SGEM#497: We Could be Heroes – Just with a Little Help from Batman

Date: Dec 17, 2025 Reference: Pagnini F, et al. Unexpected events and prosocial behavior: the Batman effect. npj Mental Health Research. November 2025 Guest Skeptic: Dr. Dennis Ren is a pediatric emergency medicine physician at Children’s National in Washington, DC. You may also know him as the host of SGEM Peds. Case: It’s been a dark, cold day in Gotham City. You’re finally on the metro heading home after a long shift. The train is packed, and you’re standing, crammed uncomfortably among all the other citizens eager to get home. Outside, you see the holiday lights and decorations, trying valiantly to shine through the flurries of snow. At the next stop, you see a visibly pregnant passenger board the crowded train car. She shuffles in and stands, holding the rail. No one around her moves. No one gets up to offer their seat. Hardly anyone even notices. You pull your coat a bit tighter around you and wonder: What’s it going to take to nudge people to be a little more helpful? Background: Prosocial behaviour is something we rely on every shift but hardly ever discuss explicitly. Psychologists typically define it as voluntary actions aimed at helping others. Examples include holding a door open, donating money, giving up your seat, or stepping in to assist a stranger. It’s an umbrella term that covers everything from simple everyday kindness to extraordinary acts of altruism. The world just saw an extraordinary example of prosocial behaviour in the Bondi Beach attack ‘hero’, Ahmed Al Ahmed. Motivations vary: empathy and concern for others, a desire to follow social norms, expectations of reciprocity, and even a wish to avoid guilt all influence prosocial behaviour. Prosocial behaviour has traditionally been studied in several primary ways. In the lab, researchers use economic games (such as dictator, ultimatum, and public goods games), staged helping tasks (such as picking up dropped pens or assisting with a “broken” computer), or vignette-based scenarios (“Would you stop to help?”). In real-world settings, classic bystander studies explore whether people intervene when someone seems in need and what situational factors (crowding, diffusion of responsibility, perceived danger) influence their decision to act or remain passive. Throughout all approaches, a key theme is that context plays a crucial role: the same individual may assist in one situation but ignore someone in another. Over the past decade, there has been increasing interest in how subtle environmental cues influence prosocial behaviour. Mindfulness research indicates that when people focus on the present moment, they may be more inclined to notice others’ needs and respond accordingly, although the evidence remains modest and not definitive. Another area of study examines "social primes." For example, images of superheroes can temporarily boost helping intentions and small acts of assistance. A related body of research on the “pique technique” demonstrates that unusual, unexpected events or requests can disrupt automatic “no” responses and increase compliance or helping, likely by pulling people out of autopilot. The “Batman effect” study we explore today extends these ideas into real-world scenarios. Could an unexpected disruption, such as a person dressed as Batman, increase a specific prosocial behaviour? For an emergency physician accustomed to crowded waiting rooms and chaotic departments, it’s an intriguing yet potentially significant question: can small, harmless environmental “shocks” encourage people to do the right thing a little more often without anyone ever having to take a mandatory module on ethics? Clinical Question: Among passengers on a crowded metropolitan subway, does the presence of an unexpected event (a person dressed as Batman) increase the likelihood that someone offers their seat to a pregnant-appearing woman, compared with no Batman present? Reference: Pagnini F, et al. Unexpected events and prosocial behavior: the Batman effect. npj Mental Health Research. November 2025 Population: Passengers in crowded cars on the Milan underground metro. Intervention: Presence of someone dressed as Batman (~3 meters away, no interaction with an experimenter pretending to be a visibly pregnant woman). Comparison: Identical set-up without Batman Outcome: Whether the seated passenger offered their seat to the (pretend) pregnant woman during a single-stop ride Type of Study: Quasi-experimental, non-randomized controlled field study. Both conditions were conducted simultaneously in different train cars and different areas of the platforms. Authors’ Conclusion: “This study suggests that unexpected events can increase prosocial behavior by momentarily disrupting automatic attention patterns and fostering situational awareness. These findings open new avenues for understanding the environmental and cognitive mechanisms underlying prosociality, and suggest potential applications for promoting kindness and cooperation in everyday settings—extending the “Batman effect” to nonsuperheroes as well.” Quality Checklist for Quasi-Experimental Studies: Is it clear what is the ‘cause’ and what is the ‘effect’? Yes Were the participants included in any comparisons similar? Unsure  Were participants in any comparisons receiving similar treatment/care other than the exposure? Yes Was there a control group? Yes Were there multiple measurements of the outcome both pre and post the intervention/exposure? No Was follow-up complete, and if not, were differences adequately described and analyzed? Yes Were the outcomes of participants in any comparisons measured in the same way? Yes Were outcomes measured in a reliable way? Unsure  Was appropriate statistical analysis used? Yes Results: They conducted 138 observations (70 control and 68 experimental with Batman). Most who offered up their seat were women (~68%) with a mean age of approximately 42 years. Key Result: Passengers were roughly three times more likely to give up their seat for a pregnant-appearing woman when Batman was present on the train compared with when he was not. Primary Outcome: Offering a seat to the pregnant woman 38% control vs 67% Batman Odds ratio (OR) = 3.39, p < 0.001 They also conducted brief follow-up interviews with the passengers who gave up their seats and asked whether they had seen Batman. Many spoke about recognizing pregnancy, social norms, education or safety. Nobody directly admitted to giving up their seat because Batman was there. In fact, 44% of those interviewed stated they didn’t see Batman at all. 1. Hawthorne Effect: When discussing this type of field experiment, it is important to mention the Hawthorne effect, which is the idea that people change their behaviour simply because they know they are being observed. In this study, a designated observer recorded whether passengers offered their seat to the pregnant woman. It is not specified how the observer tried (or did not try) to blend into the environment. It is possible that some passengers realized they were being observed, which could have influenced their actions. 2. Confounding: We applaud the authors for attempting to avoid confounding by concurrently conducting experiments with two research teams in different train cars and areas of the platform. The train cars also had to have all seats occupied with no more than five people standing between seats. Ultimately, this was not a randomized experiment. While they looked at Batman vs No Batman scenarios, that may not be the only thing that differs amongst the groups represented in their respective train cars, which was not measured or reported. In an exaggerated example, what if the car with Batman was full of nuns while the other car had a more heterogeneous smattering of society? Did people give up their seats because Batman was there or because they were or were not nuns (assuming that nuns are more charitable)? We also don’t know if the behaviour of the “pregnant” experimenter changed at all between scenarios. Did she always choose to stand in the same place? Did she make or avoid eye contact with the people around her? Were there other environmental factors that made her easier or more difficult to notice? 3. Outcome Measurement: The outcome of interest was the number of people who gave up their seats to the pregnant woman. Based on the methods, it looks like there was only one observer. The study findings may have been more robust if they had more than one observer and reported inter-rater reliability. I’m reminded of the selective attention test, where the viewer is tasked with counting how many times players pass a ball without noticing the gorilla that walks in and out of the frame. 4. Loss to Follow Up: Among the 138 observations, 70 instances of prosocial behaviour were observed. Of those, only 52 (74%) were interviewed during follow-up. This makes us wonder: What about the people who weren’t interviewed? Would they have provided us with more insight or breadth about their reasons for giving up the seat? Would some of the people who weren’t interviewed in the experimental design have stated that they saw Batman? 5. Generalizability: The experiment was conducted in the Milan metro system. There are likely differences in cultural or societal norms that make the findings of this study less generalizable to all transit systems worldwide. One example may be that the number of stops or the distance between stops varies. Shorter distances between stops and more frequent stops could encourage more people to give up their seats to pregnant women as their stop approaches. Is it actually the Batman Effect? This threat to validity hurts me a little bit to admit…because my confirmation bias says “Of course it was the presence of Batman who was responsible for people doing good!” But I mean, who is Batman?