Key Takeaways
- Bystander Effect: People are less likely to help someone in distress when other people are present. The more bystanders there are, the lower each person’s chance of helping.
- Diffusion of Responsibility: Bystanders subjectively split the responsibility to help by the number of people present, so each person feels less personally accountable.
- Kitty Genovese: The most cited real-life example is Kitty Genovese, murdered in Queens in 1964. Later historical analysis found the “38 witnesses who did nothing” story was exaggerated.
- Five-Step Model: Latané and Darley (1970) proposed bystanders must notice, interpret, take responsibility, decide how to help, and act — hesitation at any step blocks helping.
- Three Processes: Diffusion of responsibility, evaluation apprehension, and pluralistic ignorance each help explain why people fail to help when others are present.
- Modern Evidence: A 2011 meta-analysis of over 7,700 participants confirmed the effect is real but modest, and CCTV footage shows bystanders intervene in the large majority of genuine public conflicts.
What is the bystander effect?
The bystander effect is the tendency for people to be less likely to help someone in distress when other people are present (Darley & Latané, 1968).
People are more likely to help when alone than in a group. Interest in this phenomenon grew sharply after the murder of Catherine “Kitty” Genovese in 1964, and a series of experiments in the 1960s and 1970s went on to explain why.
Kitty Genovese
The Attack
On the morning of March 13, 1964, Kitty Genovese returned to her apartment complex at 3 am, after finishing her shift at a local bar.
She parked her car in a nearby lot. She then walked towards the entrance at the back of the building.
A man was waiting at the far end of the lot. She changed direction and headed for a different street. He caught up with her and attacked.
Neighbor Robert Mozer shouted from his window, “Let that girl alone!” The attacker, Winston Moseley, fled to his car.
Genovese, badly wounded, staggered to the back of her building and collapsed in a small entryway, out of view of the street. Ten minutes later, Moseley came back and found her.
He attacked her a second and final time, stabbing her repeatedly.
The first call to the police came in at approximately 3:50 am. Officers arrived within two minutes, but Genovese died on the way to the hospital. Moseley was arrested six days later and confessed.
The New York Times Report
Two weeks later, a front-page New York Times article changed how the case would be remembered.
Reporter Martin Gansberg wrote it under the headline “37 Who Saw Murder Didn’t Call the Police.” It claimed 38 witnesses watched the attack and did nothing.
When asked later why they had not intervened or called the police sooner, the neighbors gave several answers. Some said, “I didn’t want to get involved”; others said, “Frankly, we were afraid,” or “I was tired. I went back to bed.” (Gansberg, 1964).
The paper’s metropolitan editor, A. M. Rosenthal, had reworked the story after a lunch with the city’s police commissioner. He later expanded it into a book, Thirty-Eight Witnesses (1964), which fixed the “38 witnesses” version in the public imagination.
National coverage followed, with commentators citing the case as evidence of the country’s “moral decay.”
Darley and Latané set out to find an alternative explanation.
Decision Model of Helping
Latané & Darley (1970) formulated a five-stage model to explain why bystanders in emergencies sometimes do and sometimes do not offer help.
At each stage in the model, the answer ‘No’ results in no help being given, while the answer ‘yes’ leads the individual closer to offering help.
However, they argued that helping responses may be inhibited at any stage of the process. For example, the bystander may not notice the situation or the situation may be ambiguous and not readily interpretable as an emergency.
The Five Steps of the Decision Model
- Notice: The bystander must notice that something is wrong.
- Interpret: The bystander must define the situation as an emergency.
- Take Responsibility: The bystander must decide they are personally responsible for helping.
- Decide How to Help: The bystander must work out the best way to assist.
- Act: The bystander must follow through and provide the help decided on.
Figure 1. Decision Model of Helping by Latané and Darley (1970).
Why does the bystander effect occur?
Latané and Darley (1970) identified three different psychological processes that might interfere with the completion of this sequence.
Diffusion of Responsibility
The first process is a diffusion of responsibility, which refers to the tendency to subjectively divide the personal responsibility to help by the number of bystanders.
Diffusion of responsibility occurs when a duty or task is shared between a group of people instead of only one person.
Whenever there is an emergency situation in which more than one person is present, there is a diffusion of responsibility. There are three ideas that categorize this phenomenon:
- The moral obligation to help does not fall only on one person but the whole group that is witnessing the emergency.
- The blame for not helping can be shared instead of resting on only one person.
