Group Title: comparison of several different imagery instruction procedures for the treatment of test anxiety
Title: A comparison of several different imagery instruction procedures for the treatment of test anxiety
CITATION THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00102827/00001
 Material Information
Title: A comparison of several different imagery instruction procedures for the treatment of test anxiety
Physical Description: Book
Language: English
Creator: Harris, Gina M ( Gina Melanie ), 1953-
Copyright Date: 1980
 Record Information
Bibliographic ID: UF00102827
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 07240468
ltuf - AAL5426

Full Text














A COMPARISON OF SEVERAL DIFFERENT IMAGERY INSTRUCTION
PROCEDURES FOR THE TREATMENT OF TEST ANXIETY













BY


GINA M. HARRIS


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


1980


L























This dissertation is dedicated to my parents,

Caroline and Harold Harris, the sustaining force of

my life.



Gina M. Harris

October 10, 1980















ACKNOWLEDGMENTS


I would like to thank my dissertation advisor and

dissertation committee chairperson, Dr. Suzanne Bennett

Johnson, for the guidance, continued support and incisive

criticism she offered during all phases of this disserta-

tion project. I greatly appreciate the assistance offered

by members of the dissertation committee: Dr. Hugh C.

Davis, Dr. Everette Hall, Dr. Janet Larsen and Dr. Benjamin

Barger for their thoughtful reading of my manuscript. I

am greatful to Mr. Michael DeGennaro for assistance with

the data analysis, and appreciation is also expressed to

Dr. Randy Carter for statistical advice.

I wish to thank Leslie Dubin, Elizabeth Haughney,

Andrew Kerr, Mark Miller, Linda Nissenoff, Dan Rosenthal

and Kevin Stempel who served as raters in this study.


iii

















TABLE OF CONTENTS

PAGE

ACKNOWLEDGMENTS. . . . . . . . . . iii

ABSTRACT . . . . . . . . . . . vi

INTRODUCTION . . . . . . . . . . 1

The Current Use of Imagery in Behavior Therapy. 5
Should the Image Be Studied?. . . . ... 15
How Should the Image Be Studied?. . . . 18

METHOD . . . . . . . . . . . 30

Design. . . . . . . . . .. . 30
Subjects. . . . . . . . . . 31
Measures. . . . . . .. . . . 31
Treatment . . . .... . .. . . 38
General Procedures. . . . . .. . . 39
Treatment Manipulations . . .. . . 41

RESULTS . . . . . . . ... . . 45

Analysis of Treatment Effects . . . .. 45
Expectations for Improved Academic Performance. 54
Treatment Efficiency: Speed of Hierarchy
Completion . . . . . . . .55
Imagery Analyses. . . . . . . .. 55

DISCUSSION . . . . . . . . .. . 66

Treatment Outcome . . . . . .. .66
Imagery . . . . . . . . . . 72

APPENDICES

A PROPOSITIONAL UNITS OF THE EMOTIONAL IMAGE. . 78

B INSTRUCTIONS TO THE EXPERIMENTER:
INDIVIDUALIZED COPING IMAGERY TREATMENT. . 80

C INSTRUCTIONS TO THE EXPERIMENTER:
INDIVIDUALIZED COPING IMAGERY TREATMENT
COMBINED WITH RELAXATION . . . . .. 91









D INSTRUCTIONS TO THE EXPERIMENTER:
ACADEMIC COPING IMAGERY TREATMENT . . . 95

E INSTRUCTIONS TO THE EXPERIMENTER:
ACADEMIC COPING IMAGERY TREATMENT
COMBINED WITH RELAXATION. .. . . . 100

F GENERAL INSTRUCTIONS FOR THE ADMINISTRATION
OF INDIVIDUALIZED COPING IMAGERY TREATMENT. 101

G GENERAL INSTRUCTIONS FOR THE ADMINISTRATION
OF ACADEMIC INDIVIDUALIZED COPING
IMAGERY TREATMENT. . . . . . .. 103

H STUDY SKILLS THERAPIST MANUAL. . . . . 105

REFERENCES. . . . . . . . . .. . 145

BIOGRAPHICAL SKETCH . . . . . . . .. 156










































v


1















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

A COMPARISON OF SEVERAL DIFFERENT IMAGERY INSTRUCTION
PROCEDURES FOR THE TREATMENT OF TEST ANXIETY

By

Gina M. Harris

December 1980

Chairman: Suzanne B. Johnson
Major Department: Clinical Psychology

The primary purpose of the present study was to assess

the effectiveness of different imagery instruction procedures

for the treatment of test anxiety. In this study, a com-

parison of the efficacy of instructing subjects in indi-

vidualized coping imagery treatment based on nonacademic

experiences of competence and success and individualized

coping imagery treatment based on academic experiences of

competence and success was carried out. Another purpose of

the study was to assess how effective instruction in relaxa-

tion is in increasing overall treatment effectiveness of

individualized coping imagery treatment. This study also

sought to assess whether the elaborateness and content of

imagery were related to overall treatment effectiveness.

It was evident from the results of this study that all

variations of individualized coping imagery treatment









brought about significant decreases in test anxiety. In

terms of academic performance, two groups, individualized

coping imagery without relaxation and academic individual-

ized coping imagery with relaxation, were significantly

different from the control.group at posttest. Individual-

ized coping imagery treatment alone and individualized cop-

ing imagery treatment combined with relaxation brought

about significant changes from pretreatment to posttreatment

on grade point average. Academic individualized coping

imagery alone did not change significantly on this measure

of academic performance. The waiting list control group

decreased on grade point average.

In addition, the findings of this study suggest that

changing the content of emotional imagery may be an impor-

tant factor in bringing about a successful treatment out-

come. It appears that decreases in negative coping imagery

and increases in positive coping imagery figured strongly

in treatment success.


vii


1















INTRODUCTION


Psychological historians could construct a story of the

development of psychology and its critical methodological

problems based on attempts to investigate the image. Scru-

tiny of the image was one of the first areas to be explored

by psychologists and, some have argued, one of the greatest

impediments to its development as a science (Holt, 1964).

Man's fantasy and the images of his mind dominated the

philosophy and poetry of the 20th century. The stream of

consciousness, the internal monologue, was the major charac-

teristic of the important literature of the day. James

Joyce in his book Ulysses (1922) made the stream of con-

sciousness the most discussed and influential literary

technique of our era. Amy Lowell and her followers (Sanders,

Nelson, & Rosenthal, 1972) known as the imagists,composed in

vers libre and expressed ideas and emotions exclusively

through a succession of precise and clear images. The de-

velopment of Freudian psychology was also dominated by the

stream of consciousness (Robinson, 1976).

Psychology from the perspective of its early leaders,

Wundt, Kulpe, Titchener, etc., was concerned with more than

external stimuli; it embraced images, dreams, feelings,

memories--the stuff that consciousness is made of (Marx &

Hillix, 1973). Thus the basic task of the psychologist

1









was to understand the structure and content of the human

mind by means of systematic introspection. The technique

involved a carefully worked out procedure in which highly

trained subjects were to report the immediate content of ex-

perience rather than to describe events that led to produce

the experience. Writing in 1898 Titchener stated:


The primary aim of the experimental psycholo-
gist has been to analyze the structure of the
mind, to ravel out the elemental processes from
the tangle of consciousness, or (if we may
change the metaphor) to isolate the constitu-
ents of a given conscious formation . .
The aim of the psychologist is three fold.
He seeks (1) to analyze concrete (actual) men-
tal experience into its simplest components,
(2) to discover how these elements combine,
what are the laws which govern their combina-
tion, and (3) to bring them into connection
with their physiological (bodily) conditions.
(Marx & Hillix, 1973, p. 15)


Although Wundt, Titchener and Kulpe were greatly in-

trigued by the study of the stream of thought and man's

mental imagery, the introspectionists made little progress

in advancing the knowledge of psychology. "It died of

narrow dogmatism, a disease which no school of psychology

can long survive. Structuralism lacked the support of

practical application and connections to other areas of

psychology; its demise was mourned by few" (Marx & Hillix,

1973, p. 135).

The early objective behaviorists under the leadership

of John Watson not only stymied the efforts of Titchener

and his associates but offered an alternative. In contrast

to Wundt and Titchener,Watson focused exclusively on


1








objective, observable and quantifiable behavior. He saw

mental processes, images and thoughts as not scientifically

legitimate precisely because they were not objective, ob-

servable and quantifiable (Craighead, Kazdin, & Mahoney,

1976). Writing in 1924, Watson stated that consciousness

was "neither a definite nor usable concept" (p. 2). Captur-

ing the attitude of the objective behaviorists towards the

study of the elements of consciousness, Boring in 1937 wrote:


Having understood, tough-minded rigorous
thinkers will, I think, want to drop the term
consciousness altogether. A scientific psy-
chology is scarcely yet ready to give im-
portance to so ill defined a physiological
event as an awareness of an awareness. This
concept might never have come to the fore had
not people tried to interpret others in terms
of their own "private" minds--that egocentric
Copernican distortion which properly leads to
desolate solipsism. (Marx & Hillix, 1973,
p. 135)


Thus, behavioral psychologists such as Watson felt that

"mental images are ghosts with no functional significance"

(Watson, 1913, p. 158). Experimentalists of the behavior-

ist school saw only the sensation-related aspects of imagery

as suitable subjects for experimental study because they

were closest to objective stimuli.

While American psychologists were abandoning their

efforts to study covert phenomena, it was Freud and the

European psychologists who continued to explore the meaning

of our private world, our dreams, daydreams, aspirations,

images of violence and love. Psychoanalysis was essentially

Freud's attempt to scientifically and systematically study









the inner world (Holt, 1964). However, Freud's attempt to

formalize the psychoanalytic method as a means to scienti-

fically study the private world did not meet with great

success. Its dependence on the role of the therapist, his

judgment and experience, inevitably led to the growth of

many splinter groups who believed that what they experienced

in their patients was both different and more accurate than

what was experienced by their Freudian counterparts. The

development of multiple schools of therapy--the Jungians,

Adlerians, Horneyans, Humanists, etc., pointed up the basic

problems in attempting to explain components of man's inner

life using traditional clinical methods. In addition, Freud

did not create any viable way to test his theory, leaving it

open to a wide range of interpretation (Marx & Goodson,

1976).

After the study of imagery by Titchener's students

waned in the first decade of the twentieth century, few

scientific efforts were made in this area for close to 50

years. As late as 1965, prominent researchers in the area

of verbal learning were underplaying or avoiding considering

the function of images in mediating associations between

words (Paivio, 1969).

However, a review of the clinical literature of the

past 20 years reveals that imagery is very much present and

alive. This resurgence of interest in imagery, fantasy and

hallucination appears due to the efforts of, interestingly

enough, not of the Freudian psychologists, but of the





5


behavioral psychologists who previously had cast out imagery

as a topic for serious study. Thus, while behaviorally

oriented psychologists have been uncovering important

principles about reinforcement contingencies in the en-

vironment, they have also begun to explore the capacity

for imagery production in man and the uses to which this

image-making capacity can be put to alleviate emotional

suffering.


The Current Use of Imagery in Behavior Therapy


This resurgence in the use of imagery by behaviorists

is best exemplified by reviewing some of the behavioral

techniques that appear to be dependent on imagery.

Although much of behavioral treatment utilizes the

production of overt responses such as the operant condition-

ing methods derived from Skinner's investigations or

Bandura's modeling techniques, an ever increasing number

of behavioral procedures are based on the clients' use of

imagery. Indeed, the most frequently employed of all be-

havioral modification methods is the systematic desensi-

tization procedure of Wolpe (1958) in which it is claimed

that the production and use of imagery is a critical com-

ponent (Wilkins, 1971). As part of this treatment, the

patient is asked to produce a hierarchy of imaginal fear

provoking events (Wolpe, 1958). These situations are

intended to approximate real-life stimuli or performances

which currently evoke a high level of anxiety (e.g., making









a speech). A list or hierarchy of these situations is then

built in which the imaginal scenes are ordered according to

their capacity to produce anxiety. The patient then under-

goes an extensive period of training in relaxation which

promotes an increase in the potential to produce imagery

under conditions of reduced arousal (Singer, 1974).

The efficacy of systematic desensitization as a behav-

ioral procedure has received much research scrutiny. A

multitude of reports, monographs, and controlled investiga-

tions corroborate its effectiveness (Bandura, 1969; Paul,

1969; Wilson & Davison, 1971; Davison & Wilson, 1973).

However, large questions still remain concerning why it is

effective and what are the major ingredients contributing

to its success. It is generally agreed that Wolpe's counter-

conditioning explanation for desensitization's efficacy is

insufficient. Investigations have revealed relaxation to

be a useful but not critical component for treatment suc-

cess (Wilkins, 1971). In fact, a number of behavioral

psychologists (Mahoney, 1974; Paul, 1966) have claimed that

the production of imagery appears to be the crucial component

of desensitization. Mahoney (1974), for example, states

that although in vivo hierarchies have been utilized, the

predominant stimulus modality is covert.

Behavioral psychologists such as Gordon Paul (1966)

have underscored the necessity for the client to visualize

himself in the anxiety-evoking situation, not as if he were

seeing himself in a movie, but as if he were fully









participating in all the action. In his desensitization

instructions, Paul emphasized the importance of the client

producing stimuli-laden images, drawn from all the sense

modalities in an attempt to make the imaginal scene as close

to the actually experienced anxiety evoking situation as

possible. Yet Paul did not attempt to evaluate the neces-

sity for multisensory images empirically.

