University of Southern California Dissertations and Theses

Aggression As A Function Of Arousal And Friendship Ties

(USC Thesis Other)

Aggression As A Function Of Arousal And Friendship Ties

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content INFORMATION TO USERS
This material was produced from a microfilm copy of the original document. While
the most advanced technological means to photograph and reproduce this document
have been used, the quality is heavily dependent upon the quality of the original
submitted.
The following explanation of techniques is provided to help you understand
markings or patterns which may appear on this reproduction.
1.The sign or "target" for pages apparently lacking from the document
photographed is "Missing Page(s)". If it was possible to obtain the missing
page(s) or section, they are spliced into the film along with adjacent pages.
This may have necessitated cutting thru an image and duplicating adjacent
pages to insure you complete continuity.
2. When an image on the film is obliterated with a large round black mark, it
is an indication that the photographer suspected that the copy may have
moved during exposure and thus cause a blurred image. You will find a
good image of the page in the adjacent frame.
3. When a map, drawing or chart, etc., was part of the material being
photographed the photographer followed a definite method in
"sectioning" the material. It is customary to begin photoing at the upper
left hand corner of a large sheet and to continue photoing from left to
right in equal sections with a small overlap. If necessary, sectioning is
continued again — beginning below the first row and continuing on until
complete.
4. The majority of users indicate that the textual content is of greatest value,
however, a somewhat higher quality reproduction could be made from
"photographs" if essential to the understanding of the dissertation. Silver
prints of "photographs" may be ordered at additional charge by writing
the Order Department, giving the catalog number, title, author and
specific pages you wish reproduced.
5. PLEASE NOTE: Some pages may have indistinct print. Filmed a s
received.
Xerox University Microfilms
300 North Zeeb Road
Ann Arbor, Michigan 4S106
I
I
74-26,060
ZABRACK, Merle Linda, 1947-
AGGRESSION AS A FUNCTION OF AROUSAL AND
FRIENDSHIP TIES.
University of Southern California, Ph.D., 1974
Psychology, general
University Microfilms, A XEROX Company , Ann Arbor, Michigan
T U I Q n i C C C D C DCCM Mir'DrtCII >«cn r v »r>T! w «n
AGGRESSION1 AS A FUNCTION OF AROUSAL
AND FRIENDSHIP TIES
by
Merle Linda Zabrack
A Thesis Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Psychology)
August 1974
U N IVER SITY O F S O U TH E R N C A LIFO R N IA
TH E G R A D U A TE SCHO O L
U N IV E R S IT Y PARK
LOS A NG ELES. C A L IF O R N IA 9 0 0 0 7
This dissertation, written by
M jarl.e..L.inda..2a.'bra.ak............................
under the direction of h.&x.. Dissertation Com
mittee, and approved by all its members, has
been presented to and accepted by The Graduate
School, in partial fulfillm ent of requirements of
the degree of
D O C T O R O F P H IL O S O P H Y
Dean
DISSERTATION CO M M ITTEE
Chairman
TABLE OF CONTENTS
Chapter Page
I. INTRODUCTION............................. 1
II. METHOD.................................... 14
Overview of procedure
Subjects
Experimental setting and apparatus
Procedure
Salience of group membership
III. RESULTS................................. 2 3
Manipulation checks
Behavioral measures
Mean number of balls thrown
1 Content1 s core
Other behavioral measures
Multivariate analysis
Pre-experimental questionnaire
Post-experimental questionnaire
Recollections of feelings
Aversiveness of opposition
Approximation of balls thrown
Ratings of chant as it affected
siab jects
Ratings of chant as it affected
1 trainee'
Feelings toward the experiment
Rating of the 'trainee'
IV. DISCUSSION............................... 45
BIBLIOGRAPHY.................. 59
APPENDICES........................................ 62
I. Mood Assessment Questionnaire
II. Verbatim Instructions Given to Subjects
III. 'Trainee's" Speech
IV. Assessment Questionnaire for Passive
Resistance Training— Program AV 101
ii
CHAPTER I
INTRODUCTION
A large portion of the experimental research
investigating aggressive behavior has focused on
variables that instigate subjects to aggress. Thus, for
example, subjects have been insulted, annoyed and/or
frustrated and the effects of these various manipulations
are then noted in terms of subsequent aggressive ret
sponses. Typically, this research has shown that
aggressive behavior tends to increase following such
instigation (Eg. Baron, 1971; Berkowitz, 1965;
Berkowitz & Geen, 1966; Baron, 1972). Anger can be seen
as the mediating variable between these manipulations and
subsequent aggression and, in fact, investigators often
refer to their manipulations as anger-arousal operations.
However, several writers have pointed out that even though
the arousal of anger does lead to aggression with
consistency, the two (anger and aggression) may also
occur independently (Buss, 1961; Kaufmann, 1970). This
has led to recent experimentation designed to assess
whether any generalized arousal will instigate aggressive
behavior.
Before reviewing this research, it is appropriate
2
to define the term 'arousal' as it has been used in
psychological literature. Arousal, according to Hebb
(1972), is "wakefulness, alertness, vigilance,
excitation or excitability; at high levels, emotional
disturbance [p. 294]." According to arousal theory, as
described by Malmo (195 8), there are two components
involved in arousal— an intensive component that indi
cates the degree of energy expenditure and the degree of
excitation in the central nervous system; and a
directional component, that represents the orientation of
the organism towards the goal. The latter component has
been likened to 'habit-strength' and the former to 'drive'
in Hull-Spence terminology. In fact, both Malmo (1958)
and Hebb (1955), who use the term 'arousal', seem to
agree that the concept is similar to generalized drive, as
formulated by Hull (1943). A third term, which has been
used synonymously with 'arousal' and 'drive', is
'activation'. According to Geen and O'Neal (1969),
activation theory is described as follows: "Input in any
sensory modality has nonspecific effects which may
manifest themselves in behavior as a general state of
arousal [p. 2 89]."
Regardless of which term is used by psychologists,
all seem to be in agreement that a general drive or
arousal state will have an energizing effect on the
3
organism. This effect usually enhances the emission of
a dominant response (Hull, 1952; Walters & Parke, 1964;
Zajonc, 1965). The term 'arousal' will be used hence
forth.
An additional point of interest in the arousal
literature concerns its relationship to responding. An
inverted-U function has been found consistently (Hebb,
1955); that is, as arousal increases, behavioral
efficiency also increases but only up to some optimal
level, beyond which further increases in arousal lead to
disorganization and decreased behavioral efficiency. This
decrement in efficiency has been shown to affect per
formance on tasks involving perception, attention span,
problem-solving, information-processing and other
cognitive processes (Wallace, 19 71). Thus, high levels
of arousal tend to disorganize behavior. This may be due
to different effects, one of which is competing emotional
behavior (Wallace, 1971). In other words, increases in
arousal beyond the optimal level may evoke responses
which interfere with the ongoing dominant response. This
suggests that increased arousal could serve to distract
individuals or could lead to anxiety reactions, both of
which would decrease behavioral efficiency and inhibit
performance.
In an experiment investigating the relationship
between arousal and aggression, Berkowitz and Buck (196 7)
suggested that increasing arousal should "heighten re
activity to the aggression cues" that exist in the
situation (p. 416). They found that subjects, aroused by
injections of epinephrine, responded aggressively to
stimuli which had previously been associated with
aggressiveness; that is, the stimuli had aggressive cue
value for these subjects. Stimuli which had not been
paired with aggressiveness did not evoke aggressive
responses from aroused subjects. These results supported
the hypothesis that increased arousal would facilitate
aggression in the presence of cues that had aggressive
meaning for subjects. The authors also noted that
subjects did not report feeling angry.
Additional support for the above finding was pro
vided by Geen and O'Neal (1969) . They aroused half the
subjects in their experiment by having them wear head
phones over which they heard "stimulating but not
aversive” white noise (p. 290) . The other half of the
subjects heard nothing. Before this phase of the
experiment, half the subjects had seen an aggressive film
clip while half had seen a non-aggressive film. Although
most aggressive behavior occurred when subjects were
aroused and had seen the aggressive film, a main effect
for arousal was also obtained, suggesting that arousal per
5
se may affect aggression, regardless of the presence of
aggressive cues in the environment. There is further
evidence for this latter notion.
Knipmeyer and Prestholdt (1973) offered groups
of subjects the opportunity to aggress in silence, while
exposed to loud white noise or while exposed to crowd
noise. Thus, the latter condition expanded on the Geen
and O'Neal (1969) design by providing a source of arousal
which also had aggressive meaning. Based on Berkowitz's
model, Knipmeyer and Prestholdt predicted greater
aggression from subjects who heard the crowd noise than
from subjects who merely heard white noise (both 'noise'
conditions were at sound levels 45 decibels above the
ambient noise level in the room). Their results
indicated that, although more aggression occurred when
subjects had heard either white noise or crowd noise in
comparison to subjects who had not heard anything, there
were no significant differences in aggressive responding
as a function of the type of background noise. In other
words, subjects responded similarly regardless of whether
the noise had aggressive cue value (crowd noise) or not
(white noise). Similarly, Zillmann (1971) found that
annoyed subjects administered more intense shock to a
victim after watching an arousing film, whether that film
was aggressive or erotic.
6
The research described suggests that general
arousal facilitates aggression. However, none of this
research has attempted to assess varying levels of
arousal and the effects on subsequent aggression. In
addition, comparisons of the reactions of individuals
and groups to increases in arousal, or comparisons among
different types of groups to arousal manipulations could
have implications for crowd behavior and possible
aggression control. For example, will increasing levels
of arousal differentially inhibit individuals in contrast
to groups? Zajonc (1965) has shown that the presence of
others may function as a "source of arousal" which will
facilitate the occurrence of the dominant response.
Additional research has indicated that the social
facilitation effect may be dependent on evaluation
apprehension. This was confirmed for aggressive behavior
by Zabrack and Miller (1973) who found more aggressive
responding in the presence of others when evaluation
apprehension was minimal. Thus, the mere presence of
others might interact in some way with other sources of
arousal present in the situation. A further question
concerns who those others are; that is, will inhibition
occur differentially for different types of groups?
Research by Zabrack and Miller (1973) illustrated that
friends were less inhibited in their behavior and more
7