- The belief that another bystander in the group will offer help.
- Aim: Darley and Latané (1968) isolated diffusion of responsibility from social influence, testing whether the number of unseen, unheard bystanders affected how quickly someone helped.
- Method: 72 students believed they were discussing university life over an intercom. Each heard a recording of a “victim” having a seizure, believing the group held just the victim, one extra bystander, or four extra bystanders.
- Results: In two-person groups, 85% reported the seizure, falling to 62% in three-person groups and 31% in six-person groups. Response times rose from 52 to 166 seconds as group size grew, and personality traits did not predict who helped.
- Conclusion: Not pluralistic ignorance. The drop in helping came from diffusion of responsibility: bystanders could not see each other, so each felt less accountable in a larger group.
Participants who did not help were not indifferent. They showed visible nervousness, suggesting they were still struggling to decide what to do rather than ignoring the emergency.
Evaluation Apprehension
The second process is evaluation apprehension, which refers to the fear of being judged by others when acting publicly.
People may also experience evaluation apprehension and fear of losing face in front of other bystanders.
Individuals may also fear being superseded by a more capable helper, offering unwanted assistance, or facing legal consequences for offering inferior or unsafe help.
It is harder to isolate than the other two processes. It works mainly as an amplifier, strengthening diffusion of responsibility when a confident, competent-looking bystander is present. This builds on Bickman’s early-1970s research linking responsibility to a bystander’s perceived competence.
Pluralistic Ignorance
The third process is pluralistic ignorance, which results from the tendency to rely on the overt reactions of others when defining an ambiguous situation.
Pluralistic ignorance occurs when a person privately disagrees with an idea but believes everyone else accepts it. As a result, they go along with that idea too, even though no one actually believes it.
Deborah A. Prentice cites an example of this. Despite being in a difficult class, students may not raise their hands in response to the lecturer asking for questions.
This is often due to the belief that everyone else understands the material, so for fear of looking inadequate, no one asks clarifying questions.
It is this type of thinking that explains the effect of pluralistic ignorance on the bystander effect. The overarching idea is uncertainty and perception. What separates pluralistic ignorance is the ambiguousness that can define a situation.
If the situation is clear, pluralistic ignorance does not apply. In the classroom example, if someone states they do not understand, everyone now knows someone else agrees with them.
It is the ambiguity and uncertainty which leads to incorrect perceptions that categorize pluralistic ignorance.
Rendsvig (2014) proposes an eleven-step process to explain this phenomenon.
These steps follow the perspective of a bystander (who will be called Bystander A) amidst a group of other bystanders in an emergency situation.
- Bystander A is present in a specific place. Nothing has happened.
- A situation occurs that is ambiguous in nature (it is not certain what has occurred or what the ramifications of the event are), and Bystander A notices it.
- Bystander A believes that this is an emergency situation but is unaware of how the rest of the bystanders perceive the situation.
- A course of action is taken. This could be a few things like charging into the situation or calling the police, but in pluralistic ignorance, Bystander A chooses to understand more about the situation by looking around and taking in the reactions of others.
- As observation takes place, Bystander A is not aware that the other bystanders may be doing the same thing. Thus, when surveying others’ reactions, Bystander A “misperceives” the other bystanders” observation of the situation as purposeful inaction.
- As Bystander A notes the reaction of the others, Bystander A puts the reaction of the other bystanders in context.
- Bystander A then believes that the inaction of others is due to their belief that an emergency situation is not occurring.
- Thus, Bystander A believes that there is an accident but also believes that others do not perceive the situation as an emergency. Bystander A then changes their initial belief.
- Bystander A now believes that there is no emergency.
- Bystander A has another opportunity to help.
- Bystander A chooses not to help because of the belief that there is no emergency.
Pluralistic ignorance operates under the assumption that all the other bystanders are also going through these eleven steps.
Thus, they all choose not to help due to the misperception of others’ reactions to the same situation.
Other Explanations
Other theories add to these three explanations. One is confusion of responsibility.
Confusion of responsibility occurs when a bystander fears that helping could make others think they are the attacker. This fear alone can stop someone acting.
Another idea is priming. Being shown cues can influence a later, unrelated choice. For example, someone shown home-decor words is more likely to answer “chair” than “pasta” when asked for a five-letter word.
Garcia et al. (2002) found that simply thinking about being in a group lowers helping rates, even with no group actually present. Groups are linked to being “lost in a crowd” and having a “lowered sense of personal accountability” (Garcia et al., 2002, p. 845).
