VIDEOTAPE AND THE ATTRIBUTION PROCESS: REVERSING … · Journal ol Personality and Social...

11
Journal ol Personality and Social Psychology 1973, Vol. 27, No. 2, 165-175 VIDEOTAPE AND THE ATTRIBUTION PROCESS: REVERSING ACTORS' AND OBSERVERS' POINTS OF VIEW 1 MICHAEL D. STORMS 2 Yale University Two actor subjects at a time engaged in a brief, unstructured conversation while two observer subjects looked on. Later a questionnaire measured the actors' attributions of their own behavior in the conversation either to dis- pcsitional, internal causes or to situational, external causes. Similarly, each observer attributed his matched actor's behavior. Videotapes of the conversa- tion, replayed to subjects before the attribution questionnaire, provided an experimental manipulation of visual orientation. Some actors and observers saw no videotape replay, while other subjects saw a tape that merely repeated their original visual orientations. As predicted for both of these conditions, the actors attributed relatively more to the situation than the observers. A third set of subjects saw a videotape taken from a new perspective—some actors saw a tape of themselves, while some observers saw the other participant with whom their matched actor had been conversing. With this reorientation, self-viewing actors attributed relatively more to their own dispositions than observers. The results indicated the importance of visual orientation in deter- mining attributional differences between actors and observers. Pragmatically, the theoretical framework and results of the study had relevance to the use of videotape self-observation in therapy and T groups. When an individual observes a behavior and attempts to understand its causes, he is concerned with the relative importance of personal dispositions of the actor and the surrounding social and environmental con- text. Both an observer who wishes to explain another's behavior and an actor who tries to understand his own behavior attempt to make the appropriate causal attributions. There is reason to believe, however, that actors and observers do not always arrive at the same explanation of the actor's behavior. Jones and Nisbett (1971) have argued that when actors seek to explain their own behavior, they are inclined to give considerable weight to exter- nal, environmental (i.e., situational) causes. Observers, on the other hand, place consider- ably more emphasis on internal, personal (i.e., dispositional) causes of the actor's behavior. 1 The research for this article was performed as part of the author's PhD dissertation submitted to the Department of Psychology, Yale University. The author wishes to express his appreciation to Richard E. Nisbett, who served as advisor for the thesis and who has contributed many helpful criticisms of the present article. 2 Requests for reprints should be sent to the au- thor, who is now at the Department of Psychology, University of Kansas, Lawrence, Kansas 66044. Several studies (Jones & Harris, 1967; Jones, Rock, Shaver, Goethals, & Ward, 1968; McArthur, 1970, 1972; Nisbett, Ca- puto, Legant, & Marecek, 1973) have been cited in support of this general proposition, and Jones and Nisbett have discussed a vari- ety of factors which might lead to such attri- butional differences between actors and. ob- servers. These factors include (a) differences in information about the event, behavior, and context which is available to actors and ob- servers and (b) differences in how informa- tion is processed by actors and observers. Actors may have private information about some aspects of the event, including their own feelings and the historical context in which the event transpires, while observers may have more complete information about the .behavior itself. Furthermore, in the inter- ests of controlling events and predicting the future, actors may attend more to situa- tional variables in an event, and observers may attend more to variations in the actor's behavior. The present study examines a fundamental difference between actors and observers which may lead, in turn, to some of the information differences postulated by Jones and Nisbett (1971). Perhaps the most obvious difference 165

Transcript of VIDEOTAPE AND THE ATTRIBUTION PROCESS: REVERSING … · Journal ol Personality and Social...

Page 1: VIDEOTAPE AND THE ATTRIBUTION PROCESS: REVERSING … · Journal ol Personality and Social Psychology 1973, Vol. 27, No. 2, 165-175 VIDEOTAPE AND THE ATTRIBUTION PROCESS: REVERSING

Journal ol Personality and Social Psychology1973, Vol. 27, No. 2, 165-175

VIDEOTAPE AND THE ATTRIBUTION PROCESS:REVERSING ACTORS' AND OBSERVERS' POINTS OF VIEW1

MICHAEL D. STORMS 2

Yale University

Two actor subjects at a time engaged in a brief, unstructured conversationwhile two observer subjects looked on. Later a questionnaire measured theactors' attributions of their own behavior in the conversation either to dis-pcsitional, internal causes or to situational, external causes. Similarly, eachobserver attributed his matched actor's behavior. Videotapes of the conversa-tion, replayed to subjects before the attribution questionnaire, provided anexperimental manipulation of visual orientation. Some actors and observerssaw no videotape replay, while other subjects saw a tape that merely repeatedtheir original visual orientations. As predicted for both of these conditions,the actors attributed relatively more to the situation than the observers. Athird set of subjects saw a videotape taken from a new perspective—someactors saw a tape of themselves, while some observers saw the other participantwith whom their matched actor had been conversing. With this reorientation,self-viewing actors attributed relatively more to their own dispositions thanobservers. The results indicated the importance of visual orientation in deter-mining attributional differences between actors and observers. Pragmatically,the theoretical framework and results of the study had relevance to the useof videotape self-observation in therapy and T groups.

When an individual observes a behaviorand attempts to understand its causes, he isconcerned with the relative importance ofpersonal dispositions of the actor and thesurrounding social and environmental con-text. Both an observer who wishes to explainanother's behavior and an actor who tries tounderstand his own behavior attempt to makethe appropriate causal attributions. There isreason to believe, however, that actors andobservers do not always arrive at the sameexplanation of the actor's behavior. Jones andNisbett (1971) have argued that when actorsseek to explain their own behavior, they areinclined to give considerable weight to exter-nal, environmental (i.e., situational) causes.Observers, on the other hand, place consider-ably more emphasis on internal, personal(i.e., dispositional) causes of the actor'sbehavior.

1 The research for this article was performed aspart of the author's PhD dissertation submitted tothe Department of Psychology, Yale University. Theauthor wishes to express his appreciation to RichardE. Nisbett, who served as advisor for the thesis andwho has contributed many helpful criticisms of thepresent article.

2 Requests for reprints should be sent to the au-thor, who is now at the Department of Psychology,University of Kansas, Lawrence, Kansas 66044.