willing to use aversive control than were strangers.
Again, the disinhibition that seems to occur when friends
are present could interact with the effects of arousal on
behavior.
Besides the mere presence of others functioning
as a source of arousal in crowds, there are usually
forms of external stimulation that may arouse people.
Zimbardo (196 9) says, "A generalized state of arousal
also increases the likelihood that gross 'agitated' be
havior will be released, and that cues in the situation
which might inhibit responding will not be noticed
[p. 19]." He suggests that facilitative arousal tech
niques are effective since they contribute to a certain
loss of self-awareness. One of the examples he gives as
a source of arousal is the war dance which many Indian
and African tribes use. This consists of loud,
repetitive music, and simple rhythms, combined with
singing, chanting and shouting. The increased arousal is
then "channelled into directions [that are] prescribed
[p. 19]."
Another example of music being used to stimulate
aggressive behavior occurs in military and patriotic
situations. Films of rallies during Hitler's rise to
power show how German patriotic music could rouse the
masses. In present-day situations, chants have been
utilized. Often, the final impetus before the outbreak
of a riot (political, racial, etc.) consists of a chant
which is taken up by the crowd, after which destructive
behavior may ensue.
Consideration of the above examples suggests that
in a crowd situation, any stimulus or combination of
stimuli that increases arousal would increase the
likelihood of subsequent aggression. However, there is
also anecdotal evidence that increases in arousal may
inhibit aggression: in a popular bar in the Los Angeles
area, the management increases the volume of background
music when the crowd seems to be getting 'out-of-hand' or
unmanageable. The result is typically a decrease in any
forms of antisocial behavior.
The present experiment was designed to assess the
effects of various levels of arousal on the aggressive
behavior of small groups. Specifically, arousal was
manipulated by varying subjects' exposure to a background
chant, both quantitatively and qualitatively. Of the
four levels of arousal that were implemented, three were
varied in a quantitative manner. The lowest level of
arousal involved no chanting or background noise at all;
that is, aggression occurred in silence (similar to the
'no noise' conditions employed by Geen and O'Neal, 1969,
and Knipmeyer and Prestholdt, 19 73). Two additional
9
levels of arousal were produced by playing the chant as
a background stimulus, either at low intensity (67
decibels) or high intensity (80 decibels). Finally, the
fourth level of arousal was induced by having subjects
participate in the chanting, thereby adding a quali
tatively different dimension to the arousal manipulation.
Thus, the chant was played at a moderate intensity
(74 decibels) and subjects chanted along with the tape.^
The second independent variable of interest
consisted of the relationship among group members.
Speculation regarding aggressive behavior in crowds by
LeBon (1895) and Brown (1965) has led to the suggestion
that the increased anonymity provided by crowds dis-
inhibits anti-social behavior. Zimbardo (1969) and
Baron (1970) added some experimental support to this.
But, in the latter experiments, group members could not
see each other while they aggressed against a victim.
Yet, in crowd situations, members do see one another.
Moreover, it seems quite likely that within the larger
structure of the crowd, there will be smaller groups of
people who are not anonymous frofii each other, at least in
terms of their behavior. Based on this assumption,
Zabrack and Miller (1973) investigated the effects of
anonymity of actions versus nonanonymity of actions on
aggressive responding among groups of friends and groups
10
of strangers. The most aggressive responding occurred
when groups of friends could see each other's actions.
Social support processes could explain these results;
that is, friends were seen as providing social support
for each other, as being non-evaluative and accepting;
consequently, this support could serve to reduce anxiety
and subsequently, disinhibit responding. Further con
firmation of this was obtained in a second experiment
(Zabrack & Miller, 1973); the above outcome was repli
cated in conditions where evaluation fears were directly
and specifically minimized.
According to the above findings, then, more
aggressive responding occurred when groups of friends
were together in a situation in comparison to groups of
strangers. The contention in this experiment was that
there would also be a difference in the relationship
between arousal and aggression as a function of the
presence or absence of friendship ties among group
members. Specifically, the relationship between arousal
and aggression was expected to be curvilinear (an
inverted-U shape) for strangers and monotonic for friends,
despite the fact that both groups were to be exposed to
identical arousal manipulations.
Among groups of friends, the expectation was that
as arousal increased, aggressive behavior would also
increase. This expectation was based on the supposition
that several psychological processes would be operative
among friends, as derived from the research cited above.
First of all, friends should receive social support from
each other and should not fear evaluation (Zabrack &
Miller, 1973); thus, some disinhibition should occur as in
previous experiments. Further increases in arousal among
friends should lead to ever-increasing feelings of
1deindividuation' (Zimbardo, 1969), which, in turn, would
. . . . 3
be expected to facilitate aggressive responding. In
terms of the specific manipulations of arousal in this
experiment, the participatory chanting condition was
assumed to offer the greatest opportunity for
1deindividuation1, loss of self-awareness and self-
consciousness.
But, the situation is somewhat different
psychologically for strangers. Previous research has
suggested that there should be less responding from
strangers than from friends (Zabrack & Miller, 1973).
Thus, the low arousal, no chant condition could be seen
as fairly similar to conditions in previously cited
experiments. With some increase in arousal, an increase
in responding would also be expected among strangers.
However, the greatest distinction between friends and
strangers in response to the arousal manipulations was
12
anticipated in the participatory chanting condition.
Whereas this condition was assumed to be the most
'deindividuating' and thus, most excitatory for friends,
we predicted that this same manipulation would be in
hibitory for strangers. That is, since strangers would
not receive as much social support or comfort from each
other as would friends, the expectation was that competing
emotional responses (such as anxiety, uncertainty,
feelings of self-consciousness) would be evoked instead.
These responses would then function to inhibit aggressive
responding.
FOOTNOTES
Although the term 'aggression1 will be used
throughout most of this paper to denote the behavior
emitted by subjects, there may be aspects of the pro
cedure and behavior in question which could be more
accurately described as the 'role-playing of
aggression'. This distinction will become more obvious
as the procedure is described. In addition, there is a
question regarding whether the behavior is aggressive at
all. However, in order to simplify the description of
the experiment, only the term 'aggression' will appear
until the discussion section.
^it should be noted that the decibel values given
may not reflect totally 'pure' measures of sound
pressure level. First of all, the measurements were made
in a room where the noise level was 65 decibels. Thus,
the volume level of the low intensity chant, for
example, was only slightly higher than the noise level in
the room. Secondly, there were momentary changes in
volume within the tape. In order to take this into
account, the decibel levels cited might more accurately
be described as the means of the tape segments rather
than as absolute values. Finally, the fact that during
the experimental sessions, the confederate was reading
his speech, balls were bouncing, boxes rattling, etc.,
would also serve to attenuate the 'pure' figures given.
However, these values do indicate the relative volumes
employed in the arousal conditions.
^'Deindividuation' is defined by Zimbardo (1969)
as "a complex, hypothesized process in which a series of
antecedent social conditions led . . . to a lowered
threshold of normally restrained behavior........[and]
the 'release' of behavior in violation of established
norms of appropriateness [p. 12]." Further, 'dein
dividuation' is expected to occur more readily under
conditions of minimal self-consciousness and self-
evaluation, as well as under conditions when evaluation
by others is minimized.
13
CHAPTER II
METHOD
Overview of procedure
Groups of friends or strangers believed they were
participating in a study to test the merit of a passive
resistance training program. Their role was to provide
'opposition' (in the form of styrofoam balls, verbal abuse
or any other response they could think of) for a
'trainee' involved in the program. The 'trainee'
(actually a confederate) read a two minute speech during
which subjects were free to respond. In addition, during
the response period, arousal was manipulated by means of
a background chant which was: not played at all, played
at a low volume; played at a high volume; or played so
that subjects could participate by chanting along with the
taped materials.
Thus, the experiment was a 2 x 4 factorial design.
One independent variable was the presence or absence of
friendship ties among group members. The other consisted
of four levels of arousal. The major dependent variable
was the number of balls thrown.
Subjects
One hundred and sixty-eight undergraduate females
14
15
participated in the experiment in groups of three. Most
participants were enrolled in introductory psychology
courses at the University of British Columbia.
Participation was on a voluntary basis.
Experimental setting and
apparatus
The experimental room was 2.64 metres by 5.02
metres. At one end of the room, there was a desk and
chair at which the confederate-trainee sat while reading
his speech. At the opposite end, there were three chairs
placed 33 cm. apart in which the subjects sat. In front
of each chair (at a distance of 45.6 cm.), there was a
cardboard box containing 120 styrofoam balls (3" in
diameter). The experimenter's chair was 1.45 metres in
front of the subjects and to the side of the room. A
tape recorder was located on a table at the far end of
the room from the subjects.
Procedure
Subjects were recruited by telephone. In order
to recruit friends, one individual was called and asked to
bring two friends of the same sex to participate as well.
In groups to be composed of strangers, each person was
also asked to bring two friends of the same sex. If this
request was agreed to, subjects were then given a
rationale for not bringing the friends (specifically, the
schedule indicated that friends would not be required for
that week) and were told to come alone. Any person who
could not bring two friends (whether for a 'friends' or
'strangers' group) was not allowed to participate. This
procedure guarded against differential selection as an
alternative interpretation of any differences found
between friends and strangers; that is, there was no
chance that subjects recruited for strangers' groups were
peculiar in the sense that they did not have friends.
Subjects entered the experimental room at the
same time and took one of the three chairs. They
immediately filled out a short questionnaire, entitled
"Mood Assessment", indicating their feelings at the
moment and their degree of acquaintance with the other
