Thursday, 16 January 1986

Implicit Reversal of Bystander Apathy (2008)

SSRM001 Term Paper

The bystander apathy effect is a well-established empirical phenomenon in social psychology. As shown by studies after investigating the apparent apathy of bystanders in the Kitty Genovese murder, a person who faces a situation in which another person is in distress with the knowledge that others are present and available to respond is less likely and slower to respond than one who faces the same person in distress without others present (Darley & Latane, 1968). Follow-ups to this study have consistently found the presence of this effect by manipulating the presence of others in varied situations (Latane & Nida, 1981).

A study by Darley, Garcia, Weaver & Moskowitz (2002) found that merging priming methods with bystander apathy studies can demonstrate how priming a social context can lead to lessened helping behavior on a subsequent and unrelated task. The study is significant because it confirms through priming methods, various helping scenarios and a lexical decision task that diffused responsibility and unaccountability are the mediating variables that account for why individuals react less or slower when someone is in need.

What is fascinating about this study is that this opens up room for further discussion as to whether the effect of bystander apathy can actually be put to good use and be reversed such that people become less apathetic. Current literature stipulates that the immediate or imagined presence of others exerts its influence on helping because these others are involved in the situation at hand. If individuals believe that immediate or imagined others cannot possibly help, then bystander apathy will not occur and individuals will behave as if alone (Bickman, 1972; Korte, 1971).

Intuitively, this entails priming individuals about the lack of presence of others such that, in imagining that they are alone, they will be more reactive and willing to offer a helping hand, perhaps paving the way towards a more gracious society. The purpose of this study is thus an exploration of the possibility of merging priming methodology with previous bystander apathy study methods to invoke an increase in helping behavior.
The general hypothesis of this paper can hence be established: Individuals primed with the lack of presence of others will exhibit an increase in helpfulness.

Research design

To attempt to establish a link between primed ‘loneliness’ and ‘solitude’ with willingness to help, a 2x2 factorial experimental design will be utilized with 4 independent variables and the dependent variable operationalized as follows:

X1 – Loneliness photograph set

Participants will be shown the loneliness photograph set which consists of 10 photographs depicting being alone or being in a state of solitude, namely a man alone in the desert, an empty theatre, an empty stadium, an empty warehouse, an empty restaurant, a man climbing a mountain by himself, a student alone in an exam hall, an empty conference room, an empty train and a bus with only one passenger. These photographs will be taken in such a manner that exposes as much of the surrounding as possible. These photographs will be flashed by a projector onto a screen for 5 seconds each.

X2 – Neutral photograph set

Participants will be shown the neutral photograph set which consists of 10 photographs of random, unrelated things, namely a group of buildings, the sky, the horizon, an mp3 player, the periodic table, a car, perfumes, a stack of books, a soccer ball, and a set of physics equations. Object photos will be close-ups with as little backdrop as possible (non-object photos include the sky, the periodic table and the horizon). These photographs will be flashed by a projector onto a screen for 5 seconds each.

X3 – Loneliness imagination

Participants will be given the following instruction: “imagine that you are walking alone through a street and there is absolutely nobody around, and think of why the street is empty and where you would head to.” Phrasing the question this way affirms the idea of being alone and forces participants to establish and accept that they are already alone before they can cognitively process the latter part of the question. Participants will be instructed to spend 3 minutes thinking about this, as a study by Dijksterhuis & van Knippenberg (1998) found that primes get stronger as their exposure time increases.

X4 – Null imagination

Null imagination involves not being instructed to imagine about anything.

Dependent variable

The dependent variable is essentially helping behavior. This study seeks to determine if there is more willingness to help, so the dependent variable will be operationalized as a combination of 2 measures: willingness to donate and willingness to assist impromptu in a subsequent study.

In prior studies, evidence has been found that helping behavior in the form of donations and the provision of assistance in a subsequent and unexpected task can be reduced by the imagined presence of others. This is congruent with the effect of diffusion of responsibility that bystander apathy, activated in the presence of others, creates (Darley, Garcia, Weaver & Moskowitz, 2002). If the effect of bystander apathy can indeed be reversed, then we will observe an increased willingness in helping behavior.

To approach this, in the 1st segment, participants will be given a questionnaire to inform them about a potential donation drive, which will read as follows: “A donation drive may be carried out soon to raise funds to construct an exciting new lounge on campus that will have many state-of-the-art facilities. The estimated amount that would have to be raised is projected to be about $75,000. Would you be willing to contribute to this donation drive and, if so, how much would you contribute? (Please check).” At this point, participants will be able to check one of the following contribution ranges: Unwilling to donate, $25 or less, $26-$50, $51-$75, $76-$100, More than $100. Upon completion of this questionnaire, the 2nd segment comes in and participants will be verbally informed about another experiment in the room next door and will be told to indicate if they are willing to help out by stating it on the bottom of the survey.

