What can social psychology tell us about the behaviour of 'bystanders'?

What can social psychology tell us about the behaviour of ‘bystanders’?

Since events such as the Kitty Genovase (1964) murder whereby 38 witnesses failed to intervene and others, social psychologists have attempted to explain human bystander intervention/non-intervention through relevant theories and research. Social psychology is according to Ibáňez (1997) ‘the activity of a working community who are paid wages by academic/scientific institutions, government departments and private corporations to produce valid knowledge on a certain range of social phenomena’. It is a contemporary discipline based around positivistic and empirical foundations. Key assumptions of the social approach are that other individuals influence us and we may imitate other bystanders (social learning theory) or conform to bystander group norms. ‘Bystander behaviour (also known as the bystander effect) is referred to as other bystander’s presence reduces the likelihood that victims will be helped in most situations due to the diffusion of responsibility, pluralistic ignorance, etc. It is closely linked to altruism, which is placing other’s needs first with potential cost to the altruist. Batson (1987) claims humans are pro-social animals which is shown in the willingness to give to charities, comfort friends, etc, which costs us time/money. This may seem beneficial but psychologists believe motives are involved. The purpose here is to present important psychological theories and research evidence that illustrate the reasons as to why humans do/don’t get involved in emergency/non emergency situations, Latanĕ and Darley showing an inverse relationship between the number of bystanders present and the likelihood of help. However, limitations of this will be subsequently highlighted to show caution is required; we cannot assume all North American social psychological evidence is true for everyone in other countries. The main approaches which explain bystander intervention will be discussed and other general factors; the biological, social learning, and more recently cognitive and bystander-calculus perspectives.

The biological/evolutionary approach believes helping behaviour is innate from birth which promotes species survival. Bystanders may not think of this when deciding whether to intervene but it is a plausible explanation especially when helping kin members. Burnstein et al (1994) supports this where participants were more willing to help closer than distant kin. But social psychologists are heavily critical and don’t accept sociobiological explanations for bystander intervention. There is little empirical human evidence and sociobiologists reject social learning theories. (Hogg and Vaughan 2002). Cognition is a major influence in deciding whether to intervene; behaviour is influenced by norms, empathy and social pressure, factors that Latanĕ and Darley promote.

Biological’s opposition – social learning theory, assumes bystander behaviour is acquired through environmental learning during socialisation. Classical conditioning and imitation are proposed as promoting prosocial and rejecting antisocial behaviours through childhood, through cooperation play schemes. This period is instrumental in children learning important processes. Rushton and Teachman (1978) showed children imitated another’s generosity of giving tokens to others. This study illustrates reinforcement and modelling of suitable models, which Bandura supported.

Situational approaches are most important with Latane and Darley’s ‘cognitive model’ contribution and Piliavin et al’s bystander-calculus model. Following the Kitty case Latanĕ and Darley (1970) aimed to explain why bystanders do/do not intervene and the bystander effect, (probability that the more bystanders present the less likely a victim will receive help). There are five stages that must be passed through for bystander intervention to occur including - noticing there’s a need, if it’s our responsibility and if we have adequate skills to help. A ’no’ decision at any point would prevent intervention. Although, if we progress through them ...

This is a preview of the whole essay

Latanĕ and Darley conducted many empirical studies one being students discussing problems over an intercom where they heard another have a fit. Different conditions led participants to believe there were no, one or four others besides them and the victim. They found participants took longer to react (166 seconds) and a lower percentage (62%) informed the experimenter when others were present, as opposed to 85% and 52 seconds when they believed they were alone. Also, those who failed to intervene were not uncaring – on the contrary; they were trembling and sweating. However, this may be useful for studying bystander behaviour from auditory cues, it excludes visual cues, as the intercom was the only communication between experimental actors. Usage of highly artificial experimental situations must always be questioned when trying to generalise findings to real life. However, their influential research found three social processes underlying bystander’s intervention failure.

The ‘diffusion of responsibility’ explanation for non-altruistic behaviour believes humans are less likely to intervene when others are present due to the belief that ‘someone else will help’, thus responsibility is reduced amongst the group, which may result in nobody helping if everyone perceives this. Just one other person needs to be present and they don’t need to be seen for this diffusion to occur. A bystander has a lower sense of personal responsibility. This may depend on personality of the bystander; a bystander with assertive leadership traits may takeover the situation and organise others. However, in emergency ‘fight or flight’ situations, it is hard to distinguish just who will help and who won’t during physiological arousal. Also, the social responsibility norm (we should help if needed) in societies is stronger when a bystander is alone.

Finally, evaluation apprehension is important also. Anxiety felt when being assessed by others alters behaviour because we want positive evaluation. Humans aim for social approval and so will not act in a manner that makes them stand out from the group or appear incompetent; this is why people conform to other bystander’s actions and social norms (normative influence).

Since their important evidence, recent proposals by other psychologists have been made to explain bystander behaviour. ‘Ambiguous emergency situations’ are where we observe other’s actions for clarification and act accordingly; we fear social humiliation if we perceive incorrectly. This is similar to informational social influence (ISI) confirmation/intelligence; we conform to other’s actions even if they may be false norms. Eiser (1980) believes there’s two ambiguity kinds in emergency situations – uncertainty about the situation, whereby removing this doesn’t always result in immediate intervention and confusion from another bystander not acting when it’s clear they’ve noticed and are able to intervene. This may result in the perception that it isn’t important/serious enough for intervention. ‘Pluralistic ignorance’ (Miller and McFarland 1987) is where everybody is ignorant of others perceptions and beliefs which can have grave consequences in an emergency situation. If someone rushes to aid, we may also but won’t if we are still unclear.

There is the danger that social explanations alone cannot fully understand bystander behaviour – especially in emergency situations where physiological elements (arousal and adrenalin) are involved. More successful higher ecologically valid experiments have been carried out in the field. Piliavin et al (1969) illustrates this in a New York subway where the victim collapsed either as drunk or an invalid. Males helped more than females and when perceived as an invalid received more help than the drunk. This shows how the victims were stereotyped as an invalid have higher social status than a drunk. Females probably assumed a more passive role, living up to their stereotype. This is a combination of both cognitive and physiological processes as opposed to Latanĕ and Darley’s cognitive theory. An arousal/cost-reward model was proposed where bystanders calculate rewards and costs of helping/not helping e.g. physical harm, guilt, praise, etc and are physiologically aroused in distressed situations. However, it implies that sufficient arousal is required for tasks to be undertaken sufficiently, experienced emergency helpers show this isn’t required (e.g. paramedics). Piliavin et al showed helping can reduce arousal, but help won’t be offered if costs are perceived too great. The cost of not helping was too high here but in Kitty’s situation bystanders couldn’t see each other so costs of helping were perceived high. This compares to the nature versus nurture debate; social learning is equivalised to nurture with physiological/biology being nature; an interaction of both require consideration here. A bystander forming a cost-benefit analysis is reminiscent of ‘equity theory’ where they compare this evaluation with what they feel they deserve, as a result of social interaction. This is alike Weber’s (1958) ‘protestant ethic’ notion where Lerner (1976) claims we develop trust in the world that if we provide additional investments we are entitled to additional benefits. In an emergency, Eiser (1980) claims if we view others or ourselves suffering inequity we try to return the situation back to equity. In the case of inequitable situations for others e.g. an accident, we feel morally obliged to help, probably due to socialisation. In the case of Kitty the 38 non-intervening witnesses may have perceived intervention would result in an inequitable balance for themselves e.g. the murderer attacking them.

However, this is quite a selfish conception of human decision-making and individual differences should be accounted for. To assume that all individuals behave this way is highly deterministic it ignores the free will of idiosyncratic norms and values. It also cannot account for indiscriminate altruism which requires no careful judgement (Eiser 1980). A problem of equity research is resource allocation. One bystander may crave the independent variable (e.g. media recognition) but another may resent it; it is difficult to keep resource gain constant for all participants. Eiser (1980) believes equity theory can lead us to understand the motives of bystanders; this can be controlled by cognition. The perception of equity/inequity does though seem to implicate social interaction and cognition. This is quite a plausible conclusion of equity theory.