group members (see Appendix 1) . Instructions were then
given verbally by the experimenter (see Appendix II).
First of all, background information about the alleged
passive resistance program and its importance was pro
vided. Specifically, subjects were told that the program
investigators felt passive resistance techniques would
prove to be an effective means of handling opposition
from police or hecklers. This was especially true because
of increasingly negative public opinion about violence and
the fear of stringent sanctions by authority figures
towards perpetrators of violence. The training program
17
was supposedly attended by volunteer students interested
in learning techniques of passive resistance, and had
been in existence since the summer. Thus, the 'trainees'
involved in the program were now ready to practice what
they had learned.
The experimenter then explained what the group's
role would be and why. Subjects were told that before
exposing the 'trainees' to real-life situations, the
investigators felt that short, intense, induction
sessions in the laboratory were necessary. The group's
role, then, would be to provide, primarily physical
opposition with the styrofoam balls. They could use
verbal abuse but this had been found to be relatively
ineffective for this particular 'trainee'. Subjects were
also informed that the 'trainee' would earn money if he
was effective in withstanding the opposition provided.
The experimenter then added that there were a
couple of features within the experiment designed to
simulate certain aspects of a crowd situation.
Salience of group membership. Groups of friends
were told the following:
One feature is the fact that the three of
you are friends. It is often the case that
crowds consist of smaller groups of friends
or, at least, individuals who have similar atti
tudes, opinions, experiences, etc. Having glanced
at your questionnaires, it looks as if you are all
feeling similarly anyway.
18
Groups of strangers heard the following:
One feature is the fact that the three of
you don't know each other. It is often the
case that crowds consist of different types of
people who happen to converge and then, in fact,
become a crowd. Having glanced at your
questionnaires, it looks as if you are all
feeling differently so probably represent a good
cross-sample.
Arousal manipulation. When the tape was not to
be played, nothing further was said to subjects concerning
the situation. When the tape was to be played as a
background stimulus, whether at low or high volume, sub
jects were told:
While the three of you can provide a fair amount
of physical opposition with the balls, I'm going
to play a tape in the background to provide some
auditory opposition as well. The tape consists of
a group of people chanting a repetitive phrase.
Chanting is something that often does occur in
crowds.
For groups of subjects who were to participate in the
chanting, instructions were as follows:
While the three of you can provide a fair amount
of physical opposition with the balls, I'd also
like you to provide some auditory opposition by
chanting along with a tape I'm going to play. The
tape consists of a group of people chanting a
repetitive phrase— the words of which are "two,
four, six, seven." Chanting is something that
often does occur in crowds, so I'd just like you
to chant along.^
After these instructions, all subjects were told
that the experimenter would be in the room during the
session but would be rating the 'trainee' and so could
19
not be disturbed once the session began. Further,
subjects were asked not to move past the boxes during the
opposition period. Subjects were then free to ask
questions.
Following these instructions, the experimenter
left the room and returned shortly with a male 'trainee'
(actually a confederate). She instructed him to take
the seat behind the desk, turned on the tape recorder
(for three of the four conditions) and then sat with her
back to the subjects, a clipboard and pencil in hand.
The 'trainee' read a two minute speech on censorship,^
with his head down (see Appendix III). Thus, there was
no eye contact with subjects nor did the 'trainee' react
in any way to the opposition provided (that is, there
were no facial expressions, no changes in vocal
expression, tone of voice, etc.). When the speech was
over, the 'trainee' left the room and subjects were given
questionnaires entitled "Assessment Questionnaire for
Passive Resistance Training Program AV 101," and told
that the questionnaires would provide us with information
for future training programs and with their impressions
of this particular session (see Appendix IV). After
collecting the questionnaires, the experimenter debriefed
subjects and requested that they not inform other
potential subjects about the experiment.
20
The pre-experimental questionnaire provided pre-
behavioral measures of moods or feelings and a pre-
behavioral measure of reported arousal. The behavioral
dependent measures consisted of the mean number of balls
thrown, the latency of the first ball thrown, an
'accuracy' score (that is, the proportion of hits to the
total number of balls thrown) and a 'content analysis'
score for the session. This latter measure consisted of
a score assigned to each group which reflected different
degrees of uninhibited, spontaneous behavior. In fact,
the 'content' was analyzed in two distinct ways. First,
each group was scored for the frequency of spontaneous
behaviors; that is, a simple frequency count determined
'content analysis' score 1. In the second analysis, each
group obtained a score based on the following Guttman
scale of behaviors: laughing and giggling (1); talking
amongst group members (2); general noise, for example,
clapping, rattling the boxes, etc. (3); verbal remarks
directed towards the 'trainee' (4); shouting (5);
whistles, hisses and boos (6); and standing while
throwing (7).
Items on the post-experimental questionnaire
were on 21-point scales. Ten items assessed how subjects
recollected they felt during the session— five of these
were specific to mood; three were directed towards
21
feelings of arousal (and served as manipulation checks);
and two concerned feelings towards the 'trainee1. Six
items were designed to assess how aversive subjects
perceived the opposition and one question asked subjects
to approximate how many balls they had thrown. There
were five items asking subjects to rate the chanting as
it affected them and the same five items to be rated in
terms of how the chanting seemed to affect the 'trainee'.
Two items concerned subjects' feelings towards the group
(also manipulation checks) and three items assessed
subjects' feelings about the experiment in general.
Finally, subjects rated the 'trainee' on five personality
dimensions.
FOOTNOTES
^Chant content was pretested by presenting
seventy-two subjects with five different chants. These
chants were rated on three, 11-point scales to determine
which was perceived most neutrally in terms of content.
Specifically, the chant "two, four, six, seven" was
perceived as more nonhostile then hositle (mean - 8.00),
neither funny nor serious (mean - 6.00) and more boring
than interesting (mean - 3.60). Therefore, any arousal
effects of the chant would have to be attributed to
volume and intensity, rather than to content.
^Pretest results indicated that this topic— the
institution of stricter censorship rules in order to
reduce crime rates and pornography— was counter-
attitudinal (mean - 2.69 on an 11-point scale, where
1 indicated disagreement).
22
CHAPTER III
RESULTS
The behavioral results and questionnaire results
were analyzed separately in this experiment.
Specifically, analyses of variance of three of the
behavioral measures (the latency score, the 'accuracy'
score, and the 'content analysis' score) treated each
group as a unit. That is, for each of these behavioral
measures, one score was derived for each group. However,
analyses of the questionnaire data and the ball-throwing
behavior used the scores of each individual subject. In
these analyses, groups were nested within conditions and
subjects nested within groups.
Manipulation Checks
Two items on the post-experimental questionnaire
were designed to assess whether differences existed
between groups of friends and strangers. Results
indicated that friends did, indeed, like each other
better than did strangers (F = 47.76; df 1/48;
£ <.01), and felt more cohesive as a group (F = 15.90;
df 1/48; £ <.01).6
In order to assess feelings of arousal during the
session, subjects were asked to indicate how aroused,
23
24
excited and stimulated they had felt. A main effect for
levels of arousal was obtained for the latter adjective
(F = 3.50; df 3/48; £ <.05). Generally, subjects reported
feeling more stimulated as levels of arousal increased.
Comparisons of the means revealed that subjects in the
high volume, background chant condition and participatory
chanting condition reported feeling more stimulated than
subjects in the two other conditions (p <.05) . In
addition, a multivariate analysis of variance of the
three adjectives indicated a marginally significant main
effect for levels of arousal as well (F = 1.85;
df 9/144; £ < .06).
In Siam, these results suggest that the
experimental manipulations were effective in producing
the major desired differences among subjects.
Behavioral measures
As mentioned in the method section, there were
four behavioral measures: the mean number of balls thrown,
'content' scores, a latency score for the first ball
thrown and an 'accuracy score'.
Mean number of balls thrown. A 2 x 4 analysis of
variance of this measure revealed a significant inter
action between the two independent variables (F = 2.44;
df 3/48; £ <.05).^ In addition, although there was no
25
main effect for the friendship ties variable (F <1.0),
there was a significant main effect for levels of arousal
(F = 4.33; df 3/48; £ <.01). Figure 1 illustrates these
results.
According to the predictions, as outlined in the
introduction, the largest difference between groups of
friends and strangers in response to the arousal
manipulations was expected in the participatory chanting
condition. Comparisons of the individual means, using
Duncan's new multiple range test (1955), indicated that
this prediction was borne out. That is, the difference
between the score obtained by groups of friends in this
condition (mean = 43.7) was significantly different
(p < .05) from the score obtained by groups of strangers
(mean = 23.0). However, friends and strangers responded
similarly to the other levels of arousal. In other words,
none of the other comparisons between friends and
strangers were significant.
In addition, these comparisons showed that there
was not a significant statistical difference in the
number of balls thrown in the high volume, background
chant condition by strangers (mean = 34.9) and the
number thrown by friends in the participatory chanting
condition (mean = 43.7).
Trend tests were also performed on the data,
26
Friends
Mean
number
of
balls
20
Strangers
balls
thrown
No Chant Low High Participatory
Volume Volume Chant
Chant Chant
Arousal
Figure 1. Mean number of balls thrown at the 'trainee*.
27
since a priori predictions had been made regarding the
nature of the relationship between the independent
variables. These were done separately for groups of
friends and groups of strangers.8 Results indicated that
there was a significant monotonic trend for the data
obtained by groups of friends (F = 12.04; df 1/48;
p c.Ol), but none of the trend tests were significant for
groups of strangers.
In sum, these results partially confirmed the
predictions. Specifically, groups of friends and groups
of strangers did seem to respond differentially to the
arousal manipulation in one condition. For groups of
strangers, the high volume, background chant condition
