. 82
( 115 .)


who are alone are likely 40
to help in high and low
danger situations. The 30
presence of another
bystander increased
helping in the high danger
but not low danger
situation; a clear reversal
of the usual bystander
effect. High Low
Situation Danger Level
Based on data from Fischer, et al. (2006).
Chapter 11 Prosocial Behavior and Altruism 417

bystander effect. Subjects walking in groups were more likely to help (85%) than sub-
jects walking alone (65%). In this situation”a victim is clearly in need and the helping
situation is dangerous”it seems that bystanders in groups are more likely to help than
solitary bystanders (Clark & Word, 1974; Harari et al., 1985).
The bystander effect also seems to be in¬‚uenced by the roles people take. In another
study, some subjects were assigned to be the leaders of a group discussion and others
to be assistants (Baumeister, Chesner, Senders, & Tice, 1988). When a seizure was
staged, subjects assigned the role of leader were more likely to intervene (80%) than
those assigned the role of assistant (35%). It appears that the responsibility inherent in
the leadership role on a speci¬c task generalizes to emergencies as well.
Finally, the bystander effect is less likely to occur when the helping situation
confronting us involves a clear violation of a social norm that we personally care about.
Imagine, for example, you see a person throw an empty bottle into the bushes at a public
park. In such a situation you may engage in social control behaviors (e.g., confront the
offender, complain to your partner). Contrast this with a situation where private property is
involved (e.g., painting graf¬ti in an elevator in a building owned by a large corporation).
You may be less likely to engage in social control behaviors. Chekroun and Brauer (2001)
wondered if the bystander effect would operate differently in these two situations. They
hypothesized that the bystander effect would hold for situations involving low personal
implications (e.g., graf¬ti in the elevator), but not in situations involving high personal
implications (e.g., littering in a public park). In the low-personal-implication condition
a confederate of the experimenters entered an elevator in a shopping center parking lot.
As soon as the door closed, the confederate began scrawling graf¬ti on the wall with a
magic marker. This was done under two conditions: a participant alone in the elevator
with the confederate (no bystanders) or two or three naïve individuals in the elevator
with the confederate. In the high-personal-implications condition a confederate of the
experimenters threw an empty plastic bottle into some bushes in a public park in front
of one participant or a group of two or three participants. In both situations the reaction
of the participant(s) was (were) recorded on a scale ranging from no social control to
an audible negative comment. As you can see in Figure 11.5, social control was most


Percent Social Control


Figure 11.5 Social
control behaviors are more
likely if a behavior has
high personal implications
20 (littering in a public park)
than if the behavior has
low personal implications
(graf¬ti in a privately-
owned elevator).
Graffiti Littering
Based on data from Checkroun and Brauer
Deviant Behavior (2002).
Social Psychology

likely to occur when other bystanders were present in the park-littering situation (high
personal implications). Less social control was shown by the groups of participants in
the graf¬ti situation (low personal implications).

Stage 4: Deciding How to Help
The fourth stage of the ¬ve-stage model of helping is deciding how to help. In the
staged rape study, for example, subjects had a choice of directly intervening to stop the
rape or aiding the victim by notifying the police (Harari, Harari, & White, 1985). What
in¬‚uences decisions like this?
There is considerable support for the notion that people who feel competent, who
have the necessary skills, are more likely to help than those who feel they lack such
competence. In a study in which subjects were exposed to a staged arterial bleeding
emergency, the likelihood of providing effective help was determined only by the exper-
tise of the subjects (some had Red Cross training; Shotland & Heinhold, 1985).
There are two reasons why greater competence may lead to more helping. First, feel-
ings of competence increase con¬dence in oneʼs ability to help and to know what ought
to be done (Cramer, McMaster, Bartell, & Dragna, 1988). Second, feelings of compe-
tence increase sensitivity to the needs of others and empathy toward victims (Barnett,
Thompson, & P¬efer, 1985). People who feel like leaders are probably also more likely
to help because they feel more con¬dent about being able to help successfully.
Many emergencies, however, do not require any special training or competence.
Irene Opdyke had no more competence in rescuing Jews than anyone else in Ternopol.
In the Genovese case, a simple telephone call to the police was all that was needed.
Clearly, no special competence was required.

