Citation
Manipulation of attributional outcomes

Material Information

Title:
Manipulation of attributional outcomes when focus engulfs the field
Creator:
Houser, Jeffrey A
Publication Date:
Language:
English
Physical Description:
x, 122 leaves : forms ; 29 cm

Subjects

Subjects / Keywords:
Attribution (Social psychology) ( lcsh )
Behavioral assessment ( lcsh )
Attribution (Social psychology) ( fast )
Behavioral assessment ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references.
General Note:
Submitted in partial fulfillment of the requirements for the degree, Master of Arts, Sociology.
General Note:
Department of Sociology
Statement of Responsibility:
Jeffrey A. Houser.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
28286346 ( OCLC )
ocm28286346
Classification:
LD1190.L66 1992m .H68 ( lcc )

Full Text
MANIPULATION OF ATTRIBUTIONAL OUTCOMES
WHEN FOCUS ENGULFS THE FIELD
Jeffrey A. Houser
B.A., University Colorado at Denver
A thesis submitted to the
Faculty of the Graduate School of the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Master of Arts
Sociology
1.992

1


This thesis for the Master of Arts
degree by
Jeffrey A. Houser
has been approved for the
Department of
Soci ology
by
(c-lo-IZ.
Date


Houser, Jeffrey A. (M.A., Sociology)
Manipulation of Attributional Outcomes:
When Focus Engulfs the Field
Thesis directed by Professor W. I. Griffith
ABSTRACT
This study investigates whether a perceptual
focus alters current findings in attribution theory.
Research has indicated that cognitively busy person
perceivers and extraverted individuals are likely to
make dispositional attributions. This study
challenges these findings by arguing that an
observer's focus plays a larger role in
attributional outcomes than either the observer's
level of busyness or his/her personality type.
An experimental design is used to manipulate
the focus of subjects. Findings indicate that
"focused" subjects do remember more social
information than "non-focused" subjects, but this
relationship does not carry over into the area of
attributional decisions.
Explanations are offered which attempt to
provide insight into possible methodological errors
or theoretical oversights. Given the lack of


findings, an overview of possible adjustments is
given to aid future researchers.
This abstract accurately represents the content of
the candidate's thesis. I recommend its
publication.
Signed
IV


ACKNOWLEDGMENTS
I wish to express my appreciation and thanks to
Dr. W. I. "Lou" Griffith for her advice and guidance
in this study.
Thanks is extended to Dr. Kjell Tornblom and
Dr. Gary Stern for their assistance and help. And,
finally, a special appreciation is given to my wife,
Kathy, for her patience, support and proofreading.
Thanks also to my parents for the computer.
v


CONTENTS
Figures............................................j.x
Tables.............................................x
CHAPTER
1. INTRODUCTION.......................1.......1
Purpose of the Study.................... 1
Arrangement of the Thesis............... 3
2 REVIEW OF THE LITERATURE.................. 5
Attribution Theory...................... 5
Attributional Errors............... 6
Cognitive Busyness.....................J.3
A Perceptual Focus Response
to Cognitive Busyness..............20
Hypothesis #1:
Perceptual focus..............21
Ext raversion/Int rovers ion............23
Field Dependence..............24
Self-Monitoring...............25
Transsituational
Predictability................27
A Perceptual Focus Response
to Extraversion/Introversion............28
Hypothesis #2:
Reward Contingency.................29
Summary of Theoretical Rationale........30
vi


3. METHODOLOGY.............................34
Design Of The Study.....................34
Subjects...........................35
Materials.36
Procedure. 3 6
Dual Task.....................3 6
Operationalization of Hypotheses........43
Hypothesis #1.......................43
Hypothesis #2..................... .46
4. RESULTS....49
Correlation Matrix.......................52
Manipulation Check of
Hypothesis #1............................54
Findings of Hypothesis #1................37
Findings of Operational
Hypothesis #1.......................37
Manipulation Check of
Hypothesis #2............................58
5. SUMMARY, CONCLUSIONS, AND IMPLICATIONS....68
Summary..................................68
Conclusions..............................70
Implications..............................7 6
APPENDICES
A Mathematical Equations..........................83
Vll


B Transcript Of Social
Information Tape..............................85
C Instruction Letter............................93
D Social Recall Questions.......................95
E Houser Attributional Inventory................97
F Debriefing Letter............................106
G Coding of Variables for
for Computer Analysis........................110
NOTES.............................................112
REFERENCES........................................115
viii


FIGURES
Figure
2.1 Model of the Study
14
IX


TABLES
Table
4.1 Correlation Matrix............................53
4.2 One Way ANOVA for the Manipulation
Check of Hypothesis #1.......................58
4.3 Regression Analysis of
Hypothesis #1 60
x


CHAPTER ONE
INTRODUCTION
It is our needs that interpret the world. Nietszche
Purpose of the Study
The idea that motives can influence and distort
the way in which events in the world are perceived
is not new. It is, however, a view that attribution
theory has neglected (Bains 1983) To be sure there
has been a robust interest in attribution errors and
biases. However for the most part, these have
tended to be interpreted in cognitive, information-
processing terms. In part, this neglect of
motivational influence has been a reflection of a
general tendency in social psychology. But ignoring
a motivational influence also arises because the
attributional process tends to be seen as an end in
itself and rarely as a tool used by people for
certain ends. If people have reasons other than
intellectual curiosity for wanting to explain
events, then they are likely to find certain answers
more satisfying than others. If this is the case,
then perhaps the objectivity and purity of the
1


attribution process has inevitably been eroded
(Bains 1983). If 'man is a scientist,1 the desire
to hold a particular view may lead to a high
selectivity in data collection and interpretation.
Attribution provides a greater sense of control
of the environment through the understanding of
cause and effect relationships (Heider 1958, 1976;
Kelley 1971; Pittman and Pittman 1980). The more
knowledge one has of causes and effects, the more
powerful is his/her ability to attribute. However,
while knowledge itself is power, it is important to
recognize that certain forms of knowledge imply
greater power than others.
Various researchers (Gilbert and Krull 1988;
Heider 1958; Jones and Nisbett 1971a; Ross 1977)
have established that for the most part, individuals
tend to view internal factors as underlying other
people's behavior. Ross (1977) coined this
phenomenon the "fundamental attribution error" (p.
234). Investigations into this "error" have led to
several explanations for its cause. This study will
attempt to examine two of these explanations in
2


hopes of evaluating their relative worth in terms of
perceptual focus.
Attribution has long been considered an end-all
analysis. Subsequently, little consideration has
been given to whether or not the attributional
process can be manipulated. Can certain
arrangements make it more or less likely that an
individual will make either a dispositional or a
situational attribution? If the answer is yes, then
how might this knowledge alter our understanding of
the attributional process? This study was designed
to test these ideas and perhaps offer additional
insight into the process of attribution.
Arrangement of the Thesis
The arrangement of this thesis is as follows:
Chapter Two provides a discussion of the various
theoretical orientations as well as a review of the
literature on attributional biases. Chapter Two
culminates in a set of empirical hypotheses.
Chapter Three outlines the methodological format
used to operationalize and test the empirical
hypotheses. Chapter Four reports the findings of
3


testing the operational hypotheses. Chapter Five
presents some conclusions and implications which can
be drawn from the findings as well as
recommendations for further exploration.
4


CHAPTER TWO
REVIEW OF THE LITERATURE
Attribution Theory
Recently much of the focus of person perception
has been centered in the formal area known as
attribution. First proposed by Fritz Heider (1958),
attribution theory attempts to place parameters upon
the schema used by "naive" scientists when
determining the cause behind an action. Building on
the work of G. A. Kelly (1955) Heider presented the
need to attribute causality for behavior as the
primary goal of "man the scientist" (p. 4). This
ability to attribute causality, enables a person to
predict and control her/his environment.
According to Heider and others (Heider 1958;
Schank and Abelson 1977; Zajonc 1980) attribution is
defined as the inference an observer makes about
either the internal state of an actor (including
self) or the external state of the environment on
the basis of a witnessed overt behavior.
Attribution is the process by which behaviors are
5


explained and interpreted. The attribution process
is seen as a fundamental activity that enables
individuals to create organization from chaos.
Attributional Errors
The attribution process leads, in turn, to the
crucial distinction between internal and external
causation (Jones and Nisbett 1971b). Internal
causes are factors (dispositions) within the person
which account for the witnessed behavior. For
example, if one witnessed a child buying a chocolate
ice cream cone, an internal disposition would assume
the child did so because he/she liked chocolate.
External causes are factors which lie outside the
person (within the environment) to which an observed
action can be attributed. Knowing that the soda
shop was out of vanilla might lead one to make an
external attribution about why the child bought a
chocolate ice cream cone. Understanding which set
of factors to use when interpreting behavior makes
the perceiver1s world more controlled and
predictable.
6


