If yes, did this affect your smoking? 6. Did the presence of the mirror affect you or bother you at all? If so, how? 7. Did/Would the presence of another person in the room affect you or bother you at all? If so, how? 8. Other comments: 63

PAGE 71

BIOGRAPHICAL SKETCH Gordon Rexal Walker, Jr., was born in Burlington, North Carolina, on September 11, 1954, to Mary Speight Walker and Gordon Rexal Walker. He completed his secondary education at Walter Williams High School in Burlington, North Carolina, and entered Wake Forest University in September, 1972. After graduation from i lake Forest in 1976, he entered the University of Richmond, where he received the degree of Master of Arts in 1978. He entered the Graduate School at the University of Florida in September, 1978. During that time, he studied in the Personality Area of the Department of Psychology, under the direction of Dr. William J. Froming. On July 1, 1979, he was married to Jane Linda Curry of Hingham, Massachusetts. 64

PAGE 72

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. I )? lt d i~~William J. froming, Chairma Assistant P~fessor of Ps ology 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. 2: ~ n Prof so 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. ftarry A. rater Prof=::f 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. Thomas K. Assistant Srull Professor of Psychology

PAGE 73

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. Jo7i0ch ( Assistant Professor of Marketing 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. J es Algina A ociate Professo of Foundations of Education This dissertation was submitted to the Graduate Faculty of the Department of Psychology in the College of Liberal Arts and Sciences and to the Graduate Council, and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. March, 1981 Dean for Graduate Studies and Research

PAGE 74

U N I VE R S I TY O F FL O RIDA II I II IIIIII Ill Ill lllll lllll II IIIIII IIII II IIIIII IIII Ill llll llll I I 3 1262 08553 5192