seemed to elicit the most responding. Beyond this
arousal level, however, their behavior obviously became
inhibited, and ball-throwing decreased. On the other
hand, responding increased directly as the levels of
arousal increased for groups of friends. The significant
monotonic trend supported this. Furthermore, as pre
dicted, the greatest distinction between groups of
friends and groups of strangers occurred in the
participatory chanting condition which proved to be
excitatory for friends but inhibitory for strangers.
'Content' score. Analyses of variance of the
'content' scores indicated that there was a significant
28
interaction between the two independent variables
(frequency count: F = 4.53, df 3/48; £ <.007; Guttman
score: F = 5.79; df 3/48; £ <.002). Tables 1 and 2
summarize these results.
In both cases, the interaction can largely be
accounted for by the difference in scores obtained by
groups of friends and strangers in the participatory
chanting condition (p <•01). Whereas friends received
the highest score in this condition (means = 3.57 and
5.86 for frequency count and Guttman score, respectively),
strangers obtained the lowest score (means = .86 and
1.43). These results, then, also confirmed the
experimental predictions. Specifically, this condition
does seem to be the most excitatory and 1deindividuating1
for friends, given that the 'content' score represents
various types of spontaneous, impulsive behaviors and
tendencies to be disinhibited and un-self-conscious. On
the other hand, we expected strangers to find this
condition inhibitory, perhaps embarrassing and anxiety-
provoking. And, in essence, their extremely low score on
this measure supports-this notion. The low score
indicates that strangers exhibited very little spon
taneous behavior at all and, in fact, were more in
hibited in this condition than in any other. These
results add strength to the interaction obtained when
29
Table 1
Mean 'Content1 Analysis Score for Each
Condition Based on Frequency Count
No Chant Low Volume
Chant
High Volume
Chant
Partici
patory
Chant
Friends 1.71 3.14 2.00 3.57
Strangers 2.14 1.43 2 .85 .86
Table 2
Mean 'Content' Analysis
Condition Based on
Score for Each
Guttman
No Chant Low Volume
Chant
High Volume
Chant
Partici
patory
Chant
Friends 2.14 4.14 2.86 5.86
Strangers 3.57 2.43 4.14 1.43
30
ball-throwing was the measure analyzed. Here, too,
friends threw the most balls in the participatory chanting
condition, suggesting disinhibition, while strangers
threw an amount similar to that thrown under a
relatively lower level of arousal (specifically, the low
volume, background chant condition).
It is also interesting to note that the results
obtained on this measure closely parallel the results
obtained on the ball-throwing measure. In addition to
the fact that there were significant differences
obtained between friends and strangers in the partici
patory chanting condition on both measures, there was no
significant statistical difference on the 'content' score
as well, between friends in the participatory chanting
condition and strangers in the high volume, background
chant condition. That is, strangers in this latter
condition not only threw a similar number of balls (see
previous section), but also displayed spontaneous and
uninhibited forms of behavior, comparable to that shown
by friends in the participatory chanting condition.
Table 3 presents the ANOVAs for the above two
dependent behavioral measures.
Other behavioral measures. Analysis of variance
of the square root transformations of the latency scores
Table 3
Analysis of Variance of Two Behavioral Measures
'Content' Scores
Number of Balls Frequency Guttman
Thrown Count Score
df M S F MS F M S F
Presence or
absence of
friendship
ties (A) 1 82.75 < 1.00 8.64 3.83**
10.29 2.19
Levels of
arousal (B) 3 1055.74 4.33** .62 < 1.00 1.64 < 1.00
A x B 3 595.19 2.44** 10.20 4.53* 27.19 5.79*
Error 48 243.68 2.26 4.70
Total 55
* p <.01
** p <.05
u>
H
32
revealed a significant main effect for levels of arousal
(F = 2.33; df 3/48? £ <.04); that is, latencies decreased
as arousal increased. This result, then, supports the
previously cited behavioral results, in that latencies
woudl be expected to decrease as behavior becomes more
disinhibited and impulsive.
However, there were no significant differences
obtained for the 'accuracy' score. Thus, unlike
experiments investigating performance on cognitive tasks
and arousal, behavioral efficiency did not seem to
decrease with increasing levels of arousal. This can
probably be explained by the fact that ball-throwing is
not as dependent on cognitive skills as are the tasks
typically used in arousal experiments. Perhaps it is
only with these cognitive types of behavior, in contrast
to more physical behaviors, that efficiency decreases
as arousal increases beyond some optimal level. On the
other hand, it may be that it is difficult to achieve
accuracy with styrofoam balls (in contrast to rubber
balls, tennis balls, etc.) under any level of arousal.
Multivariate analysis. A multivariate analysis
of variance was performed for three of the above
behavioral measures— the ball-throwing score, the latency
score and the 'accuracy' score. No significant
statistical differences were found.
33
Pre-experimental questionnaire
Items on the pre-experimental questionnaire
attempted to estimate how subjects felt before the
experimental session began. Thus, if pre-experimental
differences did exist, these could be used as covariates
to more accurately assess the behavioral data. Four
items and a sum score derived from them asked subjects
about their mood. Results indicated that there were no
differences between experimental groups before the
experiment began. All subjects indicated that they felt
good (overall sum score mean = 14.0, where 21 repre
sented the best feeling). A final item on this
questionnaire concerned pre-behavioral arousal. Again,
no differences were obtained. All subjects reported
feeling more aroused than unaroused (overall sum score
mean = 13.1, where 21 represented 'aroused').
Post-experimental questionnaire
Recollection of feelings during the session.
Of the ten items requiring subjects to recollect their re
actions and feelings during the induction session, three
were specific to feelings of arousal and have already
been discussed under the heading of 'manipulation checks'.
Five items were specific to mood (e.g., happy, nervous,
etc.) and two were specific to feelings about the
34
'trainee' (angry, hostile). No significant differences
were obtained between experimental conditions for any of
these measures. In general, subjects reported that they
felt neither good nor bad (overall sum score mean = 10.03)
and neither angry and hostile nor peaceful and nonhostile
(overall sum score mean = 12.24). All in all, subjects
were fairly non-committal about their moods and feelings.
Aversiveness of the opposition. Subjects were
asked to rate their perceptions of how aversive the
opposition was. A main effect for the presence of
friendship ties variable was obtained for the adjective
'painful' (F = 4.70; df 1/48; £ <.05) , indicating that
friends perceived the opposition to be more painful than
did strangers. None of the other adjectives nor the sum
score of aversiveness revealed any significant effects.
Generally, subjects did not perceive that their
opposition was too aversive (overall sum score mean
= 9.26, where 21 represented the most aversive
opposition).
A multivariate analysis of the six adjectives
also revealed a marginally significant main effect for
the friendship ties variable (F = 1.98; df 6/288;
p <.07). In addition, the multivariate analysis indicated
a main effect for the arousal variable (F = 1.65;
35
df 18/288; £<.05). Inspection of the means shows that
as arousal increases, so does the perceived aversiveness
of the opposition provided by subjects. (Combined means
of the six adjectives: no chant = 7.1? low volume =
8.1; high volume = 8.2; participatory chanting = 9.5).
These results again parallel the ball-throwing data,
where a main effect for arousal was also obtained.
Approximation of how many balls were thrown.
This item was constructed in order to determine whether
subjects could accurately assess the extent of their
behavior or whether they would inflate or deflate the
amount of responding in which they had engaged. A
difference score was derived for each subject by
subtracting the guessed number of balls thrown from the
actual number of balls thrown. A significant interaction
was obtained for this measure (F = 4.91; df 3/48;
£ <.05) and the difference scores that were obtained (all
under-estimations) corresponded perfectly to the actual
scores obtained. In other words, the greater number of
balls thrown, the greater was the under-estimation.
Table 4 illustrates these results. This outcome might
merely reflect a memory decrement; that is, it would
probably be more difficult for subjects to remember
exactly how many balls they had thrown if the total
number had been large and perhaps, the more that were
36
Table 4
Difference Scores Representing the Actual
Number of Balls Thrown Less the Guessed
Number of Balls Thrown*
No Chant Low Volume
Chant
High Volume
Chant
Partici
patory
Chant
Friends 2.81 3.81 3.90 17.24
(16.7) (16.9) (27.4) (43.7)
Strangers 1.00 5.33 9.33 6.00
(13.9) (23.3) (34.9) (23.0)
*Actual number thrown are in parentheses.
thrown, the larger the under-estimation. On the other
hand, the approximation could have been either a
deliberate or unconscious attempt to deflate or deny the
actual number thrown. This denial or underestimation
would then represent a justification technique used by
those subjects who had thrown the most balls.
Ratings of the chant as it affected subjects.
Since subjects in the no chant conditions were obviously
not exposed to the chant content, they were unable to
answer these questions and so were eliminated from these
analyses. Subjects in the other three conditions were
asked to rate the chant in terms of how it had affected
them. Two adjectives (provoking and exciting) were
designed to assess the 'arousing' nature of the chant and
a 'chant arousal1 sum score was derived to denote this.
Similarly, three adjectives (annoying, not pleasant and
irritating) focused on the 'aversive' nature of the chant
and a 'chant aversiveness' sum score described this.
Analysis of variance of the 'chant arousal' stun
score revealed a significant interaction (F 4.76;
q
df 2/36; £ <.05), as did a multivariate analysis of
variance of the two adjectives (F = 4.15; df 4/166;
£ c.003). Table 5 summarizes these results. It can be
seen that these results seem to parallel the behavioral
38
Friends
Strangers
Table 5
'Chant Arousal' Sum Scores
Low Volume High Volume Participatory
Chant Chant Chant
11.29 10.1 20.62
8.1 16.38 12.71
39
data. (There was also a main effect for arousal as with
the behavioral data: sum score: F = 5.11; df 2/36;
£<.05; Multivariate: F = 3.36; df 4/166; £< .01) . In
the participatory chanting condition, where friends threw
the most balls, they also perceived the chant as most
arousing. Similarly, strangers threw the most balls in
the high volume, background chant condition and the table
indicates that they found the chant most arousing in this
condition. As with the behavioral data, the means
obtained in these two cells (friends = 20.62, strangers
= 16.38) do not differ significantly. Thus, these
subjects confirmed the experimental hypotheses in their
own words: the more balls they threw and the more
disinhibited they were, the more arousing they perceived
the chant to be! However, these results remain only
suggestive as correlational analyses of the comparable
conditions did not reveal a significant relationship
between the number of balls thrown and the 1 chant
arousal' sum score.
Analysis of the 'chant aversiveness' sum score
also disclosed a significant interaction (F = 4.52;
df 2/36; £ <.05). The multivariate analysis showed a