Several studies (Jones & Harris, 1967;Jones, Rock, Shaver, Goethals, & Ward,1968; McArthur, 1970, 1972; Nisbett, Ca-puto, Legant, & Marecek, 1973) have beencited in support of this general proposition,and Jones and Nisbett have discussed a vari-ety of factors which might lead to such attri-butional differences between actors and. ob-servers. These factors include (a) differencesin information about the event, behavior, andcontext which is available to actors and ob-servers and (b) differences in how informa-tion is processed by actors and observers.Actors may have private information aboutsome aspects of the event, including theirown feelings and the historical context inwhich the event transpires, while observersmay have more complete information aboutthe .behavior itself. Furthermore, in the inter-ests of controlling events and predicting thefuture, actors may attend more to situa-tional variables in an event, and observersmay attend more to variations in the actor'sbehavior.

The present study examines a fundamentaldifference between actors and observers whichmay lead, in turn, to some of the informationdifferences postulated by Jones and Nisbett(1971). Perhaps the most obvious difference

165

Page 2: VIDEOTAPE AND THE ATTRIBUTION PROCESS: REVERSING … · Journal ol Personality and Social Psychology 1973, Vol. 27, No. 2, 165-175 VIDEOTAPE AND THE ATTRIBUTION PROCESS: REVERSING

166 MICHAEL D. STORMS

between actors and observers is that theyhave, quite literally, different points of view.Actors cannot see themselves act; physicallythey cannot observe much of their own be-havior. They may watch the antecedents oftheir own behavior, or its consequences, orboth. But they do not normally view thebehavior itself. In addition to the physicaldifficulty of watching oneself, there are tem-poral restrictions which contribute to a lackof self-observation. There may not be enoughtime or mental capacity to contemplate pastbehavior, monitor present behavior, and planfuture behavior all at once. Finally, there aremotivational reasons for avoiding an excessof self-observation. In the interest of actingunself-consciously and maintaining controlover the immediate events taking place, theactor may learn that it is dysfunctional tobe overly concerned with his own present andpast behavior. Instead, it is reasonable toassume that most actors focus on the situa-tion in which they find themselves. They lookat, attend to, and think about variouschanging aspects of the environment in whichand to which they must respond.

While the actor is watching the situationin which he finds himself, the observer isprobably watching the actor. It is usuallyinteresting and often important to watch thebehavior of other people. Consequently, ob-servers are often visually oriented toward theactor. Although an observer can take his eyesoff the actor and view other aspects of the sit-uation, he probably sees less of the situationthan the actor does. As with actors, the ob-server's scope is also limited by time. Ob-servers cannot simultaneously watch the actorand observe as much of the situation as theactor can. Moreover, observers may find itmore efficient in terms of controlling andpredicting the ongoing event to concentrateon the actor's behavior rather than on theactor's situation. Finally, the actor is, afterall, part of the observer's situation. For thesame reasons that an actor focuses on hisown situation, the observer focuses on thebehavior of the actor, which is part of his(the observer's) situation.

Thus, we postulate that there is a simpledifference between actors and observers. Ac-tors watch their environment (which in-

cludes the behavior of other people) morethan they watch their own behavior. Ob-servers watch the behavior of the actor morethan they watch the actor's situation.

If it is true that attributions are largelyinfluenced by point of view, it should be pos-sible to change the way actors and observersinterpret a behavior by changing their visualorientations.

A test of this hypothesis requires somemeans of changing actors' and observers'orientations. Fortunately, modern technologyprovides a simple and interesting means toaccomplish this change—namely, the use ofvideotape. Videotapes of an event, takenfrom various camera angles, can be replayedto actors and observers to redirect their at-tention to other aspects of the event. Ofparticular interest is the case in which video-tape presents a new visual orientation, thatis, when actors are shown a tape of theirown behavior from the observer's perspectiveand when observers are shown a tape of somekey aspect of the actor's situation from theactor's perspective. Such reorientation shouldaffect actors and observers so as to weaken(or even reverse) their original attributionalbiases. Actors who see themselves shouldmake more dispositional attributions abouttheir own behavior. Observers who see an-other aspect of the actor's situation shouldbecome more situational in attributing theactor's behavior.

Thus, the question to be answered by thisstudy is whether actors' and observers' attri-butions can be significantly influenced, per-haps even reversed, by changing their visualorientation toward an event. The implicationsof such a question may go beyond immediatetheoretical concerns. Discrepancies betweenactors' and observers' perceptions and inter-pretations of behavior are of paramountconcern to therapists, group relations con-sultants, and T group trainers. Often suchpractitioners must attempt to bridge theinterpretational gap between actor and ob-server, patient and therapist, and individual;and group.

METHODOverview

The hypothesis was tested in an experiment ffiatfeatured a simple interpersonal event, namely a

Page 3: VIDEOTAPE AND THE ATTRIBUTION PROCESS: REVERSING … · Journal ol Personality and Social Psychology 1973, Vol. 27, No. 2, 165-175 VIDEOTAPE AND THE ATTRIBUTION PROCESS: REVERSING

VIDEOTAPE AND THE ATTRIBUTION PROCESS 167

brief getting-acquainted conversation between twostrangers (actors). In addition, two other subjects(observers) were told to watch the conversation butnot to participate in it.

Videotape replays of the conversation providedthe experimental manipulation. The design made itpossible to compare the effects of three orientationconditions: (a) one in which no visual reorientationwas attempted (no videotape), (6) one in whichvideotape was used simply to repeat the subject'soriginal orientations (same orientation), and (c) onein which videotape reversed the orientation of actorand observer (new orientation). In one set of condi-tions, actors and observers saw a videotape fromessentially the same orientation as they had hadin "real life." Actors saw a videotape replay of theother participant with whom they were conversing(actor—same orientation), and observers saw avideotape of the same actor they had been observingand about whom they would later answer questions(observer—same orientation). In another set of con-ditions, actors and observers received an entirely neworientation on videotape. Actors saw a videotape ofthemselves in the conversation (actor—new orienta-tion), and observers saw a videotape of the otherparticipant with whom their target actor had beenconversing (observer—new orientation). In addition,a set of actors and observers were run with novideotape replay.

