Let’s say you’re in a group chat and one person has become the joke. The comments were mild at first, but now they’re becoming cruel. You start to type a kind reply to break the momentum, but you pause. Maybe the person being targeted is taking it in good humor. Maybe someone who knows them better will say something. No one else seems bothered.
Of course, you can’t see the other people hovering over their own keyboards, wondering much the same thing.
While you may be positive you’d overcome the hesitation and do the right thing, people commonly don’t—and this truth offers insight into a question psychologists have studied for decades: What happens to our willingness to help when other people are present? It’s often referred to as the bystander effect. Under some circumstances, say researchers, each of us becomes less likely to act when we believe someone else could step in. It’s not easy to be a lone ranger.
But research also shows that people frequently do help, and that a group can sometimes make intervention more likely. Understanding both possibilities is perhaps more constructive than assuming crowds always bring out the worst in us and may help us actually do the thing we think we’re capable of: overcome the hesitation.
Even when people don’t act, the problem goes much deeper than a temporary freeze response in an emergency. It’s rooted in the silent social systems we build, where collective inaction slowly morphs into complicity, even among people who might, in other circumstances, be very compassionate and proactive. What are the psychological mechanisms that keep us silent and how do we interrupt them? If we’re to move toward a culture of active intervention, it helps to recognize that interrupting them is a shared communal responsibility.
Why Caring People Hesitate
In a study published in 2020, Richard Philpot and colleagues examined surveillance footage of 219 public conflicts in the United Kingdom, the Netherlands, and South Africa. Someone intervened in roughly nine out of ten incidents, often with several people helping. Having more bystanders present was associated with a greater chance that at least one would step in.
These were public conflicts captured on camera, so the findings can’t tell us how often people help in every kind of emergency. Helping also doesn’t guarantee a good outcome. Still, the study challenges the assumption that it’s natural to avoid intervening when multiple others are present.
“We argue that it is time for psychology to change the narrative away from an absence of help and toward a new understanding of what makes intervention successful or unsuccessful.”
Richard Philpot, et al., “Would I Be Helped? Cross-National CCTV Footage Shows That Intervention Is the Norm in Public Conflicts,” (2020)
It also begs two important questions. Am I less likely to act when other people are present? Is anyone else in this crowd likely to help? A group can reduce each person’s sense of responsibility while still increasing the number of potential helpers. These questions are important—certainly to the person who needs assistance. But they should matter to anyone who hopes they have the personal agency and moral clarity to resist the gravitational pull of group dynamics.
Serious investigation into bystander intervention began in the late 1960s with researchers John Darley and Bibb Latané. Through a series of foundational studies, they mapped out what’s now universally recognized as the bystander effect. This describes a paradoxical tendency of human behavior in which an individual observing an emergency alone will almost always report it immediately, while when multiple observers are present the likelihood of anyone helping drops exponentially. Their work helped focus attention on discovering which situations encourage or discourage action, rather than simply assuming the bystanders don’t care.
Darley and Latané outlined a four-step cognitive gauntlet an observer navigates before taking action: First they must notice the event, then they must interpret the event as an emergency or problem. They must feel personally responsible for dealing with it and also feel they possess the practical skills or resources to act safely.
That chain can break down at any single link. One obstacle a bystander encounters has been termed diffusion of responsibility: the feeling that someone else will take care of it. We may assume that someone closer to the scene has already called for help, or that someone with more authority knows what to do. In other words, when everyone’s responsible, nobody feels accountable.
Another is pluralistic ignorance. When a situation is unclear, we look around for clues to help us determine what we should do. If everyone else seems calm, we decide the situation must not be critical. Yet everyone else may be interpreting our own apparent calm in the same way. Concern can be widespread without being apparent.
We may also worry about how others will judge us (audience inhibition). What if we’ve misunderstood the situation? What if speaking up makes us look foolish? And in a workplace, team, or friendship group, there may be added fears about retaliation or losing our place in the group.
“From the perspective of the victim it does not matter whether help is withheld because of responsibility diffusion, because of anomie, or because bystanders doubt the emergency’s genuineness.”
Dianne M. Tice and Roy F. Baumeister, “Masculinity Inhibits Helping in Emergencies: Personality Does Predict the Bystander Effect,” (1985)
It’s important to note that explanations like these can help us understand hesitation, but they do not make harm acceptable. Nor can we read a witness’s intentions from a few seconds of video. A bystander may be frightened, unsure what is happening, seeking help out of view, or unable to intervene in the way we imagine they should. These variations are highlighted by a 2011 meta-analysis led by Peter Fischer, which found that the bystander effect was weaker in circumstances that were clearly real emergencies.