marginally significant interaction as well (F = 2.09;
df 6/164; £ <.06). These results are illustrated in
Table 6.
40
Friends
Strangers
Table 6
'Chant Aversiveness' Sum Scores
Low Volume High Volume Participatory
Chant Chant Chant
34.57 30.33 25.57
28.52 41.76 36.48
41
This outcome presents an interesting distinction
between friends and strangers in their perceptions of the
aversiveness of the chant. Among friends, the results
seem consistent with the behavioral data. The more
responding they emitted and more aroused they admitted to
being, the less aversive they perceived the chant. That
is, there was an inverse relationship between amount of
responding and perceived aversiveness. However, this was
not the case among strangers, where the relationship was a
direct one; the more balls they threw, the more aversive
they rated the chant. Again, though, correlational ana
lyses indicated that there was no statistically significant
relationship.
However, these results, in combination with those
attained for the 'chant arousal1 sum score seem to suggest
that the chant had different meaning for groups of friends
and groups of strangers. Strangers seem to be responding
to both the arousing and aversive aspects of the chant.
In fact, perhaps the very perceived aversiveness of the
chant was arousing for strangers. Yet friends seem to
distinguish between these two elements of the chant; in
conditions where they describe the chant as most arousing,
they also describe it as least aversive. A possible
explanation is that this difference in perceptions of the
chant may reflect what occurs when subjects participate
42
with their friends in comparison to strangers. Being with
friends seems to alleviate potentially negative aspects of
the situation while being with strangers may accentuate
them.
Rating of the chant as it affected the 'trainee'.
Subjects were also asked to rate the chant on the same
five dimensions described in the previous section, as it
seemed to affect the 'trainee1. Analyses of the sum
scores did not reveal any differences, but significant
effects were obtained for two of the individual adjectives.
Specifically, there was a main effect for the adjective
'annoying' (F = 5.02, df 2/36; p <. 0 5)- Subjects in both
the high volume, background chant condition and partici
patory chanting condition (means = 6.57 and 6.76,
respectively) thought the chant was more annoying to the
'trainee' than subjects in the low volume, background
chant condition (mean = 3.33).
The analysis of the adjective 'exciting' showed a
significant interaction between the variables (F = 3.91;
df 2/36; £ <.05). Again, the results paralleled the
behavioral data. Friends in the participatory chanting
condition and strangers in the high volume, background
chant condition thought the chant was more exciting to the
'trainee' than in the other conditions.
43
Feelings toward the experiment. Three items
assessed subjects' general reactions to the experiment.
There were no differences between experimental conditions
on two of these measures. Specifically, subjects
indicated that they disliked rather than liked providing
opposition (mean = 8.58) and felt more unjustified than
justified about their behavior (mean = 5.14). Thus, there
was some tendency for subjects to derogate the experiment.
However, a significant interaction appeared for
the third item (F = 3.44; df 3/48; £ <.05), where subjects
were asked if they would participate in a similar
experiment to this one again. Although in general
subjects who had responded the most (thrown the most
balls), also indicated that they would be more willing to
participate in a similar session, correlational analyses
were again statistically insignificant. Thus, this
provides only suggestive evidence that subjects may have
been trying to rationalize their uninhibited behavior
with their expressed willingness to do it all again.
Rating of the 1 trainee1. Subjects gave their
impressions of the 'trainee' by rating him on five
personality dimensions. There were no experimental
effects for any of these measures. On the whole, subjects
perceived the 'trainee' more favorably than unfavorably
(overall sum score mean = 14.51).
FOOTNOTES
6In all analyses to be reported, an alpha level
of .05 was considered to be statistically significant.
^Since specific predictions were made concerning
the results and direction of differences, analyses of
the behavioral measures are 1-tailed tests.
®The pooled error estimate from all subjects in
the experiment was used for the separate trend analyses
of friends' and strangers1 groups.
^The change in df in these analyses is accounted
for by the elimination of subjects in the no chant
conditions. Thus, the analyses were based on a reduced
2 x 3 design.
44
CHAPTER V
DISCUSSION
Before the actual results of this experiment will
be discussed, there is the necessity for establishing
exactly what type of behavior was, in fact, observed.
The first question concerns the role-playing nature of
the procedure. As with any experiment done in the
laboratory, there is always the matter of generalizability
to 'everyday' behavior. However, in the special case of
the role-playing situation, some social psychologists
feel that generalizations would be even more difficult
since role-playing experiments lack the realism and
spontaneity of behavior that may be obtainable in a
deception experiment (e.g., Aronson & Carlsmith, 1968).
Although this criticism may be valid for role-playing
experiments where subjects are merely passive observers
or reporters, it loses strength when applied to
situations where the role-player is an active participant.
According to Mixon (1972), the active role-playing
situation does allow the individual the opportunity to
engage in spontaneous speech and behavior, thereby
providing the possibility of 'realism'. The few large-
scale role-playing experiments that have been conducted
45
46
would seem to support this contention (e.g., "The
Grindstone Experiment", Olson & Christiansen, 1966; the
Prison Study conducted by Zimbardo, 1973). Mixon (1971)
further states that "there is no way to assess the
relationship between a role-playing experiment and a
conventional experiment unless one performs the con
ventional experiment [p. 27]." Similarly, at this stage
in the development of social psychology, a direct
relationship between any laboratory experiment and
'everyday' life is also indeterminable. Seen in this
context, it would seem that if the role-playing experiment
allows the subject to be an active participant, behaving
spontaneously, the question regarding role-playing as an
empirical method reduces to the question of general-
izability applicable to any laboratory experiment. Since
the present experimental procedure did allow subjects
some freedom of expressive behavior (both physically and
verbally), this author would agree with Mixon's (1972)
statement that "role playing results need not be
equivalent to anything in order to be interpreted
[p. 153]." That is not to say that eventual
generalizations would not be optimal. However, since a
sweeping generalization is not within the scope of this
paper, this author does feel that the results of the
experiment may be discussed in and of themselves.
47
A more important question to be addressed, though,
is the following: was it aggressive behavior (or the
role-playing of aggressive behavior) that was actually
being studied? There is some suggestive evidence to
support this. First of all, the experimental rationale
essentially asks subjects to role-play aggressive
behavior. They are specifically told that they are to
provide opposition and the emphasis is on behaviors that
occur in crowd settings— with mention being made of
passive resistance, hecklers, violence and chanting in
crowds. Further, the topography of the response is one
that would objectively be labeled 'aggressive" and some
writers maintain that this is sufficient reason to label
a response aggressive (e.g., Goranson, 1970).
Additionally, subjects were told that the 'trainee' would
gain money if he was effective in withstanding their
\
opposition. Thus, by flinging balls at him, there was a
chance that they could prevent him from attaining that
goal and in this sense, they were harming the victim.
Some of the questionnaire data seemed to indicate that
certain justification techniques were used. For
example, subjects tended to derogate the experiment,
indicating that they disliked providing the opposition
and felt unjustified in doing so, and underestimated the
number of balls they had thrown. This latter response
48
could have been an attempt to camouflage their actual
behavior. After all, for all they knew at that time,
their answer to this question might have been the only
means the experimenter had of assessing their behavior.
(Subjects did not know the experimenter would spend the
period after debriefing counting balls.) And, of course,
there would be no need to employ such justification
techniques if the behavior in question did not warrant
it. In sum, the above arguments lend some support to the
view that aggressive behavior did occur.
However, there is one problem with accepting this
conclusion; that is, there is no indication of subjects'
intent to harm the target. Instead, it is possible that
subjects perceived the situation as an opportunity to
provide 'helping' behavior, where helping is defined as
"the giving of assistance or aid toward a definite object
or end [Wispe, 1972, p. 4]." Since subjects had been told
that the passive resistor they were to oppose had been
engaged in a training program, their opposition would be
helpful in determining how effectively the resistor could
withstand opposition, how effective the training program
had been, etc. Furthermore, besides helping the
resistor practice his passive resistance, subjects were
also potentially helping the experimenter. Again, how
ever, there is the problem of intent and the motivation
49
behind the behavior, an issue which is as relevant to
helping behavior and altruism as it is to aggression
(Krebs, 1970). In other words, there does not seem to
be any conclusive way, given the information and data at
hand, to determine whether subjects perceived themselves
to be behaving aggressively or helpfully (or both) in this
experiment. On the one hand, if one set of features of
the experimental rationale and procedure were salient,
subjects may have been acting primarily in terms of an
aggressive motivation (for example, their behavior might
cause the target to lose money? they were role-playing a
heckler; and providing opposition as in a crowd). On the
other hand, subjects may have focused on those aspects of
the rationale which implied that they were being helpful
(for example, helping the target practice his resistance?
helping the experimenter assess the training program).
However, given that this information regarding
subjects' intent to help or harm is not available, there
is another way to view the behavior that was expressed in
this experiment which does not necessitate knowledge of
intent. Specifically, the behavior in question and the
situation in which it occurred, could be seen as 'de
individuated1 behavior. According to Zimbardo (1969)
(see Footnote 3), 'deindividuation' refers to a process
which includes antecedent social conditions, inferred
50
subjective changes and output behaviors. Some of the
antecedent conditions of 'deindividuated1 behavior (as
outlined by Zimbardo, 1969, p. 14) that were present in
this experiment include: responsibility diffusion (all
group members were responsible for the activity and
behavior that ensued), arousal (the background and
participatory chants), physical involvement in the act
(ball-throwing), and a novel or unstructured situation.
These antecedent conditions presumably lead to changes
in self-perception and perception of others. Given that