SubjectsOne hundred and twenty Yale undergraduate male

volunteers participated in 30 groups of 4. Subjectswere solicited by sign-up sheets which specified thatpeople who volunteered for the same session shouldnot be previously acquainted.

ProcedureWhen each group of four subjects arrived at the

experiment, they were told,

This is a study in an area of social psychologycalled "interpersonal dynamics." More specifically,I'm interested in what I call "getting acquainted"—that is, what happens when two strangers meetfor the first time and initiate their first conversa-tion. Two of you in this study will be having ashort, first conversation with each other. In addi-tion, this study calls for two observers.

Subjects were randomly assigned to the role of actor(actually referred to as participant in the script) orobserver. Two subjects were assigned to be actorsand to have a getting-acquainted conversation to-gether. Each of the remaining two observer subjectswas assigned to observe his matched actor duringthe conversation.

The experimenter then mentioned,

There is one thing I would like to add to theprocedure today. I've gotten hold of some video-tape equipment and I will be taping your con-

FIG. I. Setup of the experimental room.

versation. My thought was that it might be usefulto you in answering the questionnaires to see theconversation replayed on tape.

Subjects were then seated in the experimental roomas shown in Figure 1. Actors sat at one end of thetable, across from each other, with one camerafocused on each. Observers sat at the other end ofthe table, diagonally across from and facing theirmatched actors. The experimenter reiterated thatthe conversation would last about 5 minutes, thatthe actors could talk about anything they wished,perhaps starting with their names and where theylived, and that observers should silently watch theirmatched actors.

After adjusting the equipment, the experimentersignaled to the participants to begin their conversa-tion. Five minutes later, he asked them to stop andwait silently while the tapes were rewound. At thispoint, the experimental manipulation was performed.A random number table was consulted to determinewhether the session would be a control session, inwhich case the subject would not see any tape, oran experimental session. If an experimental sessionwas indicated, the experimenter continued, "I'mafraid only one camera was working very well andthe other one is just too poor to see anything. Sowe'll only be able to see one of you on the video-tape." Experimental subjects were always shown thetape of Actor 1.

Thus one actor, Actor 2, saw a tape of the sameparticipant he had just seen in real life (Actor I)and was the actor-same orientation subject. Theother actor, Actor 1, viewed the tape of himself andwas the actor-new orientation subject. Similarly, oneobserver, Observer I, saw a tape of the same actorhe had been observing in the conversation (Actor 1)and was the observer-same orientation subject. Theother observer, Observer 2, saw a tape of the par-ticipant whom he had not been observing previously(Actor 1) and was the observer-new orientationsubject. Thus each experimental session yielded onesubject in each of the four experimental cells.

If a control session was indicated, the experimentersaid the following instead: "I'm afraid this is lousyequipment. It just didn't take a good enough pictureto be worth our while looking at it. So we'll just

Page 4: VIDEOTAPE AND THE ATTRIBUTION PROCESS: REVERSING … · Journal ol Personality and Social Psychology 1973, Vol. 27, No. 2, 165-175 VIDEOTAPE AND THE ATTRIBUTION PROCESS: REVERSING

168 MICHAEL D. STORMS

have to skip the tapes and go on to the question-naire." These no-videotape control sessions producedtwo actor-no-videotape subjects and two observer-no-videotape subjects.

At this point, for control subjects, and after thevideotape replay for experimental subjects, the ex-perimenter introduced the questionnaire, stressingthat it was confidential and that the subjects wouldnot see each other's responses. When the subjectscompleted the questions, they were debriefed. At thistime, the experimenter raised the issue of experimen-tal deception, but no subject indicated suspicion thatthe videotape had been a deliberate manipulation oreven an essential part of the experiment.

MeasuresOn the postexperimental questionnaire, actor sub-

jects answered mostly questions about themselves,and observer subjects answered questions about theirmatched actor. After a few introductory filler items,a page of instructions and the key dependent mea-sures of attribution were presented. The instructionsinformed subjects that in the next part of thequestionnaire they would be asked to describe theirown (their matched actor's) behavior along fourstandard dimensions: friendliness, talkativeness, ner-vousness, and dominance. Then, for each of the fourbehaviors, subjects were to indicate how much influ-ence they thought the following two factors had incausing that behavior:

(A) Personal characteristics about yourself (yourmatched participant): How important were your(his) personality, traits, character, personal style,attitudes, mood, and so on in causing you (him)to behave the way you (he) did?

(B) Characteristics of the situation: How im-portant were such factors as being in an experi-ment, the "getting acquainted" situation, the topicof conversation, the way the other participantbehaved and so on in causing you (him) to behavethe way you (he) did?

Thus, on each of the next four pages, three ques-tions were presented. The first asked about the per-ceived level of behavior on one of the four dimen-sions, for example, "To what extent did you (yourmatched participant) behave in a friendly, warmmanner?" The question was followed by a 9-pointscale labeled extremely friendly (9) to extremelyunfriendly (1). Presented next were the two attribu-tion questions: "How important were personalcharacteristics about you (your matched participant)in causing you (him) to behave that way?" and"How important were characteristics of the situationin causing you (him) to behave that way?" Each ofthese questions was followed by a 9-point scalelabeled extremely important (9) to extremelyunimportant (1).

These last two questions, repeated over the fourbehavioral dimensions, provided the principal andmost direct measure of subjects' attributions. Thesefour dimensions were not selected on the basis of

any particular theoretical or empirical considerations,but simply because it was anticipated that subjectswould manifest behaviors along each of these dimen-sions and that subjects would be able to make judg-ments about them. Since the hypothesis was con-cerned with the relative strength of dispositionalversus situational attributions and made no distinc-tions among the four behavioral dimensions, theappropriate measure was the difference between per-ceived importance of personal characteristics andperceived importance of situational characteristics incausing the actor's behavior, summed over all fourbehaviors. This difference score was referred to asthe dispositional-situational index. A higher value onthis index indicated that a subjects' attributions wererelatively more dispositional and less situational. It isimportant to note this dual meaning of the dispo-sitional-situational index. When an effect is describedas "relatively more dispositional," it is equally validto say "relatively less situational."

