James H. Bryan Educational Testing Service and Mary Ann Test Northwestern University

Transcription

1 RB F E S E A fi e L L E T I A LADY IN DISTRESS; THE FLAT TIRE EXPERIMENT James H. Bryan Educational Testing Service and Mary Ann Test Northwestern University This Bulletin is a draft for interoffice circulation. Corrections and suggestions for revision are solicited. The Bulletin should not be cited as a reference without the specific permission of the authors. It is automatically superseded upon formal publication of the material. Educational Testing Service Princeton, New Jersey September I966

2 A Lady in Distress: The Flat Tire Experiment Abstract The effects of observing a helping model upon subsequent offers of help were assessed in a naturalistic setting. Passing motorists were exposed to a contrived situation wherein a girl stood beside an auto which had a flat left rear tire. One half of the passing motorists had previously seen a young male helping a girl in a similar situation (model condition); the remaining motorists were not exposed to such helping behaviors. The presence of a model significantly increased the helping behaviors of motorists.

3 A Lady in Distress: The Flat Tire Experiment Attention has been directed recently toward the study of altruistic behaviors; i.e., those acts entailing self-sacrifice for no apparent personal gain (Rosenhan & White ; London 8s Bower"; Berkowitz & Daniels, 1963; Midlarsky & Bryan, 1966). While such actions have generally been ignored by the psychological community, perhaps because of this community's biological orientation (Campbell, 1965), such behaviors have important practical and theoretical implications» At the very least such behaviors are useful to the survival of a variety of social institutions, man's included (Campbell, I965). Additionally, the study of such actions will further clarify the nature of reinforcement. Several recent experiments have suggested that one important determinant 2 of helping behaviors is the presence of models. Rosenhan and White have demonstrated that significantly more children will donate gift certificates, a highly valued secondary reinforcer, to fictitious orphanage after having seen an adult model do so than will E>s who were not exposed to such models. While many of the studies on modelling effects have been based on children (cf. Bandura & Walters, 1963)* several investigations have suggested that models might be effective in eliciting self-sacrificing behavior in adults. Experiments by Rosenbaum and Blake (1955) and. Roseribaum (I956) have found that college students exposed to a model who volunteered, upon the personal request of the E, to participate in an experiment would be more likely to consent than Ss not exposed to such a model or who observed a model refuse to cooperate. Pressures toward conformity in these experiments were great, however, as the request was made directly by the E and in the presence of a large number of other students.

4 -2- k Test and Bryan, under conditions designed to maintain S anonymity, but within a college laboratory, found that college students who were either helped or observed others being aided were more likely to offer help to a stooge than Ss who were provided with no model or a model who refused to help. Investigations of modelling that employ adults as Ss and that demand selfsacrifice on the part of Ss are limited in number, exploit strong pressures toward conformity, and rely upon college students as Ss. The present experiment was designed to assess the impact of models upon subsequent spontaneous offers of help by using adults other than college students in an experimental setting other than a university. Method The standard condition consisted of an undergraduate female stationed by a 196^ Ford Mustang (control car) with a flat left rear tire. An inflated tire was leaned upon the left side of the auto. The girl and the flat and the inflated tires were conspicuous to the passing traffic. In the model condition, a I9Ó5 Oldsmobile was located approximately a quarter of a mile from the control car. The car was raised by jack under the left rear bumper, and a girl was watching a male changing the flat tire. Stooges played the same role throughout the experiment. In the no-model condition, the model was absent; thus, only the control car was visible to the passing traffic. The cars were located in a predominantly residential section in Los Angeles, California. They were placed in such a manner that no intersection separated the

5 -3- model from the control car. No turnoffs were thus available to the passing traffic. Further, opposite flows of traffic were divided by a separator such that the first U turn available to the traffic going in the opposite direction of the control car would be after exposure to the model condition. The experiment was conducted on two successive Saturdays between the hours of 1:45 a QcL 5:50 Ρ Μ. Each treatment condition lasted for the time required for 1000 vehicles to pass the control car. While private automobiles and trucks, motorscooters and motorcycles were tallied as vehicles, commercial trucks, taxis, and busses were not. Vehicle count was made by a fourth member of the experiment who stood approximately 100 feet from the control car hidden from the passing motorists. On the first Saturday, the model condition was run first and lasted from 1:4-5 to 3:15, In order to exploit changing traffic patterns and to keep the time intervals equal across treatment conditions, the control car was moved several blocks and placed on the opposite side of the street for the no-model condition. The time of the no-model treatment was 4:00 to 5:00 P.M. On the following Saturday, counterbalancing the order and the location of treatment conditions,was accomplished. That is, the no-model condition was run initially and the control car was placed in the same location that it had been placed on the previous Saturday during the model condition. The time of the no-model condition was 2:00 to 300 P.M. For the model condition, the control car was placed in that locale where it had been previously during the no-model condition. The time of the model condition was 4:30 to 5:30 P.M. Individuals who had stopped to offer help were told by the young lady that she had already phoned an auto club and that help was imminent. Those who nonetheless insisted on helping her were told the nature of the experiment.,