Cultivating a Culture of Intervention
How do we transform a society of silent onlookers into a culture of active protectors? We often think of peer pressure as something that influences people toward bad decisions. But the expectations of a group can also support helping. It all depends on a number of influences.
Research undertaken by Mark Levine and Simon Crowther has pointed to the importance of the relationship among bystanders. In one study, larger groups inhibited intervention when the bystanders were strangers but encouraged it when they were friends. Their wider findings also point to the influence of group identity and what seems normal to that group in a particular setting. They argue that bystanders are more likely to help when the victim feels like “one of us.” Further, while they do look to the crowd for cues on how to act, they only care about what the crowd thinks if they consider those people part of their own group. If they view the other bystanders as “outsiders,” those onlookers won’t have much influence over them.
Interestingly, another of Levine and Crowther’s studies found that the gender composition of the group also had an impact. Both male and female participants were more likely to help after being grouped with women than after being grouped with men. And the lowest observed helping rate was in men after being grouped with other men. The authors suggested that women together may have felt greater solidarity with the woman needing help. For men surrounded by women, they suggested that expectations about being protective or stepping forward may have encouraged helping. But it’s also possible that the men may have been influenced by social beliefs about women’s standards of compassion. So, did being with women encourage men to help or did being with men suppress helping? The study didn’t really answer that question. But it’s the difference between asking, “Who feels responsible for protecting this woman?” and “What kind of response feels expected or valued among the people around me?”
“Highly masculine persons may value calm stoicism (or its appearance) in the face of crisis more than others do. . . . This fear of losing one’s cool and appearing foolish in a situation that may or may not be an emergency may have inhibited highly masculine subjects from intervening.”
Dianne M. Tice and Roy F. Baumeister, ““Masculinity Inhibits Helping in Emergencies: Personality Does Predict the Bystander Effect,” (1985)
The latter is consistent with other study findings that leave us with some crucial questions to ask about our own groups: What do people believe they’ll be supported for doing? Are we building cultural ecosystems that are interconnected enough to talk about unacceptable behavior within our neighborhoods, local sports teams, or online group chats?
For instance, let’s say a football player reports a team initiation that causes humiliation or harm to one of his teammates. If the coach dismisses it as harmless tradition (“boys will be boys”), the rest of the team has learned something about behavior expectations, as well as the cost of objecting. On the other hand, a coach who takes the concern seriously will send the very different message that cruel behavior isn’t acceptable and that those who come to him with concerns won’t be ignored.
What culture is prevalent in our schools, workplaces, and families? Are people expected to take responsibility for their actions? Are they expected to protect the group’s reputation at any cost? Can they question someone with authority? Who can they go to if the first person they tell brushes them off?
When we notice a toxic hierarchy forming in a classroom, or spot exclusionary behavior in a group of children, we’re witnessing the primitive components of the bystander effect. Adults have the golden opportunity then to step in and say, “We don’t do that here. Let’s talk about what’s actually happening and what we should do instead.”
Of course, the bystander effect doesn’t explain the escalation from witnessing harm to concealing it, encouraging it, and committing it. If we want to build the kind of wider culture that encourages intervention, it’s critical to seize these opportunities to help children identify harmful dynamics that can lead to criminal behavior. Peers have the power to interrupt poor moral reasoning in one another, although it often goes untapped. By normalizing the act of speaking up, we teach young people that loyalty to a friend doesn’t include complicity in harm.
“When bystanders step in to support someone who is calling out mistreatment or harmful behaviour it sends a strong message to onlookers that this behaviour is unacceptable. . . . Conversely, staying silent, or changing the subject to avoid awkwardness . . . undermines the efforts of the confronter.”
University of Bath, “Bystander support is crucial for tackling anti-social behaviour,” (2024)
Give People Somewhere to Turn
For parents and other adults who work with young people, conversations about intervention strategies (tailored to their age group and the situations they’re likely to encounter) should begin well before a crisis. We can talk through the awkward situations they typically face: a friend being excluded, a cruel message being passed around, an older student influencing them to join in with bullying.
Rather than simply asking, “Would you stand up for someone?” we can ask more specifically, “Who could you go to for help?” or “What could you say if you wanted to leave a risky situation with your friend?” These are ways to rehearse possibilities, with room to acknowledge that speaking up can be hard.
Adults need to pay attention to a child’s concerns if they’re to respond constructively. A child told to find a trusted adult needs adults in their network who show them adults will listen. An employee encouraged to report misconduct needs a process that takes the report seriously. We can’t ask people to be brave enough to be proactive if we’re only going to leave them exposed to retaliation. In effect, we’re asking those with the least power to bear all the risk and responsibility.