Priming matters too. It shows the bystander effect is not only about who is physically present, but also about a person’s own group-related thoughts.
Bystander Experiments
Two further experiments show how ambiguity and personal relationships shape whether bystanders act.
The Smoke Study: Pluralistic Ignorance in Action
- Aim: Latané and Darley (1968) tested whether other people’s presence stops someone interpreting an ambiguous event as an emergency, later termed pluralistic ignorance.
- Method: Male students completed a questionnaire alone, with two passive confederates, or with two other real participants, as smoke drifted in through a wall vent.
- Results: 75% of participants alone reported the smoke, half within two minutes. With two passive confederates who ignored it, only 10% reported it; among three-person groups, only 38% produced a reporter.
- Conclusion: Non-reporters were not indifferent. They had reinterpreted the smoke as steam or vapor, having looked to others’ calm reactions for cues about the ambiguous situation.
This is a clear example of pluralistic ignorance, affecting step 2 of the Latané and Darley decision model above.
Lady in Distress
A third study asked whether the effect held even when the bystander faced no personal risk.
- Aim: To test whether group inhibition still occurs when the bystander is not personally at risk. A second question was whether inhibition depends on the other bystander being a stranger rather than a friend.
- Method: 120 male Columbia undergraduates were tested alone, with a passive confederate, with two strangers, or with two friends. A hidden recording then played a market researcher next door falling and calling for help.
- Results: 70% of participants helped when alone, but only 7% did with a passive confederate present. At least one person helped in just 40% of stranger pairs, compared with 70% of friend pairs.
- Conclusion: Friends inhibit each other far less than strangers. They can read each other’s reactions accurately and talk, so they are less likely to mistake passivity for “nothing is wrong.”
The friends-versus-strangers contrast is strong evidence that misread social cues, not diffused responsibility, drove the inaction here. The sample was small, all-male, and from a single campus, so the results may not generalise widely.
Ambiguity in Domestic Violence
Genuine ambiguity, not just other people’s presence, can affect whether someone helps. Context matters. Shotland and Straw (1976) tested this in a staged domestic dispute versus a stranger attack.
Male participants watched a staged fight between a man and a woman. She shouted either ‘I don’t even know you’ or ‘I don’t even know why I married you.’ The wording alone mattered.
Three times as many men intervened when the couple appeared to be strangers than when they appeared to be a couple. Bystanders read the apparent relationship between victim and attacker as a cue to how much the situation called for outside help.
People are less likely to intervene if they believe that the incident does not require their personal responsibility.
The finding supports the model’s third step. An ambiguous relationship makes personal responsibility harder to accept, since a bystander may assume a domestic dispute is “not their business.”
Critical Evaluation
While the bystander effect has become a cemented theory in social psychology, the original account of the murder of Catherine Genovese has been called into question. By casting doubt on the original case, the implications of the Darley and Latané research are also questioned.
Two independent reviews back the correction. Journalist Nicholas Lemann’s 2014 New Yorker essay and Catherine Pelonero’s 2014 book both retraced the original reporting and reached the same conclusion.
Manning et al. (2007) did this through their article “The Kitty Genovese murder and the social psychology of helping, The parable of the 38 witnesses”. By examining the court documents and legal proceedings from the case, the authors found three points that deviate from the traditional story told.
Three points stood out.
The “38” figure appears nowhere in the police record. It was a number generated in the New York Times newsroom, and its authority came from the headline rather than from any source.
Very few people saw any part of the attack, and effectively no one saw the killing itself. The fatal, second stabbing happened inside a small vestibule invisible from almost every apartment window identified as a “witness.”
The traditional story also overstates the neighbors’ inaction. At least two people did call the police, and one man shouted from his window and drove the attacker off.
The potential inaccurate reporting has not negated the bystander effect completely. But it has called into question the case’s evidential value and narrowed the questions the field went on to ask.
Limitations of the Decision-Helping Model
Schroeder et al. (1995) believe that the decision-helping model provides a valuable framework for understanding bystander intervention.
Although developed to explain emergencies, the model has since been applied elsewhere, from preventing drunk driving to deciding whether to donate a kidney to a relative.
The model has limits, though. It does not explain why a “no” decision happens at each stage, especially once someone has already interpreted the event as an emergency.