6 Results and Discussion The dependent variable was the number of cars which stopped and from which at least one individual offered help to the stooge by the control car. Of the UOOO passing vehicles, 93 stopped. With the model car being absent, 35 vehicles stopped; with the model present, 58 halted. The difference between the conditions was statistically significant (X = 5 53> corrected for continuity, df = 1, p <.05, two-tailed). It should be noted that the time of day had little impact upon the offering of aid. Fifty vehicles stopped during the early part of the afternoon; h3, during the later hours. Likewise, differences in help offers were not great between successive Saturdays as k5 offers of aid were made on the first Saturday, HQ on the second Saturday. The results of the present study support the hypothesis that helping behaviors can be significantly increased through the observation of others' helpfulness. The data further underscore the importance of modelling for the adult S_. It is noteworthy that under conditions where conformity pressures are rather weak, that is, conditions where the direct confrontation with the requester" can be easily avoided, where requests are not made directly, where the anonymity of the actor is so easily maintained, and where self-gain is so minimal, that adults are still affected by the presence of models. While it is clear that the behavior of these motorists was not dictated by a variety of situational or social pressures usually associated with study of modelling in adults (cf. Freed, Chandler, Mouton, & Blake, 1955; Rosenbaum, 1956; Rosenbaum & Blake, 1955) or experiments in academic settings (Orne, 19^2), the

7 -5- mechanisms underlying the model effects are not obvious* Berkowitz and his colleagues (Berkowitz & Daniels, I963; Daniels & Berkowitz, 1963) as well as Schopler and Bateson (1965) have provided evidence that dependency states in others will elicit helping behavior from college students. They have argued for the existence of a "norm of social responsibility," a norm dictating the helping of dependent others. It is perhaps possible that the presence of a helping model served as a reminder of such a norm. If this were indeed the case, the model may not only serve to define normative behavior to adults (Blake, Rosenbaum, & Duryea, 1955)> "but also may serve to activate already internalized norms, Whatever the mediating events, the present study does support the contention that models are important in eliciting helping behaviors under those conditions where nonconformity would produce no interpersonal consequences and conformity would produce very little personal gain in adults other than college students in situations other than academic settings. The findings thus lend support to the generality of laboratory findings regarding the impact of models upon behavior.

8 -6- References Bandura, A., & Walters, R. Social learning and personality development. New York: Holt, Rinehart & Winston, I963. Berkowitz, L., & Daniels, Louise. Responsibility and dependency. Journal of Abnormal and Social Psychology, 1963, 66, k29-kj>6. Blake, R., Rosenbaum, Μ,, & Duryea, R. Gift-giving as a function of group standards. Campbell, D. T. Human Relations, 1955; 8, 6l-73. Ethnocentric and other altruistic motives. In D. Levine (Ed.), Nebraska symposium on motivation: I965. Lincoln: University of Nebraska Press, Pp II. Daniels, Louise, & Berkowitz, L. Liking and response to dependency relationships. Human Relations, I963, l6, ΐΑΐ-ίΑβ. Freed, A., Chandler, P., Mouton, Jane, & Blake, R. Stimulus and background factors in sign violation. Journal of Personality, 1955> 23, ^99- Midlarsky, Elizabeth, & Bryan, J. H. Training charity in children. Journal of Personality and Social Psychology, I966. (In press). Orne, M. On the social psychology of the psychological experiment: With particular reference to demand characteristics and their implications, American Psychologist, I962, 1, Rosenbaum, M. The effect of stimulus and background factors on the volunteering response. Journal of Abnormal and Social Psychology, I956, _5j>, , Rosenbaum, M., & Blake, R. Volunteering as a function of field structure. Journal of Abnormal and Social Psychology, 1955> 5_2> Schopler, J., & Bateson, N. The power of dependence. Journal of Personality and Social Psychology, I965, 2, 2ΚΊ-2%.

9 -7- Footnotes Thanks are due Cheryl Dellhoussaye, Betty Umann, Joel McNair, and Frank Siri who served so responsibly as the experimenters and stooges, and to David Rosenhan and Lawrence J. Strieker for their critical reviews of earlier manuscripts. Rosenhan, D., & White, G. Observation and rehearsal as a determinant of pro-social behavior. Unpublished manuscript, London, P., & Bower, R. K. Altruism, extraversion and mental illness. (Submitted for publication). Test, Mary Ann, & Bryan, J. H. Dependency Models and Reciprocity. Unpublished manuscript.