all subjects were asked to perform the same response, one
could assume that concern regarding social evaluation
would be minimal once responding had begun. Further, this
evaluation concern would likely be less among friends, as
suggested by prior research (Zabrack & Miller, 19 73).
This, according to Zimbardo, should lead to a "lowered
threshold for expressing inhibited behaviors [p. 14]."
Ball-throwing, chanting and heckling could certainly be
termed 'normally restrained behavior,' especially among
females. Thus, the specific experimental results that
were obtained will now be discussed as representing
'deindividuated' behavior.
First of all, the outcome of this experiment
provided supporting evidence for previous findings
(Geen & O'Neal, 1969? Knipmeyer & Prestholdt, 1973) which
51
indicated that generalized arousal, in the form of back
ground sound stimulation, would lead to increased
responding. In addition, the results of this experiment
suggest that as arousal increases quantitatively,
responding will also increase.
However, of even more interest to this investi
gator, was the fact that the results also supported the
predictions concerning differential responding from groups
of friends and strangers to at least one of the arousal
levels. Both friends and strangers reacted similarly to
the first three levels of arousal, which represented the
quantitative dimension of the arousal manipulations. That
is, there were no significant differences between the two
types of groups in terms of how many balls they threw at
the experimental victim ('trainee')• As arousal increased
over these three conditions, the amount of behavior also
increased. However, at the last arousal level, where
subjects were given the opportunity to participate in the
chanting, there were striking differences between the
responses of friends and strangers. Friends continued to
throw more balls and the number thrown in this condition
exceeded the number thrown in the high volume, background
chant condition. Strangers, on the other hand, became
decidedly inhibited in their behavior, throwing a
significantly fewer number of balls than friends did and
52
a number that was less than they had thrown in the high
volume, background chant condition.
The results obtained from analyzing the 'content'
score closely paralleled the above behavioral results and
shed further light on the processes involved. The
'content' score was designed to be a measure of
spontaneous, impulsive behavior; the more types of
behavior emitted and the higher degree of spontaneity
these responses represented, the higher the 'content'
score assigned to the group. Friends obtained the
highest score on this measure in the participatory
chanting condition. Thus, not only were friends most
uninhibited in this condition in terms of ball-throwing
but also in terms of spontaneous behavior. The
participatory chanting condition seemed to present to
friends the opportunity to be most impulsive and most
'deindividuated'. Yet this condition had totally
opposite effects for strangers, who obtained the lowest
'content' score in this condition. Apparently the high
volume, background chant condition was the one wherein
strangers could express 'deindividuated' behavior, both
in terms of ball-throwing and the emission of spon
taneous, impulsive responses.
These results had been predicted based on the
assumption that social support processes are strong and
53
salient among groups of friends. Social support seems
to function in a way that reduces anxiety and discomfort,
especially in unusual and perhaps ambiguous situations.
Accordingly, when friends are asked to participate in the
chanting, as well as in ball-throwing, the presence of
social support enables them to respond in an uninhibited,
spontaneous and un-self-conscious manner.
Strangers, however, lack this comfort and
support. Thus, the participatory chanting condition may
have been too uncomfortable, embarrassing or anxiety-
provoking. Without some assured support from others to
alleviate the anxiety and uncertainty that this condition
could potentially impose (such as that based on past
experience and confidence that support would accrue),
strangers would attempt to mitigate the discomfort in
some other way. Inhibition of responding would be one
way to avoid anxiety and embarrassment and in fact, that
is what occurred.
However, in view of this explanation, it is
somewhat unclear why there were no overall differences
in responding between groups of friends and strangers,
especially since differences had been expected based on
previous research. One possible explanation lies in
consideration of the nature of the response. Ball-
throwing is not a response familiar to females in our
54
culture. (Several subjects remarked that they had never
thrown an object at someone or hit someone, did not know
if they could do it, etc.) Given the unusual nature of
the response, it may be that friends were just as in
hibited or uncertain about engaging in the response as
strangers. In previous experiments, subjects 'aggressed'
by pressing levers which administered a noxious stimulus
to a victim, a rather indirect form of aggression (and
one which might be more socially acceptable— pressing
levers is not a response for which females have received
negative social and cultural sanctions; throwing objects
is). What this suggests is that social support processes
may only become salient and operative when the extra
dimension of participatory chanting was added to the
situation, thereby accounting for the difference between
friends' and strangers' groups. Further experimentation
is obviously needed to clarify this explanation; for
example, one of the experiments previously conducted
could be repeated using the ball-throwing response as
the dependent measure.
Much of the questionnaire data also corresponded
to the behavioral data. For example, in conditions-
where subjects (both friends and strangers) threw the
most balls and were most uninhibited, they also indicated
that they would be more willing to participate in a
55
similar experiment again, thought the chant was exciting
to the 'trainee' and also rated the chant as most arousing
and exciting to themselves. All these results seem
fairly consistent.
However, there was one inconsistency between
friends and strangers that would seem to warrant further
experimental investigation. Whereas friends seemed to
delineate the arousing and aversive nature of the chant
(the more arousing they rated the chant, the less
aversive they also rated it, per condition), strangers did
not make this distinction. Instaed, in conditions where
the chant was perceived as most exciting, it was also
perceived as most aversive. The unaswered question this
raises is whether the aversiveness per se was arousing to
strangers.
To summarize: as arousal (defined in terms of
increased volume of a background stimulus) increased,
responding (ball-throwing) increased as well, both for
groups of friends and groups of strangers. However,
when a qualitative change was made in stimulation
(specifically, the requirement to participate in the
chant), friends and strangers seemed to react
differently. Whereas friends behaved even more
impulsively, spontaneously and in an uninhibited manner,
strangers became inhibited. This suggests that the
56
request to chant may have given rise to anxiety,
embarrassment or discomfort (although the ball-throwing
did not), among strangers. Any anxiety or discomfort
that might have arisen among friends seemed to be
alleviated by social support processes.
Even though the results of the experiment provided
some interesting group differences in the study of
spontaneous, impulsive behavior, one major purpose of
this experiment was not adequately studied— that is,
aggressive behavior. Further experimentation should
place greater emphasis and take greater pains to obtain
information regarding subjects' intent and motivation.
Besides more direct and extensive post-experimental
inquiries (either in questionnaires and/or debriefing
sessions), a variation on the present role-playing
procedure might be fruitful. Specifically, the role-
playing situation could be described to subjects in such
a way as to provide subjects with a choice of behaviors.^
Subjects could be told that one role a 'crowd member'
could take is one of opposition. This role should be
chosen if the subject disagrees with the speaker's
(target's) point of view. That is, as in a 'real' crowd,
when one opposes a speaker, one may heckle, use verbal
abuse, etc. On the other hand, the other role available
to subjects is one of merely listening to the speaker.
57
This role should be chosen if the subject agrees with
what the speaker is saying. Thus, by giving subjects a
more viable choice of behaviors, they essentially self-
define their own behavior— as aggressive or non-
aggressive .
Other directions for further investigation might
include the replication of previous experiments with
friends and strangers employing the role-playing
methodology; an extension of the arousal manipulation
along the one quantitative dimension; and exploration of
variation in chant content. For example, what reactions
would be obtained in response to an overtly hostile
chant or to an overtly peaceful chant (such as "Om")?
FOOTNOTES
1(51 thank Professor James Kahan for this
suggestion.
58
BIBLIOGRAPHY
Aronson/ E. & Carlsmith, J. M. Experimentation in
Social Psychology. In G. Lindzey & E. Aronson,
The Handbook of Social Psychology, Volume 2, 196 8.
Baron, R. A. Magnitude of Victim's Pain Cue and Level
of Prior Anger Arousal as Determinants of Adult
Aggressive Behavior. Journal of Personality and
Social Psychology, 1971, 17, 236-243.
________. Aggression as a Function of Ambient
Temperature and Prior Anger Arousal. Journal of
Personality and Social Psychology, 1972, 21, 183-189.
Baron, R. S. Anonymity, Deindividuation and Aggression.
Unpublished doctoral dissertation, University of
Minnesota, December 1970.
Berkowitz, L. Some Aspects of Observed Aggression.
Journal of Personality and Social Psychology, 1965,
2, 359-369..
_ & Buck, R. W. Impulsive Aggression: Reactivity
to Aggressive Cues Under Emotional Arousal. Journal
of Personality, 1967, 35, 415-424.
_ & Geen, R. G. Film Violence and the Cue
Properties of Available Targets. Journal of
Personality and Social Psychology, 1966, 3, 525-530.
Brown, R. Social Psychology. New York; Wiley & Sons,
1961.
Duncan, D. B. Multiple Range and Multiple F Tests.
Biometrics, 1955, 11, 1-42.
Geen, R. G. & O'Neal, E. Activation of Cue-elicited
Aggression by General Arousal. Journal of
Personality and Social Psychology, 1969, 11, 289-292.
Goranson, R. E. Media Violence and Aggressive Behavior:
A Review of Experimental Research. In L. Berkowitz
(Ed.) Advances in Experimental Social Psychology,
Volume 5. New York: Academic Press, 1970.
59
60
Hebb, D. 0. Drives and the C.N.S. (Conceptual Nervous
System). Psychological Review, 1955, 62, 243-254.
Hebb, D. 0. Textbook of Psychology. Philadelphia:
W. B. Saunders, 1972.
Hull, C. L. Principles of Behavior. New York: Appleton-
Century-Crofts, 1943.
________. A Behavior System. New Haven: Yale University
Press, 1952.
Kaufmann, H. Aggression and Altruism. New York: Holt,
Rinehart & Winston, 1970.
Knipmeyer, J. & Prestholdt, J. The Influence of
Environmental Noise on Group Aggression. Paper
presented at the Southeastern Psychological
Association Convention, April 1973.
Krebs, D. L. Altruism— An Examination of the Concept and
A Review of the Literature. Psychological Bulletin,
1970, 73, #4, 258-302.
LeBon, G. Psychologie Des Foules, 1895. Transl. The
Crowd. New York: The Viking Press, 1960.
Malmo, R. B. Measurement of Drive: An Unsolved Problem
in Psychology. Nebraska Symposium on Motivation,
1958, 229-265.
Mixon, D. Behavior Analysis Treating Subjects as Actors
Rather than Organisms. Journal for the Theory of
Social Behavior, 1971, 1, #1, 19-33.
