(warms

 

2W7
lfflrltiman24'5
5936383 em

 

SCME EFFECTS OF STRESS ON ABSTMCT AND CONCRETE BEHAVIOR

by

Jacques M. Levy

AN ABSTRACT

Submitted to the School of Graduate Studies of Michigan State
University of Agriculture and Applied Science in
partial fulfillment of the requirements for
the degree of

MASTER OF ARTS

Department of Psycholog

1958

Approved by:

 

ABSTRACT

From a theoretical viewpoint which proposes that individuals
tend to cope with inadequate conditions by becoming more concrete,
two hypotheses were formulated with regard to the effects of stress.
It was expected that if individuals were given concrete an! abstract
tasks, i.e. tasks which activate these differential performances,
then the subjects under stress would find channels for the expression
of their coping behavior in the concrete task. So then, it was pre-
dicted that:

1. Stress would facilitate performance on a concrete tasks;

2. Stress would impair performance on an abstract task.

Thirty male students served as subjects for this study, fifteen
in the emerimental group and fifteen in the control group. Each 8
was given three pairs of block-design tasks to solve (one pair being
a sample, and therefore not bearing upon the results). Each pair
consisted of an abstract task (Task 1) and a concrete task (Task C).
All subjects took the sample pair and another pair (pro-Tasks A and C),
under normal conditions. Then, half the subjects were placed under
stressor conditions (introduction of false norms) and the other half
waited for a comparable time period. Subsequently, all of the subjects
took the third pair of tasks (post-Tasks A and C). Effectiveness of
the stressor conditions was ascertained by introspective data.

Using analysis of covariance technique, the subjects were all

equated on pro-Tasks A and C. The comparisons made were between groups

on both post-Task A and post-Task C. No significant differences
were found between groups on either task. Since the introspective
data seemed to indicate that the stressor conditions were effective,
the results did not substantiate either of the hypotheses.

Some possible reasons for this were advanced. The most tenable
of these seemed to be that the tasks themselves were not valid. The-
oretically, performance on concrete tasks should show no relationship
to performance on abstract tasks. Correlations between pro-Task A
and pro-Task C and between post-Task A and post-Task C for the exper-
imental and control groups indicated that one can predict, with a fair
amount of accuracy, the performance on one kind of task from the per-
formance on the other. Since this violated an essential assumption
of the study, 1.6. that abstract and concrete behaviors were being
discretely tapped, possibilities for future research on this same topic,

which would eliminate this error, were offered.

SOME EFFECTS OF STRESS ON ABSTRACT AND CONCRETE BEHAVIOR

By

Jacques M. Levy

A THESIS

Submitted to the School of Graduate Studies of Michigan State
University of Agriculture and Applied Science in
partial fulfillment of the requirements for
the degree of

MASTER OF.ARTS

Department of Psychology

1958

 

 

*-

from:

to Kurt Goldstein

Always the beautiful answer
who asks the more beautiful question *

e.e. cummings, Poems 1923-195};

ACKNGVLEDGEAENTS

The author wishes to express his sincere gratitude and appre—
ciation to Dr. Albert I Rabin, who, as both Friend and teacher, has
given consistent help and understanding in the two years that we
have known each other; and who, in his capacity as chairman of this
thesis committee, afforded the fullest measure of freedm possible.

Many thanks are also due the other numbers of the committee: Dr.
Charles Henley was of immeasurable aid in the design and statistical
analysis of this study, and Dr. Hans H. Toch contributed very helpful
observations and suggestions concerning the theoretical issues herein.

The author further wishes to thank Miss Kay Royer, who reproduced
the materials used, and Miss Gloria L. Cheek, who typed the manuscript
in her usual competent and responsible fashion.

But, as always, the greatest measure of gratitude must go to the
students vho served as subjects, and especially to those who were mem-
bers of the Experimental group. These latter encountered a situation
not of their own choosing, with regard to which the author would like,
once again, to extend his sincere apologies, and also his appreciation
of their cooperation.

TABLE OF CONTENTS

I. Introduction:

A.
B.
C.
D.
E.
F.

G.

The abstract and concrete attitudes..............
Coping with the world............................
The definition.of stress.........................
Stressor conditions..............................
History..........................................
The nature of the task...........................

HypothGBBS.......................o...............

II. materials eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeooeeeeeeewo

III. Subjects and procedure...............................

IV. Results;eeoeoeoeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee

V. .DiscuSSioneeeeeeeeeeeeeeeeeeeeeeeeeeeeeoeeeeeeeeeeeee

VI. Summary and Conclusions..............................

VII. Bibliography.........................................

PAGE

\6 \o (n to k4

15
17
19

26
32
36
37

c c o
o o I
IV e a
o I O
s a e
e e s

Table 1:
Table 1A:

Table 2:

Table 2A:

Table 3:

Table h:

LIST OF TABLES

PAGE
Comparison between groups on abstract task...... 29
Mean solution times for’Task A.................. 29
Comparison between groups on concrete task...... 30
Mean solution times for Task C.................. 30

Comparisons between groups on intrOSpective

dataOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOCOOOOOOO... 31

Correlations between Task A and Task C fer each
group in eaCh oonditlon......................... 31

soon...

see.

' D

I. INTRODUCTION
A. The Abstract and Concrete Attitudes.

Gelb and Goldstein (17), from their experiences with brain-
injured patients, first made the distinction between two modes of
behavior which they called "abstract” and “concrete“. Tears later,
Goldstein and Scheerer (21) more explicitly studied and described
these. They wrote:

The normal person is capable of assuming both,
whereas the abnormal individual is confined to but one
type of behavior - the concrete. The abstract and con-
crete behaviors are dependent upon two corresponding
attitudes vhich are psychologically so basic that one
may Speak of them as almost levels.

The abstract and concrete attitudes are not ac—
quired mental sets or habits of an indivimal, or spe-
cial isolable aptitudes, such as memory, attention, etc.
They are rather capacity levels of the total personali .
Each one furnishes the basIs Tor-'a'lI—p'eFformances per.»1
taining to a specific plans of activity. In other words,
each attitude constitutes one definite behavioral range
which involves a number of performances and responses.
These latter, when taken individually at their surface
value, may appear to be discrete entities of quite a di-
versified nature (e.g. attention, recall, retention, re-
cognition, synthesising, symbolisation, etc.). Closer
analysis reveals that these seemingly diverse performances
and responses have as a cannon basis 293 functional level
of integration, one cross-sectional attitude of the per-
sonality: either the concrete or the abstract attitude...

...Our concept of attitude implies a ca acit level
of the total personality in a specific plus of activity.

This plane can be related to the outerworld situation
or to an inner experience. One can assume either an ab-
stract or a concrete attitude towards the outerworld as
well as towards an imer experience. In assuming the one
or the other, the individual as a whole gears himself to.
a specific direction of activity which we call abstract
or concrete behavior. (p. 1f).