For all its presumed importance in the systematic de-

sensitization technique, behaviorally oriented psychologists

have devoted little empirical attention to exploring the

importance of imagery in desensitization. Like the Freudian

psychologists before them, behavioral psychologists have

used imagery as a therapeutic tool but have ignored many of

the critical questions about the nature of the imagery pro-

duced. Indeed, while other components of desensitization,

such as progressive relaxation, have undergone exhaustive

scrutiny and refinement, the scientific energy devoted to

examining the importance of imagery has commenced only

recently (Lang, 1977).

This same phenomenon has occurred with other thera-

peutic approaches presumably derived from classical condi-

tioning and reinforcement models. While systematic de-

sensitization is the behavioral technique that has been

applied most frequently and with substantial success for a

variety of fears and phobias, a number of therapeutic

approaches utilizing imagery have emerged. Emotive imagery,

covert sensitization, covert reinforcement, implosive









therapy, covert modeling have been in wide use. It is

generally assumed that imagery is an important component of

these techniques; however, little scientific energy has

been expended to explore whether imagery is crucial.

Emotive imagery, growing out of the counterconditioning

idea, was employed by A.A. Lazarus and his associates

(Lazarus & Abramovitz, 1962; Lazarus, Davison & Polefka,

1965) for the treatment of children's anxieties because of

the difficulties they experienced in teaching children pro-

gressive relaxation. In this technique, the children were

instructed to imagine "stronger and stronger phobic stimuli

woven into progressively more enjoyable fantasies" (Lazarus,

1971, p. 211). The assumption of this procedure was that

the positive feelings evoked by the fantasy would compete

successfully with the anxiety-arousing stimuli until these

stimuli would become conditioned to feelings incompatible

with fear. An example is provided by Lazarus (1971) who

treated an eight-year-old boy experiencing fears when

visiting the dentist:


The sequence consisted of having the child
picture himself accompanying Batman and
Robin on various adventures and then imag-
ining them receiving dental attention.
He was asked to picture this scene at least
five times daily for a week. Next, he was
to imagine himself in the dentist's chair
while Batman and Robin stood by and observed.
He also practiced this image several times
a day for one week. He visited the dentist
the following week, and according to his
mother, he sat through four fillings without
flinching. (p. 211)








However, in emotive imagery treatment, as in desensitiza-

tion, the importance of imagery in bringing about behavior

change has not been subjected to empirical scrutiny.

Covert sensitization, an imaginally based variant of

aversive counterconditioning, is another of those behavioral

interventions which assumes imagery to be a crucial compo-

nent. Covert sensitization has been used to modify behavior

such as sexual deviance, smoking, alcoholism and overeating

(Jones, 1969; Rachman & Teasdale, 1969). In covert sensi-

tization, images which evoke extreme disgust, unpleasantness

are employed (Cautela, 1967; Davison, 1968; Lazarus, 1958).

After having undergone progressive relaxation, the client

is instructed to imagine himself about to engage in the

reinforcing deviant behavior, e.g., smoking, drinking,

pedophilia, etc., and then to imagine that he feels both

nausea and disgust in response to previously reinforced

activities.

Cautela (1970) has employed a covert reinforcement

technique for the alleviating of fears and phobias. Al-

though purportedly having its origins in the reinforcement

model, it bears a marked similarity to the emotive imagery

techniques of Lazarus. As part of his procedure, a number

of reinforcing images such as activities the client fre-

quently enjoys doing are first elicited by the therapist.

In the next step in this treatment, the client is instructed

to imagine "a difficult but desired response sequence"

such as asking a girl for a date and then to imagine himself


1






10


in one of the reinforcing images. Covert reinforcement has

been used to modify evaluative anxieties (Cautela, 1970;

Wisocki, 1973; Kosta & Galassi, 1974; Chang-Liang & Denny,

1976), snake phobia (Marshall, 1975) and to improve the self-

concept of emotionally disturbed children and psychiatric

patients (Krop, Calhoun, &Verrier, 1971; Krop, Perez &

Beaudoin, 1973). Cautela has proposed that covert re-

inforcement enables the client to more easily imagine the

target behavior; ultimately this result is transferred to

dealing with the target situation in real life. In covert

reinforcement techniques, as in desensitization, little

attention has been devoted to determine whether imagery is

indeed a crucial component of the procedure.

Another of the behavioral techniques which relies

heavily on the use of imagery is implosive therapy. In this

technique, presumably based on the principles of extinction,

the client receives repeated exposure to intense anxiety

evoking stimuli, imagining the worst possible outcome of a

particular fear or obsession. It is assumed that anxiety

will first greatly increase and precipitously decrease as

the person is exposed to increasingly elaborate versions of

fear provoking stimuli. However, the evidence for the ef-

fectiveness of implosive therapy is only equivocal

(Morganstern, 1973), and there is little knowledge of the

possible mechanisms of behavior change. Various investiga-

tions have suggested "extinction, adaptation level, fatigue,

modeling, habituation, cognitive rehearsal and








discrimination," as possible mechanisms (Smith, Dickson, &

Sheppard, 1973, p. 358), but the imagery component has

never been empirically studied.

As a rule, investigations have not sought to discover

if the imagery component of these treatments reliably adds

something in addition to the other techniques employed as

part of the treatment. Further, these imagery based be-

havioral treatments such as covert reinforcement, emotive

imagery, covert sensitization, implosive therapy, etc.,

have usually been employed in combination with other treat-

ment procedures, e.g., study skills (Wisocki, 1970, 1973;

Cautela, 1970; Kendrick & McCullough, 1972; Lazarus, 1971;

Prochaska, 1971). Moreover, most studies exploring the

efficacy of behavioral imagery treatments have frequently

employed self-report inventories as the sole indices of

improvement, often when more reliable behavioral measures

were available. Thus, for the most part, the empirical

evaluation of these techniques has been limited. A tech-

nique that has received greater empirical evaluation is

covert modeling.

Covert modeling is a behavioral imagery based procedure

in which the client is "directed in the symbolic rehearsal

of appropriate behavior" for a difficult performance task or

a feared situation (Mahoney, 1974, p. 112). Here too, im-

agery is assumed to play a critical role in treatment. The

results of studies suggest that covert modeling is effective

in alleviating avoidance behavior (Kazdin, 1973a; Cautela,


L









Walsh, & Wish, 1971; Flannery, 1972b). In some of these

studies there was a definite effort on the part of the ex-

perimenter to instruct the subjects in the sort of imagery

that should be used (Kazdin, 1974; Cautela, Flannery, &

Hanley, 1974). In fact, studies of covert modeling are

among the few that have attempted to manipulate the content

of the imagery produced (Cautela, Flannery, & Hanley, 1974).

However, most studies have been of a laboratory analogue

nature. Therefore, we do not know how effective they are

with actual patient populations. Moreover, the active

therapeutic components are still unclear.

Covert phenomena and imagery based treatment has been

a subject of so much interest that an entirely new arena of

inquiry and controversy, cognitive behavior therapy, has

been opened (Meichenbaum, 1977). Nevertheless, while im-

agery based therapies have been used widely, carefully con-

trolled assessments and refinements of these therapies "have

lagged embarrassingly behind clinical applications"

(Mahoney, 1974, p. 117). One reason for the paucity of

controlled assessments of the importance of imagery in

these therapies may be the presence of substantial methodo-

logical difficulties and procedural inadequacies in studying

covert phenomena. The first and most obvious methodological

difficulty in studying imagery in the therapeutic process

is that images are private events, available only to human

introspection. Lang in 1977, sounding very much like Watson

in 1924, states, "As their observation cannot be shared nor









their dimensions measured by any instrument, they cannot be

data in a scientific analysis" (Lang, 1977, p. 862). The

image cannot be observed, tasted or touched.

Because it is so difficult to study the image, many

areas deserving of research scrutiny have consequently been

neglected. Singer (1974) points out that in scientific in-

vestigation of therapies incorporating imagery, investiga-

tors have not been attuned to "the fluid character of fan-

tasy, the rapidity of shifts away from the assigned image

or the various covert antagonisms that may occur in the

course of presumably attempting to produce imagery on de-

mand" (p. 235). Weitzman (1967) observes that within de-

sensitization treatment, the image is not a static entity

but is so fluid that it is often quite different from the

image called for by the specific item in the hierarchy.

Different images may impinge on the desired images. Persons

and events, frightening or not, unconnected with the present

scene to be imagined, may impose themselves on the scene

the patient is trying to imagine. Further, therapists and

investigators have tended to ignore crucial differences

between individuals' imaginative ability which could have

a significant effect on determining which patients could

benefit from imagery based treatments. Such knowledge of

differences could also signal the therapist that some

patients require added instruction and skill in producing

imagery. The possible use of pretreatment practice in

imagery, with the exception of the work of a few recent









investigations (Lang, 1977), has been too little explored.

Indeed, there has been little attempt to employ even paper-

and-pencil imagery assessment techniques to aid in the

assessment of patients.

While depending greatly upon the client's producing

images in the imagery based treatments, experimenters have

paid little attention to the characteristics the imagery

produces or whether it is being produced at all. In all

these treatments, whether systematic desensitization, im-

plosive therapy, covert sensitization, etc., the therapist

assumes that the client is actually imagining the desired

image and not just saying he is. During such treatment,

the patient can choose whether or not to cooperate with the

procedures employed. Wolpe and Lazarus (1966) cite cases

in which patients fail to cooperate with desensitization

treatments by introducing negative thoughts or unpleasant

images into the imaginal sequence. Therefore, an underlying

methodological problem appears to be the experimenter's lack

of control over the client's production of imagery.

Because of difficulties in studying the phenomena,

there has been a lack of research attention paid to the

importance of the vividness and affective intensity of the

imagery produced. These data,which could be obtained

through physiological and self-report measures, could

impart extremely vital information bearing on imagery

processing during therapy.









Should the Image Be Studied?


Given these substantial methodological difficulties,

does it make sense to study the elusive image, to system-

atically explore covert phenomena? Researchers, theorists

and social critics as diverse as Mahoney (1974, 1977),

Lang (1977, 1979) and Mumford (1967) propose that it does

indeed make sense to study imagery as a very significant

component of human thought.


To dismiss the most central fact of man's
being because it is inner and subjective is
to make the hugest subjective falsification
possible--one that leaves out the really
critical half of man's nature. For without
that underlying subjective flux, as experi-
enced in floating imagery, dreams, bodily
impulses, formative ideas, projections and
symbols, the world that is open to human
experience can be neither described nor
rationally understood. When our age learns
that lesson, it will have made the first
move toward redeeming for human use the
mechanized and electrified wasteland that
is now being bulldozed at man's expense and
to his permanent loss, for the benefit of
megamachine. (Lewis Mumford, 1967, pp. 75-76)


Despite the current interest in exploring the image as

a subject for scientific study and as a therapeutic tool,

there are a number of critics, e.g., Rachlin (1977a), who

question whether scientific energies are well spent explor-

ing covert phenomena.

This controversy concerning whether it makes sense to

study mental imagery was brought into sharp focus by a de-

bate between Rachlin and Mahoney in 1977. While Rachlin









argued that "it is not necessary to refer to thoughts or cog-

nitions at all" (Rachlin, 1977c, p. 661), seeing observable

behavior as sufficient, Mahoney took the position that the

"dichotomy between observed and inferred events is an arti-

ficial one. It implies that there are some stimuli which

register directly with our sense and these somehow are more

real, legitimate or scientific" (1976, p. 674). Mahoney sug-

gests that all stimuli, overt or covert, are mediated. He

proposes that no stimulus impinges on the sensorium without

undergoing the transforming and constructive process that is

human perception. He further suggests that the "Doctrine of

Immaculate Perception," which implies that some stimuli are

apprehended without undergoing the mediating influences of

constructive perceptual processes is "wholly untenable and

it has long since been abandoned by its staunchest philo-

sophical defenders" (p. 674). "No one has ever directly

seen a pure, unedited stimulus; human perception is a con-

structive process" (Mahoney, 1976, p. 674).

In contrast, Rachlin (1977b) argues that while per-

ception may be constructive, while thoughts and images may

occur, "the tactic of inferring events inside the organism

has not led to coherent prediction or control of behavior

or in useful therapeutic techniques" (1977a, p. 680). The

data provided by observable phenomena have the virtues of

"parsimony, simplicity and theoretical consistency," (1977b,

p. 373) will suffice. Writing in response to Mahoney,

Rachlin stated:









I would abandon my non-mediational position
when it proved inadequate to predict and
control behavior and when a mediational
position was shown to be:
(A) at least as coherent
(B) at least as applicable in the laboratory
(C) at least as applicable in every day life
(D) at least as applicable in therapy.
(1977a, p. 681)

Mahoney continues to contend that the study of covert

phenomena, images, thoughts, etc., do improve our predic-

tion, understanding and control of behavior and hence

should continue to be studied.

There are a growing number of psychologists who share

this attitude. Most investigators do acknowledge that the

image is very difficult to study. They all seem to be in

agreement that it forms "that part of human behavior which

is so private and close to the self as to belie effective

scientific scrutiny in an age of operationism" (Singer,

1973, p. 385). However, spurred on by great progress in

the fields of neurophysiology, sensory deprivation (Hebb,

1949) and dream research (Aserinsky & Kleitman, 1953) and

by the "new look at man as an information processing or-

ganism brought on by computer theory" (Singer, 1973, p. 385),

investigators from both cognitive and clinical psychology

have recently begun to direct their efforts towards a re-

evaluation of imagery as a cognitive phenomena and as a

therapeutic tool.









How Should the Image Be Studied?