A second, less direct measure of the subjects'attributions appeared later in the questionnaire. Thesubjects were asked to report their estimates of theactor's level of behavior in general on each of thefour behavioral dimensions, for example, "Howfriendly a person are you (is your matched partici-pant) in general?" Responses were made on a scalefrom very friendly (9) to very unfriendly (1). Itwas then possible to compare these answers to thesubjects' previous answers about the actor's level ofbehavior in the conversation. If a subject had per-ceived that the actor's behavior in the conversationwas due to a stable personal disposition, then thesubject would likely have predicted that the actorbehaved the same way in general. Thus, dispositionalattributions would lead to a low discrepancy betweenthe subject's perception of the actor's behavior in theconversation and his behavior in general. On theother hand, if the subject had thought that theactor's behavior was caused by the situation, hewould more likely have reported that the actorbehaved differently in general. Thus, situationalattributions would lead to greater discrepancy be-tween the subjects' perceptions of the actor's presentand general levels of behavior. The simplest measureof this discrepancy was the absolute value of thedifference between the present level-of-behaviorscores and the general level-of-behavior scores,summed over all four behaviors. This measure wasreferred to as the present-behavior-general-behaviorindex. The higher the value of this discrepancy index,the more a subject made situational (or the less hemade dispositional) attributions.

The remainder of the questionnaire contained itemsnot directly related to present concerns.

RESULTS

Dispositional versus Situational Attributionsfor Behavior

The main hypothesis of the present studyconcerns the effects of videotape reorientation

Page 5: VIDEOTAPE AND THE ATTRIBUTION PROCESS: REVERSING … · Journal ol Personality and Social Psychology 1973, Vol. 27, No. 2, 165-175 VIDEOTAPE AND THE ATTRIBUTION PROCESS: REVERSING

VIDEOTAPE AND THE ATTRIBUTION PROCESS 169

on actors' and observers' causal attributionsof the actor's behavior. Before considering theeffects of reorientation, however, it is helpfulto examine the evidence pertinent to theoriginal Jones and Nisbett (1971) hypothesisthat actors are characteristically inclined toattribute causality to aspects of the situation,while observers tend to attribute causality tothe actor's disposition. Evidence for thisproposition is found in two conditions of thepresent experiment: the no-videotape cells inwhich the subjects did not receive any video-tape replay, and the same-orientation cells inwhich the videotape merely repeated thesubjects' original visual perspectives.

The relevant data are presented in Table 1.The key dependent measure, the total dispo-sitional-situational index, reflects the relativestrength of dispositional and situational attri-butions; a higher value on this index indicatesrelatively more dispositional (less situational)attributing. A comparison of the disposi-tional-situational means for actors and ob-servers in the no-videotape and same-orienta-tion cells reveals that, in both of theseconditions, actors attributed relatively moreto situational causes than did observers(p < .12, p < .05, respectively).3 It is furthernoted from these data that a videotape whichmerely repeated the subjects' original orienta-tion had little effect on either actors orobservers. Dispositional-situational scores foractors in the same-orientation condition didnot differ from those for actors in the no-videotape condition (q — 1.79, ns), and scoresfor observers in the two conditions were alsosimilar (q — 1, ns). Thus, under conditionsof no videotape and under conditions of re-peated videotape orientation, the subject'srole as actor or observer was an importantdeterminant of attributions. Actors attributedtheir own behavior relatively more to situa-tional causes, and observers attributed the be-havior relatively more to dispositional causes.

TABLE 1DISPOSITIONAL, SITUATIONAL, AND DISPOSITIONAL

MINUS SITUATIONAL ATTRIBUTION SCORES TOTALEDOVER ALL FOUR BEHAVIOHS

8 These comparisons, and all two cell comparisonsin the present study, are based on the q statisticfrom the Newman-Keuls procedure for testing dif-ferences among several means (see Winer, 1962).The degrees of freedom, taken from the overallanalysis of variance, equal 114; « equals 20 per cell.The Newman-Keuls is a more stringent test than theusual two-tailed t test.

Attribution

Actors' attributions ofown behavior

DispositionalSituationalDispositional-

situationalObservers' attribu-

tions of matchedactor's behavior

DispositionalSituationalDispositional-

situational

Sameorientation

26.1025.95

.15"

27.1022.20

4.90b°

Novideotape

27.3525.10

2.25ab

27.3022.50

4.80bo

Neworientation

27.5020.70

6.80°

25.7524.15

1.60»"

ATo/tf.Dispositional-situational means not sharing the samesuperscript are significantly different at the .05 level or beyondby Newman-Keuls tests.

The main hypothesis of the present studycan be examined with the data presented inthe last column of Table 1. It was anticipatedthat actors who saw themselves on videotapewould become relatively less situational(more dispositional) in attributions of theirown behavior, while observers who saw avideotape of the other participant with whomthe actor had been conversing would becomerelatively more situational (less dispositional)in their attributions of the actor's behavior.Since opposite effects of videotape reorienta-tion were predicted for actors and observers,the hypothesis was properly tested by theinteraction between subjects' roles (actor orobserver) and videotape orientation. The pre-dicted Role X Videotape Orientation interac-tion was obtained at beyond the .001 level ofconfidence (F = 9.72, df = 2/114, p < .001).Neither the main effect for role, nor the maineffect for videotape orientation was signifi-cant. The interaction reflected a completereversal of the relative perspectives of actorand observer in the new-orientation condition.In the same-orientation and no-videotapeconditions, the actors' attributions were moresituational than the observers'. In the new-orientation condition, in contrast, the actorswere relatively more dispositional than theobservers. This reversed effect was significantin itself (p < .05).

Examining the simple dispositional andsituational scores also presented in Table 1,

Page 6: VIDEOTAPE AND THE ATTRIBUTION PROCESS: REVERSING … · Journal ol Personality and Social Psychology 1973, Vol. 27, No. 2, 165-175 VIDEOTAPE AND THE ATTRIBUTION PROCESS: REVERSING

170 MICHAEL D. STORMS

it is apparent that reorientation had astronger influence on the subjects' evaluationof situational factors than on their evaluationof dispositional factors. The array of meansfor attributions to dispositional causes was inthe direction of the predicted interaction,but the effect did not reach significance(77=1.38, df = 2/114, ns). The situationalattribution scores showed the expected reversepattern, and the interaction was significant(F = 5.78, df = 2/114, p < .005).

The hypothesis is thus strongly supported.Visual orientation has a powerful influenceon the attributions of actors and observers.Indeed, the data in Table 1 suggest thestrongest possible conclusion: Under somecircumstances actual role as actor or observeris unimportant, and visual orientation istotally determinative of attributions.

Two other aspects of the dispositional-situational data are noteworthy, (a) Repeti-tion on videotape of essentially the sameinformation which had been presented in reallife had little effect on either the actors orthe observers. Actors in the same-orientationcondition were only slightly and nonsignifi-cantly more situational than no-videotapeactors, and same-orientation observers wereonly slightly and nonsignificantly more dispo-sitional than no-videotape observers, (b) Thepredicted experimental effects were not ob-tained with equal strength for all four of thebehaviors on which the total dispositional-situational index was based.

The fact that videotape in the same-orientation cells had little effect on the sub-ject's attributions suggests that mere repeti-tion of information and the addition of timeto review the event did not affect the sub-ject's perceptions of the event. The subjectsappear to have absorbed all relevant dataabout the event during its real-life occurrence.Of course, one would not necessarily expectthis to be true of all events. If the episodewere more complex or of longer duration,subjects could easily miss important informa-tion in vivo. A videotape replay would fillin these informational gaps and could, quitepossibly, produce different attributions.

The most noteworthy difference among thefour behavioral dimensions was the failure ofthe dominance dimension to contribute to the

experimental effects. Considering each behav-ioral dimension separately, the Role X Video-tape Orientation interaction was significantfor friendliness, talkativeness, and nervous-ness, each at the .025 level of confidence, butwas trivial for dominance (F < 1). Com-ments by subjects during the debriefing sug-gest a possible reason for the failure of domi-nance to contribute to the experimental ef-fects. Subjects complained that dominancewas a difficult dimension on which tojudge people in the context of a simple,5-minute getting-acquainted conversation.While friendliness, talkativeness, and ner-vousness are dimensions with concrete behav-ioral counterparts (such as smiling, talking,and fidgeting), apparently dominance is amore abstract dimension and requires a higherorder of inference.

When the dominance question was excludedfrom the analysis, each of the experimentaleffects was strengthened. Across the remain-ing three dimensions, the interaction test ofvideotape reorientation was strengthened froman F of 9.72 to an F of 13.89 (df = 2/114,p < .001). Tests for the Jones and Nisbett(1971) hypothesis were also strengthened;the contrast between actors and observers inthe no-videotape condition was significant atthe .05 level, and the contrast between actorsand observers in the same-orientation condi-tions was significant at the .01 level.

Perceived Level of Behavior and PerceivedDiscrepancy from General Behavior

In addition to the two attribution ques-tions, the subjects also answered questionsabout the perceived level of behavior on eachdimension. Past experiments in this area havetypically created a specific, standardized be-havior for subjects to attribute. The presentexperiment, with its unstructured conversa-tions, did not furnish all subjects with thesame behavior. This flexibility was desirable,in that it provided a more general test of theattribution hypotheses over several, naturallyoccurring behaviors. But it also created thepossibility that perceptions of the perceivedlevel or intensity of behavior could differamong experimental conditions and thus ac-count for the different attributions. This doesnot appear to have been the case, however.

Page 7: VIDEOTAPE AND THE ATTRIBUTION PROCESS: REVERSING … · Journal ol Personality and Social Psychology 1973, Vol. 27, No. 2, 165-175 VIDEOTAPE AND THE ATTRIBUTION PROCESS: REVERSING

VIDEOTAPE AND THE ATTRIBUTION PROCESS 171

There were two ways of calculating perceivedlevel of behavior: (a) by taking the directvalue from the 9-point scale for each level-of-behavior response and (b) since the scaleswere bipolar (for example, 9 = very friendlyto 1 = very unfriendly), by taking the devia-tion of the subject's response from the mid-point of the scale (5). Neither of these mea-sures yielded significant comparisons betweenany cells in the experiment, either for eachbehavior considered separately or for all fourbehaviors totaled. Furthermore, the overallcorrelations between the total dispositional-situational measure of attributions and thetwo measures of perceived level of behaviorwere trivial and nonsignificant (r — — .049,for the direct score; r = —.021, for the scoreof deviation from midpoint). Thus, it is ap-parent that differences in perceived level ofbehavior could not account for the attributiondifferences.

Since there were no significant differencesin perceived level of behavior, it is meaning-ful to examine the second measure of subjects'attributions, the present-behavior-general-behavior discrepancy scores. This index re-flected the absolute difference between thesubjects' perceptions of the actor's presentbehavior (in the conversation) and the actor'sgeneral behavior, summed over all four be-haviors. A small discrepancy would indicatethat a subject expected the actor's presentbehavior to generalize and was thus makinga dispositional attribution. A greater discrep-ancy would indicate less generalization ofthe actor's behavior and thus a situationalattribution.

The results of the present-behavior-general-behavior discrepancy measure, presented inTable 2, corroborated the findings on thedispositional-situational measure of attribu-tions. The effects of videotape reorientation,as tested in the Role X Videotape Orientationinteraction, reached significance at p < .05,(F = 3.38, df= 2/114). Again, neither themain effect for role nor the main effect fororientation was significant. Although the di-rection of differences between the actors andobservers in the various conditions was asexpected, none of the individual compari-sons between cells reached significance onthe present-behavior-general-behavior mea-

TABLE 2PRESENT BEHAVIOR MINUS GENERAL BEHAVIOR

DISCSEPANCY SCORES SUMMED OVERALL FOUR BEHAVIORS

Subjects

ActorsObservers

Sameorientation

7.155.45

Novideotape

5.004.90

Neworientation

4.255.90

sure, even with the exclusion of the domi-nance dimension. It appears that the re-sults for the present-behavior-general behav-ior measure followed the same pattern as, butwere generally weaker than, the results forthe dispositional-situational measure. Thetwo measures were, incidentally, significantlycorrelated (overall r — .361, p < .01).

