A Summary of the Bystander Effect: Historical Development and Relevance in the Digital Age

John Darley and Bibb Latané were the first psychologists to formulate and study the bystander effect. The bystander effect, as defined by Darley and Latané (1968), is the phenomenon in which the presence of people (i.e., bystanders) influences an individual’s likelihood of helping a person in an emergency situation. Specifically, Darley and Latané believed that as the number of people who are present in an emergency situation increases, the less likely it is that any single individual will help someone in need. This was the original framework for bystander intervention that guided the researchers’ experimentation of a social behavior in a laboratory setting. In order to effectively analyze the history of research relevant to the bystander effect, it is necessary to understand the key historical events that preceded the work of Darley and Latané.

One key event that occurred in the history of psychology was the creation of the Society for the Psychological Study of Social Issues (SPSSI). Before the establishment of this organization in 1936, the field of psychology was not involved in studying and applying psychological principles and research to social issues (Benjamin, 2014). That is, most psychologists at the time did not believe that the field of psychology could be engaged in the attempt to solve social problems and still remain a scientific discipline. However, during the time of the Great Depression, the zeitgeist began to change as more people began to think that psychology could be applied to resolving the social conflict in America and abroad (Benjamin, 2014). Indeed, since the establishment of the SPSSI, psychologists have been focused on applying the science of psychology to better understand and solve social problems by looking at group dynamics (Benjamin, 2014). For example, the pioneering work of Kurt Lewin, attributed to the beginnings of research on social action and social change, occurred after 1939 and contributed to the understanding of the dynamics of groups, such as the factors that influence prosocial behaviors in groups.

Kurt Lewin arrived in America from Germany in 1933, escaping the Nazi regime (Benjamin, 2014). He is often credited as a critical figure, if not one of the forefathers, in the field of applied social psychology; his contributions to social psychology are extensively described in several books on the history of psychology (e.g., Baumeister & Vohs, 2007; Benjamin, 2014; Hogg & Cooper, 2003; Mook, 2004). Lewin, upon his arrival in the US, became highly involved in social research and its applications in the world. In fact, he was a key figure in the development of the SPSSI in 1936 (Baumeister & Vohs, 2007; Benjamin, 2014). His main area of interest was in researching the ways that groups function and influence the individual members of the group, as he thought that this was an important piece to solving social problems (Benjamin, 2014).

The situation a person finds him or herself in is a significant determinant of how one will behave under the given circumstances

Lewin, who was heavily influenced by Gestalt psychology, was also interested in studying the situational factors that influence a person’s behavior, which led to his development of field theory. He argued that the situation a person finds him or herself in is a significant determinant of how one will behave under the given circumstances (Mook, 2004). Furthermore, Lewin’s contribution to theories on the motivational influences of behavior in the 1950s has led social psychologists to speculate what motivates others to engage in prosocial behaviors (Hogg & Cooper, 2003). This focus on motivational factors is characteristic of the “cognitive revolution” that emerged in the 1950s and extended into the 1960s (Baumeister & Vohs, 2007).

The cognitive approach focuses on how people think about themselves and the world around them; this focus on cognition was critical for psychologists who were trying to understand the mental processes guiding an individual’s behavior in groups. While the social psychologists at this time were interested in the factors affecting a person’s motivation to help others, the research question would change in the 1960s to what causes a person to not provide any help to someone in an emergency situation. This shift was prompted by a tragic event in 1964 and is evident in Darley and Latané’s (1968) classic study on bystander intervention.

A major inspiration for Darley and Latané’s (1968) research was the 1964 murder of a New York City woman in which no bystander intervened to help. According to Darley and Latané, many people at the time were trying to find a plausible explanation for the inaction on behalf of all the bystanders (people viewing the violence from their apartment windows). Namely, many people believed that apathy and indifference were the causes of inaction on behalf of the bystanders, reflecting the idea that personal characteristics solely drive behavior. However, Darley and Latané focused on the social conditions, such as the number of bystanders, that may have had an influence on whether the bystanders reacted, which reflected Lewinian theories on the situational determinants of behavior. Moreover, the three reasons (i.e., diffusion of responsibility, diffusion of blame, and thinking that another person is already taking action to help) that Darley and Latané gave for believing that the presence of bystanders may influence an individual’s likelihood of helping someone in an emergency consist of thinking strategies that are representative of the cognitive era in the 1960s.