Assuming we are going to study the image, can we define

it? The problem of defining the image has been made particu-

larly difficult because "we know nothing about any form or

structure that the image might possess or reside in" Bugelski,

Kidd, & Segman, 1968). Investigators have used expressions

as diverse as "pictures in the mind," "impressions made

upon a plastic substance" (Tichener, 1912) or "the petri-

fied product of perceptual functions" (Kluver, 1932) in an

attempt to describe and define what is essentially a hypo-

thetical, implicit and nonobservable cognitive product.

Psychologists have offered a number of definitions of imag-

ery. Some investigators studying the image have defined it

from a sensory perceptual (Singer, 1974) or a constructional

perspective (Pylyshyn, 1973; Lang, 1977), while still other

investigators appear to be studying the image without offer-

ing any explicit definitions (e.g., Paivio, 1971).

Working from the sensory-perceptual perspective, psy-

chologists such as Singer (1973, 1974) propose that "the

image represents man's capacity to duplicate environmental

information in the absence of the persistence of external

signals" (1973, p. 385). In contrast to the percept, in

which the individual makes a "response to an object or stimu-

lus that continues to be within the physical scanning capaci-

ty of the sense organs for at least one second," one can

posit the existence of an image if the "original source of









stimulation is further removed in time from the observer but

is described as present or experienced in some form as part

of the subject's consciousness" (Singer, 1973, p. 385).

Cognitive psychologists working from this sensory-

perceptual perspective have focused considerable research

attention on eidetic imagery, an intriguing phenomenon found

in a small segment of the population. To be considered

eidetic, an imager after studying a scene for 30 or 40 sec-

oncs should be able to retain his perception of it for sev-

eral minutes afterward and "view" the eidetic image in as

much detail as if the picture were still there (Jaensch,

1930; Haber & Haber, 1964). Some eidetic images seem indis-

tinguishable from actual perceptions to the imagers even when

employing objective criteria. In an experiment conducted

by Leask, Haber,& Haber (1968) eidetic subjects were shown

two schematic drawings of ocean scenes. If these scenes

were juxtaposed in the imagination, the picture that would

result would be that of a bearded man. Leask, Haber, &

Haber (1968) reported that some of their eidetic subjects

reported the presence of a face after imaginally juxtaposing

these scenes. For these subjects, it appears that the image

almost functioned like a sensory impression.

Also adopting the sensory-perceptual perspective,

Sheehan (1966), Segal (1971), and Shepard (1978) have been

responsible for bringing about important methodological

refinements in the study of imagery, which in turn have

contributed to the effort to predict and control behavior.








Sheehan (1966) has developed a new procedure to obtain in-

formation on vividness and other characteristics of visual

imagery. In Sheehan's procedure, the individual, having

studied a picture of an object such as a red apple, projects

it mentally onto a screen. The subject then manipulates a

projector light for color, intensity and shape in order to

bring this objective measurable index as close as possible

to his actual image. Through this method, Sheehan was able

to determine how well memory images served in creating some-

thing similar to the original stimulus, how well memory

images capture the qualities--the shape, color, etc., of

the original stimulus.

Segal (1971) has been studying various aspects of the

Perky phenomena. In the initial experiment, Perky's sub-

jects were asked to imagine common objects while looking at

a fixation point in the middle of a window. In the back of

the window a machine projected forms that were very close

to the items to be imagined in all their characteristics

(e.g., shape, color). After the subject was told what ob-

ject to imagine, the projected form was gradually brightened

to about threshold brightness and was removed before the

subject began to describe what he had imagined. Perky found

that all subjects confused the projected forms with their

own images. All these observers were unable to discern that

an actual objective stimulus was present, believing that the

stimulus they described was totally a project of their

imaginative processes. Segal (1971) was able to successfully









replicate the Perky effect, which could be defined as an

inability to differentiate "a real from a mentally projected

image." The way that subjects decide whether a stimulus

is real or not was subjected to exhaustive empirical scru-

tiny; the expectations induced by the experimenter seemed

to be critical. If subjects were told to expect to see

some real stimuli, they were successful in discerning these.

Conversely, if the subjects did not expect to see any "real

stimuli," they did not see them (Segal, 1971).

In an attempt to dispense with what he perceives as

fruitless arguments about internal processes underlying

imagery, Shepard (1978) proposes that we can study such

images "as defined solely in relation to their correspond-

ing external objects" (1978, p. 125). In his studies,

Shepard found that "to the extent that mental images can

substitute for perceptual images, subjects are able to

answer questions about objects as well when those objects

are merely imagined as when they are directly perceived"

(Baylor, 1971; Kosslyn, 1975). Shepard has concluded that

subjects make the same judgments about objects in their

absence as in their presence: Subjects who imagine a

particular object are very quick and accurate in making a

response to related external stimuli. In addition, he found

that when called upon to spatially transform stimuli in the

imagination, subjects are accurate in carrying out these

transformations. These results are derived from many

experiments employing a wide range of objects, including









two dimensional shapes (Shepard & Chipman, 1970), spectral

colors (Shepard & Cooper, 1975), one digit numbers in such

forms as Arabic numerals, printed English names and patterns

of dots (Shepard, Kilpatrick &Cunningham, 1975), familiar

faces, and musical sounds (Shepard, 1975). In all cf these

experiments, there was no statistical difference between how

well subjects performed in the experimental condition in

which the object was directly perceived or when the object

was imagined.

Defining the image from a constructional perspective,

investigators, notably Neisser (1967) and Pylyshyn (1973),

propose that the experience of an image itself arises out

of constructive processes. These psychologists suggest that

"the units abstracted and interpreted during perception are

stored in long-term memory in an abstract propositional

format and must be acted on by processes that serve to

generate or to produce an experience of an image" (Kosslyn,

1975, p. 342). Thus, the aforementioned investigators sug-

gest that we view the image not as a picture in the head,

which the phenomenological description implies, but that we

define the image as a propositional structure, which is

more like an elaborated description of the information con-

tained in the image. Rather than being "a raw harvest of

sensory observation" (Lang, 1977, p. 864) that is implied by the

picture metaphor, the image gives us information about

objects or events, not pictures or representations of them.

Working within this perspective, P.J. Lang and others are









studying imagery and its component processes through the use

of physiological techniques. From the results of his stud-

ies, Lang (1977) has suggested that careful training and

experimenter-monitoring and reinforcement of appropriate

responses can enable subjects to increase their psycho-

physiological response to imagined situations and hence

the potential effectiveness of the use of imagery during

therapy. It appears that the way instructions to the client

are organized can be shown to control somatovisceral re-

sponses during imagery. It is Lang's contention that

psychophysiology provides "a window through which the image

can be observed and a means through which it can be al-

tered" (1977, p. 882).

Lang is not alone among experimental psychologists in

having investigated the power of imagery to produce physio-

logical changes. Simpson & Pavio (1966) observed changes

in pupillary size during imagery. May & Johnson (1973)

perceived an increase in heart rate when arousing images

were presented. Studies such as those of Barber & Hahn

(1964) and Grossberg & Wilson (1968) suggest that imagining

a painful or frightening event results in a strong physio-

logical response that can be gauged through physiological

measures such as GSR. Increases in electromyograms (EMGs)

have also been noted by several researchers (Craig, 1969;

Jacobsen, 1929, 1930; McGuigan & Schoonover, 1973).

While psychologists such as Paivio (1969), Bower (1971),

Bugelski (1968), studying imagery and its relationship









to learning and memory, have made certain assumptions about

the image, they have not offered an explicit statement of

definition of imagery with respect to memory and learning.

There has been "with memory images, as with memory in gen-

eral, an almost irresistible tendency to rely on metaphors"

(Crowder, 1976). Paivio (1971) cites three basic operations

that have been employed by cognitive psychologists to describe

and manipulate imagery: "stimulus attributes, experimental

manipulations and individual differences" (p. 253). Viewed

within the perspective of Paivio (1971) and Kosslyn &

Pomerantz (1977), it may not be necessary to assume a defini-

tion of the image. Kosslyn & Pomerantz (1977) state that the

"image is not the only construct in cognitive psychology that

lacks a unique operational definition . it is legitimate

for a scientific enterprise in its formative period to be

engaged in research on a construct whose definition has not

been precisely formulated" (p. 64). Therefore, one could

study the image in relation to its individual differences,

stimulus attributes and experimental manipulations without

resorting to a definition. For example, eidetic imagery has

been studied in relation to individual differences in the

subjects' eidetic abilities (Leask, Haber, & Haber, 1968).

The image has been described and studied through its stimu-

lus attributes, described "in terms of the image-arousing

value or concreteness of the stimulus material. This can

be viewed as a dimension ranging from objects or their pic-

tures on the high imagery end, through concrete words to









abstract words on the low end of the scale" (Paivio, 1971,

p. 253). Studies of learning and memory (Paivio & Madigan,

1970; Paivio, 1971) have shown that imagery can greatly en-

hance both these processes. Described in terms of its

stimulus attributes, the use of concrete noun images re-

sulted in significantly better paired associate recall than

the use of abstract nouns in the studies conducted by Paivio

and associates. Imagery has been found to have a facili-

tating effect not only in paired-associate learning but in

free recall, serial learning, verbal discrimination learn-

ing and the Brown-Peterson short-term (STM) memory task

(Paivio, 1969; Paivio & Rowe, 1970, Paivio & Smythe, 1971).

It has been proposed by Crowder (1976) that mediation of

retention through imagery techniques is among the most

powerful experimental effects found in the psychology of

memory (Crowder, 1976).

Within cognitive psychology, training subjects in

the use of imagery through different instructions appears

to be the most frequently used method of experimentally

manipulating imagery (Bower, 1971; Bugelski, Kidd, &

Segman, 1968). In his studies, P.J. Lang found that the

way instructions to the client are organized has been

shown to exert substantial control over the client's somato-

visceral responses during imagery.

Among the clinical treatment studies employing imagery,

only a handful of investigations conducted using covert









modeling have sought to determine whether manipulating the

type of imagery instruction employed results in differen-

tial treatment effectiveness. Covert modeling, as previously

mentioned, is the behavioral imagery based procedure in

which the client is "directed in the symbolic rehearsal of

appropriate behavior" for a difficult performance task or

feared situation (Mahoney, 1974, p. 112). These covert model-

ing studies have suggested that one can enhance the efficacy

of treatment by instructing subjects in the sort of imagery

that should be used (Kazdin, 1974; Cautela, Flannery, & Hanley,

1974). For example, Kazdin (1974) found that subjects who

were instructed to imagine a model of the same sex and age

were significantly more effective in reducing avoidance be-

havior (fear of harmless snakes) than those who were in-

structed to imagine models that were dissimilar in these

characteristics. Recent research on model characteristics

has also focused on the superiority of coping vs. mastery

models in the alleviation of avoidance behavior (Kazdin,

1973a, 1974; Meichenbaum, 1972). Both Meichenbaum and Kazdin

concluded as a result of their studies that subjects who were

instructed to imagine coping models who demonstrated initial

fearful behavior, then coping behavior and finally mastery

behavior were significantly more effective in enhancing

behavioral change than were subjects who were instructed

to imagine mastery models who demonstrated no fear, but

competence throughout. These studies suggest that manipu-

lating the content of imagery through instructions may









substantially augment the power of the imagery technique

in bringing about desired changes in behavior. Support for

this contention is found in the results of a recent study

by Harris & Johnson (1980) who compared the efficacy of

instructing subjects in individualized coping imagery based

on experiences of success and competence to self-control

desensitization for the treatment of test anxiety. The

images of competence and success used as part of this coping

technique were drawn from the individual's own repertoire

of success experiences achieved in situations other than

test taking. In this study, the type of imagery instruc-

tion employed seemed to be important. Individualized coping

imagery treatment was significantly more effective in in-

creasing grade point average and decreasing self-report of

anxiety than was self-control desensitization. A replica-

tion of individualized coping imagery treatment produced

similar results.

Building upon the results of the previous studies

(Kazdin, 1974; Lang, 1977; Harris & Johnson, 1980) which

found support for the use of instructions aimed at manipu-

lating the content of the imagery produced, the primary

purpose of the present study was to assess the effectiveness

of different imagery instruction procedures for the treatment

of test anxiety. In this study a comparison of the efficacy

of instructing subjects in individualized coping imagery

treatment based on nonacademic experiences of competence and

success and individualized coping imagery treatment based on









academic experiences of competence and success was carried

out. The images of competence and success used as part of

the initial treatment study conducted by Harris & Johnson

(1980) were essentially nonacademic in nature, using suc-

cess experiences drawn from fields as diverse as athletics

and gourmet cooking. Because the target anxiety of this

study was test anxiety, it was felt that it might prove

useful to instruct subjects in the use of individualized

coping images based upon previously realized academic suc-

cess (e.g., making an academic presentation, doing well on

a previous exam) within the desensitization framework for

the treatment of test anxiety. The use of academic images

of competence derived from subjects' academic achievements

may further promote generalization through the employment

of imagery highly relevant to test taking and other academic

stressors. For subjects in both these treatments, images

of competence were drawn from the individual's own rep-

ertoire of success experiences and the individual was a

party to the formulation of his own treatment. As in the

previous study, the subject was able to become an active

agent in a very individualized regimen, perhaps making the

approach more efficient.

Another purpose of this current study was to assess

how effective instruction in relaxation is in increasing

both imagery vividness and overall treatment effectiveness

of the individualized coping imagery techniques. Several

investigators, among them Singer (1973, 1974), Van Egeren,






29


Feather & Hein (1971), have cited the role that relaxation

instruction plays in increasing both the vividness and

overall treatment effectiveness of imagery based techniques.