If the purpose of attribution is to render the
world manageable, controllable and predictable
(Heider 1958, p. 80), it would seem to follow that
accuracy in perception would be a desired goal.
Research, however, has shown that accuracy in
attribution is not always the case. In fact, many
attribution studies point out the shortcomings and
biases which function within person perception
(Carver, DeGregorio, Gillis 1980; Fields and Schuman
1976; Forgas, Bower, Krantz 1984; Howard 1984; Ross
1977; Ross, Green, House 1977; ; Snyder, Stephan,
Rosenfield 1976; van der Pligit 1984).
While investigating one of these attributional
biases, Ross (1977) coined the term the "fundamental
attribution error" (p. 234). This error was
described as an observer's tendency to overestimate
underlying dispositions or internal traits, while
underestimating the environmental or situational
constraints. In other words, the fundamental
attribution error occurs when a dispositional
attribution is made when in fact a situational
attribution should have been made.
7


Many studies have attempted to explain why this
systematic bias occurs (Fincham and Jaspers 1980;
Jellison and Green 1981; Jones and Nisbett 1971a;
Storms 1973). The earliest explanations can be
traced back to Heider in his statement, "behavior
engulfs the field" (1958, p. 54). Behavior was seen
to have such salient properties that it stood out
against the background of the situation. To Heider,
perceiving the action of a person as the origin of
an event was the simplest and most satisfying
explanation for the event's occurrence. As such,
Heider described "persons as the prototype of
origins" (1944, p. 360). Heider's thoughts in this
area set the stage for research which investigated
why internal attributions were preferred over
external.
Further investigations into attributional
errors led Storms (1973) to note that an actor-
observer bias existed within the attribution
process. Actors and observers were said to have used
different information sets to arrive at different
conclusions (Jones and Nisbett 1971a). As an actor,
individuals knew those situational forces which were
8


constraining their behavior. Therefore the actors
were able to make situational attributions. For the
actor the environment captured his/her attention and
cause. For the observer, the actor's behavior was
the major focus of attention and therefore attracted
causal explanations. Consequently, observers made
more dispositional attributions. This work led to
an understanding that attributional focus
(dispositional versus situational) was a major
determinant for the type of attribution to be made.
Several other explanations have been posited to
account for the fundamental attribution error.
Jones (1979) suggested that observers were too ready
to assume that any single witnessed behavior was
representative of an actor's entire repertoire of
behaviors. This was particularly true if the
behavior was something the observer was not sure
she/he would have done in the same situation.
Another explanation suggests that the
fundamental attribution error stems from shared
norms in our culture favoring internal attributions
over external attributions (Fincham and Jaspers
1980; Jellison and Green 1981). A third explanation
9


holds that when observers judge a peer's behavior,
they recognize individuals act differently in the
same situation. This being the case, the observers
assume that any particular individual's behavior
reflects that individual's personal disposition
(Higgins and Bryant 1982).
Recent work in the area of attributional biases
has advanced two new perspectives of why person
perceivers tend to overestimate dispositions and
underestimate situational constraints. These new
perspectives fall under the headings: (a) Cognitive
busyness and (b) Personality traits. It was within
these new perspectives that the present study was
conducted.
The first perspective of interest to the
present study was coined "cognitive busyness" by
Gilbert and his associates (Gilbert, Pelham, Krull
1988, p. 733). The notion of cognitive busyness
implies that active* and passive2 observers differ in
their ability to assess social information
accurately. Gilbert et al.'s finding indicated that
active observers made more attributional errors than
10


their passive counterparts. (This will be
elaborated on page 12.)
The second perspective suggests that dissimilar
personality types may attribute differently. Work
in this area has indicated that extraverted
individuals, as opposed to introverts, are more
likely to make dispositional attributions
(Berscheid, Graziano, Monson, Dermer 1976; Collins,
Martin, Ashmore, Ross 1973; Witkin, Lewis, Hertzman,
Machover, Meissner, Wapner 1954) Investigations
into this phenomenon indicated extraverts were more
interpersonally focused than introverts (Witkin and
Goodenough 1977) Findings also reported that
extraverts believe people to be consistent across
situations (Snyder 1979) (This will be elaborated
on page 21.)
This author is hesitant to accept the findings
of the last two perspectives. He believes that the
findings in these studies are subject to harsh
limitations. Therefore an attempt will be made to
modify these findings in these perspectives through
the use perceptual focusers. It will be this
study's mission to interpret whether a perceptual
11


focus (as manipulated by rewards and punishments)
affects the findings indicated in the works of
Gilbert et al. (1988) and Berscheid et al. (1976).
In other words, will cognitively busy observers who
are focused on person perception still be
predisposed to making dispositional attributions?
And will extraverts who are given a perceptual focus
still show a tendency to attribute observed behavior
to an internal cause? The question to be asked is:
Would the effect of a perceptual focus modify
certain types of individual1s tendencies toward
making dispositional attributions?
It should be noted here that the dependent
variable in this study is not a fundamental
attribution error. The present study was not
designed to assess attributional errors, it was
however, designed to assess attributional outcomes.
Work in the area of attributional errors does offer
insight into an observer's attributional
predispositions. Findings surrounding the
fundamental attribution error have led to an
understanding that when viewing another person's
behavior, observers tend to look more at the
12


behavior and less at the situation. This
observational focus leads person perceivers to
overestimate dispositions while underestimating
situations when seeking causal answers. In some
cases this may well be an error but, in the present
study the point is moot since no right or wrong
answer can be given in terms of an attributional
outcome.
As an aid to the reader a model of the study
has been represented in Figure 2.1.
Cognitive Busyness
Recent work in the area of attribution expanded
the role of the observer from that, of passive to
active (Gilbert and Krull 1988). Gilbert and Krull
(1988) criticized most of the work in attribution
for being "unrealistic" (p. 193). These authors
maintained that subjects isolated in a laboratory
did not accurately represent person perceivers.
Laboratory subjects usually are only involved in
performing a single task--attributing behavior. As
such, these subjects were described as passive
observers.
13


Figure 2.1
Model of the Study-
Previous Works
EXTRAVERT INTROVERT
Field-
dependent
High Self
monitors
Field-
independent
Low Self
monitors
Transsituational
predictability
+ Direct relationship
- Indirect relationship
14


Figure 2.1
Continued
View of Present Study
0 R
T E
h s Effects of Cognitive Busyness and
E u Personality Types Negated by
s l Perceptual Focus.
1 t
z s
E
D
15


Gilbert and Krull argued that in the 'real
world' person perception is only one of many tasks
observers perform simultaneously. The example these
authors used was that of a student who met one of
her/his professors at a social gathering. The
student's perceptions of the professor were greatly
different at the party than those originally formed
in the classroom. As a passive perceiver in the
classroom, the student was able to observe the
professor without concern for her/his own self
presentation. At the party, however, the student
became an active perceiver as she/he was busy
managing her/his own impression, predicting the
professor's behavior, and evaluating other courses
of action. Of all the differences between active
and passive perceivers, one was found to be
fundamental: "Active perceivers, unlike passive
perceivers, are almost always doing several things
at once" (Gilbert et al. 1988, p. 733).
How do the complexities of engaging in a social
interaction affect the process of person perception?
Gilbert et al. suggested that the question is
traceable only if one recognizes there is no single
16


process of person perception. Rather, there are
several different processes that together constitute
the act of knowing others. Attempts have been made
to outline the processes observers go through when
making an attribution. Trope (1986) has argued that
person perception has two major components:
behavioral identification (what is the actor doing?)
and attributional inference (why is the actor doing
it?). Behavioral identification involves
categorizing an action, whereas attributional
inference involves causal reasoning about the
categorized act. With only these processes all
causal inquiries would end in dispositional
attributions. The observer perceives the child
buying a chocolate ice cream cone (behavioral
identification). The observer reasons that the
child must like chocolate (attributional inference).
Quattrone (1982) argued that this view of
attributional processes was short sighted. He
suggested that attributional inference may itself be
comprised of two minor components: Perceivers first
draw a dispositional inference about the actor, and
then adjust this inference by taking into account
17


external forces. In short, these perspectives
suggested that person perception consisted of (a)
categorization (seeing the child buy a chocolate ice
cream cone); (b) characterization (deciding the
child likes chocolate); and finally (c) correction
(realizing that the soda shop was out of vanilla).
Gilbert et al. (1988) suggested that
Quattrone's view of.the attributional process was
not entirely correct. They questioned whether the
active perceivers ever reached the correction stage
due to the demands of social interaction. Gilbert
et al. argued that characterization is generally an
overlearned, automatic process that requires little
effort or conscious attention. On the other hand,
these authors perceived corrections to be a more
deliberate, controlled process that uses a
significant portion of the perceiver's processing
resources. They contended that the additional
activities in which active observers engage may
disrupt correction without similarly disrupting
characterization. Thus, active perceivers may be
able to make dispositional inferences from the
behaviors of others. However they are less likely
18


than their passive counterparts to use situational
information to correct these inferences. This lack
of correction results because the demands placed
upon the active observers leave them unable to
correct their inferences. Gilbert et al. termed
this phenomena "cognitive busyness" (1988, p. 733).
In the Gilbert et al. paradigm observers were
'activated' (changed from passive observers to
active observers) by shifting their focus to
something other than person perception. The active
observers were given two tasks to perform. The
first was viewing a video of an individual at a job
interview. The second was responding to a series of
random letters which would appear on the video
screen. Subjects in the low busyness condition
(passive observers) were told to ignore the flashing
letters and just concentrate on the person at the
interview. Subjects in the high busyness condition
(active observers) were told to press a button
placed in front of them each time the letters R, S
and T appeared on the screen, in order to receive a
reward.
19