Goldstein (20) offers further explication of these concepts:

The concrete attitude is realistic. In this atti-
tude, we are given over and bound to the immediate exper-
ience of the given thing or situation in its particular
uniqueness. Our thinking and acting are directed by the
immediate claims made by one particular aspect of the ob-
ject or situation in the environment.

In the abstract attitude, we transgress the ined-
iately given specific aspect or sense impression; we ab-
stract from particular properties. We are oriented in our
action by a more conceptual viewpoint...We detach ourselves
from the given impression...The abstract attitude is basic
for the abilities:

1. To assume a mental set voluntarily.

2. To shift voluntarily from one aspect of the situa—

tion to another.

3. To keep in mind simultaneously various aspects.

1;. To grasp the essentials of a given whole; to

break up a given whole into parts and to isolate
them voluntarily; to synthesise.

5. To generalize; to abstract cannon properties; to

plan ahead ideationally; to assume an attitude
toward the 'msre possible,’ and to think or per-

form symbolically.
6. To detach our ego from the outer world. (p. 18f)

Because, in his writings, Goldstein sometimes speaks of a 'lack
of abstract attitude" and sometimes of ”impairment" of it (l9,pp.66-
67) there has been a great deal of confusion as to vhether this is posited
as an all-or-none phenommon. The difficulty arises from the fact that
Goldstein is sometimes not rigorous in his terminology. A general survey,
however, leads to the conclusion that although he does speak of patients
in which there is a ooglets loss of this capacity, he also conceives of
this loss in terms of greater or less extent, i.e. impairment. At the
same time, abstract and concrete are not invariable in themselves. There
are gradations of each: We have to differentiate between various degrees

of both the concrete and the abstract behavior.” (21, p. 7).

B. Coping with the World.

The normal organism under adequate conditions shows behavior
which is ”characterised by an alternation between an attitude in-
volving abstract and one involving concrete behavior, and this a1-
ternation is appropriate to the situation and the individual, to
the task for which the organism is set." (19, p. 66). One can only
understand the behavior of an individual when one takes into account
the meaning of the performance for the organism, for 'those perfor—
mances are always fulfilled which are most important for the organise."
(18, p. 202). This, however, ”presupposes a normal, adequate envir-
onment. Because these conditions are not always fulfilled, even in
normal life, the orgmism might often appear to be governed transi-
torily by a special tendency." (ibid., p. 202). In such situations,
when the conditions are inadequate for the person, '...normal reac-
tions...no longer take place. Under these coalitions, stimulus util-
izations may occur which have the characteristics of isolation.“
(ibid. p. 322). Looked at broadly, '...the organism also tends pri-
marily to avoid catastrophes, and to maintain a certain state which
makes that possible...the only form of self-actualisation which re-
mains is the mintsinence of the existent state.‘l (ibid., p. 197).

The important ideas here is that the organism's actualisation
_t__sk_gg £113 for! of self-maintainence. In order to utilise his capa-
cities to their fullest extent, the person must be presented with

channels in the environmmt for this to take place. If tlmse chan-
nels are not present (or are not perceived), then the person is
"driven to utilize his preserved capacitiu in the best possible
way," (ibid., p. 197) or he is faced with catastrophe, the disin-
tegration of the orgmima.

In an attempt to define what is meant by adequacy and inade-
quacy of conditions, a general formulation would be that a situation
is inadequate for a person when he experiences a threat to his exis-
tence, due to his subjective experience of being unable to cape with
a task to which he has addressed himself. The result of such a con-
dition would be anxiety usually, but not necessarily. That is, the
bran injured patient will one into anxiety because his capacities
are so much impaired in general that an incapacity to perform does
mean some threat to his existence. However,

"in a normal person, this will not always be the case.

He will cope with a difficult situation in some fashion
without becomirg actually endangered in his existence.
...Yet, however true it may be that difficult situations
do not involve severe anxiety, the concomitant mmtal
state is nevertheless structurally of the same type. The
only reason they do not impress us as anxiety, is because
they are less relevant to the total personality and its
existence...(However) a failure, which by itself is irre-
levant, may bring anxiety into distinct prominence, if it
occurs in a situation in which it becomes significant for
the very existence of the person concerned; for example,
anxiety during an examination." (ibid., p. 3021).

In coping with the difficult situation, the person's reactions
will, as was stated above, have the characteristics of isolation.
These are: dedifferentiation, abnormal stimulus bondage, abnormal

dependency upon the quality of the stimulus, forced responsiveness,

etc. - in short, concrete behavior. 50 then, when the individual
feels threatened, he moves to a concrete mode in order to cope
with the inadequate conditions.

This is quite similar to Rogers' view of the process of defense,
wherein he states that defense against threat keeps 'the total per-
ception of the experience cmsistent with the individual's self-
structure,... The general consequence of the process of defense,
aside from its preservation of the above consistencies, are: rigidity
of perception..." (38, p. 62). Arieti (b) has, in the same way,
characterised schizophrenia as '... a reaction to a severe state of
anxiety...(which) consists in the teleologic or motivational use
of a more or less advanced impairment of the abstract attitude....
(This impairment) has the purposes of avoiding anxiety at the same
time that it permits the uniqueness of the individual to surge."
(pp. 316-317).

Experimentally, Beier (5) found just this relationship between
avoidance of anxiety and impairment of abstract attitude. Using
Goldstein's theory and fornmlatims, he placed a group under stress
(by reading a faked Rorschach report to each subject) and found a
definite impairment of abstract behavior resulted. But, the question
remains as to what the nature of these subjects' behavior was. The
only channel cpen to the subject for expression of his responses was
the abstract task. Since their performance on this was impaired, i.e.,

they became more concrete, it would seem fairly safe to assume that

C

OI

“OQ‘

 

they did feel threatened. But, at the same time, none of the sub-
jects was reported to have entered into a catastrophic reaction.
They did cope with the situation, i.e. they were maintaining a state
of austence in order to avoid catastrophe. But the coping behavior
was not afforded a channel for expression, and therefore a seenirgly
important variable was overlooked.

In most of the studies reported in the literature, a similar
procedure is followed, and so this coping behavior is not experi-
mentally isolated. In the present study, the individual is confron-
ted with bgth an abstract and a concrete task. This is done in order
to facilitate the channeling, of the coping behavior in a measurable
way. In this manner, it was hoped that when the situation was made
inadequate (i.e. threat introduced), the subject would become more
concrete; and this would be reflected in his performance on the con-
crete task. The characterization of a task as concrete will be taken

up in more detail in the section on Materials.

C. The Definition of Stress.

Before summarising the relevant literature which bears specifi-
cally upon the effects of psychological stress on performance, the
writer would like to present some of the attempts at defining stress
as a variable.

Grinker and Spiegel (22) consider psychological stress to be

strictly analogous to the physical concept of stress. They speak of

physical stress as the restorative forces exerted by an elastic back
when an external force deforms that body. This homeostatic function
of physiological mechanisms was more elaborated by Selye (142), to
defined physiological stress as aw condition that produces the ”ad-
aptation-syndrome" - which is the reaction of the organism in return-
ing to the homeostatic state. Chiles (9) smarises the definition
of Darrow and Henry and of Haggard, who define stress as a condition
under which the subject or individual reacts with responses or actions
which are out of proportion or inappropriate to the circumstances in
which he is placed. Chiles further presents the following authors'
treatments of stress:

1. Katchmar defines it thusly: 'To sumarise...we
can say that stress is an internal process of the organism,
manifested as an equilibrium seeking response, occurring
in the psychological context when the objective situation
is cognitively evaluated as one involving a goal, the at-
tainment of which is thwarted, or interpreted as being
thwarted. '

2. Katchmar states that Lazarus and Deese (in an un-
published study) consider the basis of stress to be (a)
the thwarting of a motive (in the clinical or psychoanaly-
tical sense); (b) the thwarting, or more specifically the
threat of thwarting, must be recognised by the organism;
thus previous experience (cogrdtions, habits) plans an im-
portant role in the stress situations; (c) a third class
of variables is that of the affective aspects of stress.

3. Bass, Harder and Ellis state: 'In order to define
stress adequately it is necessary to consider the inter-
action of stimulating conditions, organismic ability and
organismic motivation. One way of doing this is to con-
ceive stress in terms of the cost of performance to the
organism... ' (measured by) energ expenditure.

h. manhunt” the position that stress is the
disruption of homeostasis , but it is not c1ear...whether
he is referring to psychological or physiological...or
both. Miller states: 'As a working definition, we may
say that stress is any marked increase or decrease in some

characteristic of the environment which affects the

individual. This may be a very strong stimulation or

threat...’ (ibid, p. h)

Lazarus, Deese and Osler (27) conceive of stress as an inter-
vening variable. Stress occurs when a particular situation threatens
the attainment of some goal. Schaffer (hl) proposes that stress
should be considered as the 'functional decortication' of the organ-
ism. i stressful situation is regarded as essentially one in which
a major disruption of the relationship between an organism and its
environment has taken place.

Amidst the man treatments of this concept, wherein much overlap
is present, a statement by Lazarus, Deese and Osler (op. cit.) seems
to summarise quite appropriately the extant situations "The psyche-
logist has no adequate way of defining the psychological condition
that corresponds to the homeostatic steady state. Consequently, the
use of the term 9-2755 must necessarily be a little looser than we
would like it to be.“ (p. 295).

Without wishing to become side-tracked on the issue of whether
the normal organism is, in fact, governed by a tendency toward hmeo-
stasis and tension-reduction, this writer will sun-arise what, in this
study, the characteristics of psychological stress were taken to be.
They are: that the person feels (not necessarily in symbolized aware-
uses) a threat to his existence; that he has a subjective experience
of being unable to cope with a task to which he has addressed himself;

that he perceives himself as being negatively evaluated; and that the

foregoing may find expression in symbolized feelings of tenseness,

sadness, and nervousness.
D. Stressor Conditions.

We now turn to the means of inducing stress. Following Vogel,
Baker, and Lazarus (h?) we shall call these the stressor conditions.
”By stressor conditions we mean those experimental conditions which
are intended by the investigator to be stressful, but which are not
necessarily responded to as such by the S. The term ‘stress' we use
to refer to a state of the organism, not a class of conditions.‘I (p.
105). The various techniques which have been used fall into two
main classes (27). These are: (a) stress induced by failure, and
(b) stress induced by working conditions and the task itself. In the
former category, the more frequently used, three mthods have been
employed: presenting the abject with an unsolvable task, interrupt-
ing the subject before he has finished, and the introduction of false
norms which indicate failure even if the performance has been adequate.
In the latter category, the methods most often used, although the var-
ieties are infinite, have been: distraction, physical discomfort, ver-
bal disparagement of the performance by the experimenter, requiring
the subject to attend to too many things or to perform too many oper-

ations at once, rapid pacing, etc.
E. HiBtOI'ye

Let us proceed, then, to have a look at some of the studies that

10

have appeared in the literature. In their review of the theoretical
issues and experimental results concerned with the effects of psycho-
logical stress upon performance, Lazarus, Deese, and Osler (op. cit.)
present studies which "have yielded varying results: some show only

a decrement in performance, others show improvement, and still others
produce both of these effects for different individuals (an increase in
variability)." (p. 310). Three years later the situation apparently
had not changed, for Farber (lS) reviewed the current literature on the
role" of motivation in verbal learning and performance and stated:

Experimental studies have usually tended to show that
when failure instructions are not applied to specific res-
ponses, but are used, instead, to characterize general per-
fo rmance, subsequent performance in verbal learning situa-
tions is impaired. As a result of such observations, it is
generally believed that statemnts emressing approval of a
learner's performance are more beneficial to his subsequent
performance than statements indicating disapproval. At the
same time, however, it is also rather generally believed that
excessive praise may decrease motivation and thus impair per-
formance, whereas some kinds or degrees of censure may increase
motivation and thus improve performance. This indecision with
reSpect to the effects of failure characterizes treatments of
the effects of frustrating events in general. (p. 316).

On this same issue, Child and Waterhouse (8), from a more theore-
tical vantage point, wrote:

Indeed, the conflict appears strikingly in some general
textbooks in psychology. In a chapter on thinking and reason-
ing, frustration is viewed as the coalition for more organized
behavior, and in a chapter on emotion, it is viewed as the con-
dition for less organized behavior. (p. 127).

Frustration may operate to increase the motivation suppor-
ting the goal-oriented activity and thereby to improve the qual-
ity of performance (p. 136),...(but) severe emotion does often
have a disorganizing effect and thus reduces the quality of
performance in the face of frustration. In part this may be

(W

 

QCOQ

IV

1‘

(’3

11

because the emotional responses themselves are to some

extent incompatible with the ongoing instrumental act-

ivity. (p. 137)

We shall first present some ,of the studies wlmrein an increment
in performance was observed following stress. Lazarus and Eriksen
(28) found that stress produced an increase in the number of errors
on the digit symbol test, but this was compensated for by a concom-
itant increase in speed. Since, however, in this task speed scores
accounted for 92% of the variance in total scores, lowered accuracy
mattered very little for it was so poorly correlated with total scores.
The authors concluded, in part, that “frequently subjects do well under
stress by increasing their spew at the expense of accuracy.“ (p. 1014).
Deese and Bowon (13) obtained initial decrement but later improveamnt
in code-learning as the result of stress due to pacing. However, Hur-
lock (23), using failure, found just the opposite relationship. Clark
and McClelland (10) reported that subjects with mild fear of mm.
performed better, particularly towards the end of a test. Using motor
coordination and color naming tasks, Gates and Rissland (16) found im-
proved performance under failure-stress. .

