When not intervening is more dangerous: the bystander effect

Reading time: 4 minutes

During the peak hours of a trafficked street, a car ignores a red light and is hit by another vehicle coming from its right. The accident, although not dramatic, may have caused injuries. And yet, despite the presence of several witnesses, nobody goes to the rescue of the two drivers, and everybody stays still and watches the scene, waiting for someone else to make the first move.

This phenomenon is called the bystander effect, that is the tendency not to act in a critical situation when other passive spectators are present. The effect has been described and studied in the Sixties, after a tragic report in New York. In 1964, Kitty Genovese was raped then killed, while some of her neighbours watched the scene, intervening only when it was too late. The case caused public uproar and led many scientists to research why this phenomenon occurs.

In 2011, a team of researchers coming from several European universities wrote a metanalysis, trying to find an order in all the literature regarding this effect, analysing almost fifty years of studies and experiments. The three questions that they wanted to find a definitive answer for were:

  1. Does the effect decrease in situations of high emergency?
  2. Are there specific situations in which the bystander can intervene since they are seen as a welcome help?
  3. Are there other theoretical and practical factors that influence the effect?

The researchers started their analysis by looking at the oldest literature.

The psychological decisional process

In 1970, Latané and Darley proposed a decisional model in five steps, according to which a person decides to act in a dangerous situation:

  • A critical situation is noticed
  • The situation is defined as an emergency
  • A sense of personal responsibility is developed
  • The competence necessary for the intervention is evaluated
  • A conscious decision is taken

However, this model can be influenced by three other psychological processes. The first one is the diffusion of responsibility, that is the tendency to evaluate your own personal responsibility by dividing the total responsibility by the number of present bystanders. The second process is the evaluation apprehension: the fear of being judged by others when you act in public, therefore the fear of making mistakes. The last one is pluralistic ignorance, the tendency to lean on overt reactions of others for the evaluation of an ambiguous situation. In this case, the maximum effect occurs when everybody stays still and watches the scene, because everyone thinks that nobody else perceives the situation as an emergency.

In una strada trafficata, durante l’ora di punta, un’automobile ignora un semaforo rosso e viene colpita da un altro mezzo proveniente dalla sua destra. L’incidente, seppur non drammatico, potrebbe aver causato dei feriti. Eppure, nonostante ci siano molti testimoni, nessuno va in soccorso dei due guidatori e ciascuno resta a guardare la scena, aspettando che qualcun altro faccia la prima mossa.

Relevant factors

The classic experiment to test the entity of this effect is making the participant witness a fake emergency, staged by the researchers, evaluating the difference in intervention between the case in which there is a single bystander and the case of a group of spectators. Since the first experiments, it was clear that this effect occurs both in case of emergencies, such as an injury or an asthma attack, and in less serious situations, such as a fake technical problem during the experiment. This phenomenon occurs even for trivial disturbances in our daily routine. How many times does it happen that the doorbell rings and no one gets up to answer it?

Subsequent studies confirmed that the effect increases when the number of bystanders gets higher or when the situation is ambiguous, therefore it is difficult to evaluate its severity. The bystander effect tends to be stronger in cities, rather than rural areas. On the contrary, an irrelevant factor is sex, both the bystander’s and the victim’s. The same goes for age, even if some studies seem to attribute to bystanders in their later years a weaker effect, maybe due to a higher competence for an intervention. A curious factor is the level of familiarity with the other bystanders: a group of friends tends to help a struggling person more, when confronted with a group of complete strangers.

It may seem strange, and yet a higher possibility of communication between the bystanders increases the entity of the effect, therefore non-intervention. The researchers suggest that this may happen because the inability to communicate with others increases the feeling of personal responsibility.

The effect in cases of emergency

More recent studies have showed how the bystander effect occurs more rarely in situations of severe danger or when the bystanders are very competent.

Analysing the literature, the researchers propose three possible explanations for the reduction of the bystander effect:

  • Arousal and cost of non-intervention
    Dangerous situations are recognized faster and the cost of non-intervention increases, if the victim is not helped. Therefore, the bystander experiences a state of arousal that can lead him to act, regardless of the number of other people that are present. The perceived danger is a direct source of physical arousal, which helps in identifying a real emergency and leads the bystander to intervene.
  • Bystanders as physical support when facing fear
    If the emergency is represented by a person who acts as a threat, the bystander can fear negative physical consequences. Dangerous situations can therefore increase the feeling of fear of being harmed if there is an intervention, but, if other bystanders are present, they can be seen as a source of physical support, which diminishes the traditional bystander effect.
  • Rational choice
    As an alternative to the explanation based on arousal, this hypothesis claims that the decision of the bystander is based on the perceived cost of intervention, on the benefit achieved by helping the victim and the perceived probability of other witnesses intervening. Generally, a high cost of intervention reduces the tendency to intervene. Having said this, there cand be the perception that some emergencies are so dangerous that they require the intervention from several bystanders, which can provide certain and effective help. Some situations can be solved only if more spectators coordinate their intervention. In addition, dangerous cases can increase the rate of individual intervention, because there is the belief that others will help too, after having evaluated the severity of the situation.

The researchers claim that recent studies seem to confirm all of these three possible explanations, even if it can seem counterintuitive. But, in their opinion, the greatest discovery of this metanalysis is that it seems that the bystander effect has decreased with the passage of time. A possible explanation is that the notoriety of this phenomenon succeeded in changing the public perception, leading more people to intervene in situations of emergency. Tragic stories such as the death of Kitty Genovese have been in fact tangible examples of how high the cost of non-intervening is in dangerous cases; everyone’s duty is to change their own decisional process before it happens again.

Carlo Sordini


  1. Fischer P, Krueger JI, Greitemeyer T, Vogrincic C, Kastenmüller A, Frey D, Heene M, Wicher M, Kainbacher M. The bystander-effect: a meta-analytic review on bystander intervention in dangerous and non-dangerous emergencies. Psychol Bull. 2011 Jul;137(4):517-37. doi: 10.1037/a0023304. PMID: 21534650.
  2. Latané, B., & Darley, J. M. (1970). The unresponsive bystander: Why doesn’t he help? New York, NY: Appleton-Century-Croft.
  3. Clark, R. D., & Word, L. E. (1972). Why don’t bystanders help? Because of ambiguity? Journal of Personality and Social Psychology, 24, 392–400. doi:10.1037/h0033717
  4. Fischer, P., Greitemeyer, T., Pollozek, F., & Frey, D. (2006). The unresponsive bystander: Are bystanders more responsive in dangerous emergencies? European Journal of Social Psychology, 36, 267–278. doi:10.1002/ejsp.297