The reason for this dual approach to quantifying helpfulness is that the donation questionnaire, though useful as a measure of the level of contribution one is willing to provide, is merely a hypothetical scenario and may result in skewed results due to its non-committal nature and low personal cost involved. With the subsequent and unexpected study assistance, we can further determine if participants are willing to assist in a less remote helping situation, which is more plausible in terms of real life situations that often are immediate and do not offer adequate time to make careful cost-benefit considerations.


A total of 120 randomly selected participants from SMU will be enrolled in this study.


As can be seen from the table, the study will be designed such that there are 4 groups and can determine any possible null, main or interaction effects. The 120 participants will be randomly assigned to each group such that every group will have 30 participants.

To begin, Group 1 and Group 2 will be given treatment X1 where participants will be shown loneliness photographs, while Group 3 and Group 4 will be given treatment X2 where participants will be shown neutral photographs.

Before moving on to the next part of the experiment, participants will answer a filler question, “what is your ideal outdoor temperature?” so as to keep participants ‘blind’ to the purpose of the study.

Group 1 and Group 3 will then be given treatment X3 where participants will be instructed to “imagine that you are walking alone through a street and there is absolutely nobody around, and think of why the street is empty and where you would head to” for 3 minutes, before concluding the experiment. Group 2 and Group 4 will conclude the experiment immediately after their first treatments, which is essentially treatment X4. All participants will then be debriefed as to the actual purpose of this experiment.

Group 4 is the control group as it will be subjected to only neutral photographs and no attempt to imagine being alone will be made, thus supposedly representing one’s neutral state. A benchmark can be made here for the other groups to be compared against.

The experiment order is the photograph treatment first followed by the imagination treatment because we wish to prime participants in a certain manner (through the photographs) such that the imagination process may be guided in a certain way, specifically reinforcing the idea of being alone. This design also allows us to determine whether the photographs or imagining takes precedence in the strength of priming. If, for example, imagining is a stronger prime than looking at photographs depicting being alone (i.e. effect of X3 is greater than effect of X1), then the amount of help rendered will be expected to be highest for Group 1, 2nd highest for Group 3, 3rd for Group 2 and lowest for Group 4.


As the imagination treatment is expected to be stronger (especially as a reinforcer) than the photograph treatment, the expected order of willingness to help will be in ascending order as follows: Group 4, Group 2, Group 3 followed by the most willing to help, Group 1.
If the expectations are met, the experiment shows that we can indeed prime the lack of presence of others such that it is not necessary to be physically alone for an increased sense of responsibility and accountability, resulting in an inclination to offer help. This confirms the hypothesis: Individuals primed with the lack of presence of others will exhibit an increase in helpfulness.


Firstly, it is uncertain as to whether or not the images shown in the neutral photograph set, X1, are truly neutral. In either systematic or random fashion, certain subjects in the photographs may, for example, unintentionally invoke sensations of accountability which will result in greater willingness to help if the mediating variable (heightened responsibility and accountability) is indeed true.

Secondly, it is assumed that the photographs will prime participants in such a manner that guides their subsequent imagination process (for the imagination treatment group) and as such the experimental treatment order is the photograph treatment first followed by the imagination treatment. The results should determine whether this order of treatment is valid, but does not entirely rule out the possibility of other mediating variables coming into play. Furthermore, there is a possibility that participants may interpret the sequence in a certain way, leading to participant bias as they may suspect what is demanded of them. This is especially so for Group 1 where there are more variables to draw inferences from. The filler question is designed to prevent this as far as possible.

Lastly, current literature has gathered evidence that bystander apathy will not occur if individuals know that immediate or imagined others cannot possibly help. Individuals will then behave as if alone (Bickman, 1972; Korte, 1971). An alternative way to confirm if the mediating variable is true is to use photographs and imagination treatment depicting the inability of others to assist in a situation where someone is in need of help.

Future directions

If the reversal of bystander apathy works and the hypothesis: individuals primed with the lack of presence of others will exhibit an increase in helpfulness is confirmed, then we could possibly this to a more practical, real world context, such as in promoting a gracious society marked by a greater sense of helpfulness through marketing campaigns. At this point in time it sounds rather absurd to ‘tell’ people that they are alone in advertisements, drives or public speeches just to elicit helping behavior, but with more thought and further experimentation, especially with the mediating variable more clearly understood, we could perhaps design creative methods of promoting helpful behavior.

There is also the possibility of employing subliminal priming. Albeit ethically questionable, it is worth further study to determine the lasting effect of subliminal messages to prime heightened responsibility and accountability and subsequently willingness to help, and whether or not it is possible to combine these messages with marketing efforts to create campaigns promoting a gracious and kinder society.