However, their conclusions emerge mainly from case reports,

therapist observations, etc., and have not been subjected

to controlled assessment.















METHOD


In order to compare the efficacy of individualized

coping imagery based upon academic and nonacademic images

of competence and to assess the value of relaxation train-

ing, students seeking treatment for their test anxiety were

randomly assigned to one of four treatment conditions in a

2x2 factorial design. Subjects in the first treatment group

were given training in the use of nonacademic coping images

to reduce test anxiety. The second group received both

training in progressive relaxation and the use of non-

academic coping images. Subjects in group three were

trained in the use of individualized coping imagery based

on academic success experiences. The fourth treatment group

received both individualized coping imagery based on aca-

demic experiences of competence and relaxation training.

Because no amount of anxiety relief can compensate for in-

adequate study skills, a study skills package was adminis-

tered to all treatment groups (Mitchell & Ng, 1972). In

addition, in order to assess whether any of the treatments

produced an effect, a waiting list control group was

established.









Subjects


Sixty-three subjects were selected from a group of

individuals meeting the following criteria: (A) volunteered

in response to announcements of a study for the treatment

of test anxiety made in several core university premedical

courses (e.g., organic chemistry, chemistry, biology), in

which multiple examinations are given; (B) scored above 30

on the debilitating anxiety scale of the Achievement Anxiety

Test (Alpert & Haber, 1960); (C) had a cumulative grade

point average (GPA) under 3.5; (D) were willing to partici-

pate in all of the 8 treatment sessions and 2 assessment

sessions; (E) granted permission for the investigator to

obtain grade records from the Registrar's office. Students

who had the highest degree of subjectively measured anxiety

and the lowest cumulative grade point average were con-

sidered first for treatment.


Measures


Treatment outcome. Two different sets of measures

were used to assess the efficacy of the different treatments

for test anxiety: five self-report measures of test and

evaluative anxiety and academic performance as reflected

in change in grade point average.

The self-report anxiety assessment battery, adminis-

tered at pretest and posttest, consisted of:








(A) The Suinn Test Anxiety Behavior Scale (STABS)--

this is a 50 item inventory describing situations related

to test taking (Suinn, 1969).

(B) The Alpert-Haber Achievement Anxiety Test (AAAT)--

this instrument is designed to measure both Debilitating

Anxiety (DA) and Facilitating Anxiety (FA) in test situa-

tions (Alpert & Haber, 1960).

(C) The Test Anxiety Scale (TAS)--the TAS is a 37 item

true/false inventory which attempts to assess the subjective

emotional reactions experienced by students in test situa-

tions (Sarason, 1972).

(D) The Social Avoidance and Distress Scale (SADS)--

this scale consists of 28 items which appear in a true/false

format and relate to how much subjects deliberately avoid

social situations (Watson & Friend, 1969).

(E) The Fear of Negative Evaluation Scale (FNES)--is

comprised of 30 items which appear in a true/false format;

this inventory measures how fearful subjects are of negative

evaluation (Watson & Friend, 1969).

Grade point averages were collected for the quarter

prior to treatment, the quarter during which the study was

conducted and the quarter following treatment.

Expectations for improvement. In the postassessment

session, subjects in all five groups were asked to estimate

what their expected grade point average for that treatment

quarter would be, given their academic performance up to

that point in the quarter. This expected grade point








average was designed to assess whether there were different

expectations for improvement across treatment groups.

Treatment efficiency: Speed of hierarchy completion. In

order to assess one aspect of treatment efficiency, the num-

ber of trials it took for subjects to complete the hierarchy

was tabulated.

Imagery measures. Subjects who met the initial criteria

for inclusion in the study underwent an additional assess-

ment procedure designed to assess the images the subjects

produced. This assessment was conducted both before and af-

ter treatment. Scenes were selected by dividing the de-

sensitization hierarchy employed by Harris & Johnson

(1980) in their test anxiety treatment study into three

portions reflecting low, medium and highly anxiety evoking

items. One item was randomly chosen from each anxiety

level of the hierarchy. Thus, one scene was chosen from

the least anxiety evoking section of the hierarchy (items

1-9), a second scene was randomly chosen from items (10-18),

and a third scene was chosen from the most anxiety evoking

third of the hierarchy (items 19-27). These same target

scenes were employed for subjects in all five groups.

All imagery assessments were conducted individually in

a dimly illuminated carpeted room in which distracting

sounds were minimized. Subjects were presented verbally

with four situations to imagine and describe. Before imag-

ining and describing the three actual test taking target








scenes, the subject was given a pretest practice scene

which was to imagine that s(he) is taking a walk on a sunny

spring day.

For this practice scene, as in the subsequent three

scenes to be imagined, the subject was asked to imagine this

scene as clearly as s(he) could, and when s(he) imagined

the scene as clearly as s(he) could, to raise a finger of

the right hand. At that point, the subject was asked to

describe what was imagined in detail. In these preliminary

instructions to the subjects, it was emphasized that the

purpose of this procedure was to help the therapists in

preparing for treatment.

After the subject described the practice scene, s(he)

was told that "I have a few situations related to tests and

test taking here--just a few that I'd like you to imagine

and describe to me in detail." The same procedure was fol-

lowed for the imagined test-related scenes as was followed

for the practice scene. Thus, after imagining the scene,

the subject was asked to describe the scene in detail. A

tape recorder was used to record both the experimenter's

instructions and the subjects' descriptions of the imagined

hierarchy items.

To control for order effects, all six combinations of

scene presentation were employed. Thus, one-sixth of the

subjects were asked to imagine the least anxiety evoking

scene first, the second most anxiety evoking scene second

and the most anxiety evoking scene last. A second sixth of









the subjects in all treatment conditions were asked to im-

agine the most anxiety evoking scene first, the second most

anxiety evoking scene second and the least anxiety evoking

scene last, etc. Thus, within each treatment group of

twelve subjects, two participants followed the same order

of scene presentation.

This same procedure employing the same practice scene

and three different scenes chosen from the same test

anxiety hierarchy was repeated at posttreatment assessment.

Although different test anxiety scenes were employed, each

subject was asked to imagine one scene from the low, medium

and high anxiety levels of the hierarchy.

From these pre- and posttest scene descriptions, four

imagery measures were taken:

(A) Elaborateness--a behavioral measure of scene

elaborateness was defined as the number of words employed

by the subject to describe each of the target scenes s(he)

was asked to imagine. Numerical tabulations of the number

of words in each scene were compiled by the author and

checked by a volunteer university student.

(B) Latency--this measure was defined as the length of

time it took for a subject to signal by raising a finger

that s(he) was ready to begin his (her) description of a

target scene, after having received instructions from the

experimenter to imagine and describe these scenes. This

time was measured with a stop watch to allow for an

accurate reading.









(C) Content: Stimulus/Response--stimulus content of

a scene is that aspect of the imagined situation which re-

fers to the physical details of the object or situation

described. The response content of the imagined scene re-

fers to the somatovisceral or emotional responses that the

subject described him/herself making in the imagined target

situations.

In order to provide for reliability, seven raters, un-

associated with the study, coded the imagery descriptions

given by subjects in response to the target scenes. Each

sentence was classified by a rater according to whether it

primarily reflected stimulus attributes or response at-

tributes.

In advance of the actual coding procedure, all identi-

fication was removed from subjects' imagery descriptions.

Packets of equal numbers of imagery descriptions were then

randomly selected. Neither the experimenter nor the raters

knew from which treatment group these descriptions were

drawn.

Raters received pretest training in order to afford

them sufficient practice in coding the sentences of the

scenes, according to the criteria employed. Thus, as part

of the pretest training the principles governing the differ-

entiation of predominantly stimulus and response sentences

were first discussed and examined. After raters clearly

understood stimulus vs. response characteristics, they then

received both examples and sample scenes to code along these









dimensions. Difficulties experienced in coding these scenes

were explored and any questions were answered by the experi-

menter.

Raters were checked for reliability every five scenes.

One of the scenes in each packet which had been randomly

chosen from all imagery descriptions was the scene on which

each rater was evaluated for reliability. Reliability was

calculated for each test scene employing the following

formula:

agreements
agreements + disagreements

If a rater's reliability fell below 80% on any test scene,

this rater received retraining on additional scenes to en-

able him/her to better understand and employ the criteria.

The rater then was required to rerate the packet on which

s(he) fell below the reliability standard. The experimenter

served as the reliability standard for all raters.

(D) Content: Coping--in addition to coding the imagery

descriptions according to whether they reflected predomi-

nantly stimulus and response characteristics, raters were

asked to evaluate and classify the same imagery descrip-

tions as to whether each sentence of each description given

in response to the target scene reflected predominantly

positive or negative coping. For those statements which

raters saw as neither positive nor negative coping in

character, they were asked to code as neutral.









As part of this procedure, the seven raters were

assessed for reliability throughout their coding of the

data, with the experimenter serving as reliability standard.

All procedures to assess reliability for these variables

were identical to those employed in the coding of stimulus-

response data. Thus, if reliability fell below 80%, raters

received additional training in the coding principles and

procedures and were instructed to code that set of descrip-

tions and were checked again.


Treatment


After measures of general and test anxiety were admin-

istered and imagery assessments conducted, subjects were

randomly assigned by sex to one of five groups: individual-

ized coping imagery treatment based on nonacademic experi-

ences of competence and success (N=12), individualized cop-

ing imagery treatment based on nonacademic experiences of

competence and success combined with relaxation training

(N=12), individualized coping imagery treatment based on

academic images of competence and success (N=12), individual-

ized coping imagery treatment based on academic images of

competence and success combined with relaxation training

(N=12), and a waiting list control group (N=15).

Therapists. The author and one female, advanced gradu-

ate student in clinical psychology served as therapists for

this study. Each of the therapists conducted two of the

four types of treatment so that each therapist had contact

with the same number of subjects.









Treatment setting. All group treatment took place in

a dimly illuminated carpeted room in which distracting

sounds were minimized.


General Procedures


All treatments were conducted in groups for the eight

weekly, 60 minute sessions and were given the same rationale,

anxiety hierarchy, study skills training and homework assign-

ments.

Rationale. Individuals taking part in these groups

were given a rationale that emphasized that they had re-

sponses already in their repertoire, success experiences

that could be used to cope with anxiety (see Appendix B).

Test anxiety hierarchy. In session two of treatment,

the construction of a group hierarchy was carried out by

employing the results of the preassessment measures in con-

junction with subject suggestions to create a hierarchy

specific to test taking situations. In order to control for

the divergent elements of the treatment procedures, the same

hierarchy was employed for all four treatment groups. For

all treatment groups, each item on the hierarchy was pre-

sented for a minimum of two times. The length of item pre-

sentation was either 60 seconds or the amount of time for

all individuals in the group to imagine the hierarchy scene

for a 20 second anxiety free period. However, if a subject

signaled anxiety after the second exposure, the item was

repeated until no subject signaled anxiety when imagining

this scene.









Study skills. Subjects in all four treatment groups

received training in study skills employing a variety of

study skills techniques. Throughout all sessions, as part

of the study skills portion of the treatment, the basic

focus was on helping subjects to examine their present be-

havior and to try new and more productive methods for meet-

ing the demands of their current academic environment. In

these sessions it was emphasized that more efficient prepara-

tion would in turn result in fewer feelings of anxiety prior

to and during examinations. In session one, discussion

centered on how students arrange their study time, examin-

ing for high and low efficiency areas. In addition, an

attempt was made to elicit subject suggestions concerning

specific behavioral steps which might reduce study interrup-

tions and more adequately meet course related demands on

their time.

In the second session, study skills training focused

on different methods of increasing study behavior such as

systematically studying in the same place and time, re-

ducing study interruptions and more adequately meeting

course related demands on their time. The student was

helped to construct and modify proposed time scheduled

during the following weeks.



1These study skills techniques were the same ones used
by Harris & Johnson (1980) and were based on a study
skills package developed by Allen (1971).









In session three the SQ3R method of study (Robinson,

1961) was taught to subjects. This method was used to help

students select what they are expected to know, comprehend

these ideas more rapidly, fix these ideas in memory, and

review efficiently for examinations.

Session four of the study skills portion of the treat-

ment was concerned with techniques of good note taking,

with an emphasis on organization. In addition, the SQ3R

method was reviewed. Sessions five and six focused on

preparation for examinations--reviewing, outlining, repeti-

tion and additional memorization techniques. Session seven

centered on the use of response management techniques to

increase study behavior. The final session served to re-

view the principles covered during the entire study skills

part of treatment. Study skills training took approximately

20 minutes of each treatment session.

Homework. Subjects received a homework assignment

based on the study skills techniques examined during each

treatment session.


Treatment Manipulations


In a 2(academic vs. nonacademic coping images) x

2(relaxation vs. no relaxation) design, each treatment

group differed in the type of imagery and relaxation in-

structions employed.

Individualized coping imagery treatment based upon

nonacademic images of competence. For two treatment groups,









highly personalized, previously established images of com-

petence and proficiency achieved by the subjects in stress-

ful environments other than test taking were paired with

visualization of anxiety eliciting scenes that comprised the

test anxiety hierarchy. Individuals taking part in this

treatment were given a rationale that emphasized responses

already in their repertoire as current resources that could

be used to cope with test anxiety. The first session

focused on a group discussion of possible success experiences

of the subjects that could be transformed into images of

competence for use in this covert imagery procedure. In

sessions three through eight, the actual treatment proce-

dure was carried out. Using these previously established

images of competence (e.g., running a radio broadcast,

tending a busy bar, playing in a recital), subjects were

presented with the test anxiety hierarchy items. In ad-

vance of the presentation of each item, subjects were

instructed (1) to imagine themselves performing well in

their personalized coping image and (2) then to transpose

the same confident and competent person of the personalized

coping image to the imaginal anxiety-provoking situation

presented in the hierarchy. Thus, the subject was instructed

to imagine this same competent person of his individualized

coping image now contending with test situations as

they appeared in the hierarchy.