Gilbert et al. hypothesized that the amount of
busyness incurred by a perceiver would have a direct
effect upon their making a dispositional
attribution. As the busyness of an active perceiver
increased, their likelihood of making a
dispositional or internal attribution would also
increase. The reason behind this was the
diminishing cognitive resources left for person
perception.
In summary, Gilbert and his colleagues argued
that social interaction usurped the cognitive
resources of active observers. Therefore active
observers took the 1 easy way out1 in person
perception by making dispositional attributions
which did not require correction.
A Perceptual Focus Response
to Cognitive Busyness
The present study was designed in part to
examine whether perceptual focus affects cognitive
busyness and its relationship to attributional
outcomes. The question being raised is purely
speculative since no known work asks such a
question.
20


This author believes that Gilbert confused
cognitive busyness with perceptual focus. It is
curious in the present author's mind that Gilbert et
al. never assessed cognitive busyness with subjects
whose focus was on person perception. This study
argues that to those subjects in Gilbert et al1s
high busyness condition, the social information3
presented to them was little more than noise. In
other words, why would a subject put forth the
effort to remember social information, let alone
make causal inferences about some ambiguous actor
when the subject was focused upon gaining rewards?
The question which springs to mind for the present
author is: Are cognitively busy subjects able to
concentrate on person perception when it is
important for them to do so (i.e. when they are
rewarded to do so)?
Hypothesis #1: Perceptual focus. Can the
question of cognitive busyness and its direct
correlation to dispositional attributions be reduced
to one of perceptual 'focus'? The concept of
perceptual focus is not new to the social sciences.
Broadbent (1958) proposed that a person's focus
21


works in a manner similar to a radio receiver. When
the focus of an observer is finely tuned she/he is
able to screen out all other sensory stimuli.
However, when needed the band of attention can be
wide enough to take in multiple sources. In a sense,
focus works as a selective filter screening out
distraction and noise. Broadbent concluded that as
a persons needs for selectivity changes they will
use their focusing abilities to narrow or widen
their attentive band.
The present study proposes that perceptual
focus may prove to be a limitation on Gilbert et
al.1s notion of cognitive busyness. This study
maintains that those subjects in the high busyness
condition had merely restricted their perceptual
focus. Had the subjects in the high busyness
condition been rewarded for social recall as well as
for letter identification, then perhaps they would
have opened up their perceptual awareness to include
situational constraints. In order to test this
assumption a design must be employed in which all
subjects are equally cognitively busy and it is
22


their perceptual focus that is manipulated. The
hypothesis to be tested is as follows:
(1) Given identical levels of cognitive
busyness; active observers who are rewarded for
focusing on person perception will make
significantly fewer dispositional attributions
than active observers who are not rewarded for
focusing on person perception.
Should this hypothesis be correct, it could then be
said that it was the type of perceptual focus that
mediated when a dispositional attribution occurred,
rather than the level of busyness as Gilbert et al.
(1988) seemed to suggest.<
Extraversion/Introversion
The second perspective tested by this study was that
of personality traits. Research in this area has
found that different types of individuals do not
attribute the same (Adorno 1950? Snyder 1979). Of
interest to the present study was work done on the
trait of extraversion-introversion. Findings
23


indicate that extraverts are more likely to make
dispositional attributions than introverts*
(Berscheid, Graziano,. Monson, Dermer 1976) In
order to understand why this was the case, it was
necessary to draw a composite sketch of the
extraverted individual.
Field Dependence
Early studies in perceptual attention noted
that individuals varied in their capacity to
differentiate complex environments (Witkin, Lewis,
Hertzman, Machover, Meissner, Wapner 1954) .
Gardner, Jackson, Messick (1960) followed up on this
work by outlining a typology of the differences
found by Witkin et al. The Gardner et al. typology
consisted of two types of individuals: field-
dependent and field-independent. Field-dependent
individuals were defined as those who make use of
external cues as a guide to their behavior. Field-
independent individuals were defined as being
partially able to rely on internal cues.
Witkin and Goodenough (1977) later expanded
these definitions of field-dependent and field-
24


independent individuals. These authors found field-
dependent people to be more attentive to social
cues, have an interpersonal orientation, and more
socially skilled than field-independent individuals.
To their benefit, field-independent persons were
found to have greater cognitive analysis skills.
Morris (1979) brought this awkward terminology into
the jargon of social psychology by correlating
field-dependent individuals with extraverts; and
field-independent with introverts.
Self-Monitoring
Snyder (1979) expanded the notion of field-
dependence/independence to include aspects of self-
presentation theory. In doing so Snyder (1974,
1980) developed the concept of self-monitoring.
Self-monitoring was defined as modifying one's
behavior to meet the expectations of others. Snyder
concluded that individuals differed in their desire
to control these modifications. He developed a
typology which outlined these differences. High
self-monitors (HSM) were defined as individuals who
vary with the situation. They were represented as
25


social chameleons; always changing to fit their
surroundings. Low self-monitors (LSM) were defined
as individuals who hold true to an internal sense of
self. As such, low self-monitors care little about
changing their self-presentation in accordance with
the environment. Of interest to this study was
Snyder's findings that scores on a self-monitoring
inventory correlated with the personality trait of
extraversion-introversion. High self-monitors
paired with extraverts and low self-monitors paired
with introverts. In terms of attribution theory,
findings indicate that high self-monitors make more
dispositional attributions than-their low self-
monitoring counterparts (Berscheid, Graziano,
Monson, and Dermer 1976). Berscheid et al.
concluded that a high self-monitor1s need to quickly
assess their surroundings in order to 'know' how to
behave facilitated this effect. These authors
speculated that the dispositions of other people
become the barometer by which high self-monitor1s
gauge their own behavior. Due to their
interpersonal focus, it was not surprizing to find
26


that high self-monitors overattribute to
dispositional causes.
Transsituational Predictability
Work by Collins, Martin, Ashmore, Ross (1973)
helped broaden the picture of an extravert. These
authors found evidence that extraverts believed in a
"transituational predictability of behavior" (p.
473). By this they meant that extraverts believed a
person's behavior was predictable across situations.
This belief was found to hold for self as well as
others. What is gained by this work is an
understanding that extraverts feel the actions they
witness are reflective of the actor's internal
dispositions. Given Collins et al.'s findings it is
possible to see why extraverts may ignore
situational constraints thus making an internal
attribution.
The combined works of Witkin et al., Snyder et
al., and Collins et al. completed the composite of
an extraverted individual. Extraverts can be seen
as field-dependent individuals who alter their self-
presentation in accord with the expectations of
27


others (high self-monitors). Additionally,
extraverts believe that a person's actions are
consistent across situations (transsituational
predictability). Each of these aspects of an
extraverted individual's personality were found to
increase their likelihood of making dispositional
attributions.
A Perceptual Focus Response
to Extraversion/Introversion
The present study was designed in part to
determine whether personality differences affect the
number of dispositional attributions people make.
The question this study raised was: Can a
perceptual focus alter the attributional
predispositions of extraverted and introverted
individuals?
Unlike the response made to cognitive busyness
(cf. Houser 1992, p. 17); work has been performed to
investigate the effects of rewards and punishments
in relation to extraversion-introversion.
Personality theorists such as Eysenck (1967) and
Gray (1970, 1972, 1978, 1981) have established
28


extraverts are more concerned with gaining rewards,
while introverts are more interested in avoiding
punishments.6 Unfortunately, the work produced by
these authors is inherently untestable in the
traditions of positivism. Science as we know it
today has no way of establishing whether or not
systems within the cortex are able to create
differences in personality types let alone correlate
them with reward contingencies. At best these
researchers are walking the tightrope of a tautology
when they interpret findings which were arrived at
by empirical hypothetico-deductive methods. Suffice
it to say that the use of this material is purely
speculative at best.
Hypothesis #2: Reward contingency. A summary
of the attributional work on personality types led
to an understanding that extraverts make more
dispositional attributions than introverts. These
works maintained that extraverts were
dispositionally focused. Incorporating the work of
Eysenck and Gray the present study attempted to
modify this relationship. These authors maintained
that extraverts were strongly motivated to seek
29


rewards, while introverts were strongly motivated to
avoid punishments. Combining these two perspectives
the present study was able to test whether a
perceptual focus affected an individual's
attributional predisposition. The empirical
hypothesis generated was as follows:
(2) extraverts who are rewarded when focusing
on person perception will make significantly
fewer dispositional attributions than
extraverts who are rewarded when not focusing
on person perception.
Conversely; introverts who are punished when
focusing on person perception will make
significantly fewer dispositional attributions
than introverts who are punished when not
focusing on person perception.
Although it appears that these are two
hypothesis, in fact they are not. Within each are
the independent variables of perceptual focus and
reward contingency; the blocked variable of
30