Increments in performance after stress have often been interpreted
in light of the energizing aspects of motivation. Lazarus, Deese and
. Color (27 ) quote two studies wherein an increase in motivation, brought
about through stress, was accompanied by 31 increased output in perfor-
mance. "Miller and Wickert both suggest that fear, produced in a stress-

ful situation, may actually be beneficial to performance. They cite as

12

evidence the fact that ALF personnel reported that under the stress

of combat they performed more efficiently.” (p. 310). Animal studies
have been corducted by Miller (31;), Brown and Jacobs (7), and Amsel
(2), nong others, to show that anxiety does exhibit both the rein-
forcing and energizing properties of a drive. Experiments by Brown
(6), and by Rosenbaum (to) indicated that increases in the hunger
drive result in the gemral elevation of performance measures of stim-
ulus generalization. Rosenbaum (39) found this same elevation as a
result of experimentally induced anxiety. Using motor responses to
visual stimuli, he concluded that "experimentally induced anxiety shows
the energizing function of a drive..." (p. ’42). The role of electric
shock as a reinfa‘cer is well known, and its generalizing effect has
been demonstrated in numerous studies (e.g., 3, 36, 1:3, hh).

m the other hand, the literature is replete with studies which
indicate decraents in performance due to stress. Earlier (p. 8), the
study by Beier was mentioned. Cowen (11) found that stress elicited
increasing amounts of problem-solving rigidity on the Luchins water-
jar problem. Using a similar technique, the same author (12) success-
fully replicated his results and further showed that praise produces an
”extinction” of rigidity. Osler (37) applied failure-stress and found
depressed performances on various intellectual tasks (arithmetic, non-
sense syllables). llontague (35) found that increase in anxiety caused

anxious Ss to perform less well on a difficult list of nonsense syllables

13

to be learned. A similar study by Lucas (31) arrived at the same
conclusion. Deese, Lazarus, and Keenan (114) found that low anxious
Ss suffered a decrement in nonsense syllable learning due to stress,
and in a repeat of the study, varying the difficulty of the list of
syllables, both high and low anxious Ss showed the decrement (26).
Using nine-year old boys, Lents (214) obtal ned impairment of Stanford-
Binet scores following failure. Further examination by Lents of the
various subtests indicated that tadcs requiring visual or rote memory
were not affected, while those involving reasoning or thinking were.
Decrements following stress have been found in production in a senten-
ce-formation test (1), learning in arithmetic problems (115), digit
span on fifth-and sixth-grade children (1:9), recall arri relearning of
nonsense syllables (50), digit symbol (h8), visual discrimination (32),
card sorting (33), reaction times for completing pictures (h6), and
many others.

Many of the authors do not provide a clear differentiation among
concepts such as frustration, emotion, motivation, drive, stress, and
the like. Apparently, howeVer, such differentiations are implicit in
their usage of the terms. It would not seem productive to attmnpt, in
this paper, to provide an adequate theoretical model concerning this
issue. Let it suffice to say that these terms are usually used in two
groupings: on the one hand, frustratiar, stress, and motion are thought

of as short-term, indmed states of the organism, while on the other hand,

1h

motivation and drive are considered as longer-term organismic
qualities. The induced states affect these long-term qualities
by either raising or lowering their levels. That is, an organin
is not always frustrated, stressed, or emotional, but is always
motivated or driven to a greater or lesser extent. The concept
of "learned drives” (314) would not easily fit into the two group-
ings offered here, but this writer tends to feel that drive, in
this case, is a misnomer, and might better be referred to as frus-
tration or stress.

More recently, Lazarus and various co-authors have begun to
advance more sophisticated hypotheses. Interaction effects are
being sought after in their fairly complex experimental designs.
Vogel, Baker, and Lazarus (It?) report a study that appears to be
first in what promises to be a long line of experiments attmpting
to tease out the relationdlip between stress and motivation. Using
a perceptual-motor tad: and accompanying physiological reactivity as
dependent variables from which behavior under stress is obtained, the
authors proceeded to manipulate motive states in two ways: through
induction by use of ego-involving its tructions; and through infer-
ences from TAT material, academic accomplishment, and study habits.
Although the results of the study are, for the most part, not sig-
nificant, they did obtain some interesting interactions between in-

duced motivation and physiological reactivity. Their conclusion:

15

"The findings of this experiment indicate the complexity of the
role of motivation in determining response to stressor conditions
and highlight the methodological difficulties involved in the ...
investigation." (p. 111).

This experimentation involving the systematic use of motive
states as the independent variable is a product of two earlier papers
by Lazarus and Baker (214, 25) which presented a theoretical arr! meth-
odological framework for the studies to come. Their conception is
that the motivational state of the irflividual in relation to the
stressor conditions imposed upon him determines whether or not a

state of stress occurs.
F. The Nature of the Task.

It is truly surprising that, with all of the work that has been
done in this area, there is none reported wherein a controlled study
of the relationship between stress and the nature of the task has been
made. In the early review (27), under a section irich outlined what
appeared to be the six most obvious questions for future research, the
nature of the task was listed first.

Farber '(15) too concluded, in part, that "just as the associative
mechanisms attending failure may either benefit or harm performance,
so might an increase in drive, in and of itself, affect behavior either
favorably or adversely, depending on the; flecific patina; of; Erie Basic

and the experimental conditions involved." (p. 317, underlining ours).

16

Child and Waterhouse (8) had said practically the same thing when
they concluded that: "The extent to shich heightened drive can lead
to improved performance, and the extent to which other responses are
incompatible and produce interference, may vary with the exact nature
of the task or activity in which quality of performance is being jud-
ged.” (p. 137). And, for Lazarus, Deese, and Hamilton (26) along with
Osler (37), the exact nature of the task appeared to be an extremely
important variable which they had not taken into consideration.

It, is just this variable which the present study d_o_e_s take account
of, although not in and for itself, but in order to be able to make in-
ferences about the kinds of performnces which are needed to cope with
the task presented. It is extremely difficult to extract from the many
studies quoted above just what differences, if any, in performance are
related to the nature of the task. Broadly conceived, however, there
does seem to be a consistency that emerges at times - when performance
of a task requires the subject to function at a high level of abstrac-
tion (e.g. water-jar problem, complicated memory tasks, difficult arith-
metic), then stress impairs performance; on the other hand, when the
task requires only the lowest level of concrete functioning (e.g. simple
motor tasks, conditioning of part processes, color naming, animal be-
havior, )then stress facilitates performance.

Strictly Speaking, the present study is focused upon the function-
ing of the individual in terms of the requirements of the task, rather

than on the nature of the task, per as. It is not quite accurate to

,‘I

 

17

define a task as concrete or abstract. It is the individual's
approach and behavior which‘can be so defined. But, it is possi-

ble to "devise Special tests in which we can ascertain unequivocally
the corresponding approach required for each task; i.e., the solu-
tion must presuppose a definite attitude and the experimental arran-
gement preclude Success by means of any other @proach...Tasks which
presuppose the activation of the abstract performance-level cannot

be genuinely solved as long as the subject operates within the con-
fines of the concrete level." (21, p. 22). It is of utmost importance
that this not be misunderstood. The differences betweer abstract and
concrete attitudes is NOI‘ merely a difference in difficulty or com-
plexity. ”The greater difficulty connected with the abstract approach
is not simply one of greater complexity, measured by the number of sep-
arate, subservient functions involved. It denands ratler the behavior
of the new, emergent quality, generically different from the concrete.”