Individualized coping imagery treatment based upon

academic images of competence. Subjects in the two other









treatment groups were instructed to employ individualized

coping imagery based upon academic images of competence.

Highly personalized, previously established images of com-

petence and proficiency achieved by the subjects in aca-

demic activities (e.g., making presentations, performing

difficult problems in class, successful test taking experi-

ences in the past) were paired with the visualization of

anxiety scenes from the test anxiety hierarchy. Individuals

taking part in this treatment were given a rationale that

emphasized responses already in their repertoire and cur-

rent resources that could be used to cope with test anxiety

and that these current resources could be derived from their

own experiences of competence and success in academic ac-

tivities. It was assumed that even test anxious college

students would have some history of academic achievement

and success (albeit an intermittent one) in order to have

gained college admission. The first session focused on a

group discussion of possible success experiences of the

subjects that could be transformed into academic images of

competence for use in this covert imagery procedure. In

sessions three through eight, the actual treatment procedure

was carried out. Using these previously established aca-

demic images of competence, subjects were presented with the

test anxiety items. In advance of the presentation of each

item, subjects were instructed (1) to imagine themselves

performing well in their personalized coping image and

(2) then to transpose the same confident and competent









person of the personalized academic coping image to the

imaginal anxiety-evoking situation presented in the hier-

archy. In this treatment, too, the subject was instructed

to imagine this same, competent, confident person of his

individualized coping image now contending with test situa-

tions as they appear in the hierarchy.

Progressive relaxation instructions. Subjects in two

of the treatment groups received training in progressive

relaxation following the format developed by Dorothy

Susskind (1970). The relaxation procedure consisted of

the alternate tensing and relaxing of various muscle groups.

Subjects were told that the relaxation training facilitated

the lowering of their tension level, making it easier for

them to cope with test taking situations both as they

imagined them in the session as well as when they were

confronting them outside. Subjects were instructed to

practice their relaxation at home between sessions. The

other two treatment groups received no specific instructions

in progressive relaxation.

Waiting list control group. Fifteen subjects made up

a fifth experimental group of waiting list controls. After

the posttreatment assessment, individuals of the waiting

list control group were offered the opportunity to partici-

pate in the test anxiety treatment proven to be most ef-

fective.















RESULTS


Analysis of Treatment Effects


All subjects, eighteen males and fifteen females, in

the four treatment groups who completed the eight one-hour

treatment sessions and two assessment sessions were in-

cluded in the data analysis. Three subjects dropped out

of each of the two relaxation treatment groups, leaving

nine subjects in each. For the group involved in individu-

alized coping imagery treatment without relaxation, four

subjects withdrew in the course of treatment. Five sub-

jects withdrew from the academic coping imagery treatment

without relaxation. Of the control subjects, thirteen out

of fifteen completed the postassessment measures. A multi-

variate analysis of variance was conducted on the pretreat-

ment assessment measures completed by the dropouts to de-

termine if they differed significantly from nondropouts.

This analysis revealed no significant differences. Conse-

quently, data from dropouts were discarded from all

subsequent analyses. The intercorrelation matrix for the

scores on the seven dependent measures used to assess the

efficacy of test anxiety treatment is presented in Table 1.

As expected, the correlations between self-report measures









Table 1

Intercorrelation Matrix of Scores on the Seven Dependent
Measures at Pretreatment



Measure 1 2 3 4 5 6 7


1. GPA -.19 -.02 .024 -.20 -.02 -.06

2. AAT-DA -.63 .488* .449* .12 .075

3. AAT-FA -.29 -.19 -.04 .19

4. STABS .38* .22 .30

5. TAS .12 .21

6. SADS .25

7. FNES


Note: GPA = Grade Point Average; AAT-DA = Debilitating
Anxiety subscale of the Alpert-Haber Achievement Anxiety
Test; AAT-FA = Facilitating Anxiety subscale of this same
instrument; STABS = Suinn Test Anxiety Behavior Scale;
TAS = Test Anxiety Scale; SADS = Social Avoidance and
Distress Scale; and FNES = Fear of Negative Evaluation
Scale.

p < .05.



were low to moderate. There was no significant correlation

between any self-report measure and grade point average.

Since multivariate analysis of variance of pretreat-

ment scores revealed significant pretreatment differences

for one variable, FNES, this variable was subjected to a

separate analysis of covariance. A multivariate analysis

of the pretreatment scores for the remaining six self-

report and performance measures was nonsignificant,









F(24,102) = .83, p< .6913, indicating initial equivalence of

the four treatment and control groups on these measures.

To assess the relative efficacy of the four treatment

and one control conditions, a multivariate analysis of

variance was run on the six repeated measures (Service,

1979). Since the groups were significantly different at

the multivariate level F(24,90) = 2.25, p < .0032, further

analyses were performed. Separate repeated measures analy-

ses of variance were conducted for each of the dependent

measures. When appropriate, Duncan's Multiple Range Test

for pairwise comparisons was used to assess posttreatment

differences between groups. Pairwise T-tests were employed

for assessing pre- to posttreatment changes.1

Self-report measures of test anxiety. Repeated meas-

ures analysis of variance performed on the pretest and

posttest Debilitating Anxiety subscale scores of the

Alpert-Haber Achievement Anxiety Test revealed a highly

significant Group x Time interaction F(4,40) = 4.71, p <

.0047, indicating differential improvement among the groups

on this measure.



1Anadditional 2(Relax) x 2(Image) x 2(Time) Manova was
conducted to assess the possible main effects of relaxation
and imagery used on the treatment outcome measures. No
main effects nor interactions were found. A multivariate
analysis of variance was run to determine if treatment was
differentially effective for men and women taking part in
the study. The multivariate analysis was nonsignificant
indicating that there was no interaction between sex and
treatment for any of the dependent measures.









The mean Debilitating Anxiety and Facilitating Anxiety

scores for each of the four treatment and one control con-

ditions are shown in Table 2.

While all treatment groups decreased on this measure

of test anxiety, Table 2 reveals that the magnitude of

change from pretest to posttest appears to be greatest for

individualized coping imagery treatment alone, followed by

individualized coping imagery treatment combined with re-

laxation and academic coping imagery combined with relaxa-

tion. The Duncan's Multiple Range Test for pairwise com-

parisons employing Kramer's (1956) adjustment for unequal

sample sizes conducted on posttest scores indicated that all

four treatment groups scored significantly lower than the

waiting list control group, p < .05, but were not signifi-

cantly different from each other. Results of pairwise

T-tests indicated that all four treatment groups showed

significant change from pre- to posttreatment, p < .04.

No substantive change occurred for the waiting list control

group.

Repeated measures analysis of variance on the Facili-

tating Anxiety subscale at the AAT did not yield a signifi-

cant Group x Time interaction. The main effects for group,

F(4,40) = 12.88, p < .0001, and time, F(1,40) = 9.49,

p < .0045, were significant. Duncan's test showed that

each treatment group reported more facilitating anxiety

than the control group, p < .05. There was an overall in-

crease in pre- to posttreatment scores (see Table 2),

p < .05.















Ln
Ln


CN










N

r4l
N




































0 4-n






N rNd

S 0
C *H
< C


H -1-1 4
-I

- M C!
ac *a


cN


00co
N

N N


tn 0


OXd
0 4


N




II
or >

> -,-4


H -4


>1








0
-H -





0 rd
. i ci






C0
rQ


0 0
O X

aS
< I


>1

(1)rd

-4




0-4
o ^




4-J



< J -)


O




4
-I
0C
0




4-1f
*HO
cm
-I .
,-I









Repeated measures analysis of variance performed on

the Suinn Test Anxiety Behavior Scale (Stabs test) re-

vealed a highly significant Group x Time interaction

F(4,40) = 6.63,p < .0006 (see Table 3).



Table 3

Suinn Test Anxiety Behavior Scale Means


Pretest


Individualized coping imagery
with relaxation (N=9)

Academic coping imagery with
relaxation (N= 9)

Individualized coping imagery
--no relaxation (N= 9)

Academic coping imagery--no
relaxation (N= 7)

Wait list control (N= 13)


140


133.22


161.125


134.71

143.53


Posttest


106.55


100


118


100.142

169


Individualized coping imagery treatment without

relaxation showed the largest decrease in test anxiety,

followed by individualized coping imagery treatment

combined with relaxation, academic coping imagery alone,

and academic coping imagery combined with relaxation.

The waiting list control group increased on this

measure of test and evaluative anxiety. At posttest,

all four treatment groups were significantly lower









on this measure than was the waiting list control group,

p < .05. All other group comparisons were nonsignificant.

In addition, at posttest, pairwise T-tests revealed that

all four treatment groups were significantly lower on this

measure than at pretest, while the waiting list control

group showed a significant increase, p <..04.

Results of the Test Anxiety Scale, presented in Table 4,

show that all treatment groups decreased more than the con-

trol group on this measure.



Table 4

Sarason Test Anxiety Scale Means


Pretest Posttest

Individualized coping imagery
with relaxation (N= 9) 23.77 14.66

Academic coping imagery with
relaxation (N= 9) 23.88 14.33

Individualized coping imagery
--no relaxation (N= 8) 26.25 14.50

Academic coping imagery--
no relaxation (N= 7) 24.0 16.71

Wait list control (N= 13) 22.53 23.23




The magnitude of change from pretest to posttest appears

to be largest for individualized coping imagery treatment

alone followed by academic coping imagery combined with









relaxation, individualized coping imagery combined with

relaxation and academic coping imagery alone.

Repeated measures analysis of variance performed on

this measure revealed a highly significant Group x Time

interaction, F(4,40) = 3.47, p < .0196. At posttest all

four treatment groups scored significantly lower than the

waiting list control group, p < .05, but were not signifi-

cantly different from each other. Pre- to posttreatment

changes as measured by pairwise T-tests were significant

for the individualized coping imagery group combined with

relaxation, the academic coping imagery group combined with

relaxation and the individualized coping imagery treatment

without relaxation, p < .005. Pre- to posttreatment changes

on this variable for the academic coping imagery group

approached significance, p < .0680. The waiting list con-

trol group showed a slight, nonsignificant increase on this

measure.

Self-report measures of evaluative and social anxiety.

Repeated measures analyses of variance of the scores ob-

tained at pretest and at posttest on the Social Avoidance

Distress Scale (Watson & Friend, 1969) indicated no signifi-

cant Group x Time interaction on this measure. The analysis

did reveal significant group F(4,40) = 16.16, p < .0001 and

time F(1,40) = 16.90, p < .0002, main effects. However,

subsequent tests using Duncan's revealed no significant

differences between groups, although all subjects scored

significantly lower on this measure at posttest that at

pretest.









Because of initial pretreatment differences on the

FNES (Fear of Negative Evaluation Scale), an analysis of

covariance was conducted. This analysis revealed no dif-

ference between groups.

Academic performance. Mean grade point averages ob-

tained from the University of Florida Registrar's Office for

both the pretreatment and posttreatment quarters are pre-

sented in Table 5.



Table 5

Grade Point Average


Means

Pretreatment Posttreatment

Individualized coping
imagery combined with
relaxation (N= 9) 2.0 2.62

Academic coping imagery
combined with relaxa-
tion (N= 9) 2.32 2.96

Individualized coping
imagery without
relaxation (N= 8) 2.38 3.00

Academic coping imagery
without relaxation
(N= 7) 2.32 2.54

Waiting list control
group (N= 13) 2.66 2.08




Repeated measures analysis of variance performed on

the pretreatment and posttreatment grade point averages









revealed a highly significant Group x Time interaction

F(4,40 = 3.57, p < .0173, indicating differential improve-

ment among the groups on this measure of academic perform-

ance. Duncan's test revealed that at posttest none of the

four treatment groups were significantly different from

each other. Only the academic coping imagery group combined

with relaxation and the individualized coping imagery group

alone were significantly different from the control group.

From pre- to posttreatment, both individualized coping imag-

ery treatment combined with relaxation and the individual-

ized coping imagery group without relaxation showed a

significant increase in grade point average, p < .05. Pre-

to posttreatment changes in grade point average for the

academic coping imagery group combined with relaxation

approached significance, p < .089. Academic coping imagery

without relaxation made a small nonsignificant increase in

academic performance from the pretest to posttest. The

waiting list control group significantly decreased on this

measure of academic performance, p < .0437.


Expectations for Improved Academic Performance


In the postassessment session, which occurred before

summer term examination week, subjects in all four treatment

groups were asked to estimate what their grade point aver-

age for that quarter would be, given their academic per-

formance up to that point in the quarter. No significant

differences in group expectations emerged from these

analyses.









Treatment Efficiency: Speed of Hierarchy Completion


An analysis of variance performed on this measure re-

vealed a highly significant difference between groups

F(3,76) = 4.80, p < .0042. Further analyses in the form of

Duncan's Multiple Range Test showed that individualized

coping imagery combined with relaxation, academic coping

imagery combined with relaxation, and individualized coping

imagery alone were significantly more efficient treatments

in terms of the number of trials taken for subjects to com-

plete each item of the hierarchy than the academic coping

imagery--no relaxation treatment.