introversion-extraversion; and, the dependent
variable of dispositional attributions. Should this
hypothesis be correct, it could then be said that it
was perceptual focus that mediated when an
dispositional attribution occurred, rather than the
personality type as suggested by Witkin and
Goodenough (1977), Berscheid et al. (1976), and
Collins et al. (1973).
Summary of Theoretical Rationale
The rationale behind the present study can be
summarized in the following fashion. Attribution
theory (Heider 1958) set the overall context in
which the study was based. Within attribution,
biases have been recorded which distort the
information received in a social interaction (Storms
1973). One of these biases was termed the
fundamental attribution error (Ross 1977).
Gilbert et al. (1988) proposed this bias may be
due to the cognitive demand placed upon person
perceivers. These authors' concept of cognitive
busyness suggested that active person perceivers
31


made more dispositional attributions than passive
person perceivers.
The present study maintained that Gilbert et
al.'s interpretation was not entirely correct. In
response to Gilbert this study introduced the
concept of perceptual focus. In an attempt to test
this response an empirical hypothesis (cf. Houser
1992, p. 19-20) was generated. This hypothesis
maintained that fewer dispositional attributions
will be made by active observers whose focus is on
person perception as compared to active observers
whose focus is not on person perception.
A second line of thought followed works which
proposed that personality types may play a
significant role in attributional biases (Berscheid
et al. 1976; Collins et al. 1973 ; Witkin and
Goodenough 1977). It was found that extraverted
individuals were more likely to make dispositional
attributions than their introverted counterparts.
In response to these findings the present study
reviewed the works of Eysenck (1967) and Gray (1970,
1972, 1978, 1981) in order to set the stage for the
second empirical hypothesis (cf. Houser 1992,p. 26).
32


This hypothesis proposed that given varied reward
contingencies dissimilar personality types would
differentially focus; and, within this framework,
those subjects whose perceptual focus was on person
perception would make fewer dispositional
attributions than those subjects whose focus was not
on person perception.
In summary, the argument behind this study
contends that person perception can be a motivated
behavior. And, due to differences in focus,
attentional outcomes can be manipulated. It is
hoped that the results of the present study will
increase our understanding of the attribution
process.
33


CHAPTER THREE
METHODOLOGY
Design of the Study
The present study developed issues in line with
formal theoretical analysis in the social sciences.
This study's primary objective was theory testing.
No attempt will be made to generalize any findings
beyond the highly controlled environment of the
laboratory. The focus of this study was more
concerned with internal validity than external
validity. In a study where the focus is on internal
validity, an experimental design is usually selected
as means of testing the hypotheses. The
experimental method allows for greater control than
other methods over relevant variables. The
experimental method also is the best research method
for establishing causal relationships.
The independent variables examined in this
study consisted of: perceptual focus (Social Focus
and Math Focus) and reward contingency (Reward Only,
Punishment Only, and simultaneous Reward and
34


Punishment). The dependent variables are: correct
social recall answers and the number of
dispositional attributions made by subjects.
Additionally, the non-manipulable, naturally
occurring variable of extraversion-introversion, was
treated as a blocked variable. This variable was
assessed in order to examine the differences in
attributional strategies made by alternative
personality types. In methodological jargon the
present study composed a two by three by two
completely randomized factorial, blocked
experimental design (Fisher 1971).
Subjects
Subjects in this study consisted of 143 (87
females and 56 males) undergraduate students aged 19
to 27. Subjects were recruited from introductory
sociology and psychology courses at the University
of Colorado at Denver during the summer and fall
semesters of 1991. Subjects participated in the
study in exchange for extra credit points. Data
from groups of up to six subjects was collected at a
time. The data collection took place within a
35


typical university laboratory setting. Data was
collected between May and September of 1991.
Materials
Materials used in this study included a
questionnaire packet, a tape recording of
biographical data, and a standard tape player. The
questionnaire packet contained: an instruction
sheet; twenty moderately complex mathematical
equations; twenty recall questions; fifteen
attributional assessment items; and the Eysenck
Personali ty Inventory.
Procedure
Dual task. In order to assess the hypotheses
developed by the proposed study, an experimental
design was needed which would address each of the
empirical hypotheses separately and cumulatively. A
dual task design seemed to be the best fit for the
required scenario; one in which extraverted and
introverted subjects were cognitively busy while
engaged in person perception and under a
differential system of perceptual focus.
36


The actual dual task required subjects to
answer a series of moderately complex mathematical
equations (the mathematics equations can be found in
Appendix A), while listening to a taped monologue of
bogus biographical information. Subjects were led
to believe that they were participating in a
psychological experiment where cognitive
strategizing was being assessed (e.g. do individuals
use the same strategy or factors in betting a horse
to show instead of to win). The recording of social
information lasted ten minutes. (See Appendix B for
a transcript of the tape recording.) Subjects were
informed that their accuracy in answering on both
the math problems and a set of recall questions
pertaining to the tape (which would be administered
later) would heighten their probability of winning a
lottery. The grand prize in the lottery was posted
as twenty dollars. The instruction letter attached
to the questionnaire packet explicitly informed the
subjects that their maximum output was needed in
order to establish the integrity of the study's
results (see Appendix C for a copy of the
instruction letter). Manipulation of the
37


independent variables of perceptual focus and reward
contingency was carried out within the instruction
letter.
The instruction letter outlined which focus
condition the subjects were assigned to. The
independent variable of perceptual focus was
manipulated by randomly assigning subjects to one of
two focus conditions: Social Focus and Math Focus
(nominal level data). In the Social Focus condition
subjects were informed that the social recall
section of their questionnaire was weighted three
times higher than the math section. In the Math
Focus condition, subjects were informed that the
math items were weighted three times more than the
social recall questions. Subjects in the Math Focus
condition closely resembled subjects found in the
original works of Gilbert et al. (1988) in that the
social information they were asked to process was a
distraction.
Within the instruction letter subjects were
also informed of the reward contingency which they
would be operating under. The independent variable
of reward contingency was manipulated by randomly
38


assigning subjects to one of three different
contingencies: Reward Only, Reward and Punishment,
and Punishment Only. Subjects in the Reward only
condition were led to believe that only correct
answers on both the math and recall questions would
be positively rewarded and tallied, and incomplete
or incorrect answers would be ignored. Subjects in
the Reward and Punishment condition were led to
believe correct answers on both the math and the
recall items would be positively rewarded and
tallied, as well as incorrect and incomplete answers
negatively rewarded and tallied. Subjects in the
Punishment Only conditions were led to believe that
only incomplete and incorrect answers on the math
and recall items would be negatively punished and
tallied, and correct answers ignored. Since there
was no a priori reason to believe that this variable
was in anyway rank ordered, the manipulation of
reward contingency results in at best a nominal
classification.
Subjects were allowed to work on the math
problems only for the duration of the taped message
(ten minutes total). Upon completion of the tape,
39


subjects were then asked to respond to the best of
their ability to a set of recall questions (see
Appendix D). These questions asked subjects to
recall specific statements made by the person on the
tape. Again, subjects were given up to ten minutes
to answer the recall questions; however, it was
found that it was not necessary to provide this much
time. Subsequent runs of the experiment were
conducted so that each group of subjects were given
adequate time to answer the recall questions. On
average, subjects required no more than four minutes
to complete the recall section of the questionnaire.
The gathering of data on this measure, though
not listed in the empirical hypotheses, would serve
as a manipulation check of the empirical hypotheses.
In other words, assessment of correct answers on the
recall test would serve to establish whether or not
the independent variables truly caused subjects to
focus differentially. The dependent variable of
correct answers on the recall items was scored in
the following manner. Subjects were led to believe
that their overall probability of winning the
lottery was dependent upon not only the number of
40


correct answers on this measure but also the
weighting system within which they were operating.
However, this was not the "true" calculation of this
measure. Each subject, regardless of focus
condition or reward contingency, was given one point
for each correct recall answer (total of twenty (20)
points possible).
In order to respond to the hypotheses generated
by this study is was necessary to determine the
attributional outcomes of subjects by condition.
The dependent variable of dispositional attributions
was assessed as follows. Once subjects had
completed the math and recall items they were
instructed to offer their perspective of the person
on the tape. Subjects were given an additional
questionnaire (see Appendix E) called the Houser
Attributional Inventory (HAI). The HAI directed the
subjects to answer yes/no regarding fifteen (15)
selected attributes (e.g. anxiousness) of the
individual speaking on the tape. Subjects were then
directed to assess if they felt the speaker was
"anxious" in general (thus making a dispositional
attribution) or due to the fact of being recorded
41