(ibid, p. 22).
G. Hypotheses

So then, what we shall do is present each subject with two tasks —
one which presupposes the activation of the abstract approach, and one
which presupposes the activation of the concrete approach. Since, as
we have indicated, iniividuals tend to cope with. inadequate conditions
(stress) by maintaining their existent state through becoming more con-

crete, our erqaectations are that this will be reflected in their differ-

18

ential performance on the two kinds of tasks.

That is, we hypothesize that, under stressful conditions:

1. Performance on a task which requires an abstract approach
will become impaired, i.e. a decrement will be seen;

2. Performance on a task which requires a concrete approach will

be facilitated, i.e. an increment will be seen.

19
II. MATERIALS

The task of the subject was to reproduce colored designs with
blocks. Both the designs and the blocks are those used in the Gold-
stein-Scheerer Cube Test (21). Two kinds of reprodrctions were re-
quired of the subject.-

1. A colored design was placed before him, and along with it
was placed a modification of this design. This modification is the
exact size of a four-block model (approximately four inches 2: four
inches), whereas the original colored design is the size of one block;
on the face of the modification two intersecting lines are drawn, which
divide the design into four equal sections. Both the original and the
modified designs face in the same direction (one higher than the other),
and aside from the afore-mentioned changes, they are identical in color,
shape, and proportion. The subject's task was to build a model using
four blocks which will reproduce the design shown to hill.

2. Only the original colored design was placed before the subject.
His task was to build a model again - but, this time there are three
important changes: (a) the model was to be built using sixteen blocks;
(b) the model was to be built rotated 90 degrees, one full turn, either
to the left or to the right - it could not be built first and then rota-
ted; (c) the model was to be built with the original colors inter-
changed, i.e. where there was red on the original, white was to be put,

and vice versa.

20

The first reproduction (four-block) will be referred to as
Task C, standing for concrete. The presentation of the modified
design follows the procedure of the Goldstein-Scheerer Cube Test.
In their test, they have set up sixlevels on an abstract-concrete
scale. The presentation used in.our'study is at step four (one being
most abstract, six being most concrete). These authors state that
"the experimental.findings on this test suggest that there are degrees
of lesser and stronger concreteaess in a subject, which correspondp
ingly require more or less concrete aids. A subject who needs only
facilitation with regard to the difference in.size (step two) is in
all probability less concrete than a subject who fails with this help
and requires the line division (step four).“ (ibid, p. 57). These mod-
ifications of the presented.designs consist of stages of diminiShing
discrepancy between the design to be copied and the block model to be
composed. In summary,

The modifications introdmed represent aids of a concrete,

palpable nature; they supplant in the phenomenal realm.of

perception the lack of abstract approach which the subject

does not or cannot assume at that moment. It would be ab-

solutely erroneous to suppose that these aids help to ini-

tiate a process of abstraction in an individual who lacks

the prerequisites for that act. These aids simply render
that act unnecessary. (ibid., p. 55, underlining ours):—

 

So then, if what has been said up to now is actually the case,
there is every reason to believe that Task C fulfills the requirements
as a task which presupposes the activation of the ccncrete approach

mainly. Putting it another way, this task is a simple, copying task

21

and as such does not require conceptualization for attainment of
solution.

The second reproduction (sixteen-block, rotated, colors inter-
changed) will be referred to as Task A, standing for abstract. Due
to the demands which this task makes upon the subject, it would ap—
pear to have face validity as a high-level performnce. The subject
is continually faced with the necessity of mnipuleting three sets
(size, position, color) at one time, and he has no model from which
to copy except his own cmceptual one. A series of these tasks has
been used successfully to tap the functions of analysis and synthe-
sis (29), both of which are intrinsically related to the abstract
attitude.

Task A and Task 0 comprise a pair of block designs which attenpt
to make available to the subject channels through which he can express
both abstract and concrete approaches to problems. In the present

study, three such pairs were used, one of them being a sample.

22
III. SUBJECTS AND PROCEDURE

Thirty male, undergraduate students at Michigan State Univerb
sity were the subjects in this study. All volunteered, and each
subject was seen individually, the experimental sessions lasting
from twenty minutes to one hour. The subjects had no previous knows
ledge about the experiment, except that it had “something to do with
problem-solving.” The same experimenter stamined all of the subjects.

It is well known.that wide variation exists among individuals in
ability to perform on block design tasks. In order to isolate the
treatment effects an analysis of covariance design was used. (30).

This would allow for the extractiOn of variance due to "block design
ability" by equating all subjects on a.pre-test. In.this way, what-
ever intellectual and.personality factors go into muting up this abil-
ity would be controlled for. No other controls were placed upon the
subjects. Selection of subjects for-both Control and Experimental
groups was done in a fortuitous manner; nevertheless, before am sib-
ject was seen, he had been assigned to one or the other group.

All subjects received three pairs of tasks. First, a fairly simple
sample pair was presented to each subject. If he encountered difficulty
in either'Task C or‘Task A, sparing help was offered; however no state-
ments regarding methods of approach.or underlying principles was made.
The importance of the sample tasks was to insure that the subject under-
stood what was to be required in the tasks which followed. Then, pre-

TaSk C and.pre-Task A.were administered.1 Following this, depending on

 

1 preJTask refers to the tasks administered before stressor conditions
or intenpolated time interval. Post-Task refers to those administered
after - see schematic design.

23

whether the subject was a Control or an Experimental one, either
stressor conditions were induced or a comparable time period was
allowed to elapse. In the latter case, conversation was limited
and no questions regarding the experiment were answered- the exper-
imenter busied himself in one way or another. This time period
was never more than 30 seconds. Then, post-task C and post-Task A
were administered. Finally, introspective data was collected with
regard to the effectiveness of the stressor conditions.

In all trials, Task C was presented before Task A. Therefore,
if an order effect had arv importance it was probably held constant
(barring interactions).

Schematically, the design is:

 

Single Pre-Tasks Post-T asks
Experimental C-A C-A Stress C-A
Control C—A G—A C-A

The directions for the tasks were as follows:

I am going to show you some colored designs and I
will ask you to reproduce them with these blocks. (indi-
cate the blocks) As you can see, these blocks are all
identical - they have a red side, white side, and so
forth. Now here is the first design you are to repro-
duce. (place both original and modified design in front
of subject and indicate that they are both representa-
tions of the same design). I'd like you to make it out
of four blocks (separate them from the rest), just the
way you see it here in front of you. Please tell me
when you are finished. Go ahead. (after 8 has comple-
ted his model, brush up the blocks, remove the modified
design - the original remairs in front of the subject)
Now I'd like you to make this design out of all sixteen
blocks - but, I'd like you to build it so that it is

2h

rotated 90 degrees to the left or right - that is,
one full turn to the left or the right. You must
build it rotated - you cannot build it and then turn
it. And, at the same time, I'd like you to inter-
change the colors as you build it - where there's
red, put white; and where there's white, put red.

