Investigation of the Effect of Bystander Behaviour on Helping Behaviour in a Non-Emergency Situation

Abstract

One model explaining whether people offer assistance in emergency situations is diffusion of responsibility: the greater the number of bystanders present, the less personal responsibility is felt by each bystander. A second model, the normative theory, suggests that people comply with social norms, being more likely to help if an appropriate helping response is modelled by others. To investigate which model best accounts for helping behaviour, a naturalistic study was conducted in which participants were able to offer help in a minor emergency. The participants were 1122 undergraduates from Monash University. Most helping was demonstrated when no bystanders were present. The results supported the diffusion of responsibility theory more than the normative theory.

In 1964, a New York city woman Kitty Genovese was brutally stabbed to death outside her apartment block. Of the 38 witnesses, none offered assistance, not even the simple act of notifying the police (Latane & Nida, 1981). This failure to help stimulated research into understanding why bystanders often fail to give assistance in emergency situations.

Latane and Darley (1970) wondered why people were unwilling to offer assistance in emergencies when they were quite happy to help in non-emergencies. They suggested a five-step model explaining what a bystander must do in order to help.

Stage 1. Recognise that an event is occurring.

Stage 2. Interpret the event as an emergency.

Stage 3. Decide that he or she has a personal responsibility to help.

Stage 4. Consider what form of help he or she can give.

Stage 5. Execute the helping response.

Helping behaviour is action that is intended to assist or benefit another person. This study is most concerned with Latane and Darley’s Stage 3. Completion of Stage 3 might be affected by whether the bystander considers the victim deserving of help, whether the bystander feels competent to help, the relationship of the bystander towards the victim, and whether responsibility is shared with other bystanders. Diffusion of responsibility is when the psychological costs of not helping are shared amongst bystanders. Diffusion of responsibility explains the bystander effect, which states that the more bystanders there are, the less likely it is than any one of them will help.

Explanatory theories for helping behaviour can be tested by creating an emergency situation and examining the unprompted helping behaviour given by naïve participants. To model the Kitty Genovese case, Latane and Darley (1970) created an emergency situation in which a number of bystanders were all able to provide help, but were unable to communicate with other bystanders or know the behaviour of the others. The results of this experiment suggested that as the number of bystanders present increased, the probability of any one bystander trying to help decreased (Latane & Darley, 1970). Many further studies by Latane, Darley and others, conducted with a wide variety of emergency situations, participants and levels of communication between bystanders, support this diffusion of responsibility hypothesis (Latane & Nida, 1981). When participants can communicate with each other, pluralistic ignorance can result – a majority of people in a group privately reject the group norm, but erroneously assume that most other people accept the norm (Brehm & Kassin, 1996). In an emergency situation, people look around, and may take the hesitancy of other bystanders as indicating that the situation isn’t an emergency. Asch (1999) suggests that our desire for social conformity is cause for concern, because individuals are prepared to disregard their own sound judgement and take on the majority view even when the majority is obviously wrong.

This is a preview of the whole essay

Bryan and Test (1967), however, suggest that the presence of others does not necessarily inhibit helping. In an emergency situation, whether people help is influenced by how others react, not solely by whether others are present. They suggest that the likelihood of helping behaviour can be increased if an appropriate helping response is modelled by others. In their naturalistic study, a woman was unable to change a tyre on the side of a highway. In the first situation, participants had observed a man helping a different woman change a tyre earlier along the road. In the second situation, no such model was provided. Bryan and Test found that a greater proportion of participants helped the woman in the first situation. This suggested that an observer will feel more social obligation or responsibility to help if he or she witnesses an observer helping than if there were no other bystanders. In fact, a greater feeling of social responsibility resulted. An observer seems likely to help when a model responds to a possible emergency situation by helping, following a social norm for responding to the situation (Bryan & Test, 1967).

Cialdini, Reno, and Kallgren (1990) conducted research which demonstrated that social norms impact considerably on littering behaviour. People take the behaviour of others as cues as to how they ‘should’ feel or behave in certain situations.