(thus making a situational attribution). One (1)
point was tallied each time the subject marked the
HAI within the range between moderately and
extremely in regard to the dispositional attribution
question.
Additionally, it was necessary to assess the
subject's personality in terms of extraversion-
introversion. Since a subject's placement on a
extraversion-introversion continuum was a naturally
occurring phenomena it could not be randomized. In
accordance with the rules of experimental
methodology, this naturally occurring variable was
blocked" in order to allow for the interpretation
of its effect within the design of the experiment.
This blocked variable was assessed by means of the
Eysenck Personality Inventory (EPI) (Eysenck and
Eysenck 1968), which was also included in the packet
of questionnaires given to the subjects.
Once the subjects had finished with the math,
recall, and attribution sections of the
questionnaire packet, they were instructed to fill
out the EPI, at their leisure. It was projected
that the first three sections in the questionnaire
42


would require approximately thirty (30) minutes to
complete. Problems associated with fatigue or
demand were not expected since the EPI is a self-
evaluative measure with a high reliability quotient
(r = .90) Subjects with scores on the EPI which
were at least one half standard deviation above the
median were classified as extraverts. Subjects with
scores on the EPI which were at least one half
standard deviation below the median were classified
as introverts. Individuals whose scores on the EPI
fell between +/- one half standard deviation from
the median were classified as ambiverts. Ambiverts
were excluded from statistical interpretation since
the rationale behind the study only attempted to
make claims about predominantly extraverted and
introverted individuals.
Operationalization of Hypotheses
Hypothesis # 1
The first empirical hypothesis (p. 21)
generated by the present study maintained that it
was a person's perceptual focus and not their level
of cognitive busyness which affected the number of
43


dispositional attributions they would make. The
independent variable in this hypothesis was the
different focus conditions the subjects were under
(Math Focus or Social Focus).
The dependent variable in this hypothesis was
the number of dispositional attributions made by
each subject. (Situational attributions were not
interpreted since no theoretical perspective was
offered why subjects choose external causes over
internal causes. The literature on attributional
errors only pertains itself to the predominance of
dispositional outcomes.) In order to examine this
directly, it was necessary to establish what the
subjects focus truly was. That is to say that it
cannot be assumed subjects would focus merely
because of the manipulation of the independent
variable. Therefore to examine where a subject's
focus lay, it must first be determined how many
right answers they made of the social recall
questions. It is reasonable to assume that people
who recall more social information, will have more
social information at their disposal. And, if this
fits into a subject's perceptual schema, then it is
44


likely that they would make less dispositional
attributions merely because they 'knew' more about
the social environment.
A manipulation check was run to establish where
a subject's focus truly lay. The dependent variable
in this check was the number of correct answers on
the recall questions. The manipulations check was
run to provide a measure of control on memorization
processes which could affect the attributional
decisions of the subjects. By controlling for the
number of correct recall answers, that is
controlling for informational levels, hypothesis #1
could then be tested to examine whether or not
subjects in the two focus groups attributed
differently. The operational extension of empirical
hypothesis number #1 is as follows:
#1 Given identical levels of cognitive
busyness (i.e. identical tasks) and controlling
for score on the recall questions; subjects in
the Social Focus condition will score
significantly lower on the HAI than subjects in
the Math Focus condition.
45


Hypothesis #2
The review of the literature leading up to
empirical hypothesis #2 (p. 28) maintained that
extraverted individuals would make significantly
more dispositional attributions than introverted
individuals. This study generated a response which
stated that this need not be the case. If as argued
by Eysenck (1967) and Gray (1970, 1972, 1978, 1981)
these types of individuals focus differentially
(given a system of rewards and punishments); then
those individuals who are focused on person
perception would make significantly fewer
dispositional attributions than those subjects who
are distracted by person perception. The
independent variable in this hypothesis was the
perceptual focus (given the reward contingency) of
the subjects. Additionally, data was collected by
means of the EPI to assess the blocked variable of
extroversion-introversion.
The dependent variable in this hypothesis was
the number of dispositional attributions made by
each subject. Once again, in order to examine this
46


directly, it was necessary to establish what the
subjects focus truly was. That is to say that it
cannot be assumed subjects would focus merely
because of the manipulation of the independent
variable. Therefore to examine where a subject's
focus lay, it must first be determined how many
right answers they made of the social recall
questions.
A manipulation check was run to establish where
a subject's focus truly lay. The dependent variable
in this check was the number of correct answers on
the recall questions. The manipulations check was
run to provide a measure of control on memorization
processes which could affect the attributional
decisions of the subjects. By controlling for the
number of correct recall answers, that is
controlling for informational levels, hypothesis #2
could then be tested to examine whether or not the
different personality types within the focus
conditions attributed differently. The operational
extension of empirical hypothesis number #2 is as
follows:
47


#2 In the Reward Only condition when
controlling for score on the recall questions;
extraverts in the Social Focus condition will
score significantly lower on the HAI than
extraverts in the Math Focus condition.
In the Punishment Only condition when
contolling for score on the recall questions;
introverts in the Social Focus condition will
score significantly lower on the HAI than
introverts in the Math Focus condition.
Subjects were debriefed through a letter
stating the actual nature of the study and
announcing the code number of the winner of the $20
lottery (see Appendix F).
48


CHAPTER FOUR
RESULTS
The main concern of this study was to
investigate the relationship between an individual's
perceptual focus and their attributional outcomes.
The theoretical framework concentrated around the
assumption that an individual's focus was the
dominant force when choosing between internal or
external causes. This study argued that if an
observer was focused on person perception they would
make fewer dispositional attributions because they
opened up their perceptual awareness to include
situational variables.
A review of the literature established that
other authors had identified certain types of
individuals who were predisposed to making
predominantly dispositional attributions. Most
notably, Gilbert et al. (1988) argued that
cognitively busy observers had a tendency to make
these types of decisions. Similarly, Berscheid et
al. (197 6) presented arguments which described
49


extraverted observers as holding a predisposition
toward internal attributions.
The present study argued that the above
relationships may be spurious in light of a
perceptual focus. As such this study developed
hypotheses which would attempt to modify the results
found in the works of Gilbert et al. and Berscheid
et al.
To specify the expected relationship between
the independent and dependent variables, two
hypotheses were proposed. The first as a response
to Gilbert et al.; the second to Berscheid et al.
The hypotheses are:
1. Given identical levels of cognitive
busyness; observers who are rewarded for
focusing on person perception will make
significantly fewer dispositional
attributions than active observers who are
not rewarded for focusing on person
perception.
50


f
2. Given that reward contingencies
differentially focus extraverts and
introverts; those subjects who are
rewarded to focus is on person perception
will make fewer dispositional attributions
than subjects who are not rewarded for
focusing on person perception.
In other words, Hypothesis #1 proposes that it
is the focus of the subjects which will determine
the type of attributional outcomes they will make,
not the amount of busyness. Hypothesis #2 projects
that" again it is the focus of the sub-jects which
will determine the type of attributional outcomes
they will make and not their personality types.
Before proceeding with the analysis of the
findings relating to the above hypotheses, the
characteristics of the subjects in the study will be
discussed. All of the 143 subjects were
undergraduate students at the University of Colorado
at Denver. Eighty seven subjects were female and 56
were male. Subjects ranged from 19 to 27 years of
age.
No effects of sex were found in this study.
51


Correlation Matrix
Before proceeding with the analysis generated
by the two hypotheses of this study, an initial
investigation of the data was run to examine the
correlational arrangements among all of the
independent (including the blocked variable) and
dependent variables. Variables were coded for
computer analysis as listed in Appendix G. The main
reason for implementing a correlational analysis was
to determine if any of the variable combinations
were providing redundant information. For example
if FOCUS and CNTNGNCY were highly correlated,
(r = .90 or higher), then a condition know as
multicollinearity would exist where both variables
were measures of the same phenomenon.
Multicollinearity would cause both logical and
statistical problems. The logical problem is that
it would not be a good idea to include redundant
variable in the same analysis, as they would not be
needed. And, because redundant variables would
reduce the degrees of freedom for error, they would
weaken the analysis. The statistical problem
52


created by multicollinearity is that it would render
unstable matrix inversions which will be needed in
the regression analyses to be performed in the
study.
The results of the correlational analysis can
be found in Table 4.1.
Table 4.1
Correlation Matrix
Correlation. Covariance, 1-tailed Sig, Croas-Product:
SOCIAL ATTR CNTNGNCY FOCUS INTRBXTR
SOCIAL 1.000 12.095
1717.455
ATTR -.098 1.000
-1.515 19.823
.122
-215.182 2814.811
CNTNGNCY .117 .022 1.000
.329 .079 .654
.082 .397
46.727 11.217 92.825
FOCUS -.249 -.023 -.080 1.000
-.434 -.052 -.032 .251
.001 .391 .172
-61.636 -7.392 -4.587 35.706
INTRBXTR .041 -.036 .088 .063 1.000
.118 -.132 .059 .026 .692
.315 .337 .147 .228
16.727 -18.706 8.441 3.720 98.210
The findings listed in Table 4.1 indicated that
multicollinearity between the independent variables
was not a concern. The only significant correlation
found to exist was between FOCUS (focus condition
53


the subject was under) and SOCIAL (the number of
correct answers on the social recall test); (r = -
.249, p < .001), but this was weak at best.
Manipulation Check of Hypothesis #1
The first operational hypothesis (p. 42)
generated by the present study was in response to
Gilbert et al. (1988). These authors maintained
that a person's degree or amount of cognitive
busyness was the driving force behind their
attributional outcomes. The present study argued
that Gilbert et al. had not provided a fair test of
this assumption. Further, this study presented
arguments which speculated that what Gilbert et al.
had witnessed was not cognitive busyness but
perceptual focus. In order to test this speculation
Hypothesis #1 was generated.
Hypothesis #1 maintained that it was a person's
perceptual focus and not their level of cognitive
busyness which affected the number of dispositional
attributions they would make. The independent
variable in this hypothesis was the different focus
54


conditions the subjects were under (Math Focus
versus Social Focus; nominal level data).
The dependent variable in this hypothesis was
the number (ratio level data) of dispositional
attributions made by each subject. In order to
examine this directly, it was necessary to establish
in which direction the subjects were focused.
Therefore to examine where a subject's focus lay, it
must first be determined how many right answers they
made on the social recall questions (ratio level
data). It was determined that a manipulation check
for hypothesis number one was necessary in order to
verify if the two focus groups did in fact
differentially focus. This analysis required that
an analysis of variance (ANOVA) procedure be run
using the SAS program's general linear model (GLM)
procedure to examine the effect of perceptual focus.
The Type III sums of squares generated by the SAS
program was chosen for interpretation since it
accounted for the unbalanced cell n's found in this
study.
The findings of this procedure are listed in
Table 4.2.
55


As can be seen in Table 4.2, the main effect of
FOCUS was significant within the manipulation check
of Hypothesis #1. What this meant was differences
in perceptual focus did manifest themselves into
significant differences in the number of right
answers on the recall questions. Subjects in the
Social Focus condition (perception that recall
questions were weighted three times higher than the
math questions) did score significantly higher than
those subjects in the Math Focus conditions
(perception that math questions were weighted three
times higher than the recall questions), Flw l4l = 9.31;
p < .0027. Further interpretation of Table 4.2
shows that on average subjects in the Social Focus
condition correctly answered 8.35 recall questions
to only 6.62 correct recall answers for subjects in
the Math Focus condition. These findings indicated
that the manipulation check of Hypothesis #1 was
successful. Due to these results, it was determined
that the manipulation check could be used to control
for information levels while analyzing Hypothesis
#1.
56


Findings of Hypothesis #1
Since the main effect of perceptual focus was
found to be significant in the manipulation check it
was subjected to further data analysis. This
further testing involved isolating the effect of
perceptual focus in order to analyze if it carried
over into the area of attributing causality. The
operationalized version of hypothesis #1 stated:
given identical levels of cognitive busyness;
subjects in the Social Focus condition would make
significantly fewer dispositional attributions than
those subjects in the Math Focus condition. In
order to test the operationalized version of
hypothesis #1 it was necessary to control for the
number of correct answers on the social recall test,
thus controlling for memorization effects.
Results of the regression analysis performed to
investigate the operationalized version of
hypothesis #1 are outlined in Table 4.3.
Findings of Operational Hypothesis #1.
As can be seen in Table 4.3, the prediction
that subjects who were focused on person perception
57


Table 4.2
One Way ANOVA for the Manipulation
Check of Hypothesis #1
The SAS System
General Linear Models Procedure
Dependent Variablei SOCIAL
Source DF Sum of Squares Mean Square F Value Pr > F
Model 1 106.39696614 106.39696614 9.31 0.0027
Brror 141 1611.05757932 11.42594028
Corrected Total 142 1717.15454545
R-Square c.v. Root MSB SOCIAL Mean
0.061950 45.34452 3.38022784 7.45454545
Source DF Type III SS Mean Square F Value Pr > F
FOCUS 1 106.39696614 i06.39696614 9.31 0.0027
The SAS System
General Linear Models Procedure
Level of ------------SOCIAL-----------
FOCUS N Mean SD
1 69 8.34782609 3.54312128
2 74 6.62162162 3.22108934
(Social Focus condition) would make significantly
fewer dispositional attributions than subjects in
the Math Focus condition was not supported. In
fact, when
controlling for the number of correct answers on the
recall test, no significant differences existed
between any of the perceptual focus conditions.
Manipulation Check of Hypothesis #2
The second operational hypothesis (p. 44)
generated by the present study -was in response to
58


Berscheid et al. (1976). These authors maintained
that the personality dimension of extraversion-
introversion mediated attributional outcomes. The
present study was designed in part to evaluate these
assumptions in combination with the concept of
perceptual focus. Hypothesis #2 maintained that it
was a person's perceptual focus and not their
personality type which affected the number of
dispositional attributions they would make.
Research in the area of personality theory
offered insight into one possible strategy which
could be used when attempting to focus different
types of personalities. Work by Eysenck (1967) and
Gray (1970, 1972, 1978, 1981) suggested that reward
contingencies affected the focus of different
personality types.
The independent variables in Hypothesis #2 were
the different focus conditions the subjects were
under (Math Focus versus Social Focus; nominal level
data) and the different reward contingencies the
subjects were under (Reward Only, Reward and
Punishment, and Punishment Only). Additionally,
data was collected on the subject's personality type
59


Table 4.3
Regression Analysis of Hypothesis #1
All subjects regardless of focus condition: N of Cases = 143
----------- Variables in the Bquation -------------------------------------------------------
Variable B SB B Beta Correl Part Cor Partial F Sig F
SOCIAL -.125291 .107296 -.097867 -.097867 -.097867 -.097867 1.364 .2449
(Constant) 7.290632 .882050 68.319 .0000
EQUATION GENERATED Y = 7.29
Only subjects in Social Focus condition: N of Cases = 69
---------------,----------------Variables in the Bquation------------------------------------
Variable B SB B Beta Correl Part Cor Partial F Sig F
SOCIAL -.089182 .163148 -.066633 -.066633 -.066633 -.066633 .299 .5864
(Constant) 7.208244 1.477892 23.789 .0000
EQUATION GENERATED Y = 7.21
Only subjects in the Math Focus condition: N of Cases = 74
---------------------- Variables in the Equation----------------------------- ------------
Variable B SB B Beta Correl Part Cor Partial F Sig F
SOCIAL -.200435 .151623 -.153934 -.153934 -.153934 -.153934 1.748 .1904
(Constant) 7.583964 1.115033 46.261 .0000
EQUATION GENERATED Y = 7.58
60


by means of the Eysenck Personality Inventory (EPI).
Personality type was treated as a blocked variable
in accordance to the rules of research methodology.
The dependent variable in this hypothesis was
the number (ratio level data) of dispositional
attributions made by each subject. Again, a
manipulation check was first performed using the
number of correct recall answers as the dependent
variable.
The manipulation check for Hypothesis #2 was
run to determine if either the independent variable
or the blocked variable had any effect on the
subjects' focus. This manipulation check consisted
of a series of t-tests pitting the means of one
treatment by block condition against another. All
possible combinations of treatment conditions were
analyzed. The dependent variable in all of the fc-
tests was the number of correct answers on the
recall questions. This variable was chosen as the
dependent variable for the manipulation check since
it would accurately assess how attentive the subject
was to the taped information. Results of the t-
tests are as follows:
61


Significant results:
1. Introverts in the Social Focus condition
made significantly more correct answers on the
recall questions than did introverts in the
Math Focus condition (t4I = 2.04; p < .05) .
2. Extroverts in the Social Focus condition
made significantly more correct answers on the
recall questions than did extraverts in the
Math Focus condition (t56 = 1.90 ; p < .03).
Non-significant results:
1. Introverts did not significantly differ
from extroverts in terms of number of correct
recall questions (t99 = -. 65 ; ns)
2. Punished subjects did not significantly
differ from rewarded subjects in terms of
number of correct recall questions (t91 = 1.49;
ns)
62


3. Punished subjects did not significantly
differ from subjects in the Reward and
Punishment condition in terms of number of
correct recall questions (t = 1.67, ns).
4. Rewarded subjects did not differ from
subjects in the Reward and Punishment condition
in terms of number of correct recall question
(t = 1.11; ns)
5. Punished introverts did not differ from
punished extroverts in terms of number of
correct recall question (t36 = -1.51; ns).
6. Rewarded introverts did not significantly
differ from rewarded extrovert in terms of
number of correct recall answers (= -.21,
ns) .
7. Reward and Punished introverts did not
significantly differ from Reward and Punished
extraverts in terms of the number of correct
recall answers (t26 = .18; ns).
63


In general these results indicated that the
manipulation check for Hypothesis #2 was
unsuccessful. No differences were found between
introverts and extroverts in any pairing of reward
contingencies. Implied from this finding is that
different personality types are not differentially
focused due to rewards and/or punishments. In
short, Eysenck's (1967) theory was not confirmed for
this experiment.
The only manipulation check to show significant
differences was that associated with the main effect
of perceptual focus. Since this effect had been
investigated thoroughly in Hypothesis #1 it was not
included in Hypothesis #2.
Even though the manipulation checks of
Hypothesis #2 failed to show a differential focusing
between the two personality types, further analysis
of this hypothesis was conducted. Hypothesis #2's
main question lay in the analysis of whether or not
different personality types reach different
attributional outcomes. In order to test this
question a second series of t-tests was run with the
number of dispositional attributions as the
64


dependent variable. The findings of these tests are
listed below.
Significant findings:
No significant findings.
Non-significant findings:
1. Punished introverts did not significantly
differ from punished extraverts in terms of the
number of dispositional attributions made (t36 =
-.12} ns) .
2. Introverts in the Reward and Punishment
condition did not significantly differ from
extroverts in the Reward and Punishment
condition in terms of the number of
dispositional attributions made (t26 = -1.07;
ns) .
3. Rewarded introverts did not significantly
differ from rewarded extraverts in terms of the
number of dispositional attributions made (t3J =
.41; ns).
65