Stage 5: Implementing the Decision to Help
Having passed through these four stages, a person may still choose not to intervene. To
understand why, imagine that as you drive to campus, you see a fellow student standing
next to his obviously disabled car. Do you stop and offer to help? Perhaps you are late
for your next class and feel that you do not have the time. Perhaps you are not sure it is
safe to stop on the side of the highway. Or perhaps the student strikes you as somehow
undeserving of help (Bickman & Kamzan, 1973). Or perhaps the place where the help
is needed is noisy (Moser, 1988). These and other considerations in¬‚uence your deci-
sion whether to help.

Assessing Rewards and Costs for Helping
Social psychologists have found that peopleʼs evaluation of the rewards and costs
involved in helping affect their decision to help or not to help. There are potential rewards
for helping (gratitude from the victim, monetary reward, recognition by peers) and for
not helping (avoiding potential danger, arriving for an appointment on time). Similarly,
there are costs for helping (possible injury, embarrassment, inconvenience) and for not
helping (loss of self-esteem). Generally, research indicates that the greater the cost of
helping, the less likely people are to help (Batson, OʼQuin, Fultz, & Vanderplas, 1983;
Darley & Batson, 1973; Piliavin & Piliavin, 1972; Piliavin, Piliavin, & Rodin, 1975).
In a study of this relationship, Darley and Batson (1973) told seminarians taking
part in an experiment at Princeton University that a high school group was visiting the
campus and had requested a seminarian speaker. Half the subjects were told they had
little time to get across campus to speak to the high school group, and the other half
were told they had plenty of time. Additionally, some subjects were asked to speak
Chapter 11 Prosocial Behavior and Altruism 419

about the meaning of the parable of the Good Samaritan. The seminarians then left the
building to give their talk, and lo and behold, while walking down a narrow lane, they
saw a young man collapse in front of them. What did they do?
Now, do you recall the story of the Good Samaritan? A traveler is set on by robbers
and left by the side of the road. A priest and a Levite, people holding important positions
in the clergy of the time, walked by swiftly without helping. But a Samaritan, passing
along the same road, stopped and helped. We might say that, for whatever reasons, helping
was too costly for the priest and the Levite but not too costly for the Samaritan.
What about the seminarians? The “costly” condition in this experiment was the
tight schedule: Stopping to help would make them late for their talk. Was helping too
costly for them? Yes, it was. Subjects who were in a hurry, even if they were thinking
about the story of the Good Samaritan, were less likely to stop and help than were sub-
jects who were not in a hurry.
In an attempt to “capture” the effects of various costs for helping and nonhelping,
Fritzsche, Finkelstein, and Penner (2000) had participants evaluate scenarios containing
three costs for helping (time required to help, the discomfort involved in helping, and
the urgency of the help) and three costs for not helping (victim responsibility, ability
to diffuse responsibility, and victim deservingness). Participants read the scenarios in
which these six variables were manipulated and were instructed to play the role of the
individual receiving the request for help. For each scenario, the participant indicated
his or her likelihood of helping the person making the request for help.
Fritzsche et al. (2000) found con¬rmation for the effects of cost on helping. In the
scenarios where costs for helping were high, participants expressed lower willingness to
help. Fritzsche et al. evaluated the importance of each of the six variables in determining
willingness to give help. They found that the cues varied in importance with respect to
helping. There was no signi¬cant gender difference in how the variables affected will-
ingness to help. The following list shows the importance of the six variables (in order
starting with the most important one):
1. Victim responsibility
2. Urgency of the help
3. Time required for help
4. Diffusion of responsibility
5. Discomfort involved in helping
6. Victimʼs deservingness
As is the case in decision-making research, there was a discrepancy between what
participants believed would be important in determining helping and what actually turned
out to be important. Participants believed that victim deservingness, time required to
render help, and ability to diffuse responsibility would be the most important factors
driving willingness to help. However, as you can see from the previous list, only one
of those factors was near the top of the list (time required for help). Finally, there was
a gender difference in this ¬nding. Males were more accurate than females in identify-
ing the importance of the variables.