Living in cities encourages people to withdraw from strangers - even from family. This failure to interact daily with others is one reason why people are less likely to intervene both to strangers as other bystanders and victims (Milgram 1970). Also, if people rely on media representations and perceive society to be highly dangerous than it actually is, they are also less likely to help, as they are scared of the stranger, the unknown bystander, the victim or the victim’s attacker. A further factor is the bystander’s perception of those involved. Shotland and Straw (1976) conducted two conditions in an artificial squabble and found when bystanders perceived it to be between strangers, 65% intervened while only 19% did as a married couple perception. This could be a factor in Kitty’s non-intervention – that people dislike involvement in a domestic situation. This seems plausible as society generally aims to keep domestic issues private and so intervention from the public sphere is unnecessary. James Bulger’s murder shows this as researchers found witnesses didn’t intervene as they thought the murderers were his brothers. Bickman (1971) showed the ability to offer help is important, whereby distance from the victim and other bystanders matters. 74% reacted to the victim when closer than another bystander, but this fell to 40% when further away. Bystander’s skills are also important where the assumption is that if a bystander has first aid skills in emergency situations they are more likely to help. Although not psychological evidence, this can be seen on films where doctors, nurses, etc rush to distressed victim’s aid. The bystander effect can also be explained in terms of prejudice and discrimination. Gaertner and Dovidio (1977) showed bystanders (female white participants) were more likely to help a white victim (confederate) than in the black victim condition and this occurred when there were two other helpers rather than alone. This shows race discrimination as a strong bystander effect was shown against the black victim but much weaker when white – in the presence of white others.

In conclusion, there is wide research conducted in the bystander behaviour field, distinguishing between old and contemporary research. Highly controlled experimental and quantitative analysis have given way to qualitative methods such as discourse analysis, interviewing, etc, providing better understanding of bystander’s intervention/non-intervention. Social psychology was often attacked on methodological grounds due to the difficulty of reproducing social phenomena in artificial laboratory environments. Highly controlled experiments are useful, but are more valid if used in conjunction with other research methods; they suffer from low ecological validity due to the social phenomena investigated. Many participants were students, which questions whether samples are representative both to other international students and other ages. Eurocentric bias is evident as most published research is American; a full understanding cannot be gained from this alone as self-interest is the dominant ideology (highly individualistic assumptions are made); this differs in collectivist cultures. To enable a fully universalistic understanding of bystander behaviour, further cross-cultural research with native researchers is required, although some has been published there’s not an equal balance. Britain and America are now multicultural so research within cultures also needs addressing. Ethical considerations are paramount; the use of deception and no informed consent were evident with most experiments. However, due to the phenomena under investigation, this is difficult in practice.

Rose (1985) claimed psychology was used in the 1960’s-70’s for state regulatory control. Its promise was to progress and alleviate problems in the social sphere e.g. hostility, etc. This seems plausible, as Kitty’s incident was 1964; most of the influential research was produced at this time due to public outcry at the non-intervention. Her screams of help were ignored and the implication here led psychologists to believe screaming ‘fire’, etc, should be employed. Although, it should be noted that this case was an extreme example of bystander behaviour and everyday situations aren’t like that. The higher educated/higher class maybe more prone to helping - apparently not in Kitty’s case where the attack occurred in a respectable residential area. Maybe further research could investigate. Bystander research suggests people are subjective as opposed to objective in their decision-making in situations. Lataně & Darley’s evidence is important as it shows a group can undermine individual rational psychological functioning but Piliavin et al (1981) refutes this, as groups are just as helpful. So maybe the laboratory context is more important. One use of the research is social reform to understand and promote helping behaviour on a daily basis – to reshape how individuals think with social interaction. It could be argued that the two main costs are time and effort and the two main benefits are recognition and personal gratification. But the idea that some bystanders do intervene simply because their actions benefit others (altruism), shouldn’t be disregarded. Overall, bystander behaviour is context dependent and the nature of individuals involved is also important. It depends if the action is convincing enough to elicit a bystander response and in reality humans are more helpful to each other. Those who’ve heard of bystander studies are more likely to help.

(2231 words)