Imagery Analyses


The same individuals who completed the treatment out-

come measures completed the imagery assessment procedure at

pretreatment and posttreatment. However, one of the male

subjects, who completed the individualized coping imagery

treatment combined with relaxation and all treatment out-

come measures, failed to complete the posttreatment imagery

assessment.

For each of the imagery measures (i.e., elaborateness

or words, latency, stimulus-response, and coping) separate

analyses of variance were conducted on the pretreatment

scores on these variables to assess whether the dropouts

differed significantly from nondropouts. As these analyses

revealed no significant differences, data from dropouts were

discarded from all subsequent imagery analyses.









Scene elaborateness. A behavioral measure of scene

elaborateness was defined as the number of words employed

by the subject to describe each of the target scenes s(he)

was asked to imagine. In order to assess whether the four

treatment and one control group were different on this

behavioral measures of scene elaboratensss, a 5(Groups) x

2(Time) x 4(Scene) anova was conducted. This analysis

revealed no significant interactions but did yield main

effects for Time F(1,37) = 4.43, p < .04 and Scene

F(3,123) = 12.21, p < .0001. Further analysis in the form

of Pairwise T-tests indicated that at posttreatment, sub-

jects used more words to describe these target scenes than

at pretest. Duncan's revealed that all subjects used more

words to describe the three anxiety evoking scenes than

the practice scene, p < .05.

A further 2(Relax) x 2(Imagery) x 2(Time) x 4(Scene)

anova was performed in order to assess whether the addition

of relaxation or the type of imagery used differentially

contributed to scene elaborateness. This analysis also re-

vealed no significant interactions or main effects between

relaxation and imagery on this scene elaborateness measure.

Latency. This measure was defined as the length of

time it took for a subject to signal by raising a finger

that s(he) was ready to begin his(her) description of a

target scene (e.g., practice scene, anxiety evoking scenes)

after having received instructions from the experimenter to

imagine and describe these scenes. A 5(Groups) x 2(Time) x









4(Scene) analysis of variance was conducted to assess

whether the four treatment and one control groups were

different on this measure. This analysis revealed no

interactions or main effects.

To determine the effects of relaxation and image type

on latency, a 2(Relax) x 2(Imagery) x 2(Time) x 4(Scenes)

anova was also performed. All interactions and main effects

proved to be nonsignificant.

Scene content: Stimulus and response. Stimulus

content of a scene is that aspect of the imagined situation

which refers to the physical details of the object or situa-

tion described. The response content of the imagined scene

refers to the somatovisceral or emotional responses that

the subject described him/herself making in the imagined

target situations.

As described previously, seven individuals unassociated

with the study served as raters. Each rater received an

equal number of imagery descriptions to code. Each sentence

was classified by the raters according to whether it pri-

marily reflected stimulus attributes or response attributes.

The experimenter served as reliability standard for

all raters. If a rater's reliability fell below 80% on any

test scene, this rater received retraining on additional

scenes to enable him/her to better understand and employ the

criteria. The rater was then required to rerate the packet

on which s(he) fell below the reliability standard. Only

one of the seven raters required additional training on









the criteria. Reliability was calculated for each test

scene employing the following formula:


agreements
agreements + disagreements

Reliability between the raters and the experimenter (re-

liability standard) ranged between 80-100% and averaged

85% for all test scenes.

In order to control for differing scene length and to

normalize the data, all stimulus and response data were

subjected to Arcsin Transformations. Analysis of either

stimulus or response statements through a 5(Groups) x

2(Time) x 4(Scene) repeated measures analysis of variance

on the transformed scores was essentially equivalent,

yielding a highly significant Scene x Time interaction

F(3,102) = 4.3, p < .006 for both of these variables.

Duncan's revealed that at pretest subjects used signifi-

cantly more stimulus elements (statements) in the practice

scene than they did in any of the three anxiety scenes,

p < .05. Subjects also used more stimulus elements in

their descriptions of medium anxiety scene than they used

in the most anxiety scene, p < .05. The number of stimulus

statements evoked by the low and medium anxiety scenes

were not significantly different from each other. At

posttest, this pattern was again apparent, with the number

of stimulus elements used in the practice scene exceeding

those used in the anxiety scenes, p < .05. In addition,

the low anxiety scene exceeded the medium and high anxiety









scene in the number of stimulus elements used, p < .05.

The medium and high anxiety scenes were not statistically

different from each other.

Follow-up Duncans on the response variable revealed

that at pretest, significantly more response statements

were used to describe the three anxiety scenes than were

used in the description of the practice scene, p < .05.

More response statements were used to describe the high

anxiety scene than were used to describe the medium anxiety

scene, p < .05.

At posttest, descriptions of anxiety scenes again

contained significantly more response elements than made

up the descriptions of the practice scene, p < .05. There

was no difference in the number of response statements

elicited by the medium and high anxiety scenes. The number

of response elements evoked by the low anxiety scene was

lower than these two scenes, p < .05.

Table 6 shows both the percentages of each of the

anxiety scenes and practice scenes that were made up of

stimulus and response statements as well as the trans-

formed scores.

To determine the effects of relaxation and image type

used as part of treatment on both the stimulus and response

content of the scenes, 2(Relax) x 2(Imagery) x 2(Time) x

4(Scene), analysis of variance were performed on both the

stimulus and response variables. These analyses were











































co 0 r- o
Cm "









\D 0


m 0 C inO
: ) co mo


r- 0 CO
Ln Hr


oa


nU)

03




dp g

-U
m


0)


p 0

U)






i-l
,-

U)



Hd


a,

a,
U
U)


Ln







>1
4-)

, -


M:









essentially equivalent and yielded no significant inter-

actions or main effects for either of these variables.

Pairwise T-tests were conducted to ascertain if each

scene elicited more stimulus or response oriented content.

As can be seen in Table 6, a pattern of greater response

orientation compared to stimulus orientation was true for

all anxiety scenes, p < .001, at both pretest and

posttest. In contrast, for the practice scene, there was

no difference between stimulus and response orientation at

pretest. At posttest, the practice scene elicited signifi-

cantly more stimulus content, p < .03.

Scene content: Coping. Positive coping content of a

scene refers to those statements made by the subjects in

response to the target scenes which would indicate that

they were functioning effectively, adopting a problem-

solving orientation or attitude that would lead to a satis-

factory outcome. Negative or "bad coping" was the converse

of the above.

The same procedures for calculating reliability were

employed for coping/negative coping scene content as were

employed for stimulus and response content. Only one of

the seven raters required additional retraining on the

criteria in order to increase his level of reliability.

Reliability for coping/negative coping ranged from 80 to

100% and averaged 86%.

In order to control for differing scene length and to

normalize the data, all coping and negative coping data









were subjected to Arcsin Transformations. A 5(Groups) x

2(Time) x 2(Type of Coping) x 4(Scene) repeated measures

analyses of variance was run on the transformed scores to

determine if different treatments made subjects more likely

to describe themselves coping successfully or unsuccessfully

with the situations presented to them in these target

scenes. These analyses revealed a significant Scene x

Type of Coping x Time interaction F(3,136) = 5.53, p < .002.

Further Duncan's indicated that at pretest there were more

positive coping elements used for the three anxiety scenes

than for the pretest practice scene, p < .05 (see Table 7).

At posttreatment more positive coping statements were made

by subjects to describe the three anxiety scenes than were

used to describe the practice scene, p < .05.

Because of the pre-post increase in the percentage of

positive coping elements used for both the medium and high

anxiety scene, additional follow-up pairwise T-tests were

conducted. This analysis revealed a highly significant

change for the medium anxiety scene from pre- to post-

treatment, p < .0001.

Additional pairwise T-tests were employed to determine

if there might be differences between the treatment groups

in the number of positive coping statements used to describe

the medium anxiety scenes. Every treatment group showed

significant increases in the number of coping statements

employed from pre- to posttreatment. This did not occur

for the control group.















oo
C0 r-
c-'


,a




ol

E-i4
4-







0
o



4-4

(r
E-4






dP








dp


i-l m 1-1 i^








H r I H Vf
Nr) N



Ln H~ H


0)
E <


>1

-H


H <


0)
4
>0
> u

0i -
m cr

0
0
U



0 O
> 0
> u
-H c


O C-
0 -r-H









0
0)
> 0'
















0)

rto
a()

>0
4-r




u


O N m )N
Hl -l (N







C) O rH
I ID Ln










O r-


0-
O H
4- 41

CCO




QI



c,
40

(a







0
a4

0r r








Similar analyses were conducted on the negative coping

statements. Duncan's revealed at pretest that all anxiety

scenes elicited more negative coping statements than did the

practice scene, p < .05. The high and medium anxiety scenes,

though not significantly different from each other, elicited

more negative coping than the low anxiety scene, p < .05.

At posttest, all anxiety scenes again evoked more nega-

tive coping statements than the practice scene, and the high

anxiety scene evoked more negative coping than both the

medium and low anxiety scenes. These two were not signifi-

cantly different from each other.

Because of the pre-post decrease in the percent of

negative coping elements used in the low, medium and high

anxiety scenes, additional follow-up pairwise T-tests were

conducted. This analysis revealed a highly significant

decrease for the medium and high anxiety scenes from pre-

to posttest, p < .01. The pre- to posttest change for the

low anxiety scene approached significance, p < .07.

Additional pairwise T-tests employed to determine if

there might be differences between the treatment groups in

the number of negative coping statements used to describe

the three anxiety scenes showed significant decreases on

the medium anxiety scene for both the individualized cop-

ing imagery group with relaxation and the academic coping

imagery group with relaxation, p < .05, and a decrease ap-

proaching significance for the individualized coping imagery

group without relaxation, p < .09. A significant decrease

occurred for the individualized coping imagery group combined









with relaxation on the high anxiety scene, p < .02). No

significant pre- to posttest decrease was made for the

academic coping imagery group without relaxation or the

control group on any of the scenes.

Pairwise T-tests were also conducted to ascertain

whether different scenes elicited more positive coping or

negative coping content. At pretest, more positive coping

statements compared to negative coping statements were pro-

duced by subjects responding to both the low and medium

anxiety scenes, p < .06. A similar pattern emerged at post-

test with positive coping orientation dominating in the

practice scene and the low and medium anxiety scenes,

p < .03.

A 2(Relax) x 2(Type of Imagery) x 2(Time) x 4(Scene)

analysis of variance was conducted to assess whether the

addition of relaxation or the type of imagery used differ-

entially contributed to the number of positive coping and

negative coping elements used. This analysis revealed a

significant Relax x Time interaction, p < .04. However,

follow-up analyses in the form of Duncan's and pairwise

T-tests revealed no significant differences between groups

or changes from pre- to posttest on this variable.















DISCUSSION


Treatment Outcome


It was evident from the results of this study that all

variations of individualized coping imagery treatment

brought about significant decreases in test anxiety as

measured by self-report anxiety questionnaires. In terms

of academic performance, two groups, individualized coping

imagery without relaxation and academic coping imagery com-

bined with relaxation, were significantly different from

the control group at posttest. The individualized coping

imagery treatment combined with relaxation and individualized

coping imagery treatment without relaxation brought about

significant changes from pretreatment to posttreatment

assessment on grade point average. Subjects receiving

academic coping imagery treatment alone manifested substan-

tial improvement on some measures of evaluative anxiety,

but subjects in this treatment were not able to achieve

significant increases from pretest to posttest on academic

performance and were not significantly better than waiting

list controls. While preventing subjects from deteriorat-

ing in academic performance, it appears that academic cop-

ing imagery treatment alone was not enough to bring about









consistently significant pretest to posttest increases in

grade point average.

Academic coping imagery treatment alone was clearly

the least efficient of all of the four treatments. At the

outset of the academic coping imagery treatment, subjects

experienced much difficulty employing their academic images

of competence to counter the anxiety evoking situations of

the hierarchy. They required significantly more hierarchy

scene exposures than subjects taking part in the other

three treatment groups. Initially, subjects reported they

had trouble producing and imagining these academic images

of competence clearly. Thus, while the decreases in test

anxiety were similar to those of other treatment groups,

it did not appear to be as efficient to institute and carry

out. It was also noteworthy that this group had the high-

est dropout rate of all four treatment groups.

When all variables are considered, individualized

coping imagery with or without relaxation appears to be an

effective treatment across the domains of anxiety self-

report measures, academic performance and treatment effi-

ciency. Similar consistent positive results were obtained

with academic images only when they were paired with

relaxation training.

The waiting list control group showed an increase in

self-report anxiety on two of the primary self-report

measures of evaluative anxiety and a significant decrease

in grade point average from pretreatment to posttreatment









assessment which is consistent with the findings of several

previous investigators (Harris & Johnson, 1980; Holroyd,

1976; Paul & Shannon, 1966). It appears that subjects who

are part of a waiting list control group run the risk of

further deterioration in academic performance and increases

in anxiety.

It is noteworthy that differences that were obtained

in both treatment efficiency and academic performance were

not reflected in differential expectations for improvement

at posttest; apparently, all treatments were viewed by

participants as equally credible and convincing.

The fact that there was no Group x Time interaction on

both the Social Avoidance and Distress Scale (Watson &

Friend, 1969) and the Fear of Negative Evaluation Scale

(Watson & Friend, 1969) is consistent with the findings of

Harris & Johnson (1980) and Spielberger, Anton, & Bedell

(1975), who found that while test anxiety treatment may

produce reduction in anxiety measures that were specific to

testing situations, such treatment may not result in gen-

eralized significant decreases on self-report indices of

social and evaluative fear.

The low correlations achieved between grade point

average and self-report anxiety measures in this study

points to the need for multiple measures in treatment

outcome studies of this type.