The aim of the present study is to determine which of the two competing theories, the diffusion and normative theory, best accounts for helping behaviour.

To test the diffusion theory, participants either were alone or with another person when given an opportunity to provide help in a mild or ‘non-emergency’ situation. To simultaneously test the normative theory, there were two conditions in which the participant was in a situation with another bystander; one in which the bystander helped (the helping-confederate condition) and the other in which the bystander did not help (the non-helping-confederate condition).

In Bryan and Test’s (1967) study, participants were present at two separate events, and so it did not allow for the diffusion of responsibility to come into play. In the current study, comparison of helping behaviour when the bystander is present or not present allows us to test the diffusion theory. The comparison of helping behaviour when the bystander helps or does not help allows us to test the normative theory.

According to the diffusion of responsibility theory, there will be more helping behaviour in the no-confederate condition than in the confederate conditions. According to the normative theory, the helping-confederate condition will produce more helping behaviour than the non-helping-confederate condition, and also more than the no-confederate condition.

Method

Participants

1122 students from Monash University campus grounds (Caufield, Clayton, Gippsland, Peninsula and Sunway campuses) participated. Opportunistic sampling was used. The ages and genders of participants were not recorded.

Materials
100 A4 pages of loose-leaf paper.

Procedure

A naïve participant, who was not engaged in any activity requiring a high level of concentration (such as walking across the campus), had the opportunity to help the experimenter pick up loose pieces of papers that the experimenter dropped 3-4 metres away from the participant, and more than 10 metres away from any other people.

In the confederate conditions, the confederate (3-4 metres away from the experimenter) provided a model for ‘appropriate behaviour’ by either helping or not helping the experimenter to retrieve the paper. In the no-confederate condition, only the experimenter and participant were present. Experimenters were randomly allocated to one of the three conditions, and only tested one participant.

If helping, the confederate did so in such a way that the participant’s assistance would reasonably be considered by the participant to be desirable to the experimenter. The confederate was engaged in an activity requiring a low level of attention (such as looking at his or her mobile phone or looking through his or her bag), so that the confederate’s behaviour appeared natural. The papers were dropped such that the papers were spread out and close to the participant’s and confederate’s feet. The experimenter immediately began to pick up the papers, and did not request assistance. The experimenter did not interact with the participant or any confederate. An observer standing away from the site recorded whether helping behaviour was exhibited by the participant. Helping behaviour was defined by whether the participant picked up one or more pages within 30 seconds of the papers being dropped. The observer also determined when the trial could not be included in the results, for example if a third person helped or if the participant was suddenly unable to help.

Design

The experiment had a single factor, between groups design. The independent variable was the presence and behaviour of confederate, and consisted of three levels: no-confederate, helping-confederate, and non-helping-confederate. The dependent variable was whether or not the participant displayed helping behaviour.

Results

The raw scores are presented below along with their percentages. The percentages express the proportions of participants who helped the experimenter for the no-confederate, helping-confederate and non-helping-confederate conditions separately.

Table 1

Number and Percentage of Persons Exhibiting a Helping Response as a Function of Confederate Condition

As shown in Table 1, the amount of helping was greatest in the no-confederate condition, then the helping-confederate condition, and the least helping was shown in the non-helping-confederate condition.

Discussion

The diffusion theory predicts a decrease in helping behaviour when a bystander is present. The results supported the diffusion theory, as more participants demonstrated helping behaviour in the no-confederate condition compared to the two confederate conditions. The results suggest that the presence of bystanders in a mild emergency situation inhibits the likelihood of a helping response.

The normative theory predicts an increase in helping behaviour when a bystander offered help as compared to when the bystander did not offer help (Bryan & Test, 1967). The normative theory was partially supported, in that participants in the helping-confederate condition showed more helping behaviour than the participants in the non-helping-confederate condition. This suggests that if one bystander is present, the likelihood of helping behaviour from someone witnessing a mild emergency situation is increased if the bystander is helping rather than not helping.

However, a higher proportion of participants helped in the no-confederate condition than in the helping-confederate condition. This result may challenge the normative theory, which suggests that the participant would be more likely to help after seeing a model of helping behaviour than if no bystander were present (Bryan & Test, 1967).

The observation that there was a greater incidence of helping behaviour in the no-confederate condition is consistent with the findings of Latane and Darley (1970). That there was more helping in the helping-confederate condition as compared to the non-helping confederate condition is consistent with the findings of Bryan and Test (1967). As mentioned, the greater proportion of participants helping in the no-confederate condition as compared to the helping-confederate condition is not consistent with Bryan and Test’s findings.

However, dropping papers is far milder than the emergency situations used in traditional experiments and studies investigating the diffusion of responsibility and normative theories. In the helping-confederate condition, it may be that no further help by the participant was likely to be of any use – the task of picking up the papers could be easily completed by two people: the confederate and the experimenter. In such circumstances, the participant would not feel compelled to help despite seeing a model of helping behaviour.

The internal validity of the experiment might be affected by subject variables such as gender and age, which were not controlled or recorded. Also the methodology may not have been consistent, as different experimenters conducted the trials. One example of this is the skill level of the experimenter in dropping the papers. Participants’ reactions may differ according to the manner in which the papers were dropped; for example, a spectacular dropping of papers spanning a wide area may affect participants differently than if the accident was more localised. Also the level of distress or embarrassment portrayed by the experimenter may not have been constant.

The use of opportunistic sampling is likely to have compromised external validity because the participants chosen may not have been representative of the university population. For example, since it was preferable that no other people other than those privy to the experiment and the participant were present, people walking on their own were selected as participants. This may have resulted in an overrepresentation of certain personality traits in the participants.

It would be interesting to see whether the differences between the three conditions were similar at the Australian and Malaysian campuses. The Malaysian (Sunway) campus may have a more collectivist culture than Australian campuses. Poor consistency between campuses would indicate low external validity.

This experiment could also be improved by changing the situation so that in the confederate conditions, extra help by the participant would be of significant benefit.

The implication of the results is that helping behaviour may not be governed by our moral characters, but by simple social mechanisms. With this in mind, perhaps we can better recognise why people fail to comply with moral expectations. Experiments like these may promote greater tolerance of the behaviour of others, if we are able to recognise that a large part of a person’s response is based on their situation, rather than on their character traits (Harman, 1999).

The aim of this experiment was to see which of the diffusion of responsibility theory and the normative theory better accounts for helping behaviour. It was found that the diffusion theory better accounted for helping behaviour, in that the effect associated with the diffusion of responsibility was larger than the effect associated with seeing helping behaviour modelled by others.

References

Asch, S. E. (1999). Opinions and social pressure. In E. Aronson (Ed.), Readings about the social animal (8th ed., pp. 17-26). San Francisco, CA: Freeman.

Brehm, S. S., & Kassin, S. M. (1996). Helping others. In Social psychology (3rd ed., pp. 270-277). Boston: Houghton Mifflin.

Bryan, J. H., & Test, M. A. (1967). Models and helping: Naturalistic studies in aiding behavior. Journal of Personality and Social Psychology, 6, 400-407.

Cialdini, R. B., Reno, R. R., & Kallgren, C. A. (1990). A focus theory of normative conduct: Recycling the concept of norms to reduce littering in public places. Journal of Personality and Social Psychology, 58, 1015-1026.

Harman, G. 1999. Moral philosophy meets social psychology: Virtue ethics and the fundamental attribution error. Proceedings of the Aristotelian society, 99, 315–331.

Latane, B., & Darley, J. M. (1970). The unresponsive bystander: why doesn't he help?. New York: Appleton-Century-Crofts.

Latane, B., & Nida, S. (1981). Ten years of research on group size and helping. Psychological Bulletin, 89, 308-324.