To summarize the historical context briefly, the violent murder of a woman in New York City along with all of the aforementioned historical events prompted the research on the bystander effect: the formation of the SPSSI in 1936 normalized and made it more acceptable to study social issues in psychology; Lewin’s research on group dynamics and social determinants of behavior provided the groundwork for future psychologists (i.e., Darley & Latané, 1968); and cognitive psychology emphasized the role of thinking and perception, which was used to explain certain behaviors in group settings.

In their classic study, Darley and Latané (1968) proposed that the number of individuals present in an emergency situation influences how quickly, if at all, any individual responds. They hypothesized that the more people present at an emergency, the smaller the chance that a person will intervene and help or the more time it will take for him or her to do so. In order to test this, they created a simulation of an emergency situation. Participants were deceived so that they would not know they were going to be in a fake emergency; they were told that they would take part in a study that would require them to discuss their problems pertaining to college.

In the experiment, an individual participant was placed into a room with a microphone. Darley and Latané manipulated their independent variable by telling the participant one of three things: there would be one other participant in a different room communicating with the participant; two other participants in separate rooms; or five other participants in separate rooms. However, these other “participants” were only prerecorded voices. At some point in the “discussion,” the participant would hear someone speak who started to have what sounded like a seizure–the victim’s voice would begin to break and the participant heard cries for help, indicating that the participant was having a seizure. The dependent measure was the time it took for each participant to respond to the emergency. The results supported Darley and Latané’s prediction.

Most of the participants who thought they were alone with the victim (in the two-person group) responded to the emergency whereas only 31% of the participants who thought they were with four bystanders (in the six-person group) notified the experimenter of the emergency. Also, participants in the two-person condition responded at a faster speed than the participants in the six-person condition. Darley and Latané concluded that those who thought they were alone with the victim intervened when the victim was having a seizure because they felt the most pressure to help as the consequences of not helping (feeling guilt and shame) were all on their shoulders; therefore, they resolved their conflict quickly. The researchers also argued that people who felt they were not alone in witnessing the situation were not as pressured to help and, because of this, they were less likely or slower to react. Their results strongly suggested that personality factors of apathy and indifference were not causing the participants’ decision to not intervene, as was previously believed.

Psychologists often evaluate the quality of an empirical study by assessing the internal, external, and construct validity of the research (e.g., Morling, 2012). In assessing the internal validity of the Darley and Latané (1968) study, or their ability to draw a cause-effect relationship from their results, it is important to recognize that their description of the study’s design suggests that they randomly assigned the participants to one of the levels of the independent variable (i.e., number of bystanders believed to be present). This eliminated individual differences such as personality characteristics as a likely explanation for their results. Furthermore, they made an effort to keep other factors constant. For example, they prerecorded the voices they would play for each participant. This would lessen the chance that an extraneous variable, such as change in tone of voice, would have affected the speed at which participants responded.

They also varied the order in which the voices were played. Darley and Latané noted that participants’ nervousness, surprise upon finding out the true nature of the experiment, and comments made during the experiment indicated that the seizure was perceived as real. Therefore, their internal validity is very high. However, their cause-effect conclusion might not generalize to other settings (i.e., external validity) because the participants were placed in a situation in which they only heard, but did not see, the other bystanders. The emergency situation itself, which involved someone having a seizure, is quite different from witnessing someone stab a person to death, as occurred in the 1964 murder case. Nonetheless, Darley and Latané were able to create a simulated and controlled laboratory experiment that mimicked an emergency situation. Many future scientists conducted research that replicated Darley and Latané’s results and expanded the research on bystander intervention.

After Darley and Latané’s (1968) classic study was published, many researchers became interested in the bystander effect and its impact. A search of a popular database of psychological research, PsycINFO, revealed a plethora of studies on bystander intervention published since 1968. A review of all this research is well beyond the scope of this article. Instead, I will summarize a few representative examples of research related to Darley and Latané’s classic study. Twelve years after Darley and Latané’s research on the bystander effect, two studies appear to be representative of the research on bystander intervention in the year 1980. One of these studies (i.e., Valentine, 1980) attempted to establish the ecological validity of the Darley and Latané findings on bystander intervention and other factors that may affect whether the bystander effect could be diminished. Another study conducted by Schwartz and Gottlieb (1980) also reflected psychologists’ attempt to study the factors that could influence the occurrence of the bystander effect; specifically, these researchers examined anonymity as a mediating variable in the bystander effect. Both of these studies represent an effort in the 1980s to further test bystander intervention by manipulating factors, other than the number of bystanders, which may influence prosocial behaviors.

Valentine’s (1980) goal was to investigate the factors that could potentially weaken the bystander effect. She argued that if a positive relationship is formed between a bystander and the victim, then the bystander may feel more compelled to help the victim. According to Valentine, the best way to establish a relationship between bystander and victim is by implementing an interpersonal gaze, in which eye contact is established between bystander and victim. Therefore, Valentine hypothesized that the gaze by the victim would increase helping behaviors in participants regardless of whether bystanders would be present. In order to test the influence of gaze on the bystander effect, Valentine conducted an experiment outside of the laboratory.

She instructed either one confederate (no bystander condition) or two confederates (two bystander condition) to approach random women at designated bus stops in New York. One confederate would “accidentally” lose all of her change from her pocket while the other one stood nearby reading a newspaper. Afterward, the victim who dropped the coins would either gaze at the participant or stare at the ground for five seconds before picking up the dropped coins. Helping occurred when the participant would pick up the dropped coins or point to where they were on the ground. Valentine found that gaze did influence helping in participants as expected–participants in the gaze condition were more likely to help and help more quickly (regardless of the number of bystanders) than participants who were not gazed upon.

The study by Valentine (1980) differed from Darley and Latané’s (1968) study in that she did not test the bystander effect in an emergency situation. Instead, she used a natural setting (field experiment) and used dropped coins to indicate help needed by the victim. Her goal was to weaken the bystander effect by introducing the intervening factor of an established relationship between victim and bystander, as represented by gaze. Moreover, Valentine did not use more than two bystanders. Although it may be difficult to imagine that a simple stare could result in forming a relationship between a bystander and participant, the goal behind this implementation was to determine whether a gaze could elicit a feeling of obligation toward the victim, which would compel the participant to engage in helping.

Keeping these aspects of Valentine’s study in mind, the study can be assessed in regard to its validities. In terms of construct validity, the clearly manipulated independent variables (gaze vs. no gaze and one confederate vs. two confederates) reflect high construct validity in the study—she was accurately manipulating the theoretical constructs. Because this was a field experiment as opposed to one conducted in a laboratory, the researcher could not control all of the extraneous variables, which means her study had low internal validity. For instance, there was no way for her to make sure that no one else would come up to the bus stop as the experiment was taking place, and thus, introduce a confounding variable.

Despite this issue, Valentine trained her confederates to act practically identically in front of the participants, indicating her attempt to keep things constant as much as she could. With regard to the external validity, the study was weakened because of the fact that only white female participants were chosen. It does not generalize to other participant groups, such as males and people of other racial/ethnic backgrounds. Yet, because this was a naturalistic setting, there was high ecological validity as the experiment took place in real life as opposed to being conducted in a laboratory. By and large, this study revealed that the bystander effect is less likely to occur when a victim makes some form of contact that acknowledges the bystander.

During the same year that Valentine (1980) published her results, Schwartz and Gottlieb (1980) published their investigation of other factors that influence the occurrence of the bystander effect. Schwartz and Gottlieb proposed that perceived anonymity of a bystander might affect whether he or she helps a victim. The authors claimed that in addition to the diffusion of responsibility and blame that Darley and Latané (1968) described, another force that could influence helping is evaluation apprehension. This explanation pertains to whether the bystander knows if other bystanders and the victim are aware of his or her presence. Schwartz and Gottlieb reasoned that a bystander who feels that he or she is anonymous is less likely to help the victim due to having less evaluation apprehension (i.e., less fear of being judged by others at the scene).

In order to test the prediction that an individual’s perceived anonymity makes it less likely that the individual will provide help in an unambiguous emergency situation, Schwartz and Gottlieb performed two complicated experiments with very elaborate procedures. Their second experiment essentially replicated the results of their first experiment; and for the sake of brevity, I will only describe their first experiment. The actual participant in their first experiment arrived in a room in the social science building and was instructed to monitor another student’s extrasensory perception (ESP) transmissions by viewing him on a television screen. The student on the television screen was actually a professional actor. Ostensibly, the actor was “transmitting” ESP to another student who was supposedly in another room receiving the ESP messages and not visible to the participant. About 7.5 minutes into the ESP experiment, the student shown on the screen became a victim when he was physically assaulted by “a roughly dressed stranger” (who was also an actor).

Two independent variables were manipulated: the presence of a bystander and anonymity. Schwartz and Gottlieb manipulated the presence or absence of another bystander witnessing the crime by leading the participants in the alone conditions to believe the student receiving the ESP messages had shown up late and was not watching the victim at the time of the crime. In addition, Schwartz and Gottlieb manipulated whether the participant believed he or she was either anonymous or known. In the anonymous conditions participants were led to believe that the other students (i.e., the victim and the ESP “receiver”) were unaware that the study involved multiple participants; in the known conditions participants were told they would meet with all the other students (i.e., the victim and/or the other ESP receiver) after the ESP experiment was over.

Since this study employed a factorial design, each participant was randomly assigned to one of four conditions: (1) in the alone/anonymous condition the participants believed they were alone with the victim and that the victim was not aware they were present; (2) in the alone/known condition participants believed they were alone with the victim and the victim knew they were present; (3) in the bystander present/anonymous condition participants believed another bystander was present and the participant was anonymous to both the victim and the other bystander; (4) in the bystander present/known condition participants believed another bystander was present and the participant was not anonymous. Helping was measured by the time it took for the actual participant to respond by calling the experimenter using a phone in his or her room or leaving the room to help the victim after the stranger entered the victim’s room and began the assault.

Schwartz and Gottlieb (1980) replicated Darley and Latané’s (1968) results because they found that the presence of a bystander lessened participants’ likelihood of providing aid. Schwartz and Gottlieb also found that, contrary to their prediction, participants’ perceived anonymity in terms of the victim (i.e., the alone/anonymous condition) did not influence participants’ likelihood of helping the victim. However, in support of their prediction, participants helped less frequently and more slowly when they believed they were anonymous to the victim and another bystander (i.e., the bystander present/anonymous condition). Schwartz and Gottlieb argue that these results are consistent with their claim that evaluation apprehension, as well as diffusion of responsibility, contributes to bystander intervention in emergency situations. Schwartz and Gottlieb extended Darley and Latané’s research by manipulating perceived anonymity, or an individual’s perception that no other bystander knows about his or her existence, and found that anonymity moderates the bystander effect.

Schwartz and Gottlieb (1980) manipulated anonymity well in their study as shown by participants’ responses to post-experimental questionnaires that were used as a manipulation check. For example, 96% of their participants correctly answered questions about whether or not they expected to interact with the other students after the ESP experiment. Their study had low external validity, similar to Darley and Latané’s (1968) study, because it was not a representative sample as it only included undergraduate students. However, their study had good ecological validity because it was conducted in a setting in which participants both heard and saw the emergency situation unfold on a television screen, unlike only voices heard from a tape recording in the Darley and Latané experiment. Lastly, the study had high internal validity since the researchers meticulously controlled for extraneous variables and used random assignment to the different experimental conditions.

Valentine (1980) and Schwartz and Gottlieb (1980) are two prominent studies from the 1980s that represent psychologists’ efforts to investigate the bystander effect under different conditions. They established that other variables such as gazing and perceived bystander anonymity affect bystander intervention. Twenty years from the time these studies were conducted, researchers began to pursue more applied research goals. For example, they began to apply the bystander effect to social issues prevalent in society. As technology advanced, bullying on social media platforms increased accordingly (Brody & Vangelisti, 2016). The year 2000 marked the beginning of research on bystander intervention as it relates to online situations. For instance, Markey (2000) conducted a study on prosocial behaviors in online chat sites. This preceded the work of Brody and Vangelisti (2016) 16 years later, who studied the influence of the bystander effect on cyber bullying.

Markey (2000) observed hundreds of chat groups on the chat site Yahoo! Chat. He posted various requests for help in solving computer questions (e.g., how to look at someone’s profile online); some requests were made to all chat group members whereas some were directed at particular members by referencing their name. He then monitored who responded to the questions and provided help. Markey found that members took longer to respond when there were more people logged onto the chat group. However, this was reversed when the question was directed at a specific member of the group. While Markey did not conduct an experiment dealing with an emergency situation like Darley and Latané (1968) did, this study revealed a critical boundary of the bystander effect in the cyber world; aiming a question directly at another member by specifying the member by name makes it more likely that a person will respond. Therefore, it inhibits the bystander effect. This can be tied back to the Valentine (1980) study in which the researcher found that gaze, or acknowledgment of the bystander, made it more likely that the bystander would intervene and provide help. While this was not an emergency situation, future researchers took on the task of studying how the bystander effect may exist in situations involving cyber bullying.

One of the more prominent social issues that psychologists became interested in since Markey (2000) has been cyber bullying. In the 16 years since Markey’s research, Brody and Vangelisti (2016) showed that cyber bullying is a social issue relevant in our modern society. These researchers conducted studies on the relationship between the number of bystanders and perceived anonymity by asking participants in a survey to describe a friend’s past online bullying experience that they witnessed. Brody and Vangelisti found a negative correlation between the victim’s perception of the number of bystanders and the likelihood of intervention (as measured by a Likert-type rating scale). That is, as the number of perceived bystanders increases, the likelihood of intervention decreases. They also found that bystander anonymity was negatively associated with likelihood of helping the victim; when bystanders were anonymous they were less likely to help. These correlations are not experimental findings, however, and should only be interpreted as associations—and not as causal relationships. Nonetheless, these findings can be viewed as a continuation of the research on the bystander effect and anonymity (and reduction of the bystander effect), as was demonstrated by Darley and Latané (1968) and Schwartz and Gottlieb (1980) respectively.

To conclude, in this article I describe the historical context surrounding the well-known phenomenon of the bystander effect. Historically, the formation of the Society for the Psychological Study of Social Issues, Kurt Lewin’s research on group dynamics and influences on prosocial behavior, and the emergence of the cognitive revolution paved the way for research on why no bystander intervened to help one woman in New York City in 1964 as she was stabbed to death. The bystander effect first demonstrated experimentally by Darley and Latané in 1968 was a classic study that changed future research on prosocial behavior. In particular, the research that came after Darley and Latané investigated the other variables that influence helping behaviors. Such variables as gaze (Valentine, 1980) and anonymity (Schwartz and Gottlieb, 1980) were studied in the 1980s. Furthermore, the beginning of the 21^st century marked a time of increased awareness of the relevance of psychological research to contemporary social issues, as evidenced by research on the bystander effect in situations like online chat rooms and social media cyber bullying. All things considered, it is clear that Darley and Latané’s classic study on the bystander effect is still highly relevant to the field of modern psychology.

---

References

Baumeister, R. F., & Vohs, K. D. (2007). Encyclopedia of social psychology. Thousand Oaks, CA: Sage Publications.

Benjamin, L. T. (2014). A brief history of modern psychology. Hoboken, NJ: Wiley.

Brody, N., & Vangelisti, A. L. (2016). Bystander intervention in cyberbullying. Communication Monographs, 83(1), 94-119. doi:10.1080/03637751.2015.1044256.

Darley, J. M., & Latane, B. (1968). Bystander intervention in emergencies: Diffusion of responsibility. Journal of Personality and Social Psychology, 8(4, Pt.1), 377-383. doi:10.1037/h0025589.

Hogg, M. A., & Cooper, J. (2007). The SAGE handbook of social psychology. London: SAGE.

Markey, P. M. (2000). Bystander intervention in computer-mediated communication. Computers in Human Behavior, 16(2), 183-188.

Mook, D. G. (2004). Classic experiments in psychology. Westport, CT: Greenwood Press.

Morling, B. (2012). Research methods in psychology: Evaluating a world of information. New York, NY: Norton.

Schwartz, S. H., & Gottlieb, A. (1980). Bystander anonymity and reactions to emergencies. Journal of Personality and Social Psychology, 39(3), 418-430. doi:10.1037/0022-3514.39.3.418.

Valentine, M. E. (1980). The attenuating influence of gaze upon the bystander intervention effect. Journal of Social Psychology, 111(2), 197.