Consistent with previous findings, there were no dif-

ferences in the efficacy of treatment for men and women









taking part in this study. Although previous investigations

have found that women admit to more anxiety than men (Hersen,

1973), in this particular group of subjects who sought treat-

ment, there were no differences.

A significant finding to emerge from the results of

this study was that manipulating the type of imagery in-

struction employed does appear to result in differential

treatment effectiveness at least on some measures. This is

consistent with the findings of Kazdin (1973a, 1974) and

Lang (1977, 1979) that the content of imagery is very sensi-

tive to instruction. The hypothesis that the use of aca-

demic images of competence alone, derived from subjects'

academic achievements, would further promote generalization

through the employment of imagery highly relevant to test

taking and other academic stresses was not supported by

the results of the study. The magnitude of change for

individualized coping imagery based on nonacademic experi-

ences of competence and success was greater than that made

by academic coping imagery alone on grade point average.

While academic coping imagery treatment alone led to small

nonsignificant gains from pretest to posttest, in grade

point average, nonacademic individualized coping imagery

treatment produced a significant increase in academic per-

formance. It had been hypothesized that because academic

images of competence were highly relevant to test-taking,

it might therefore be a more efficient treatment. The

reverse appears to be the case.









An explanation for this finding may lie in the nature

of the coping imagery instruction itself. Critical to the

individualized coping imagery treatment is the direct

harnessing of highly individualized past performance ac-

complishments as coping skills and using them within the

context of the desensitization procedure to reinforce the

belief that highly test anxious individuals can cope ef-

fectively in stressful situations such as test taking.

Their past performance accomplishments are used to provide

evidence of this. By their nonacademic nature and their

lack of association with the stressor, nonacademic experi-

ences of competence and success evoke feelings of competence

and success, not fear and avoidance. In contrast, academic

experiences of competence and success are part of a class

of situations (academic or test taking) that for these

subjects have a potential to evoke fear and avoidance.

Thus, subjects may have more difficulty producing and

maintaining academic images of competence as part of the

desensitization procedure; as such, images are too closely

associated with the experience of test taking and its

concomitant anxiety. It is not surprising, given these

factors, that both treatment efficacy and efficiency should

be impaired.

As reported in the study conducted by Harris &

Johnson (1980), test anxious subjects have often indicated

in the course of their treatment program that their negative

thoughts are, for them, more vivid and perseverant than









their images of competence and proficiency. These negative

cognitions appear to have been overlearned and may be more

directly and parsimoniously countered by the overlearning

of highly individualized positive self-thoughts, images of

competence and proficiency that comprise the coping tech-

nique employed in the nonacademic individualized coping

imagery treatment. Thus, because of the close association

with the stressor, academic coping imagery may not be in-

herently as able to alter anxiety to the same degree as

treatments employing nonacademic coping images.

However, it appeared that the addition of relaxation

to the academic coping imagery treatment augmented both its

treatment efficacy in terms of grade point average and

treatment efficiency. Subjects required no more scene

presentations to complete the desensitization hierarchy

than participants in the two nonacademic coping imagery

treatment groups.

Previous investigators, among them Singer (1973, 1974);

Van Egeren, Feather & Hein (1971), have proposed that

relaxation plays an important role in increasing both the

vividness and overall treatment effectiveness of imagery

based techniques. Their conclusions emerged mainly from

case reports and therapist observations and had not been

subject to controlled assessment. It may be that because

relaxation training reduces distracting physical sensation

and thought (Lang, 1979); it may, in turn, enhance the

ability of subjects to approach stressful situations,









including imaging even potentially stressful coping re-

sponses. Thus, as a result, subjects may be able to ef-

fectively use this response in the subsequent anxiety evok-

ing hierarchy scenes, unencumbered by fear or avoidance.

In contrast, the addition of relaxation did not substan-

tially add to the power of nonacademic individualized cop-

ing imagery treatment.

As a result of this study, the basic hypothesis that

different imagery instructions could differentially affect

test anxiety and performance was confirmed. Indeed, changes

in the instructional content of the images employed showed

these treatments to be differentially effective in increas-

ing academic performance and treatment efficiency. Such

instructions appeared to be sensitive to experimental manipu-

lation with variations in these instructions (e.g., the

differences in instructions for individualized coping

imagery vs. academic coping imagery) producing changes in

overall treatment efficacy as measured by academic per-

formance and efficiency.


Imagery


Results of the imagery analysis suggest the absence of

any significant relationship between the use of relaxation

as part of treatment and the production of highly elaborate

hierarchy scenes as measured by the number of words used by

subjects to describe the target scenes. In addition, data

analysis did not uncover differences in this elaborateness










measure for any of the five treatment groups. It is to be

noted, also, that relaxation training did not increase

subjects' use of stimulus, response, positive coping or

negative coping elements as part of their scene descriptions.

While treatment did not appear to influence the elabor-

ateness of imagery produced, it appears that the target

scenes the subjects were asked to imagine and describe were

discriminative stimuli, evoking different patterns of im-

aginal responding. Subjects not only used more words in

describing the anxiety evoking scenes than they used in

describing the practice scene but used significantly more

response elements as well. Similarly, more positive and

negative coping statements were used to describe the

anxiety evoking scenes than the practice scene. These

findings are consistent with the results of desensitization

studies (Lang, Melamed, & Hart, 1970) which demonstrated

that subjects respond to anxiety scenes with increased

physiological arousal as measured by GSR, heartrate in-

creases, etc. Anxiety scenes may not only produce dif-

ferences in physiological responding but may lead to dif-

ferences in the subject's imagery content as well.

While the use of different imagery treatment instruc-

tions appeared to have given subjects increased control

over their anxiety and academic performance, it was found

that imagery based treatment alone did not differentially

change the number of stimulus or response elements subjects

used in their scene descriptions. It was found on all








variables that there was no significant difference for the

four treatment and one control groups in the elaborateness or

stimulus-response content of the images subjects produced.

On the other hand, there were significant differences

in content for both the number of positive coping statements

and negative coping statements subjects used to describe

these anxiety scenes. From pretreatment to posttreatment,

subjects significantly increased their use of positive cop-

ing statements for the medium anxiety scene. Further statis-

tical analysis revealed that this change could be accounted

for by significant increases in positive coping statements

employed by all four treatment groups. This increase in

positive coping elements was absent for the control group.

Similar analyses made for negative coping statements

revealed highly significant decreases in negative coping for

both the medium and high anxiety evoking scenes from pre-

to posttreatment. The decrease for the low anxiety evoking

scene approached significance. Further analysis demon-

strated significant decreases in negative coping on the

medium anxiety scene for individualized coping imagery and

academic coping imagery combined with relaxation and de-

creases approaching significance for the individualized

coping imagery treatment without relaxation. A highly sig-

nificant decrease in negative coping statements was made on

the high anxiety scene by the individualized coping imagery

group. There were no significant changes for either aca-

demic coping imagery treatment alone or the waiting list

control group.









An examination of the content of the images at pre-

treatment revealed more positive coping statements compared

to negative coping statements were produced by subjects

responding to both the low and medium anxiety evoking scenes.

At posttreatment, a similar pattern emerged with positive

coping orientation predominating for the practice and the

low and medium anxiety scenes.

Because positive coping imagery significantly increased

and negative coping imagery significantly decreased, this

finding suggests that it may be that changes in image con-

tent may also figure in the successful outcome of treatment.

Unlike the report of Lang and his associates (Lang,

1977, 1978) which suggests that increasing the response

orientation (increasing the number of response sentences)

of the individual is critical to imagery based treatment,

the results of this study suggest that changing the content

of the image to incorporate successful coping strategies

may be a more significant feature. More importantly, the

finding that significant decreases in negative coping

elements on the medium anxiety scene for the treatment

groups who attained the largest increase in grade point

average suggest that decreasing negative coping imagery may

even be more crucial to the successful outcome of imagery

based treatment than increasing positive coping imagery.

These findings are in line with the theoretical model

proposed by Wine (1971) and Ellis (1962) that negative

cognitions such as self-statements lead to performance









decrements. In Wine's view, it is the modification of

negative self-statements and the substitution of positive

ones that serves to decrease anxiety and enhance perform-

ance.

Examining imagery data is only one way of attempting

to assess behavioral and attitudinal change as a result of

treatment. One cannot be completely sure that images pro-

duced in an assessment situation are adequate reflections

of the emotional content of subjects' imagery in the natural

environment (such assessments are subject to the demand

characteristics of the situation). Yet this type of assess-

ment may offer some insight into the process of changing

the emotional content of imagery and its possible relation-

ship to effective behavior change.

Bandura, Adams, & Beyer (1977) and Averill (1973) have

proposed that in dealing with potentially aversive events,

it is the perceived inefficacy and incompetence in coping

with them that makes them so fearsome. This study has used

the vehicle of changing the content of the emotional image

to impart a significant sense of personal control.

This study has addressed a number of critical questions

about the nature of imagery produced and how man's image

making capacity can be put to the use of alleviating emo-

tional suffering. Enlarging on the work of other investi-

gators into covert phenomena, the individualized coping

imagery technique employed in this study not only afforded

significant relief from anxiety but also demonstrated that






77


concrete behavior change for the individuals involved in

this study could be achieved. While the use of imagery has

become an established therapeutic tool, the careful study

of the image may permit us a better understanding of the

process by which a cognitive treatment yields behavioral

or emotional change.


















APPENDIX A
PROPOSITIONAL UNITS OF THE EMOTIONAL IMAGE


A. Stimulus propositions (auditory, visual, tactile,
cutaneous, olfactory, vestibular, kinesthetic)
1) physical details of the object or situation
2) changes in object configuration
3) object movement (approach or withdrawal)
4) physical place or general location
5) presence or absence of others as observers or
participants
6) comments made by others
7) pain--location on the body; sharp, dull, etc.

B. Response propositions
I. Verbal responses
1) overt vocalization--out loud comments or
expressive cries
2) covert verbalizations
a) emotional labeling
b) self-evaluative statements, e.g., feelings
of inferiority
c) attribution of attitudes to others

II. Somato-motor events
1) muscle tension
2) uncontrolled gross motor behavior
3) organized motor acts--freezing, approach,
avoidance

III. Visceral events
1) heart rate and pulse
2) body or palmar sweat
3) vascular changes--blanching or flushing
4) pilomotor response
5) salivary response--mouth dry
6) respiratory change
7) intestinal upset
a) vomiting
b) incontinence
8) urinary dysfunction

IV. Processor characteristics
1) perception unclear or unusually vivid,
or distorted










2) loss of control over thoughts, cannot think
clearly
3) disoriented in time or space

V. Sense organ adjustments
1) general postural changes
2) eye and head movements







































Note: All hierarchy scenes described by the subjects
were coded for extensiveness and elaborateness of verbal
description, employing the image taxonomy presented above
and developed by Lang (1977).















APPENDIX B
INSTRUCTIONS TO THE EXPERIMENTER:
INDIVIDUALIZED COPING IMAGERY TREATMENT


This is the manual for the treatment of test anxiety

through the use of individualized coping imagery combined

with study skills training. This treatment is designed to

build an individualized structure which employs the individ-

ual's perception of self-competence as the primary coping

technique. In this treatment images of perceived self-

competence are incorporated into a covert modeling procedure

for the treatment of test anxiety. Six steps are involved

in the use of this technique.

1) Exploration of degree and extent of test anxiety of

subject

2) Presentation of rationale for treatment

3) Group discussion of possible success experiences of the

subject that can be transformed into images of competence

for use in this covert imagery procedure. This step also

includes discussion of treatment goals and problems

4) Construction of test anxiety hierarchy

5) Employment of covert imagery treatment

6) Study skills training









Treatment Schedule


Session One

1) Group introduction and exploration of degree and extent

of test anxiety for each subject (10 minutes)

2) Rationale described and treatment explained (10 minutes)

3) Beginning group discussion of possible success experi-

ences that could be transformed into images of competence

to be used as part of this covert imagery procedure (15

minutes)

4) Beginning construction of test anxiety hierarchy (10

minutes)

5) Study Skills step I--see Study Skills Manual (15

minutes)

Session Two

1) Brief review of treatment goals and questions--homework

assignment discussed (10 minutes)

2) Open discussion of problems, e.g., difficulty in coming

up with success experiences (15 minutes)

3) Completion of hierarchy of relevant anxiety evoking

situations, also training in visual imagery (20 minutes)

4) Study Skills Package step II (15 minutes)

Sessions Three through Eight

1) Brief review of homework assignment and any problems--

additions to the hierarchy should be incorporated (10

minutes)

2) Instructions in the use of individualized coping imagery

(25 minutes)









3) Homework assignment made (5 minutes)

4) Group discussion (5 minutes)

5) Study Skills training (15 minutes)


Procedures to be Followed


Exploration of degree and extent of test anxiety.

This beginning exploration should serve to establish a

warm relationship between therapist and subject and also

allow subjects to become acquainted with one another.

This procedure should help the therapist learn the duration,

extent and the severity of the subject's test anxiety. In

addition to using the material contained in the self-report

anxiety questionnaires, the therapist should seek to deter-

mine whether there are other interpersonal or evaluative

situations that make the subject anxious.