Do you understand what you are to do? (if S is at
all confused, explain more fully). All right, go
ahead, and please tell me when you are finished.
(after 3 has finished, remove the design and brush
up the blocks) This is the way we will proceed on
the following designs. Do you have any questions?
(in the following tasks repeat the relevant sections
of the above each time.)

The stressor conditions were the introduction of false norms,
i.e. failure-stress. An important, although extremely difficult
to control, variable here is the degree of realism achieved by the
experimenter, however the general impression was that these evalua-
tions were seriously taken. After the subject finished the pre-
Tasks, he was told:

Well, I guess I don't have to tell you - you

did prettylousy on those. What we're doing here

is trying to make up an intelligence test, and we're

trying to get some averages for it. I'm aving these

same tests to a lot of ,different students now, but

I've already given them to a bunch of high school

seniors - and you did worse than they did. I mean

that, on the average, the high school kids did better

on this than you did. (pause) Well, do you want to

try another one? .

After the subject finished the last block desigi, the introspec-
tive data was collected. All subjects were asked to try and think back
upon how they felt just before the beginning of the last pair of de-

signs. Three forced-choice questions were asked and the subject was

instructed to choose one or the other alternative:

25

1. Did you feel tense or relaxed?

2. Did you feel happy or sad?

3. Did you feel nervous or at ease?

Each subject was given a brief description of the true nature

of the study at the end of the session and was asked not to'communi-

cate this information to anyone for obvious reasons.

26
IV. RESULTS

. The measures obtained for the groups were solution times for
the various block designs presented to them. As so often is the
case, the distribution which resulted was skewed; and so, in order
to proceed with the analysis of results, a reciprocal transformation
was done.

Two analyses of coveriance were run to test our hypotheses.
According to hypothesis one (see p. 18), under stressor conditions
we should expect a decrement in performance on the abstract task.
More specifically, this predicts that the experimental group will
take longer on post-Task A than will the control group. Table 1
presents the results of the covariance analysis. The F-ratio of .27
is not significant, and, in fact, the mean solution times (before
adjustment) for the groups indicate that the Control subjects took
longer to solve postJTask A than did the Experimental subjects. These
means are presented in Table l A.

Table 2 presents the results of the test of significance on hy-
pothesis two (see p. 18). This hypothesis gives rise to the expec-
tation that the experimental group will take less time on post-Task
C than will the control group, i.e. performance on a concrete task
will be facilitated by stress. The F-ratio of .73 is again not sig-
nificant. A look at the mean solution times (Table 2A) reveals that
the control group solved post-Task C a little bit faster than did

the experimental subjects.

27

The introSpective data (see p..1fi)'were used to allow some
assessment of the effectiveness of the stressor conditions. In
answering the three questions asked, those subjects who did exper-
ience stress would be expected to respond that they felt tense, sad,
and nervous. Control subjects should respond that they felt relaxed,
happy, and at ease. A chi-square test of significance was used for
each of the questions. Table 3 contains the results of these tests.
A11 are in the expected direction, two of them (tense-relaxed and
nervous-at—ease) are significant at below the .05 level, and the other
(happy-sad) places at below the .15 level of significance. One-tailed
tests were used because specific predictions were made in accordance
with our definition of stress (see p. 8 ).

Since a major variable in the study was the validity of the tests
used in testing our hypotheses, an attempt was made to gain some in-
sight into their character. One important aSpect of this is whether

they are, in fact, tapping the independent functions of abstract and

 

concrete behavior. If this were the case, then one should expect
there to be a negative correlation, or no correlation at all, between
the abstract and concrete tasks. That is, from.Goldstein's statement
that “even in its simplest form...abstraction is separate in principle
from.concrete behavior; There is no gradual transition from the one
to the other. The assumption of the abstract attitude...is a totally

different activity of the organism" (19, p. 60), if our tasks are

28

measuring what we designed them to measure, one should not, in prin-
ciple, be able to predict performance on one type of task from per—
formance on the other. Table h presents results which show that such
does not seem to be the case. Four product-moment correlations were
run between the performance on Task A and Task C for the two groups
during the pre- and post-trials. The correlations ranged from .318
(experimental group before stress) to .619 (control group during the
same period).

Summarizing the various results presented, neither of our hypo—
theses was borne out although it appeared that the stressor conditions
were effective. Further, it seems that the tasks used to represent
abstract and concrete behaviors were not measuring the independent

functions fer which they were designed.

29

TABLE 1

COMPARISON BETWEEN GROUPS ON ABSTRACT TASK

Analysis of Covariance for Task A

 

sum of products 6393.h0 6005.h0 388.00
sum.of squares: X 9066.30 8752.67 313.63
sum of squares: I lh037.87 13557.87 h80.00
df 29 28 1
correlation coefficient .56? .551 ...
df for r 28 27 ...
bxy value ' . .hSSh .hh29 ...
adjusted sum of squares 615k .h9 6092.60 61.89
df 28 27 1
Note: X is post-Task A N = 30 F = .27, N.S.
I is pre-Task A
TABLE 1A

Mean Solution Times for Task A

 

Control Egperimental
pre-Task A 5.92 5.38
post-Task A 7.0h 5.67

 

Nets: times presented in minutes and decimals

N'= 30

30

TABLE 2

COMPARISON BETWEEN GROUPS ON CONCRETE TASK

Analysis of Covariance for Task C

 

 

£91391 Within m
sum of products 7257.00 7356.67 -99.67
sum of squares: X 186h8.97 18631.33 17.6h
sum of squares: Y 11668.00 11122.67 hh5.33
df 29 28 l
correlation coefficient .h92 .511 ...
df for r 28 27 ...
bx? value .6220 .661h ...
adjusted sum of squares 1h135.h2 13765.5h 369.88
df 28 27 1
Note: X is post-Task C N = 30 F i .73, N.3.
I is pre-Task C
TABLE 2A

Mean Solution.T1mes for Task C

 

Control Experimental
pro-Task C :29.9 :16.7
post-Task C :21.1 :17.9

 

Note: Times in seconds and decimals.

N I 30

31

TABLE 3

COMPARISONS BETWEEN GROUPS ON INTROSPECTIVE DATA

 

 

 

 

tense- relaxed happy - sad nervous - at ease
Control 6 9 9 6 6 9
Experimtl. 12 3 5 10 12 3
Chi-sq. 3.h75 1.206 3.h75
p less than .05 less than .15 less than .05

 

Note: df = l in all cases, p values for one-tailed tests. N'8 30

TABLE 14

CORRELATIONS BETWEEN TASK A AND TASK C FOR EACH
GROUP IN EACH CONDITION

.2.

Control................pI‘e-' 0619
post- .513

Experimental. 0 e e e s ee 0 e .pI‘e- 0318
post- .577

 

Note: N'= 15 for each of the four groups. Product-moment r used in
each case.