It also leaves out emotional factors like anxiety or fear, and focuses mainly on why people fail to help rather than why they do.
Piliavin and colleagues offered an alternative. Their cost–reward arousal model treats helping as a trade-off between the costs and rewards of acting.
Whether one helps depends on weighing the costs and rewards of helping. The costs include effort, time, loss of resources, risk of harm, and a negative emotional response.
Rewards matter too. They include fame, gratitude from the victim and relatives, and the self-satisfaction of having helped. Costs and rewards can differ between people, and even for the same person from one occasion to the next.
Accountability Cues
According to Bommel et al. (2012), the negative framing of the bystander effect undermines its potential positives. Crowds are not always bad news.
Their study, “Be aware to care: Public self-awareness leads to a reversal of the bystander effect,” shows crowds can actually increase help given to a victim under certain conditions.
One problem with bystanders in emergencies is that responsibility can be split, diffusing it across the group. But “accountability cues” can reverse this. These are specific markers, like a visible camera, that signal a bystander’s actions are being watched.
The researchers created an online forum for people seeking help with “severe emotional distress.” Participants responded to visitors’ messages and rated how visible they felt on the forum.
Visibility was the key manipulation.
With no accountability cues, people gave less help and rated themselves as less visible. With cues present, such as a webcam and the visitor’s name displayed, more people helped and rated their presence on the forum as higher.
The results matched these predictions. Targeting a person’s reputation through accountability cues can increase helping, showing the bystander effect can be reversed.
Neuroimaging Evidence
Brain imaging tells a similar story. Activity drops in the regions that support helping when other people are present.
These regions include the pre- and postcentral gyrus and the medial prefrontal cortex, both linked to preparing to act (Hortensius & de Gelder, 2018). At the same time, activity rises in the anterior insula, a region tied to self-focused threat response.
Fear typically wins first.
Physical signs of distress, like a racing heart or freezing up, often override the slower, sympathy-linked response that involves the temporoparietal junction. Whether someone ultimately helps depends on which system wins: more sympathy than distress leads to helping, more distress leads to freezing.
The pattern is consistent.
This suggests the decision to help is “reflexive” rather than “reflective” (Hortensius & de Gelder, 2018). The classic five-step model may describe what people report afterwards, not what their brains do in the moment.
The researchers argue for a more personalised view, one that weighs a person’s own empathy and disposition rather than a single explanation for everyone.
Contemporary Research
Two large studies published after 2010 have reshaped this picture. One confirms the bystander effect statistically. The other says it’s far rarer outside the lab.
The 2011 Meta-Analysis
- Aim: Fischer et al. (2011) ran the first modern meta-analysis of the bystander effect, testing whether it survives in dangerous as well as ambiguous emergencies.
- Method: The researchers pooled 105 effect sizes from over 7,700 participants across five decades. They tested moderators including danger, a perpetrator’s presence, the type of cost involved, and the bystanders’ sex and type.
- Results: The overall effect was small-to-moderate (g = -0.35). It shrank significantly in dangerous emergencies, when a perpetrator was physically present, and when the cost of helping was physical rather than emotional or financial.
- Conclusion: The effect is real but modest, far smaller than the dramatic 1960s lab contrasts. Fischer’s team linked this to the arousal:cost–reward model: danger raises arousal fast enough that extra bystanders can help rather than hinder.
Real-World Evidence from CCTV Footage
Philpot et al. (2020) analyzed CCTV footage of 219 real public conflicts in Amsterdam, Cape Town, and Lancaster. At least one bystander intervened in 90.9% of incidents, with an average of 3.76 people stepping in per event. Each additional bystander made help slightly more likely, not less.
This suggests the classic “unresponsive bystander” pattern is largely a feature of the ambiguous, low-stakes laboratory paradigm. In real public emergencies, the norm is intervention, not inaction.
Applications and Legacy
The bystander effect is not just a laboratory finding. It has reshaped public-safety training and left a lasting mark on law and policy.
Public-Safety Training
The applied lesson is consistent across the classic and modern studies. The effect weakens once responsibility falls on a named individual, ambiguity is reduced, and a specific action is rehearsed in advance.
- Emergency response: CPR and first-aid instructors teach bystanders to point at one person and give a direct order, like “You, call 999 now.” This turns shared responsibility into one person’s job.
- Basic Life Support: Medical bodies build this into a “delegate one person” principle: one bystander fetches a defibrillator, another calls the emergency number, and a third clears space.
