The news all too often reports of crimes being witnessed by bystanders who do not intervene on behalf of the victim, providing neither direct help nor a call to emergency. This phenomenon, known as the Social Inhibition of Helping, has been well-documented by social psychologists since the 1970’s.
According to the social psychologists pioneering this research (Latane and Darley, 1970; 1976), the bystander to an emergency must engage in a five-step decision-making process if help is to be provided. Bystanders must (1) notice the event, (2) interpret it as an emergency, (3) decide on their degree of personal responsibility, (4) decide what method of intervention is most appropriate, and (5) implement the intervention.
Helping is believed to be inhibited by three primary psychological processes engaged by the presence of others: social influence, audience inhibition, and the diffusion of responsibility. These processes operate either alone or together so as to either slow decision-making or prevent the bystander from reaching the fifth stage of the decision-making process: implementing an intervention.
For instance, because emergencies are unexpected and often novel situations, they often possess a certain amount of ambiguity. As a result, bystanders may use the reactions of others as cues to define the situation. Greater ambiguity would yield greater reliance on others and consequently, greater opportunity for social influence. This would serve to suppress helping if others' reactions (or lack thereof) do not permit a clear interpretation of the situation as an emergency (i.e., negative social influence).
Even when the situation is interpreted as an emergency, however, the degree of personal responsibility felt by the bystander in a group may be lessened as each person may "diffuse" personal responsibility for helping onto others. Even if diffusion of responsibility does not occur, the individual may still be inhibited by the presence of others due to a fear of "standing out" or looking foolish while attempting to help. The social inhibition of helping is believed to increase in strength and likelihood of occurrence the larger the group of bystanders.
This view or model of helping implies that increasing the number of bystanders will slow or otherwise reduce the probability that the individual will intervene. In addition, the model implies that certain characteristics of bystanders (e.g., perceived personal and others’ competence) as well as the relationship and communication among bystanders will determine the extent of inhibition and whether or not helping will take place.
In other words, (a) the presence of "incompetent" or "incapacitated" others may focus responsibility on the self, (b) friends should be less likely to both misinterpret each other's reactions and diffuse responsibility than strangers, and (c) full communication among bystanders, so that each can see and be seen by others, increases the likelihood that audience inhibition, social influence, and the diffusion of responsibility will operate together to yield the lowest rate and probability of helping.
Interestingly, the laboratory and field research generated to assess this model demonstrates, with few exceptions, that the presence of others and the characteristics of both bystanders and the situation determine helping behaviour. Latane and Darley pioneered almost 4 decades of research able to show how events that often boggle the mind, such as witnessing but not reporting a crime, actually stem from psychological processes that become engaged when people are in the presence of others.