________. Instead of Deception. Journal for the Theory
of Social Behavior, 1972, 2, #2, 145-177.
Olson, T. & Christiansen, M. The Grindstone Experiment:
31 Hours. 1966.
Wallace, J. Psychology: A Social Science. Philadelphia:
W. B. Saunders, 1971.
Walters, R. H. & Parke, R. D. Social Motivation,
Dependency and Susceptibility to Social Influence.
In L. Berkowitz (Ed.) Advances in Experimental Social
Psychology, Volume 1. New York: Academic Press,
1964.
61
Wispe, L. G. Positive Forms of Social Behavior: An
Overview. Journal of Social Issues, 1972, 28, #3,
1-19.
Zabrack, M. & Miller, N. The Influences of Social Support
and Evaluative Fears on the Use of Aversive Control.
Unpublished paper, 1973.
Zajonc, R. B. Social Facilitation. Science, 1965, 149,
269-274.
Zillmann, D. Excitation Transfer in Communication-
Mediated Aggressive Behavior. Journal of
Experimental Social Psychology, 1971, 7, 419-434.
Zimbardo, P. The Human Choice: Individuation, Reason and
Order Versus Deindividuation, Impulse and Chaos.
Nebraska Symposium on Motivation, 1969.
Zimbardo, P. G. The Psychological Power and Pathology of
Imprisonment. In E. Aronson & R. Helmreich, Social
Psychology. New York: D. Van Nostrand Co., 1973.
62
APPENDIX I
MOOD ASSESSMENT
PLEASE INDICATE YOUR FEELINGS, MOOD AND PRESENT STATE
MIND ON THE FOLLOWING DIMENSIONS BY CIRCLING THE APPRO
PRIATE DOT OR X.
X ..............
Tense Neutral Relaxed
X ..............
Sociable Neutral Unsociable
X ..............
Anxious Neutral Secure
X ...............
Happy Neutral Sad
X ...............
Aroused Neutral Unaroused
PLACE A CHECK MARK ON THE LINE THAT CORRESPONDS TO YOUR
ANSWER TO THE FOLLOWING QUESTION:
Do you know any of the other participants in this
experiment?
Neither __________
One __________
Both
#1 Very close friends
Fairly close friends
Acquaintances only
Have seen the individual,
but we've never spoken.
#2 Very close friends
Fairly close friends
Acquaintances only
Have seen the individual,
but we've never spoken.
64
APPENDIX II
VERBATIM INSTRUCTIONS GIVEN
TO SUBJECTS
What a group of us have been investigating for
soem time now is techniques in passive resistance. There
has been a growing feeling among many action groups and
others that passive resistance techniques would be an
effective means of handling opposition from both police
and hecklers. This has become so primarily because of
increasingly negative public opinion about violence and
because of the fear of stringent sanctions by authority
figures if violence does occur.
So we've gathered together a group of volunteer
students from across campus who were interested in
learning passive resistance techniques and they've been
involved in a training program since the summer under the
direction of one of the other investigators involved.
We've reached the stage in our program where we're ready
to have our trainees face some tests and let them
practice the passive resistance techniques that they've
learned.
Now, our plan is to work into a real-life
situation gradually. In other words, before sending our
trainees out into the real world, we want to have some
65
short, intensive, induction sessions here in the
laboratory. And that's where you come in. Your role is
going to be to try and make our trainee lose his
composure in the shortest time possible. He will be
coming in here and reading a short speech on censor
ship— it's about two minutes long. You have each been
provided with a box of styrofoam balls with which you
can provide physical opposition. Now, we've found that
this particular trainee seems fairly immune to verbal
abuse, that is, it does'nt seem to have much effect on
him. You can use verbal abuse if you think it will help
you get into it, but we're primarily interested in how he
deals with the physical opposition.
In addition, to ensure that our trainees are
motivated to try during these sessions, they've been
promised a sum of money. The more effective he is in
withstanding your opposition, the more money he stands to
earn.
We've also tried to add a couple of features to
the situation to try and simulate in some way a crowd and
its mood, despite the fact that there are only three of
you and we're in a laboratory.
(Friends)
One feature is thefact that the three of you are
friends. It is often the case that crowds consist of
66
smaller groups of friends, or, at least individuals who
have similar attitudes, opinions, experiences, etc.
Having glanced at your questionnaires, it looks as if
you are all feeling similarly anyway.
(Strangers)
One feature is the fact that the three of you
don't know each other. It is often the case that crowds
consist of different types of people who happen to con
verge and then, in fact, become a crowd. Having
glanced at your questionnaires, it looks as if you are
all feeling differently so probably represent a good
cross-sample.
(Background chant conditions)
Secondly, while the three of you can provide a
fair amount of physical opposition with the balls, I'm
going to play a tape in the background to provide some
auditory opposition as well. The tape consists of a'
group of people chanting a repetitive phrase.
Chanting is something that often does occur in crowds.
(Participatory chanting condition)
Secondly, while the three of you can provide a
fair amount of physical opposition with the balls, I'd
also like you to provide some auditory opposition by
chanting along with a tape I'm going to play. The tape
consists of a group of people chanting a repetitive
phrase— the words of which are "two, four, six, seven."
Chanting is something that often does occur in crowds,
so I'd just like you to chant along.
Now, I will be in the room during the session
rating the trainee so once we begin you won't be able to
ask me any questions or talk to me. I'd also like to ask
you not to move past the boxes during the session. Are
there any questions before we begin?
68
APPENDIX III
'TRAINEE'S1 SPEECH
Censorship should definitely be used in many of
the media since it effectively reduces crime rate and
pornography. Not only is the public too easily swayed
by what it sees on TV and in the movies and by what it
reads, but many individuals are strongly influences by
pornographic movies and literature. The result has been
an increase in sex crimes. For instance, where else do
young adolescents and juveniles get ideas for activities
that will provide them excitement and sexual release?
It is far too easy to obtain pornographic literature and
to sneak into x-rated films and the result is inevitably
negative for the many innocent victims of sex crimes in
this country.
An obvious and practical solution is censorship.
I realize that there are already boards that censor and
rate movies but there is a need for far more stringent
control. Certain films should be totally banned, not
just forbidden to minors. Furthermore, magazines and
books should be subject to strict censorship rules too.
Areas in big cities that have become known for the
availability of 'peep shows' and pornographic literature
have become crime-ridden, dangerous and peopled with
69
degenerates. A first and easy step to take in cleaning
up the major cities would be to close down these movie
houses and book stores. It seems to me that these
measures, imposing stricter censorship rules on several
of the media and closing down the establishments that
peddle pornography, can only be beneficial to the
public at large. Censorship aids in the elimination of
pornographic material which, in turn, should reduce crime
rates substantially.
Several specialized censorship boards could be
set up— one to censor movies, one to censor books, one
for magazines and so on. There are too many crimes
committed in our country which could easily be avoided
with the application of control measures. So, in my
opinion, censorship boards should be set up as soon as
possible in order to rid the country of pornography.
This would be at least one effective measure to take to
help reduce crime rates, an issue of vital concern to all
responsible citizens.
APPENDIX TV
ASSESSMENT QUESTIONNAIRE. FOR PASSIVE
RESISTANCE TRAINING— PROGRAM AV 101
A. PLEASE INDICATE THE REACTIONS AND FEELINGS YOU
EXPERIENCED DURING THE INDUCTION SESSION ON THE
FOLLOWING DIMENSIONS:
X ........
Tense Neutral Relaxed
X ........
Excited Neutral Not excited
X ........
Unenjoyable Neutral Enjoyable
X ........
Unaroused Neutral Aroused
X ........
Happy Neutral Sad
X ........
Comfortable Neutral Uncomfortable
X ........
Peaceful Neutral Angry
X ........
Nervous Neutral Calm
X ........
Stimulated Neutral Not stimulated
X . .
Hostile
. X . .
Neutral
X
Nonhostile
71
B.
1. Do you think the individual you saw was effective in
resisting your opposition?
X ......... X ••••••••• X
Very effective Moderately Not at all
effective effective
2. Rate your perceptions of the opposition you provided
on the following scales. The opposition was:
X .......
Very annoying
. . X ........
Moderately
annoying
. X
Not at all
annoying
X
Very
• • • • • • •
unpleasant
. . X ........
Moderately
unpleasant
. X
Not at all
unpleasant
X
Very
• • • • • • •
frustrating
• . X ........
Moderately
frustrating
. X
Not at all
frustrating
X
Very painful Moderately
painful
. X
Not at all
painful
X.......
Very tension-
producing
. . X ........
Moderately
tension-producing
. X
Not at all
tension-
producing
X .
Very
! • • • • • •
disruptive
. . X ........
Moderately
disruptive
. X
Not at all
disruptive
3. Can you remember approximately how many balls you
threw?
(Fill in)
72
4. Rate the chanting in the background on the following
dimensions as it seemed to affect the passive
resistor.
X • • . . .
Very annoying
. X . . . .
Moderately
annoying
X
Not at all
annoying
X ........
Very pleasant
• X • • •
Moderately
pleasant
X
Not at all
pleasant
X • i
Very provoking Moderately
provoking
X
Not at all
provoking
X .....
Very exciting
. X . . . ,
Moderately
exciting
X
Not at all
exciting
X ......
Very irritating
. X . . . .
Moderately
irritating
X
Not at all
irritating
5. Rate the chanting in the background as it affected
you.
X .....
Very annoying
. X . . . *
Moderately
annoying
X
Not at all
annoying
X ..........
Very pleasant
. X . . . i
Moderately
pleasant
X
Not at all
pleasant
X ..... t
Very provoking
. X . . . ,
Moderately
provoking
X
Not at all
provoking
X ........
Very exciting
. X . . . <
Moderately
exciting
X
Not at all
exciting
X ......
Very irritating
. X . . . i
Moderately
irritating
X
Not at all
irritating
73
6. To what extent do you feel your behavior was
anonymous from the experimenter?
X .................X .................. X
Completely Moderately Not at all
anonymous anonymous anonymous
7. What were your feelings about providing opposition?
X .................X .................. X
Liked it Neither liked it Disliked it
very much nor disliked it very much
8. How would you describe your feelings toward the
other people who provided opposition with you in
this session?
X ..................X ................. X
Like them Neither like them Disline them
very much nor dislike them very much
9. How cohesive or close did you feel as a group?
Xa*»**>ataX»«t*»a*«aX
Very cohesive Moderately Not at all
cohesive cohesive
10. How justified did you feel in providing opposition?
X ......... X ......... X
Very justified Moderately Not at all
justified justified
11. Would you participate in a session similar to this one
again if given the chance?
X .................. X .
Definitely would Maybe
X
Definitely
would not
74
C. ACCORDING TO YOUR OBSERVATIONS OF THE PASSIVE
RESISTOR, PLEASE RATE HIM ON THE FOLLOWING
DIMENSIONS:
X ..................X ................. X
Intelligent Average Unintelligent
X ..................X ................. X
Serious Average Frivolous
X ..................X ................. X
Hostile Average Friendly
X ..................X ................. X
Unlikeable Average Likeable
X ..................X ................. X
Nervous Average Calm

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Diagnostic indices of hyperactivity and an investigation of verbal mediation training as an alternative to pharmacotherapy

PDF

The Subjective Utility Function As An Estimator Of Optimal Test Weights For Personnel Selection

PDF

Changes In Memory As A Function Of Age

PDF

Factorial Stability As A Function Of Analytical Rotational Method, Type Of Simple Structure, And Size Of Sample

PDF

The Relationship Of Multidimensional Scaling Spaces Of Trait Adjectives For Different Reference Persons

PDF

Age, Sex, And Task Difficulty As Predictors Of Social Conformity: A Search For General Tendencies Of Conformity Behavior

PDF

The Effects Of Repression-Sensitization, Race, And Levels Of Threat On Extensions Of Personal Space

PDF

The Effect Of Nonspecific Factors On The Modification Of Smoking Behaviorin Treatment Follow-Up

PDF

A Multiple Investigation Of Child-Rearing Attitudes

PDF

Differences Between Cues In Effectiveness As Retrieval Aids

PDF

Personality Variables And Intellectual Abilities As Determinants Of Concept Learning

PDF

Weight Reduction As A Function Of The Timing Of Reinforcement In A Covertaversive Conditioning Paradigm

PDF

Ucr Diminution As A Function Of Isi

PDF

Classical Discrimination Conditioning As A Function Of Probability Of Reinforcement

PDF

Death Anxiety In Leukemic Children

PDF

Influences Of Interoperative Experience And Age On Recovery Of Visual Function Following Two-Stage Lesions Of The Striate Cortex

PDF

The Psycho-Social World Of The Leprosy (Hansen'S Disease) Patient: Assessment Of Community Reaction To The Disease And Patient Psychologicalfunctioning

PDF

Modality and serial position effects in immediate and delayed recall

PDF

The Effect Of Exercise Training On Systolic Time Intervals In Elderly Men

PDF

Hypnotic Susceptibility, Achievement Motivation, And The Treatment Of Obesity

Asset Metadata

Creator Zabrack, Merle Linda (author)

Core Title Aggression As A Function Of Arousal And Friendship Ties

Degree Doctor of Philosophy

Degree Program Psychology

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag OAI-PMH Harvest,psychology, general

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Miller, Norman (committee chair), Kahan, James Paul (committee member), Warren, Robert (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-727737

Unique identifier UC11356670

Identifier 7426060.pdf (filename),usctheses-c18-727737 (legacy record id)

Legacy Identifier 7426060.pdf

Dmrecord 727737

Document Type Dissertation

Rights Zabrack, Merle Linda

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA