466 Altruism and Prosocial Behavior
punishments received following helping (or not), but also the
relative rewards—the benefits minus the costs. You should
consider observational learning or modeling that comes from
watching the actions of others. You should consider self-
rewards. Much research has supported a social learning expla-
nation of prosocial behavior (for reviews, see Bandura, 1977;
Rushton, 1980). Integrating and coordinating social learning
principles, Cialdini, Baumann, and Kenrick (1981) proposed a
three-step developmental sequence: (a) In the young child
prosocial behavior is a product of material rewards and pun-
ishments; (b) in the preadolescent it is a product of social as
well as material rewards and punishments; and (c) in the ado-
lescent and adult it is a product of internalized self-reward, as
well as social and material rewards and punishments.
Mood Effects
Building on the idea that helping can be a basis for self-
reward, Cialdini, Darby, and Vincent (1973) proposed a
negative-state relief hypothesis:that adults are more likely to
help when they feel bad. The reason is that adults have
learned that they can reward themselves for helping and so
feel better.
Not only does helping have reward value for people who
feel bad, but it also seems rewarding for people who feel
good. Indeed, the effect is even clearer for good mood.
Across a range of studies (e.g., Isen & Levin, 1972; Weyant,
1978), people induced to feel good have been more likely to
give help to good causes.
What accounts for this pervasive reward value of helping
for people in a good mood? One possibility is a desire to
maintain the good mood. Seeing another person in need can
throw a wet blanket on a good mood, so one may help in
order to shed this blanket and maintain the mood (Wegener &
Petty, 1994). Isen, Shalker, Clark, and Karp (1978) suggested
a second possibility: Being in a good mood may bias one’s
memories about and attention to the positive and negative as-
pects of various activities, including helping. When in a good
mood, a person is more likely to recall and attend to positive
rather than negative aspects of life. Applied to helping, a
good mood makes people more likely to remember and
attend to the positive, rewarding features and less likely to
attend to the negative features, such as the costs involved.
General Assessment
Social learning theory finds itself in an awkward position in
contemporary social psychology. There seems little doubt
that the theory is in large measure correct. However, perhaps
because of its relatively straightforward explanation of
behavior, without the ironic twists and the revelations of sub-
tle faux pas for which cognitive explanations have become
renown, social learning theory generates little excitement.
The direct focus on behavior and reinforcement history
seems almost unpsychological in its lack of nuance. Even
with the added emphasis on self-reward, cognitive represen-
tation, self-regulation, and reciprocal determinism (Bandura,
1977, 1991), social learning theory seems bland. Still, were
one forced to choose a single theory to explain why people
do—and do not—act prosocially, social learning theory
should almost certainly be the choice. “As Einstein has em-
phasized, the goal is to account for the most facts with the
fewest principles” (Dollard & Miller, 1950, p. 6). Social
learning theory has probably come closer to this goal than has
any other theory in the history of social psychology.Tension ReductionTension reduction has long been a popular explanation of
why people help others in need, especially others in obvious
pain or distress. The general idea is that people find it upset-
ting to see another person suffer and that preferring not to be
upset, they relieve the other’s suffering.
Perhaps the best way to describe the relationship between
tension reduction, which is a form of motivation, and social
learning is to say that they are related by marriage. Social
learning can exist without tension reduction, as in the pure
operant theories descendant from Watson and Skinner.
Tension reduction can exist without social learning, as in
reactions to pain, extreme temperatures, hunger, thirst, and
other physiological needs. Yet social learning and tension
reduction lived together for many years in relative harmony,
housed within Hull’s (1943) general learning theory and its
descendants, including Dollard and Miller’s (1950) version
of social learning theory. In response to the current cognitive
zeitgeist, social learning theory has of late been less attached
to tension reduction, showing more interest in cognitive
processes (Bandura, 1977, 1991). Whether this philandering
is grounds for divorce is hard to say. In any case, tension
reduction has also been seen stepping out without operant
processes by its side, most notably in dissonance theory—at
least as originally conceived by Festinger (1957).
Why should the suffering of others upset someone? Most
straightforward is the answer proposed by J. A. Piliavin et al.
(1981), among others. They suggested that witnessing an-
other’s distress evokes vicarious distress that has much the
same character as the victim’s distress, and the witness is mo-
tivated to escape his or her own distress. One way to escape
is to help because helping terminates the stimulus causing the
distress. Of course, running away may enable the witness to