The Effect of Mood on Helping
Likelihood of helping can even be affected by the bystanderʼs mood. The research of
Isen (1987) and her coworkers has shown that adults and children who are in a positive
mood are more likely to help others than people who are not. People who had found a
Social Psychology

dime in a phone booth in a shopping mall were more likely to pick up papers dropped by
a stranger than people who had not found a coin. Students who had gotten free cookies
in the library were more likely to volunteer to help someone and were less likely to
volunteer to annoy somebody else when asked to do so as part of an experiment.
Although positive mood is related to an increase in helping, it does not lead to more
helping if the person thinks that helping will destroy the good mood (Isen & Simmonds,
1978). Good moods seem to generate good thoughts about people, and this increases
helping. People in good moods also are less concerned with themselves and more likely
to be sensitive to other people, making them more aware of other peopleʼs needs and
therefore more likely to help (Isen, 1987).
Music, it is said, can soothe the wild beast. Can it also make you more likely to help?
North, Tarrent, and Hargreaves (2004) investigated this question. Participants in a gym
were exposed to either soothing or annoying music during their workout periods. After
the workout, participants were asked to help in a low-cost (sign a petition) or high-cost
(help distribute lea¬‚ets) situation. North et al. found that when the soothing music had
been played during the workout, participants were more likely to help in the high-cost
situation than if the annoying music had been played. There was no difference between
the two types of music for the low-cost helping situation.

Gratitude and Helping
Another factor that can affect helping is whether an individual received help when he
or she needed help. Gratitude is an emotional state that has three functions relating to
prosocial behavior (McCullough, Kilpatrick, Emmons, & Larson, 2001). First, grati-
tude acts as a sort of “moral barometer,” indicating a change in oneʼs state of mind after
receiving help. Second, gratitude can function as a “moral motivator,” impelling the
recipient of help to reciprocate to his or her benefactor or strangers. Third, gratitude
can serve as a “moral reinforcer.” When someone expresses gratitude after receiving
help, it increases the likelihood that the recipient of the gratitude will engage in proso-
cial behavior in the future. Taken together, these three functions suggest that gratitude
will increase helping. But does it?
The answer to this question is yes. A feeling of gratitude tends to enhance helping
(Bartlett & DeSteno, 2006; Tsang, 2006). In Bartlett and DeStenoʼs experiment, par-
ticipants were led to believe that they would be performing a group task with another
participant. Actually, the “other participant” was a confederate of the experimenter. The
real participant and confederate performed tasks on separate computers. In the “grati-
tude” condition, after completing a task and while waiting for scores to be displayed,
the confederate surreptitiously kicked the real participantʼs monitor plug out of a power
strip. The confederate then “helped” the participant by ¬nding and ¬xing the problem.
In the “amusement” condition, participants watched a brief, amusing video clip (to in-
duce positive affect unrelated to gratitude) after completing the task (the confederate
did not kick out the plug nor offer help). In the “neutral” condition the confederate did
not kick the plug out and only carried on a brief conversation with the real participant.
Sometime later the confederate approached the participant and asked the participant to
complete a long and tedious problem-solving survey. As shown in Figure 11.6, Bartlett
and DeSteno found that participants were more willing to help in the gratitude condi-
tion than in either the amusement or neutral conditions. Thus, it was the gratitude itself
and not just positive feelings that might be generated by receiving help that increased
helping. Bartlett and DeSteno conducted some follow-up studies to determine if grati-
tude merely activates the norm of reciprocity (you should help those who help you),
Chapter 11 Prosocial Behavior and Altruism 421



Mean Helping

Figure 11.6 Gratitude
and not just positive
5 emotions increase helping.
Gratitude seems to have
special qualities that
increase helping.
Gratitude Amusement Neutral
Based on data from Bartlett & DeSteno
Emotion Condition (2006)

thus leading to an increase in helping. Based on their results, Bartlett and DeSteno con-
cluded that it was, in fact, the feeling of gratitude experienced by the real participants
that increased helping, and not the norm of reciprocity.

Characteristics of the Victim
A decision to help (or not to help) also is affected by the victimʼs characteristics. For
example, males are more likely to help females than to help other males (Eagly &
Crowley, 1986; West, Whitney, & Schnedler, 1975). Females, on the other hand, are
equally likely to help male and female victims (Early & Crowley, 1986). Physically
attractive people are more likely to receive help than unattractive people (Benson,
Karabenick, & Lerner, 1976). In one study, a pregnant woman, whether alone or with
another woman, received more help than a nonpregnant woman or a facially dis¬gured


. 82
( 115 .)