Rationale

It is essential that all individuals involved in this

treatment program comprehend and accept what treatment will

involve. In an attempt to clarify both the principles and

procedures behind the treatment interventions used, a brief

overview should be provided. The following format should

be used. The experimenter should say the following to the

subject:

I'd like to give you a picture of
the rationale for this treatment to ex-
plain the concept of anxiety and how anx-
iety relates to the test taking situation
specifically. Then I will describe all
the facets of the treatment program. We
are working under the basic premise that









the present anxiety you feel taking tests--
and in any sort of evaluative situation--
probably is linked (has a lot to do with) to
your past experiences in these situations.
You have probably have had bad experiences
in taking tests and these unpleasant ex-
periences and memories may have been rein-
forced by hearing other people describe
their unpleasant experiences with tests.
Along with these negative test taking
experiences you may have come to believe
that tests and doing well on them is very
important. Getting into Med School or
Grad School may be uppermost in your mind
and makes tests very important. In fact,
your bad experiences with test taking
might only have served to support this
notion.
Thus your emotions and feelings sur-
rounding test taking and evaluative situa-
tions often result in feelings of anxiety
that are probably out of line with the
situation and interfere with your func-
tioning on tests. Since your perceptions
of these anxiety arousing situations take
place within you--inside your head--we
can work with these reactions and percep-
tions during these sessions.
Many times in the course of your life
you are called on to do things that make
you tense and anxious. Someone says to
you--'here, try my new moped' and you've
never been on one before. You do have
an idea how it works, but mostly you know
that you're a darn good bike rider and you
call on the skills you have as a bike rider
to help drive the moped. (Tell other
example) You come into a chemistry class
and you see the beakers and all the chemi-
cals and you're scared, thinking how am I
going to mix the right amounts and come up
with the right results, but then you re-
member that you did a very good job of
mixing paints and colors (for an art class)
so you called on these skills and you did
measure all the chemicals out, and you did
complete the experiment without blowing
up the laboratory. Everyday you call on
competencies that you have established in
other areas and use them in new ways in
areas that might originally be anxiety pro-
voking. What we are going to do in these
sessions to help you alleviate test anxiety
is something like that. We're going to









develop your ability to use your previously
established competencies to handle the problem
of test anxiety. We'll examine your test
anxiety--when it occurs, how often, what
makes it worse, what makes it better. We'll
make a special kind of list of all the test
taking situations that make you react with
anxiety--it's called a hierarchy. You'll
learn to cope with these test taking situa-
tions--each one of you will develop several
individualized coping images and you will
be taught how to use these coping images
in contending with these test taking situa-
tions. This will be accomplished by using
your imagination. We know that the nervous-
ness, the anxiety, the tension involved in
taking tests is the same as the tension you
have mastered in the other skills in which
you are competent. In this treatment, we
want to show you that test taking and the
handling of the anxiety associated with it
is just a series of skills like anything
else. As you see yourself coping and
succeeding with trying difficult situations,
you'll imagine a series of graded situa-
tions that have to do with taking tests and
the anxiety related to test taking. This
technique has been shown to be successful
in helping people--students, like yourselves--
who are test anxious. Do you have any
questions?
In addition to working with this pro-
cedure we will spend some time in each
session talking and working out ways to im-
prove study skills. While you all study--
most of you would probably like to study
more efficiently--get more out of the time
you put in. Of course, more efficient
preparation for these courses will probably
result in fewer feelings of tension and
apprehension prior to and during exams.
During the next eight weeks, as part
of the study skills training we will follow
this approach.
During our first session, we will dis-
cuss exactly what study is and present a
method found useful in helping you estimate
the time you need for accomplishing the
objectives set forth in your courses, and
several other techniques useful in deter-
mining whether you are using your time
efficiently.
Week two will be spent detailing a
method of shaping study behavior and









techniques useful in increasing motiva-
tion to study. Week three will be devoted
to presenting the SQ3R method (Scan, Ques-
tion, Read, Recite, Review Method) of how
to read a book. Week four will cover the
efficient taking down and use of lecture
notes. Week five will be spent discussing
specific techniques useful in studying for
an exam, as well as procedures helpful in
avoiding the emotional turmoil sometimes
felt as you enter the examination situation.
Week six will be devoted to getting more
study in less time by the application of
stimulus control and response management
techniques. Weeks seven and eight will
serve to review the entire study counseling
program.
For each session, we will spend a brief
period discussing any problems or successes
you had applying the principles and tech-
niques discussed and learned the previous
week, including the process you make,
your use of the study principles, etc.
During this time, feel free to discuss any
material which was unclear to you or prob-
lems with any of the above topics.

Open Discussion Period

An allotment of time should be made at both the begin-

ning and end of each session for discussion about any prob-

lems concerning treatment. In the course of treatment, if

any questions about the hierarchy arise, they can be answered

during the discussion session. A homework assignment should

be given at each session which in the first two sessions

would consist of the subjects' recording their success ex-

periences for use in the covert imagery procedure. Assign-

ments in the last six sessions will center on the practice

in coping imagery and gaining proficiency in study skills.

Development of Anxiety Hierarchy

The development of the anxiety hierarchy is one of

the most essential parts of this treatment. Anxiety evoking









situations related to taking tests should be organized to

extend from those evoking very small, manageable amounts

of anxiety to those that arouse the most. These items

making up the hierarchy should be drawn from the subject's

responses to self-report inventories and from subject sug-

gestions during the sessions.

Sample test anxiety hierarchy. The following is a

possible sample of a temporal spatial hierarchy that could

be employed for these test anxious subjects:

1) Studying material for a final exam two weeks before the

exam

2) Discussing approaching test with friends a week before

it is to be given

3) Listening to the professor describe what is going to

be on the exam a week before it is to be given

4) Remembering how you felt about a test you took in the

past

5) Studying for the test in your room the night before

6) Reviewing study materials the morning before the test

7) Eating, getting dressed, etc. just before leaving to

take the test

8) Walking over to the classroom on the morning of the test

9) Entering the room the day of the exam

10) Waiting for the test to be handed out--you hear other

students talking about how hard the test will probably be

11) Reading the first question on the test

12) Seeing that the test is longer than expected









13) Seeing a question that you can't answer

14) Seeing test questions on material that you didn't study

Any hierarchy that is developed should incorporate

all anxiety provoking elements relating to tests and test

taking that the group shares. This hierarchy should not

be excessively long with a maximum of 25 items contained

within in.

Training in Visual Imagery

Before work in using the individualized coping imagery

begins at the start of the third session, some preliminary

training in visual imagery should be conducted probably at

the end of the second session. A sample stimulus item to

imagine vividly could be offered such as:

Visualize yourself shopping in Publix or
Winn Dixie. Tell me what you see. Is
your image clear--is it black and white
or is it in color. Where are you in the
store--on the frozen food aisle, or near
the soda or cheese.

It might be useful to have subjects think of their images

as if they were watching a movie in which he/she is the

principle actor. It should be emphasized that they should

not see themselves as in a mirror image, rather as if they

are there--as active participants in all the action.

While some subjects may have no difficulty in imagin-

ing these stimulus scenes clearly, others may experience

difficulty. The therapist should emphasize that this

imagining is a skill, like relaxation, and improves with

practice. Before proceeding with the actual use of the

individualized coping imagery, the therapist should










present a few more nonanxiety evoking scenes to imagine,

e.g., eating an ice cream cone or lying on a beach.

Upon embarking upon the actual use of the individu-

alized coping imagery in the third session, subjects

should again be informed of what will occur during this

process in order to clear up any uncertainties the subjects

may have.

Individualized Coping Imagery Treatment

This group should receive the same study skills

training as the other groups. In this treatment group,

an attempt should be made to build an individualized

structure which employs the subject's perception of self-

competence as the primary coping technique. For this

treatment group, highly personalized previously estab-

lished images of competence and proficiency achieved by

the subjects in stressful environments other than test

taking should be paired with the visualization of anxiety

eliciting scenes that comprise the test anxiety hierarchy.

The first two sessions should be focused on a group

discussion of possible success experiences of the subjects

that can be transformed into images of competence for use

in this covert imagery procedure. There should be no

training in progressive relaxation as part of this treat-

ment. In sessions one and two, hierarchy construction

should be carried out in this group as it was in the other

three groups. In sessions three to eight the actual










treatment procedure should be carried out. Using these

previously established images of competence, subjects

should be presented with the test anxiety hierarchy items.

With each item subjects should be instructed to: 1) imag-

ine themselves coping with their tension by seeing them-

selves performing well in their personalized coping image

and 2) then to transpose the same competent and confi-

dent person of the personalized coping image to the im-

aginal anxiety provoking situation presented in the

hierarchy.

The subject should be instructed to imagine this

same competent person of his individualized coping image

now contending with test situations as they appear in the

hierarchy.

In the individualized coping imagery treatment,

each item on the hierarchy should be presented for a

minimum of two times. The length of item presentation

would either be 60 seconds or the amount of time for all

individuals in the group to imagine the item for a 20

second anxiety free period. However, if a subject signals

anxiety after the second exposure, the items should be

repeated until no subject signalled anxiety when imagin-

ing this item. In order to control for the divergent

elements of the two treatment procedures, the same

hierarchy of anxiety evoking scenes should be






90


employed in all variations of individualized coping imagery

treatment.















APPENDIX C
INSTRUCTIONS TO THE EXPERIMENTER:
INDIVIDUALIZED COPING IMAGERY TREATMENT
COMBINED WITH RELAXATION


This is the manual for the treatment of test anxiety

through the use of individualized coping imagery combined

with relaxation and study skills. This treatment is de-

signed to build an individualized structure which employs

the individual's perception of self-competence as the pri-

mary coping technique. In addition, relaxation training is

employed to determine if it enhances treatment effects.

Seven steps are involved in the use of this technique:

1) Exploration of degree and extent of test anxiety of

subject

2) Presentation of rationale for treatment

3) Group discussion of possible success experiences of

subject that can be transformed into images of competence

for use in this covert imagery procedure. This step also

includes discussion of treatment goals and problems.

4) Construction of test anxiety hierarchy

5) Progressive relaxation training

6) Employment of covert imagery treatment

7) Study skills training

With the exception of the use of progressive relaxation

training, all aspects of this treatment are identical to









that of individualized coping imagery treatment without

relaxation (see Appendix B). The progressive relaxation

instructions which follow were adapted from Susskind (1970)

to be used in combination with individualized coping imagery

treatment.

Instructions in relaxation. The room in which instruc-

tion in progressive relaxation takes place should be at a

pleasant even temperature. Before relaxation begins, the

lights should be dimmed, subjects should be seated in com-

fortable chairs. Subjects should then be asked to close

their eyes. The Muscle-Relaxant Method is as follows:

Please lean back in your chairs. Place
both feet on the floor. Close your eyes.

Now, stretch your legs as far as they
can go. Turn your toes under and tighten
the muscles, very, very tight. Hold it. And
now also tighten the muscle in your calves
and those in your thighs. Make your entire
leg tight as a drum, and hold it, hold it,
hold it. And now, let your legs sink down
and relax all the muscles in your calves,
all the muscles in your thighs. Let your
legs go completely limp. And now, feel
that wonderful relaxation coming up from your
toes, up your calves, and your thighs. Feel-
ing wonderfully relaxed, beautifully relaxed,
very calm, very relaxed. Feeling beautiful,
just beautiful, wonderfully relaxed.

Now I want you to stretch out your hands.
Make a fist, Feel the tightness, and now
make it tighter, tighter, tighter. Hold it.
And now also tighten the muscles in your
wrist, in your forearm, in your upper arm.
Hold it. Hold it. And now, let go, just let
go, let your arms sink down and get that
wonderful feeling of relaxation, right through
your fingers, your hands, now through your
forearm, and upper arm. Let your arms go
completely limp. Feeling wonderfully relaxed,
beautifully relaxed, very calm, very relaxed
and beautiful, just beautiful. Now, I want
you to arch your back backwards, raise your









chest, and tighten your stomach muscles,
and those in your neck. Make them as tight
as you can, tighter, tighter, tighter.
Hold it, hold it, hold it. And now, let
go, just let go, and you get that wonderful
feeling of relaxation. Just feel the muscles
relax from your back, from your neck, from
your chest, from your stomach, all over your
back, all your muscles are feeling wonder-
fully relaxed.

And now, I want you to tighten the mus-
cles in your face, around your mouth, the
muscles in your chin, around your eyes and
your forehead. Make them tighter, tighter,
tighter, hold it, hold it, hold it. And
now, let go, just let go, let go and get that
wonderful feeling of relaxation, from all
the muscles in your forehead, the muscles
around your eyes, the muscles of your cheek,
the muscles of your chin and the muscles
around your mouth, feeling wonderfully relaxed,
beautifully relaxed, very calm, very relaxed,
wonderfully relaxed.

Now, I want you to take a very deep breath
and hold it, hold it, hold it. Now, slowly,
let it out, and you're letting out all your
tensions, your frustrations, your anxieties,
feeling wonderfully well, wonderfully well,
wonderfully well.

Once again, take a deep breath, a very deep
breath, hold it, hold it, hold it. Now, slowly,
slowly, let it out, relax your tensions, your
frustrations, your anxieties, feeling wonder-
fully well, wonderfully well, wonderfully well.

Now as I count down from 10 to 1, think
of the scene that makes you feel calm, that
makes you relaxed, and that gives you a feel-
ing of well-being.
(pause 10 seconds)
Now, with your eyes closed, see that scene, in
all its details, and as I'm counting down from
10 to 1, you are going to find yourself deeper
and deeper relaxed and you will have a feeling
of well-being. Calm and relaxed, and wonder-
fully well, just relax.

I'm going to count, 10...., 9...., very deep,
8...., 7...., deeply relaxed, 6...., 5...., very,
very, deep, deeply relaxed, 4...., 3...., very




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

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - Version 2.9.9 - mvs