4. Rewarded subjects did not significantly
differ from punished subjects in terms of the
number of dispositional attributions made (t
=.17; ns).
5. Rewarded subjects did not significantly
differ from subjects in the Reward and
Punishment condition in terms of the number of
dispositional attributions made (t,2 = .87; ns)
6. Punished subjects did not significantly
differ from subjects in the Reward and
Punishment condition in terms of the number of
dispositional attributions made (fc = .13; ns)
7. Introverts did not significantly differ
from extroverts in terms of the number of
dispositional attributions (t = -1.35; ns).
In general, these findings indicate that
neither the reward contingencies, personality type
nor the interaction of the two create any
differences in the number of dispositional
66


attributions made by the subjects. In short,
Berscheid et al.'s (1976) relationship was not
confirmed for this experiment, nor was the
hypothesis that perceptual focus could negate
Berscheid et al.'s relationship.
67


CHAPTER FIVE
SUMMARY, CONCLUSIONS, AND IMPLICATIONS
Summary
The purpose of this study was to investigate
the relationship between two sets of variables:
(1) attributional outcomes as a dependent variable;
and (2) Perceptual focus as an independent variable.
Previous studies had indicated that certain types of
individuals were prone to making dispositional
attributions (Berscheid et al. 1976; Gilbert et al.
1988) The attempt of this study was to evaluate
whether these individuals could override their
predispositions by focusing on person perception in
order to gain a reward. It was assumed that the
reward offered to the subjects (the $20 lottery
prize) would be a sufficient enough incentive to
create a state of intrapersonal drive toward goal
attainment.
The present study was conducted within the
framework of attribution theory. One of the main
characteristics of human thought includes a tendency
68


to seek out causal explanations for a witnessed
behavior (Heider 1958). Cognitions about some
person or event tend to be organized and integrated
into a meaningful whole; giving some structure,
meaning, and stability to everyday experience.
Attribution theory has tended only to stress
woman/man's cognitive motivation. Human desire is
restricted exclusively to wanting to test and
structure reality. Attributional analysis has
overlooked the many other motives that affect person
perception. This study was designed to investigate
attribution theory's oversight in terms of
perceptual focus.
Studies within attribution have shown that an
observer's decision making process is not always
accurate (Jones and Nisbett 1971a; Carver et al.
1980). As such, observers are prone to arriving at
biased attributional outcomes (Ross 1977). Recent
investigations of attributional biases have
established that cognitively busy (Gilbert et al.
1988) and extraverted individuals (Collins et al.
1973) are predisposed to making dispositional
attributions. The present study was designed to
69


test these empirical findings in light of a
perceptual focus.
The only significant finding generated by this
study was the relationship between a cognitively
busy subject's perceptual focus and the number of
correct answers on the recall questions. Generally,
the data did not support the hypothesis relating
attributional outcomes and perceptual focus.
Conclusions
From the results presented and within the
limitations of the study, it can be concluded that:
1. The findings were generally inconsistent
with the hypothesis relating to the relationship
between a subject's perceptual focus and the number
of dispositional attributions they would make.
Results did indicate that the perceptual focus of a
subject did affect their ability to recall
information about the tape recording. Subjects who
believed that the social questions were weighted
three times heavier than the math questions did
answer significantly more recall questions correctly
than those subjects who believed that the math
70


questions received the higher weighting. However,
this effect did not carry over into the area of
attributional outcomes. It was hypothesized that
those subjects who were motivated to pay closer
attention to the social.information would make fewer
dispositional attributions. This hypothesis was not
supported by the findings of the study.
Four explanations for this result are offered.
First, the possibility exists that the attributional
outcome instrument (Houser Attributional Inventory,
HAI) did not accurately assess the dependent
variable. Subjects were asked if, in general, they
perceived the target person to be any or all of
fifteen trait descriptors. While it may be true
that the subjects were, in fact, making an
attribution about the target person, they were not
specifically seeking to find the causal antecedents
behind his actions. For example, subjects were
asked if they felt the target person was friendly,
in general: Yes or no?. Merely offering this type
of trait assessment (about someone they would never
meet) may not have created a strong enough desire
for the subjects to be accurate or attentive in
71


their attributional assessments. If however,
subjects had been asked to decide if the target
person deserved to be fired from his job, then
perhaps a more intensive attributional search would
have been made.
Secondly, questions have been raised as to
whether or not making one type of attribution
precludes an individual from additionally making
another type of attribution for the same behavior.
In other words, can an observer make both a
dispositional attribution and a situational
attribution for the same witnessed behavior? In
Heider's (1958) original work he argued that within
the attribution process a hydraulic relationship
existed between internal and external causes. The
attribution of more causality to one set of factors
entailed the attribution of less causality to the
other. White (1990) has criticized this assumption
noting that "internal and external (causes) are not
opposites on a single dimension" (p.266). If
White's notion of attribution is true then it would
not be surprising that differentially focused
72


subjects could come to a two-fold attribution of
causality with equal ability.
Third, it is possible that no relationship
between perceptual focus and attributional outcomes
exists. It could be that one's ability to make
causal inferences in not impaired just because of a
lack of concentration on an actor. Human perception
may be sophisticated enough that even a minimal
amount of social information is sufficient to make
attributions.
Fourth, it may be the case that individuals who
do have more social information choose not to use
this knowledge when making attributions. Perhaps it
is too costly to some inner sense for the observer
to "give the benefit of the doubt" to the actor. A
prejudiced person who holds that the actions of
persons of color (i.e. rioting) are due to their
internal disposition (lack of respect for the
system) and not a situational factor (the Rodney
King verdict) will dispose her/him to reaffirm
prejudiced beliefs. Ignoring the situational
information allows the prejudiced person to maintain
her/his view of reality (minorities are sub-white).
73


(Additional implications of this perspective will be
given under the heading of Implications).
2. The findings failed to support the second
hypothesis. No difference was found between
rewarded extraverts (punished introverts) who
focused their attention on person perception when
compared to similar individuals not focused on
person perception. In fact, no differences were
found between any combination of reward
contingencies or personality type (except those
already explained in Hypothesis #1). In short,
Gray's theory was not confirmed for this experiment.
As noted earlier, Gray's theory at. best borders
on a tautology. Research was performed which
exposed a correlation; then the correlation was
explained in terms of the research from which it
arrived. Manipulations of this nature rarely give
rise to findings which generalize beyond their
original setting. So it is not surprising that
Grays assumptions did not carry over into the area
of attributional outcomes.
Although the results demonstrated very little
relationship between the two sets of variables, it
74


may be an error to conclude that no relationship
exists. It is possible that the manner in which the
questions were asked failed to detect this
relationship.
Also, it is possible that the assumption that a
$20 lottery prize would serve as a sufficient
motivator could be in error. No post-test questions
asking the subjects about their internal drive
toward the prize were administered.
Additionally, the weighting system which
subjects believed they were operating under may have
nullified any possible results. Subjects who
believed that the math questions were weighted three
times higher than the social recall questions, but
who perceived themselves to be weak in mathematics
may have concluded that they had no chance of
winning the prize. If this were the case these
subjects may have prematurely terminated their
interest in the study. Such premature loss of
interest would constitute a threat to the internal
validity of the experiment due to a systematic
mortality of subjects. No post-test questions
asking subjects their assessment of their own
75


probability of winning the $20 were administered.
Questions which assessed this dimension could have
given light to a truer relationship between a
subjects's intrapersonal motivation and their
attributional outcomes.
Implications.
Generally the data collected for this study
showed little in the way of a perceptual focus
mediating an individual's attributional tendencies.
However, it did raise questions about this
relationship which need to be pursued in subsequent
investigations.
If it had been found that a perceptual focus
could alter the attributional tendencies of an
observer, then it is reasonable to assume that these
tendencies would carry over into other aspects of
social interaction. An example might be that of a
jury trial. Within legal circles a defense attorney
would hope that the jurors' perceptual focus was
intense enough to allow them to consider all
situational variables pertinent to the case. An
inattentive (unfocused) juror may overlook important
76


situational variables key to acquitting the
defendant. On the other hand, an attentive
(focused) juror may pick up on the fact that the
defendant was only acting in self defense, or that
the circumstances warranted the behavior of the
actor, no matter how undesirable. To be sure, the
system of jurisprudence is founded within
attribution theory. Guilt or innocence is as much a
matter of intent (dispositional attribution) as it
is of circumstance (situational attribution). It
may be the case that the manipulation of
attributional outcomes is best performed by the
lawyer and not the scientist.
Situations could also be conceived where an
increased knowledge from a perceptual focus could be
used as the means to some end. Talented con men by
and large possess one vital skill--knowing just a
little more about the "mark" (victim) than the
"mark" her/himself does. Keen perception of an
actor's circumstances may make the "hunt" for a
victim all the easier. Perceiving that the small
town boy is lost in the swirl of the big city
lights, may allow those on the grift (con-men) to
77


play on the confusion surrounding the farm boy.
Soon the naive ruralite may find that he has been
duped by the city slicker. By focusing on the farm
boy's naivete, the grifter may 'know' that what his
victim needs is a calming influence in the sea of
chaos. A steady voice, a friendly smile, and a
country twang may be just what is needed to set the
hook in the farm boy's mouth. With knowledge gained
from a perceptual focus, the grifter has acquired
the upper hand, and the mark is well on his way to
being taken.
Implications based upon the notion that an
observer may have situational knowledge and choose
not to use it could also be given merit. In fact,
implications based upon this assumption may have
more of a social psychological veneer. Failing to
attribute an actor's shortcomings to her/his
surroundings may lead to a reinforcement of
stereotypical thinking. For example, a prejudiced
person may conclude that inner city, minority
children are intellectually inferior to their
suburban, majority counterparts based only upon IQ
scores. In so doing, the prejudiced person creates
78


a stereotypical, dispositional attribution. If,
however, this observer were to include the fact that
inner city schools pale in comparison to their
suburban counterparts, then s/he might realize that
all children should be given a chance for an equal
education. A perceptual focus which blames the
system and not the child, could dispel the falsities
of prejudice and discrimination. Unfortunately,
those who hold a biased view of society do so for a
purpose.
Research has shown that a perceptual bias
within attribution serves an ego-defensive function.
Work by Weiner, Frieze, Kukla, Reed, Rest, and
Rosenbaum (1971) has shown that the attribution of
success and failure are diametrically opposed when
comparing one's self with others. When attributing
one's own failures, one stresses the external forces
(difficult task, unlucky) precipitating the
deficiency. But, when observing another's
shortcomings, one is most likely to attribute them
to the actor's internal inadequacies (lack of
ability, lack of effort). Conversely, one's
successes are more likely attributed to internal
79


qualities (wealth of ability and effort), while the
successes of another are due to situational factors
(i.e. luck, easy task). From this typology it is
easy to see how a change in perceptual focus allows
for a stroking of one's ego. At every turn the self
is reinforced. Given this systemic bias, one's
self-image should never fall below the pinnacle of
superiority.
Lerner's (1980) concept of a "just world" may
also be the result of a perceptual focus. Holding
the notion that good things happen to good people,
and bad things happen to bad people protects one's
self from the chaotic nature of reality. Only by
believing that there is a just world can one
confront the physical and social environments as if
they were stable and orderly. Society tolerates the
suffering of a good many disadvantaged peoples. It
permits sick children to go untreated and the
homeless to go unfed. Class differentiations are
entrenched in this belief of a prevailing justice.
Many comfortably suited Americans condemn the
unemployed, the lower classes, and the power
minorities as "idlers", and "welfare abusers." In
80


order to reify a hierarchy of privilege--upper
classes, males, whites and capitalists reinforce
this hegemonic concept of equity. Holding true to
this perceptual bias is but a stay in a gender, race
and class based society. Lerner maintains that
dominant group members seeing the disadvantaged
circumstances of minority-group members, may find it
easy to conclude the they (the minorities) only got
what they deserved. "..once the rejection (of the
disadvantaged) is accomplished, the observer can
again rest easy--his world is just, and he need not
feel impelled to act to reestablish justice" (Lerner
1974, p. 344). It may well be that maintaining a
perceptual focus is the blinder used to ignore the
problems of the ghetto. The great social experiment
surrounding us can continue, unfettered and
unchanged.
Ultimately attribution theory may in itself be
a self-fulfilling prophecy. One's need to manage,
predict and control one's environment may preclude
one from assessing information which is not
manageable, predictable and controllable. Perhaps
Nietzsche was right; it is our needs that interpret
81


the world: our need to control, our need to
predict, and our need for a egocentric, self-
sustaining focus.
82


APPENDIX A
MATHEMATIC EQUATIONS
83


Answer here
1. Multiply .067 and 7. ___
2. Round 1/9 to the nearest whole number. _________
3. 2/3 + 3/4 + 1/8 = ? (Express your answer in
exact decimal form.)
4. 3 1/5+ 13/4 = ? (Express your answer in
fraction form.) ___
5. 8.27 divided by 1/20 equals________? ___
6. 6 x V64 = _?
7. Six percent of 8.50 equals_____? ___
8. 3.3 x 3 V81 = _?
9. Solve for y. y/90 = 22. ___
10. Divide .3 by 81. ___
11. 720 : 6! = _?
12. 2.22 = _?
13. 16 is to 4, as 625 is to____? ___
14. (1.1 x 10) (6.0 x 10) = _?
15. 11/ (3/33) = _?
16. Divide 24.33 by 2/3. ___
17. Find 84% of 1437.25. ___
18. Add the following scores: 8.0, (-4), -(-6) and
210.
19. Divide 727 by .0825. ___
20. 4/27 + 1/3 + 6/54 = ?
84


APPENDIX B
TRANSCRIPT OF SOCIAL INFORMATION TAPE
85


Well, no one has ever asked me to talk for ten
minutes before, especially about myself. I mean I
can talk for ten but I don't know what I'm going to
tell you about myself.
I guess I'll start with the basic boring stuff.
My name is Bill Walters. I'm a student at UC-D, a
senior in the Sociology department. I'm 31 years
old. I was born and raised around Columbus, Ohio,
actually a little town about twenty-five miles south
of Columbus called Circleville; which nobody's ever
heard of. The area is mostly farm country. I grew
up on a farm.
I'm married. I've been married for about, oh,
six months now. Boy, my wife would kill me if I
didn't know that! Yeah, six months. We don't have
any kids. We are not planning on having any kids
right away. With me being in school, we think that
would be too much of a responsibility. Plus, you
know, I may be a little too selfish for kids,
because they take up so much time. You've got to
take care of them, watch over them and all of that
kind of stuff. Getting married as late as I did;
this is my first marriage, my only marriage, same
86


with my wife; getting married this late, you've got
to make a decision pretty quickly about having kids.
Right now we are going to wait until I get out of
school and then think about it.
My wife is a physical therapist. It is kind of
interesting how we met. We both grew up in Ohio.
She grew up around Toledo, but I didn't meet her
until I moved out here to Colorado. I was skiing,
two years ago, at Winter Park and I was coming down
one of the runs and I got ran over by a kid on a
snowboard. He tore up my knee really bad. I had to
have surgery on it. Anyway, after the surgery, I
went into therapy, physical therapy. That's where I
met my wife. She is a physical therapist for Rose
Medical Center. After I got out of therapy,
actually the last day I was in therapy and they were
going to discharge me, I asked her to go out with
me. She said yes as long as I wasn't a patient
anymore, she could do that. We went out and dated
for about a year and then got engaged and then got
married six months ago. It's been great, so far.
We moved up to Evergreen. That is where I live
now and drive down everyday--drive down to go to
87


school and go to work. I do a couple of things
besides go to school. I work for a catering
company, part time. I used to work for a little
deli right up the street here, but I got fired--two
weeks age--for being late. I was five minutes late!
The manger said it was the third time I was late in
like a month and a half. So he said I wasn't
dependable. I don't know, I don't think he liked
me. It wasn't my fault I was late, coming from
Evergreen on Fridays, I don't have any classes so
when I come in I have to fight rush hour. You know,
there was just too much traffic and I couldn't get
in on time. That was just between him and me. The
other two times I was late, I has the ignition
switch go out in my car. So I had to ride the bus.
If you know how the buses are, if you've ever tried
to ride a bus in from Evergreen--it is just crazy.
It takes you an hour and a half to get in from up
there.
That's enough about that. So I'm looking for a
job right now. To make some extra cash. How much
time have I used? About five minutes now? O.K.
88


My hobbies, besides skiing, I'm not so
enthusiastic about skiing as I was because I'm
afraid that my knee is pretty fragile. I don't want
to mess it up again. I don't think it could survive
a second surgery. I do a lot of stuff to try and
strengthen it. My wife being a physical therapist,
she is real concerned that it doesn't tighten down
and I lose any flexibility. So, we do a lot of
things like, we cross-country ski now--which is
really good for your legs. We also do a lot of
mountain biking. Living up in Evergreen is really
great, because there are lots of trails and lots of
parks to which you can go and bike through. Last
weekend, not the weekend it snowed but the one
before that, we went to Elk Meadow Park, which is
this huge open space across from where we live in
Evergreen. We hiked in there for about an hour and
a half, to the top of Bergin Peak. It was very
pretty, the aspens were turning and it was really
neat.
That's one of the things I miss about Ohio.
The falls are kind of whimpy compared to the falls
back in Ohio. There the trees turn different
89


colors--oranges and reds--all youve got out here is
gold. I guess back in Ohio they don't have any
mountains so it makes up for it.
Besides sports, my wife and I--my wife's name
is Kathy--we enjoy going to the movies. Not renting
a VCR, but actually going to the theater and buying
the popcorn. That whole bit. The last movie we saw
was "Ghost", with Demi Moore and Patrick Swayze. I
thought it was a really cool movie. He dies and
he's the ghost that follows her around and tries to
help solve his murder. I don't want to tell you the
story in case you might want to see it. I liked it
a lot. I think that kind of stuff is pretty neat.
My wife, she is not too keen on the occult or
afterlife, whatever you want to call it. She didn't
think it was too cool. She cried, whatever.
Besides participating in sports, I like
watching sports. I like football, college football
better than pro football. Baseball. I'm a big
baseball fan. Growing up in Ohio, you know
everybody is a Reds fan. I just thrilled to death
about the World's Series. Can you believe the Reds
won in four games. Everybody though it would be the
90