DISCUSSION

The present study demonstrates that visualorientation has a powerful influence on theinferences made by actors and observersabout the causes of the actor's behavior.When videotape was not presented and sub-jects were left to assume their own orienta-tions, or when videotape reproduced subjects'original orientations, actors attributed theirbehavior relatively more to situational causesthan did observers. This finding supports theJones and Nisbett (1971) hypothesis thatactors' attributions are typically more situa-tional than observers'. But under conditionsof reorientation, when subjects saw a newpoint of view on videotape, the attributionaldifferences between actors and observers wereexactly reversed. Reoriented, self-viewing ac-tors attributed their behaviors relatively lessto situational causes than did observers. Thiseffect was obtained on two very differentmeasures of attribution across a variety ofbehavioral dimensions in an unstructuredsituation.

Mechanisms of Videotape Reorientation

Two important issues arise concerning thepossible mechanisms by which video orien-tation affected attributions. The first issue,one crucial to any laboratory social psychol-ogy experiment, concerns experimenter de-mand characteristics. Demand characteristicscould have influenced the results of the

Page 8: VIDEOTAPE AND THE ATTRIBUTION PROCESS: REVERSING … · Journal ol Personality and Social Psychology 1973, Vol. 27, No. 2, 165-175 VIDEOTAPE AND THE ATTRIBUTION PROCESS: REVERSING

172 MICHAEL D. STORMS

ent study if the hypotheses had been com-municated to subjects either by the experi-menter's behavior or by the fact the subjectsviewed only one videotape. Both of thesepossibilities depend on subjects' developingthe expectation that videotape had impor-tance for how they should respond. The pos-sibility of communicating the hypotheses wasavoided by leading subjects to believe thatvideotape was not an essential part of theexperiment and that the experimenter hadwanted to show both tapes but could not,due to circumstances beyond his control.During debriefing, subjects were questionedon their reactions to this hoax; they reportedno suspicion that the videotape breakdownhad been intentional or important. Moreover,if subjects had been responding to the attri-bution questions out of desire to supportthe experimenter's hypotheses, it is un-likely they could have produced the results ofthe indirect present-behavior-general-behav-ior measure. This index was derived from theabsolute value of the difference between thefour level-of-behavior questions and the fourgeneral-behavior questions. These questionswere widely separated in the questionnaire,and subjects would have had to perform arather elaborate calculus to produce these re-sults deliberately. Thus, it does not seemlikely that the reorientation effects can beaccounted for by experimenter demandcharacteristics.

The second issue involves the possiblemechanisms by which videotape caused thepredicted attributions. This study was de-signed to demonstrate that a global manipu-lation (visual orientation) affects actors' andobservers' attributions of the actor's behavior.The study was not designed to separateout the many possible mechanisms by whichthis might occur. However, some informedspeculation is possible.

Jones and Nisbett (1971) proposed severalfactors that contribute to attributional dif-ferences between actors and observers, in-cluding differences in the information avail-able about an event and differences in howthat information is processed. These two cate-gories are not mutually exclusive, and video-tape orientation may have affected aspects ofboth information availability and information

processing. When actors or observers saw avideotape of an event from a different pointof view, they may have received some totallynew information. The actor may have real-ized, for the first time, some new aspects ofhis own behavior; the observer may have seennew aspects of the situation or of the otherparticipant. These new facts could have con-tributed to changes in subjects' inferencesabout the cause of behavior. Second, thesalience of already available information mayhave changed for reoriented subjects. Changesin the salience of information have beenshown to affect people's perceptions of thereasons for their behavior. For example,Kiesler, Nisbett, and Zanna (1969) foundthat subjects tended to adopt as explana-tions of their own behavior motives that weremade salient by a confederate. Similarly, sub-jects in the present study might have formu-lated their attributions about the actor's be-havior on the basis of potential causes whichhad just been made salient by the videotape.Finally, videotape reorientation may haveproduced new response sets for subjects. Ac-tors who viewed themselves on tape may havebeen put into a "self-discovery" frame ofmind and thus led to think about their ownpersonality as revealed in their behavior.Similarly, observers who saw a videotape fromthe actor's point of view may have developedan "empathic" set, imagining themselves tobe in the actor's shoes.

It is also of interest to consider the exactnature of the attributional changes evoked byvideotape. Changes on the key dependentvariable, the dispositional-situational index,were accounted for mostly by changed evalua-tions of situational causes. Actors assigned agreat deal of causality to the situation unlessvideotape forced them to look away from thesituation and toward their own behavior. Ob-servers originally assigned less causality tothe situation unless videotape impressed situ-ational factors on them. Differences in attri-bution to dispositional causes, although in theexpected direction, were much weaker thanthese differences in attribution to situationalcauses. It may be that the relatively greateramount of change on the situational dimen-sion reflects people's general way of viewingthe role of dispositions in causing behavior.

Page 9: VIDEOTAPE AND THE ATTRIBUTION PROCESS: REVERSING … · Journal ol Personality and Social Psychology 1973, Vol. 27, No. 2, 165-175 VIDEOTAPE AND THE ATTRIBUTION PROCESS: REVERSING

VIDEOTAPE AND THE ATTRIBUTION PROCESS 173

People may characteristically assign fixed andfairly high importance to personal responsi-bility for behavior. Consequently, they maybe left with only one means of modifyingtheir relative assignment of causality and re-sponsibility, namely by varying their evalua-tions of the situation. In line with this pos-sibility, there may have been a ceiling effectfor dispositional attributions in the presentstudy; the overall mean importance assignedto dispositional causes equaled nearly 7 outof a possible 9 scale points. Subjects werethus left with little room to express enhanceddispositional influences.

Up to this point, discussion has been lim-ited to information-related variables whichmay be modified by video exposure and mayin turn affect attributions. Undoubtedly,motivational variables, such as the need tomaintain self-esteem and particular self-con-cepts, could also be affected by videotapeobservations. One might expect the self-view-ing actors in particular to be influenced bysuch motivations. It is important to note,however, that the present findings were ob-tained in a situation which was, in manyrespects, low-key. The behaviors elicited inthe gettingiacquainted conversations wereroutine and probably not highly relevant toactors' self-concepts, the interaction betweensubjects was fairly unemotional, and actorsand observers did not have the opportunityto discuss their potentially opposing views ofthe actor's behavior. It is therefore importantto consider whether the present findingswould generalize to situations where actorsand observers are more emotionally involved,such as in psychotherapy and T groups. Thereis reason to believe that the present findingshave some applicability to the use of video-tape even in such emotionally charged set-tings.

Videotape in Therapy and T Groups

There has been a recent and dramatic in-crease in the application of videotape feed-back in therapy and human relations training.Alger and Hogan (1966a) asserted that"videotape recording represents a technologi-cal breakthrough with the kind of significancefor psychiatry that the microscope has hadfor biology [p. 1]." In clinical practice,

videotape is frequently used to increase apatient's knowledge of his own behavior (cf.Bailey & Sowder, 1970; Holzman, 1969), andthis apparently leads to therapeutic gain. Rei-vich and Geertsma (1968) reported increasedaccuracy in patients' knowledge of their ownbehavior after videotape self-observation. Theymeasured the disparity between a patient'sself-ratings on clinical scales and the ratingsgiven him by psychiatric nurses. After video-tape self-observation, the ratings of the actorpatient came to agree more with the ratingsof the observer nurses. Alderfer and Lodahl(1971) found that videotape playback in Tgroups increased subjects' "openness." Open-ness was defined as willingness to explore theinternal meaning of and accept personal re-sponsibility for an attitude or behavior. Fi-nally, case studies in marital therapy (Alger& Hogan, 1966a; Kagan, Krathwohl, & Mil-ler, 1963) have reported that one or bothmarriage partners are more willing to assumethe blame for a poor relationship after seeingthemselves on videotape.

On the other hand, some negative conse-quences of self-observation have also beenreported. For instance, Carrere (1954) usedvideotape to show alcoholics how they be-haved when intoxicated, but he found it nec-essary to edit the more shocking scenes. Thefull presentation of their behavior whendrunk was too stressful for many of his pa-tients. Parades, Ludwig, Hassenfeld, andCornelison (1969) similarly reported the low-ering of alcoholic patients' self-esteem afterviewing their own drunken behavior on tape.Leitenberg, Agras, Thompson, and Wright(1968) gave behavioral feedback (althoughnot video) to phobic patients undergoing be-havior modification. These authors found thatfeedback to patients about successful progressspeeds their cure, but information about tem-porary setbacks interferes with the therapy.Finally, Geertsma and Reivich (1965) re-ported that some self-viewing depressive pa-tients become more depressed, some schizo-phrenic patients engage in more bizarre be-havior, and some neurotics show an increasein the symptoms characteristic of their par-ticular disorder.

Research to date on the use of videotape intherapy is insufficient to indicate how and

Page 10: VIDEOTAPE AND THE ATTRIBUTION PROCESS: REVERSING … · Journal ol Personality and Social Psychology 1973, Vol. 27, No. 2, 165-175 VIDEOTAPE AND THE ATTRIBUTION PROCESS: REVERSING

174 MICHAEL D. STORMS

with whom it is a beneficial therapy adjunct.It may be possible, however, to apply thefindings and the theoretical framework of thepresent study to the issue of videotape use intherapy. The present study demonstrates thatself-observation can change the causal inter-pretation a person gives to his own behavior.The self-viewing actor (and possibly theself-viewing patient) is more likely to acceptpersonal, dispositional responsibility for hisbehavior and is less likely to deflect responsi-bility to the situation.

This attributional consequence of self-ob-servation may help to account for some ofthe effects of videotape in therapy. For ex-ample, the increased openness of T groupparticipants after self-observation may reflecta tendency for each group member to assumemore personal, dispositional responsibility forhis behavior in the group. Similarly, in mari-tal therapy, the husband or wife who seeshimself or herself on videotape may realizefor the first time his or her own behavioralcontribution to the marital conflict and maybe more willing to place a dispositional blameon himself or herself. Finally, the reportedincrease in agreement between a patient'sclinical self-ratings after videotape self-obser-vation and the ratings of observing psychi-atric nurses closely parallels the present find-ings. Self-observation increases an individual'sdispositional attributions, thus bringing himmore in agreement with the observer's built-in bias for dispositional attributions.

It seems likely that this increase in dispo-sitionality of a patient's attributions wouldprove to be sometimes therapeutic and some-times distherapeutic. Successful therapy nodoubt usually involves making a patientaware of his own behavior and convincinghim to accept personal responsibility for thatbehavior. Self-observation apparently aidsthis process and, to that extent, should betherapeutic. However, two potentially nega-tive outcomes of this process might be sug-gested. First, in becoming more dispositionalabout their own behavior, individuals whosee themselves on videotape may actuallyunderestimate real and viable situational ex-planations for their behavior. Actors in thepresent study who saw their own behavior on

videotape had a higher mean for dispositionalattributions and a lower mean for situationalattributions than any other group of subjects.This suggests the possibility that self-viewingactors may have been "undersituational" inattributing their own behavior. That is, video-tape may have reoriented these actors somuch that they perceived situational causesfor their behavior to be even less importantthan did others who viewed them. And if, asJones and Nisbett (1971) have suggested, ob-servers are themselves inclined to under-attribute to the situation, this poses a dis-turbing possibility for therapy. Ironically, thetherapist and the self-viewing patient couldreach complete agreement about the patient'sbehavior, yet this agreement could result froma mutual underestimation of the importanceof the patient's situation in causing his be-havior. This collaborative illusion betweenpatient and therapist could be especially harm-ful if the patient blames himself for behaviorthat is in fact due to some aspect of his en-vironment.

Past research on attribution processes hasuncovered another area where attributions tothe self can have distherapeutic results.Storms and Nisbett (1970) and Valins andNisbett (1971) have suggested that negativeself-labeling which results from attributinguncomplimentary behaviors to dispositionswithin oneself often lead to a loss of self-es-teem and an actual increase in the patho-logical behavior. For example, insomniacs whoattribute their sleeplessness to some negativestate within themselves may increase theiranxiety and thus aggravate their original con-dition. Storms and Nisbett proposed that suchexacerbation may result whenever self-attri-butions of a negative disposition increase theindividual's anxiety and when anxiety is anirritant to the pathology, such as in impo-tence, stuttering, and other neurotic condi-tions. This exacerbation phenomenon may beoccurring in some of the therapy cases wherenegative results have followed the use ofvideotapes. The finding that self-observationlowers the self-esteem of alcoholic patientsmight be an instance of this. An alcoholic pa-tient who sees a tape of his own drunkenbehavior may become quite upset and de-

Page 11: VIDEOTAPE AND THE ATTRIBUTION PROCESS: REVERSING … · Journal ol Personality and Social Psychology 1973, Vol. 27, No. 2, 165-175 VIDEOTAPE AND THE ATTRIBUTION PROCESS: REVERSING

VIDEOTAPE AND THE ATTRIBUTION PROCESS 175

pressed about himself. Such a traumatic ex-perience may only increase the likelihood thatthe patient will drink to excess. Whenever apathology is caused or influenced by a poorself-concept, self-observation of extremelyuncomplimentary behavior may serve to re-tard therapeutic progress.

Research on attribution processes may helpto create a theoretical framework for the areaof videotape self-observation in therapy set-tings. The present study suggests that self-observation increases an individual's disposi-tional attributions of his own behavior andthat this brings interpretation of his behaviormore in line with an observer's interpreta-tion. In most cases, this should be advan-tageous to the therapy process, but in cer-tain cases self-attributions could lead to anexacerbation of the original pathology. Thera-pists would therefore be well advised to lookcritically at the potential consequences ofself-observation. It seems especially impor-tant to consider whether a personal, disposi-tional attribution of the pathological behavioraids the patient to become aware of his prob-lem and to deal with it, or whether self-attribution increases the patient's anxiety tothe point of exacerbating his problem.

REFERENCES

ALDERFER, C. P., & LODAHL, T. M. A quasi experi-ment on the use of experiential methods in theclassroom. Journal oj Applied Behavioral Science,1971, 7, 43-69.

ALGER, I., & HOGAN, P. The use of videotape record-ings in conjoint marital therapy. Paper presentedat the meeting of the American PsychoanalyticAssociation, Atlantic City, N.J., May 1966. (a)

ALGER, I., & HOGAN, P. Videotape: Its use and sig-nificance in psychotherapy. Paper presented at themeeting of the Society of Medical Psychoanalysts,New York Academy of Medicine, New York, Sep-tember 1966. (b)

BAILEY, K. G., & SOWDER, W. T. Audiotape andvideotape self-confrontation in psychotherapy.Psychological Bulletin, 1970, 74, 127-137.

CARRERE, M. J. Le psychochoc cinematographique.Annales Medico-Psychologiques, 1954, 112, 240-245.

GEERTSMA, R. H., & REIVICH, R. S. Repetitive self-observation by videotape playback. Journal ofNervous and Mental Disease, 1965, 141, 29-41.

HOLZMAN, P. S. On hearing and seeing oneself.Journal of Nervous and Mental Disease, 1969,148, 198-209.

JONES, E. E., & HARRIS, V. A. The attribution ofattitudes. Journal of Experimental Social Psy-chology, 1967, 3, 1-24.

JONES, E. E., & NISBETT, R. E. The actor and theobserver: Divergent perceptions of the causes ofbehavior. Morristown, N.J.: General LearningPress, 1971.

JONES, E. E., ROCK, L., SHAVER, K. G., GOETHALS,G. R., & WARD, L. M. Pattern of performance andability attribution: An unexpected primacy effect.Journal of Personality and Social Psychology,1968, 10, 317-340.

PAGAN, N., KRATHWOHL, D. R., Si MILLER, R. Stimu-lated recall in therapy using videotape: A casestudy. Journal of Counseling Psychology, 1963,10, 237-243.

KIESLER, C. A., NISBETT, R. E., & ZANNA, M. P. Oninferring one's beliefs from one's behavior. Jour-nal of Personality and Social Psychology, 1969,11, 321-327.

LEITENBERG, H., AGRAS, W. S., THOMPSON, L. E., &WRIGHT, D. E. Feedback in behavior modifica-tion: An experimental analysis of two cases. Jour-nal of Applied Behavioral Analysis, 1968, 1, 131-137.

McARTHUR, L. Appropriateness of the behavior andconsensus and distinctiveness information as de-terminants of actors' and observers' attributions.Unpublished manuscript, Yale University, 1970.

MCARTHUR, L. A. The how and what of why: Somedeterminants and consequences of causal attribu-tion. Journal of Personality and Social Psychol-ogy, 1972, 22, 171-193.

NISBETT, R. E., CAPUTO, G. C., LEGANT, P., &MARECEK, J. Behavior as seen by the actor and asseen by the observer. Journal of Personality andSocial Psychology, 1973, 27, 154-164.

PARADES, A., LUDWIG, K. D., HASSENFELD, I. N., &CORNELISON, F. S. A clinical study of alcoholicsusing audio-visual self-image feedback. Journal ofNervous and Mental Disease, 1969, 148, 449-456.

REIVICH, R. S., & GEERTSMA, R. H. Experiences withvideotape self-observation by psychiatric in-patients. Journal of Kansas Medical Society, 1968,69, 39-44.

STORMS, M. D., & NISBETT, R. E. Insomnia and theattribution process. Journal of Personality andSocial Psychology, 1970, 16, 319-328.

VALINS, S., & NISBETT, R. E. Some implications ofattribution process for the development and treat-ment of emotiotial disorders. Morristown, N.J.:General Learning Press, 1971.

WINER, B. J. Statistical principles in experimentaldesign. New York: McGraw-Hill, 1962.

(Received February 24, 1972)