What about our own responses? When someone brings us a concern, do we become defensive about a colleague or institution we admire (“He wouldn’t do that—maybe you misunderstood”)? If so, we leave that person feeling dismissed and shut down. Do we listen long enough to understand the concerns they’re sharing? Do we give real consideration with an open mind to what we’re told? It’s in these interactions that the sincerity of a community’s commitment to responsibility becomes clear.
Let’s return to that group chat we opened with. You may still feel uncertain about what you might say in such a scenario, but you could begin by checking on the person being targeted. Ask a friend or mentor to help you respond in a way that makes it clear you’re uncomfortable with the comments. You don’t always need the perfect words.
For someone on the receiving end, discovering another person has noticed and is willing to help can change what happens next. There may be many more who feel as you do and have been waiting for someone to show it’s okay to intervene. But encouraging a culture that supports intervention requires more than simply showing it’s “okay” to intervene. It requires building social systems where mistreatment isn’t tolerated, even when the target isn’t “one of us.” It requires normalizing the act of speaking up—and it requires interrupting the psychological mechanisms that keep us silent.
In the Critical Moment
“Do something” is a difficult instruction to follow when you’re alarmed and uncertain. A more useful question is: What can I safely do here?
Certainly, formal programs that explore this question can help make intervention more likely. One campus program that has expanded to other settings—Bringing in the Bystander—teaches participants to recognize situations that can escalate to sexual violence and consider ways to intervene safely. A review of the program has found it promising, but most evaluations of the program measured participants’ attitudes, confidence, and intentions to help rather than examining whether they actually stepped in. That, of course, would be the critical question. Helping people feel more confident about intervening is a positive outcome, but such studies don’t tell us whether someone actually would have intervened or whether violence would have been prevented.
It can be hard to find the right words when someone is being mistreated in front of us. Sometimes an ordinary question offers a place to begin. Asking the person for directions or drawing them into a conversation may interrupt harassment without directly confronting the person responsible. You can also enlist someone nearby. “Could you get the bus driver?” gives another bystander a clear way to help, especially when they may be just as uncertain as you are.
What you do next depends on the situation and what seems safe. If someone is in immediate danger, rule one is to call for emergency assistance. If you decide to address the person causing harm while you wait for emergency responders, keep it brief and avoid getting drawn into an argument. Recording may be useful but first make sure someone is helping the person affected. Nothing says “We recorded this for social media attention” like a crowd of onlookers holding their phones out camera-first with no sign of help on the way.
Once the immediate incident is over, there’s still room to help. “I saw what happened. Would you like me to stay with you?” acknowledges the targeted person’s experience and lets them decide what they need. If you recorded the incident, ask what they want done with the footage, and don’t share it without their permission. After an experience in which someone else disregarded their wishes, being given a choice matters.
Online, we can take similar actions. You might report abusive comments without adding to their circulation. The aim is to find a response that’s helpful within your abilities, even while you may feel nervous or unsure. A private message can offer a similar opening. They may welcome company, help making a report, or simply someone willing to listen.
Key References
Jessica Bouchard, Jennifer S. Wong and Chelsey Lee, “Fostering College Students’ Responsibility as Prosocial Bystanders to Sexual Violence Prevention: A Meta-Analysis of the Bringing in the Bystander Program,” Journal of American College Health (2024).
John M. Darley and Bibb Latané, “Bystander Intervention in Emergencies: Diffusion of Responsibility,” Journal of Personality and Social Psychology (1968).
Peter Fischer et al., “The Bystander-Effect: A Meta-Analytic Review on Bystander Intervention in Dangerous and Non-Dangerous Emergencies,” Psychological Bulletin (2011).
Ernest N. Jouriles et al., “Bystander Programs Addressing Sexual Violence on College Campuses: A Systematic Review and Meta-Analysis of Program Outcomes and Delivery Methods,” Journal of American College Health (2018).
Bibb Latané and Steve Nida, “Ten Years of Research on Group Size and Helping,” Psychological Bulletin (1981).
Mark Levine and Simon Crowther, “The Responsive Bystander: How Social Group Membership and Group Size Can Encourage as Well as Inhibit Bystander Intervention,” Journal of Personality and Social Psychology (2008).
Richard Philpot, et al., “Would I Be Helped? Cross-National CCTV Footage Shows That Intervention Is the Norm in Public Conflicts,” American Psychologist (2020).
Roni Porat, et al., “Preventing Sexual Violence: A Behavioral Problem Without a Behaviorally Informed Solution,” Psychological Science in the Public Interest (2024).
Right To Be, “The 5Ds of Bystander Intervention,” RightToBe.org.
Dianne M. Tice and Roy F. Baumeister, “Masculinity Inhibits Helping in Emergencies: Personality Does Predict the Bystander Effect,” Journal of Personality and Social Psychology (1985).
University of Bath, “Bystander support is crucial for tackling anti-social behaviour,” Press Release (January 25, 2024)