32
V. DISCUSSION -

In contrast to the large body of previous research in the
area, in the present study psychological stress had no signifi-
cant effects on performance. TWO parallel hypotheses were advan-
ced with regard to the expected effects of stress, and both had
to be rejected.

Following a line of reasoning within Goldstein's theoretical
framework, it became evident that one of the most important var-
iables in relation to stress effects is the nature of the task.
This was not a new theoretical discovery, for it had often been
emphasized.by writers in this area. At the same time, however,
very little work had been done attempting to isolate this variable
in order to assess its meaning. It seemed consistently to be men-
tioned in the conclusions of studies done, but not in the experi-
mental design proper. The present study set out to isolate the
variable in terms of its function as a channel allowing the expre-
ssion of certain performances. That is, this writer felt that varya
ing the task along known dimensions would enable the behaviors in-
trinsically related to such dimensions to emerge. The results did
not bear this out.

Furthermore, not only were the predictions not substantiated,
but a curious turnabout took place. It was predicted that the ex-

perimental group would be impaired in performance on the abstract

33

task and facilitated on the concrete task. The assumption went
along with this that these differences would be relative to a gen-
erally unchanging control group. A look at Tables 1A and 2A reveals
that, to all intents and purposes, the control group behaved the way
the experimental group should have (according to the hypotheses),
and vice versa. The differences are not significant ones, and they
become even smaller when the groups are equated on their pre-Task
performances. on the surface, however, it might appear that, if

the hypotheses were in fact valid, the control group was the one
which experienced threat. Introspective evidence points to the con—
trary though, for the experimental subjects reported feelings asso-
ciated with stress significantly more often than did the controls.
Taking this into consideration, the reversal of roles appears to be
a chance phenomenon.

Does this lead us to the conclusion that the theory which gener-
ated the hypotheses is invalid? This is, of course, a possibility,
but not necessarily so. The counterbquestion which must be asked is
whether this study adequately went about testing the hypotheses.

Since a great deal of control was obtained over individual dif-
ferences through the covariance analysis (the correlations ranged from
.h9 to .57 as shown in Tables 1 and 2), it seems unlikely that indivi-
dual variation;was an important source of error. This is further sub-

stantiated by the fact that, although the within variance differences

3h

from pre- to postJTask are in opposite directions on Task A and
Task C performances (variance becomes greater on task A, smaller
on Task C - see Tables 1 and 2), both experimental and control
groups react in the same direction. That is, both the control and
experimental groups vary less on post-Task C than they did on pre-
Task C, and both groups vary more on post-Task A than they did on
pre-Task A.

Such a finding places suspicion on the tasks themselves. It
is here that the source of extrinsic error seems likely to be. Be-
cause of the posited independence between abstract and concrete be-
havior (see above, pp. 1, l7, and 27), it follows that any attempt
at devising a task which should activate 223 or the other of these,
'mugt not activate both. To the extent that this is violated it will
be impossible to estimate which of the two performances, abstract or
concrete, is actually taking place. The question which is here being
raised concerns the validity of the task.)

Herein lies the special importance of the correlations presented
in Table h. There is no doubt that'rasks A.and C are correlated with
each other. Contrary to theoretical expectations, one can predict,
with a fair amount of accuracy, the performance on one task from a
knowledge of the performance on the other. This indicates that the
tasks chosen in this study to present pure measures of abstract or

concrete behavior fail to do so.

35

So then, aside from the possibility that the theoretical model
is erroneous, two other possibilities now arise. One is that abstract
and concrete behavior are impossible to definitely discriminate. In
this case, further research on the general questions asked.by this
study would profit from a reformulation of the problem. Perhaps task
conmonents should be studied, rather than characterizing the task as
an entity. Very little work has been done in this area, and it would
probably prove fruitful. The main implimentation of such an approach
micht be carried out through factor analytic technicpes. The other
possibility is that it is possible to discriminate between the two be-
haviors, in which case one must design a task that will tap either
behavior, but,not both. ‘Were this study to be repeated, this‘writer
would favor‘the latterf. possibility asitheffirst. step. Various tasks
would have to be administered to samples in order to ascertain a lack
of correlation between two tasks which would still seem to have face
validity in terms of the performance in question. It is not.difficult
to find two uncorrelated tasks, but the important issue along with that
is the tasks should "look like" they are measuring abstract and concrete

behavi ors .

36
VI. SUMMARY AND CONCLUSIONS

From a theoretical viewpoint which proposes that individuals
tend to cope with inadequate conditions by becoming more concrete,
two hypotheses were formulated with regard to the effects of stress.

It was expected that if individuals were given concrete and abstract
tasks, i.e. tasks which activate these differential performances,
then the subjects under stress would find channels for the expression
of their coping behavior in» the concrete task. ‘80, it was predicted
that: '

1. Stress would facilitate performance on a concrete task.

2. Stresswould impair performance on an abstract task.

The results did not substantiate either of these hypotheses. Some
possible reasons for this were advanced. The most tenable of these
seemed to be that the tasks themselves were not measuring what they were
purported to measure. Some evidence within the experimental data lend
further support to this notion. Possibilities for further research ave-
nues on this same topic, which would attempt to eliminate this error,

were discussed briefly.

1.

2.

3.

h.

5.

6.

7.

8.

9.

10.

37

VII. BIBLIOGRAPHY

Alper, T.G. Memory for completed and incompleted tasks as a
function of personality: An analysis of group data. :1.
abnorm. soc. Psychol., 19h6, hl, hOB—th.

Ansel, A. The combination of a primary appetitional need with
primary and secondary emotionally derived needs. 3. &.
PEXChOIQ, 1950, ’40, 1-114.

Ansel, A. and Maltzman, I. The effect upon generalized drive
strength of emotionality as inferred from the level of
consumnato'ry reaponse. g. SE° PsEol” 1950, ho, 563-569.

 

Arieti, S. Inte retation of Schizophrenia, New York: Robert
Brunner, 1955.

Beier, E.G. The effect of induced anxiety on flexibility of in-

tellectual functioning. ngchol. Monogr" 1951, 61;, No. 9,
WhOle N0. 3260

Brown, J.S. The gmeralization of approach responses as a func-
tion of stimulus intensity and strength of motivation. _J_.
coup. Psychol., 19152, 33, 209-226.

Brown, J.S. and Jacobs, A. The role of fear in the motivation
and acquisition of responses. ,1. gig. Psychol” 19149, 39, 7h7-759.

Child, LL. and Waterhouse, I.K. Frustration and the quality of per-
formance: II. A theoretical statement. Psychol. £91., 1953, 60,
127‘139e

Chiles, W.D. Psychological stress as a theoretical concept, WADC
Technical Report 57-1457, ASTIA Document No. AD 1309M, July,1957.

Clark, R.A. and McClelland, D.C. A factor analytic integration of
imaginative, performance and case study measures of the need for
achievement. Unpublished research, cited in Lazarus, Deese, am
Osler.

Cowen, E.L. Stress reduction and problem-solving rigdity. g. 321;!-
sult. Psychol., 1952, 16, 1:25-1:28.

Cowen, E.L. The influence of varying degrees of psychological stress
on problem-solving rigidity. _._I_. Abnorm. Egg. ngchol., 1952,
’47, 512-519.

15.

16.

17.

18.

19.

21.

22.

23.

211.

21:.

25.

38

Deese, J. and Bowen, H.M. The effect of task-induced stress
on code learning. Unpublished research, cited in Lazarus,
Deese, and Osler.

Deese, J., Lazarus, R.S., and Keenan, J. Anxiety, anxiety reduc-
tion, and stress in learning. :1. 9352. Psyghol., 1953, 1:6,
55—60.

Farber, I.E. The role of motivation in verbal learning and per-
formance. Psychol. Bull., 1955, 52, 311-325.

Gates, G.S. and Rissland, L.Q. The effect of encouragement and
discouragement upon performance. g. educ. Psychol., 1923,
11;, 21.26.

Gelb, A. and Goldstein, K. Psycholo ische Analysen hirnpathologischer
Faelle, Barth: Leipzig, 1920. Iiartially translated in: Ellis,
ource Book 9_f Ge_s_____ta1t P chol , Harcourt, Brace: New
York, 1938)

 

Goldstein, K. '_I'_h_e_ 0rganism,A1nerican Book Company: New York, 1939.

 

Goldstein, K. Human Nature in th__e_ Li ht _of Psychopatholog, Harvard
University Press: Tambridge,l 1951.

Goldstein, K. Methodological Approach to the Study of Schizophrenic
Thought Disorder, in: Kasamin, J.S., ed. Language___ and Thought
i__n Schizophrenia, University of California ress: Berkeley,

1955.

Goldstein, K. and Scheerer, M. Abstract and Concrete Betavior: an
Experimental Study with Special Tests. Psychol. Monogr., 191d,

Grinker, R.P.. and Spiegel, J.P. l_l_e_q finder Stress, Blakiston: Phila-
delphia, 19h5.

Hurlock, E.B. The value of praise and reproof as incentives for
children. Arch. 13821101., New York, 1921;, 11, No. 71.

Lantz, B. Some dynamic aspects of success and failure. Psychol.
Monogr. 19175, 59, No. 1, Whole No. 271.

Lazarus, 11.8. and Baker, R.W. Personality and psych010gical stress:
a theoretical and methodological framework. Psychol. Newsltr.,
1956, 8, 21—36.

Lazarus, 11.3. and Baker, R.W. Motivation and personality in psycho-
logical stress. Psychol. Newsltr., 1957, 8, 159-183.

26.

27.

28.

29.

33.

31.1.0

35.

36.

37.

39

same, R.S., Deese, J., and Hamilton, R. Anxiety and stress
in learning: the role of intraserial duplication. :10 3E.

Lazarus, R.S., Deese, J., and Osler, S. The effects of psycho-
logical stress upon performance. Psychol. Bull., 1952, 1:9,
293'3170

Lazarus, 11.5. and Eriksen, C.W. Effects of failure stress upon
skilled performance. :1. exp. Psychol., 1952, 143, 100-105.

Levy, J.M. and Rokeach, M. The organization of new percqytual sys-
tems. Chap. 11: in Rokeach, M. '_I‘_1_1_e_ Organization 33 Belief
S stems, Basic Books: New York, to be published.

Lindquist, E.F. Design and Analysis _o_f Experiments in P chol
and Education, Houghton Mifflin Company: Boston, 1953.

 

Lucas, J.D. The interactive effects of anxiety, failure, and intra-
serial duplication. Amer. g. ngchol., 1952, 65, 59-66.

Marquart, D.I. The pattern of punishment and its relation to ab-
normal fixation in adult human subjects. ,1. gen. ngchol.,
191:8, 39, 107-1111;.

McClelland, D.C. and Apicella, F.S. Reminiscence following exp-
erimentally induced failure. 3].. 933. PsEhol" 191:7, 37,
159-169.

Miller, N.E. Studies of fear as an acquirable drive: I. Fear as
motivation and fear-reduction as reinforcement of new res-
ponses, g. exp. Psychol., 191:8, 38, 89-101.

Montague, E.K. The role of anxiety in serial rote learning, «_I,
exp. Psychol., 1953, h5, 91—101.

Nagaty, 15.0. The effect of reinforcement on closely following
S-R connections: I. The effect of a backward conditioning
procedure on the extinction of conditioned avoidance. ,1.
exp. Psychol., 1951, h2, 239-2h6.

Osler, S.F. Intellectual performance as a function of two types
of psychological stress. _q. gig. Pfl0h01., 1951:, 1:7, 115-121.

to

38.. Rogers, C.R. A theom 9_f_' therapy, personality, and interpersonal
relationships, as develgped _ig the client—centered frame-
work, unpublished manuscript, 1956.

 

 

39. Rosenbaum, G. Stimulus generalization as a function of level of
experimentally induced anxiety, :1. Qflh Psychol., 1953, 1:5,
BS‘hBO

ho. Rosenbaum, G. Temporal gradients of response strength with two
levels of motivation. :1. e_xp. Psychol., 1951, 1:1, 261-267.

 

 

1:1. Schaffer, E.B. Behavior under stress: a neurophysiological hypo-
thesis. szcho . 533., 19st, 61, 323.333.

1:2. Selye, H. Stress. Acta: Mantreal, 1950.

1:3. Siegel, P.S. and Brantley, J.J. The relationship of enotionality
to the consummatory response of eating. _1. fl. Psychol.,
1951, h2, Bob-306.

141:. Spence, K.W., Farber, I.E., and Taylor, E. The relation of elec-
tric shock and anxiety to level of performance in eyelid

h5. Thorndike, E.L. and Woodward, E. The influence of the relative
frequency of successes and frustrations on intellectual ach-
ievement, J. educ. Psychol., 1931., 25, 2141-250.

 

1:6. Verville, E. The effect of emotional and motivational sets on the
perception of incomplete pictures. _«1. gene . chhol, 191:6,
69, 133‘1h50

147. Vogel, 71., Baker, RM. and Lazarus, 11.5. The role of motivation in
in psychological stress. :1. abnorm. egg. Psychol., 1958, 56,
105-112.

1:8. Williams, M. An experimental study of intellectual control under
stress and associated Rorschach factors. _.1. consult. Psychol.
19h7, ll, 2lp29.

119. Zander, A.F. A study of experimental frustration. Psycho . Monogru
191.111, 56, NO. 3’ “013 N0. 256.

50. Zeller, A.F. An experimental analogue of repression: II. The
effect of individual failure and success on memory measured
by relearning. J. exp. Psychol., 1950, 1.0, 1:11.122.

 

Note: The author wishes to thank Mr. Robert J. Cohen for his help in
preparing the bibliagrapw.

1"

,1

'“.'1 V
.| r

80811 138E E.L..n

 

 

9'":