- Workplace and campus programmes: Bystander-intervention training for harassment and bullying, including the Green Dot and Bringing in the Bystander programmes, teaches the “3 Ds”: direct, distract, and delegate.
- Digital bystanders: Diffusion of responsibility also applies online. The more people who see a harmful post, the less likely any viewer is to report it, unless the platform shows accountability cues.
Legal and Cultural Legacy
The Genovese case shaped more than psychology. Two lasting changes in law followed.
First, it strengthened the political case for a single national emergency number. Callers in the United States previously had to know their local precinct’s number. AT&T announced 911 as the national emergency number in 1968. The first US 911 call was placed in Haleyville, Alabama, that February.
The UK’s 999 service, set up in 1937, expanded during the same period and was cited as a model.
Second, the case fuelled a lasting legal debate about the duty to rescue: whether a stranger is obliged to help another stranger in danger. There is traditionally no such duty in common law.
A minority of US states, including Vermont, and several European countries have “Bad Samaritan” statutes. These make it an offence to withhold reasonable help from someone in grave danger.
As Stanley Milgram observed at a 1984 conference on the Genovese case, the murder “struck to the very heart of the nature of human fear.”
Both threads point to the same lesson. Reducing ambiguity and assigning responsibility, whether by law, training, or a shouted instruction, is what turns bystanders into helpers.
References
Darley, J. M., & Latané, B. (1968). Bystander intervention in emergencies: Diffusion of responsibility. Journal of Personality and Social Psychology, 8, 377–383.
Fischer, P., Krueger, J. I., Greitemeyer, T., Vogrincic, C., Kastenmüller, A., Frey, D., Heene, M., Wicher, M., & Kainbacher, M. (2011). The bystander-effect: A meta-analytic review on bystander intervention in dangerous and non-dangerous emergencies. Psychological Bulletin, 137 (4), 517–537.
Gansberg, M. (1964, March 27). 37 who saw murder didn’t call the police. The New York Times, p. 1.
Garcia, Stephen M, Weaver, Kim, Moskowitz, Gordon B, & Darley, John M. (2002). Crowded Minds. Journal of Personality and Social Psychology, 83 (4), 843-853.
Hortensius, Ruud, & De Gelder, Beatrice. (2018). From Empathy to Apathy: The Bystander Effect Revisited. Current Directions in Psychological Science, 27 (4), 249-256.
Latané, B., & Darley, J. M. (1968). Group inhibition of bystander intervention in emergencies. Journal of Personality and Social Psychology, 10, 215–221.
Latané, B., & Darley, J. M. (1970). The unresponsive bystander: Why doesn’t he help? New York, NY: Appleton-Century-Croft.
Latané, B., & Darley, J. M. (1976). Help in a crisis: Bystander response to an emergency . Morristown, NJ: General Learning Press.
Latané, B., & Nida, S. (1981). Ten years of research on group size and helping. Psychological Bulletin, 89, 308 –324.
Manning, R., Levine, M., & Collins, A. (2007). The Kitty Genovese murder and the social psychology of helping: The parable of the 38 witnesses. American Psychologist, 62, 555-562.
Philpot, R., Liebst, L. S., Levine, M., Bernasco, W., & Lindegaard, M. R. (2020). Would I be helped? Cross-national CCTV footage shows that intervention is the norm in public conflicts. American Psychologist, 75 (1), 66–75.
Prentice, D. (2007). Pluralistic ignorance. In R. F. Baumeister & K. D. Vohs (Eds.), Encyclopedia of social psychology (Vol. 1, pp. 674-674). Thousand Oaks, CA: SAGE Publications, Inc.
Rendsvig, R. K. (2014). Pluralistic ignorance in the bystander effect: Informational dynamics of unresponsive witnesses in situations calling for intervention. Synthese (Dordrecht), 191 (11), 2471-2498.
Shotland, R. L., & Straw, M. K. (1976). Bystander response to an assault: When a man attacks a woman. Journal of Personality and Social Psychology, 34 (5), 990.
Siegal, H. A. (1972). The Unresponsive Bystander: Why Doesn’t He Help? 1(3), 226-227.
Van Bommel, Marco, Van Prooijen, Jan-Willem, Elffers, Henk, & Van Lange, Paul A.M. (2012). Be aware to care: Public self-awareness leads to a reversal of the bystander effect. Journal of Experimental Social Psychology, 48 (4), 926-930.