  1. Bickman, L. (1972). Social influence and diffusion of responsibility in an emergency. Journal of Experimental Social Psychology, 8, 438–445.
  2. Dijksterhuis, A., & van Knippenberg, A. (1998). The relation between perception and behavior, or how to win a game of trivial pursuit. Journal of Personality and Social Psychology, 74, 865–877.
  3. Darley, J. M., Garcia, S. M., Weaver, K. & Moskowitz, G. B. (2002). Crowded minds: Implicit bystander apathy. Journal of Personality and Social Psychology, 83, 843–853.
  4. Darley, J. M., & Latane, B. (1968). Bystander intervention in emergencies: Diffusion of responsibility. Journal of Personality and Social Psychology, 8, 377–383.
  5. Korte, C. (1971). Effects of individual responsibility and group communication on help-giving in an emergency. Human Relations, 24, 149–159.
  6. Latane, B., & Nida, S. (1981). Ten years of research on group size and helping. Psychological Bulletin, 89, 308–324.


Part 1

Age: _________________________
Sex: _________________________
Race: _________________________
Religion: _________________________
Nationality: _________________________
Faculty: _________________________

Circle the mood that best describes you at this moment:
Very unhappy Neutral Very happy
1 2 3 4 5 6 7

Part 2

What is your ideal outdoor temperature?

Part 3

A donation drive may be carried out soon to raise funds to construct an exciting new lounge on campus that will have many state-of-the-art facilities. The estimated amount that would have to be raised is projected to be about $75,000.
Would you be willing to contribute to this donation drive and, if so, how much would you contribute? (Please check).

 Unwilling to donate
 $25 or less
 $26-$50
 $51-$75
 $76-$100
 More than $100.
If more than $100, how much more?

Part 1 gathers participant information according to 6 different population demographics. This is important as participants can be classified accordingly, especially when there is the possibility that belonging to certain demographics causes participants to behave in a certain manner. For example, females may be more helpful than males. This may result in systematic error and determining participants’ demographics may be useful in identifying such errors. Further experimentation can then be carried out to determine why this is so.

Part 2 is a filler question designed to prevent any participant bias that may occur from participants suspecting the purpose of the study as the experiment moves from the photograph treatment set to the imagination treatment, thus improving the study’s reliability.

Part 3 of the questionnaire deals with measuring the construct, which is the participant’s willingness to help. By operationalizing helpfulness in the form of monetary donations to one’s college, we can determine how helpful one is willing to be by the amount of money one is willing to consider donating.

The most important segment of the questionnaire with regards to measuring the construct is essentially Part 3. Part 3’s measurement method of willingness to help is based on the study by Darley, Garcia, Weaver & Moskowitz (2002) and is operationalized as monetary contributions to the school, which has established itself to have content validity. A good approach that the study took which could be applied to this experiment is the use of a lexical decision task test. Having identified the diffusion of responsibility as the mediating factor between priming the presence of others and a decline in helping, Darley, Garcia, Weaver & Moskowitz (2002) proceeded to confirm it by getting participants to do a lexical decision task test involving unaccountability synonyms, such as unaccountable, innocent and exempt. Results show that participants primed with the presence of others indeed responded faster to words related to unaccountability. The experiment discussed in this paper can improve its content validity by getting participants to undergo a lexical decision task test after they have been primed with the lack of presence of others to see how fast they respond to words related to being accountable, such as accountable, responsible, liable and answerable.

Though somewhat unrelated to the questionnaire, the content validity of the experiment is refined through the use of the subsequent experiment assistance request where participants were asked if they were willing to help out in another experiment happening just after concluding the current one. This improves content validity as it represents a different aspect of the construct.

Construct validity is also fundamentally based on the study by Darley, Garcia, Weaver & Moskowitz (2002). There, 5 studies were carried out in a theoretical-experimental manner, as each study was developed after an aspect of the relationship between variables was confirmed in a way that was consistent with the theory. Riding on the construct validity of the imagination treatment in their study, the experiment in this paper modified the imagination scenario to be one that depicts absolute loneliness instead of crowds or having one other friend with oneself.
It is uncertain, however, as to whether the treatment photograph set of pictures connoting being alone truly evokes a heightened willingness to help, or affects helping behavior for that matter. However, there is a degree of face validity that would be motivational enough for its accuracy in measuring the construct to be tested.

A total of 120 randomly selected participants will be enrolled into this study, of which will be further split up into 4 group conditions of 30 each, all randomly assigned. It is adequate enough to obtain this sample from a single college or university and determine with reasonable confidence the external validity of the sample and the experiment and hence its generalizability to the population. This is because the study, which is also basically an exploratory one, is attempting to find a relative difference in terms of willingness to help. As long as we can establish that individuals are indeed predisposed to offer more assistance once they believe that there is no one else around them to offer help (based on the idea that the mediating variable of a heightened sense of accountability is operating behind the psychological mechanism), then the experiment has achieved its objective.

No comments: