The Psychology of Helping

Theory and research on altruism attempt to illuminate a fundamental characteristic of human nature. Philosophers throughout the ages have debated whether humans actually intend to perform actions beneficial to others and costly to themselves without any clear resolution (Cialdini et al., 1997). It is notable that “the golden rule,” i.e. “do unto others as you would have them do unto you,” is found in many systems of proscriptive norms, such as Christianity, Judaism, Hinduism, Taoism and Confucianism (Shroeder, Penner, Dovidio & Piliavin, 1995). This would seem to imply that helping others when the need arises is an action encouraged by the rules of society.

Psychologists have studied helping from many different perspectives. For example, Latane and Darley (1970) have developed a five-step cognitive model of bystander intervention. These steps consist of: (a) noticing the event, (b) interpreting the event as requiring help, (c) assuming personal responsibility, (d) choosing a way to help, and (e) implementing the decision. This model has been shown to be applicable not only to emergencies, but also to actions such as preventing someone from drunk driving and deciding whether to donate a kidney to a relative (Darley, Latane, Batson, Ferrari & Leippe, 1994). Behaviorists have demonstrated that helping behaviors can be increased by direct reinforcement and modeling, and social psychologists have shown that helping is more likely to occur when the rewards of helping outweigh the costs (Kenrick, Neuberg, & Cialdini, 1999). Psychologists have also uncovered characteristics of the target that increase the chances of helping, such as: attraction based on physical appearance, friendly behavior or personal qualities, and similar racial characteristics (Shroeder et al., 1995).

The Empathy-Altruism Hypothesis

Helping behaviors that incur more costs than benefits to the donor are considered to be altruistic. Some theorists have attempted to extend the definition of altruism to include helpful actions with no cost to the helping individual (e.g., Batson, 1997). The debate continues among social psychologists over whether “true altruism” exists in reality. One side of the debate supports psychological egoism, the thesis that people try to act in ways that benefit themselves. On the other hand, Batson and Shaw (1991) have challenged this assumption that the motivation for all intentional action is intended to benefit the self. Although altruist advocates admit that human motivation is frequently for self-benefit, they see the need for a pluralistic explanation of helping behaviors that includes both altruism and egoism. Batson et al.’s (1997) empathy-altruism hypothesis proposes that truly altruistic motivation can be evoked by empathic concern towards another person for whom the benefit is directed.

Actions based solely on the motivation to benefit another are proposed to result from a series of cognitive events (Batson and Shaw, 1991). In the enabling stage, the observer takes the perspective of the needy target. Batson, Turk, Shaw and Klein (1995) hold that perspective taking can be stimulated by perceived similarity between oneself and the other, by instructions to take the other’s perspective, or by an attachment such as kinship, friendship, or prior contact. This leads to an emotional response of empathic concern, feelings of sympathy, warmth, tenderness, and compassion. Feeling this empathic concern for the other results in a desire to improve the other’s welfare, rather than one’s own welfare (See Figure 1).

Studies supporting the empathy-altruism hypothesis have systematically varied whether individuals can only obtain egoistic goals by helping, or whether they can escape from the situation and obtain the egoistic goals without helping (Batson & Shaw, 1991). Aversive arousal reduction has historically been the most popular egoistic explanation for helping behaviors, with arguments proposed by philosophers such as Aquinas and Hobbes (Batson, 1991). Egoistic explanations of helping behaviors include the benefits of receiving public praise or escaping guilt and shame (one form of aversive arousal reduction) for not helping.

Figure 1. Batson et al.’s model of altruism. From D. Kenrick, S. Neuberg, & R. Cialdini, Social psychology: Unraveling the mystery (p. 344). Copyright 1999. All rights reserved. Reprinted by permission of the author and Allyn & Bacon.

In response to these explanations, Batson and Shaw (1991) conducted a study in which a co-worker, who was actually a confederate, was said to be receiving electric shocks. Participants were given the opportunity to take the place of the co-worker receiving the shock, and half were also allowed to stop watching the situation. It was presumed that individuals could reduce their aversive arousal levels by either helping or escaping, and that those who experienced high empathy would chose the helping option. The authors found that although individuals experiencing a predominance of personal distress did choose to escape, those experiencing a predominance of empathy did not try to escape. The authors concluded that at the empathy-altruism hypothesis was supported for at least a portion of the sample.

To examine the possibility that an egoistic motive of escaping public shame for not helping influenced participants to help, the authors manipulated whether the experimenter and target knew the participant’s choice. This manipulation did not affect the outcome of the decision, leading Batson & Shaw (1991) to suggest that obligation to help learned through socialization was not an influence. Because participants still chose to help at the same rate, even when the experimenter supposedly did not know of their decision, the authors concluded that empathy was the driving force behind helping decisions in some participants. The authors also concluded that these results suggested that obligation to help learned through socialization was not an influence (Batson & Shaw, 1991).

The conclusion that helping norms learned through socialization were not responsible for the helping actions assumes that these norms are not internalized within the individuals. It is possible that norms, which describe socially acceptable behaviors, are internalized as personal values, and thus socialized personal values may influence helping intentions independent of empathy. It is also possible that individuals may act in a way they see as ethical, even when they do not experience a change in affect. Batson & Shaw’s (1991) experiment actually demonstrated that compliance with respected others who may hold helping norms was not responsible for the helping actions. Other possible egoistic motives that researchers have attempted to erode with empirical research include the relief of sadness (Batson et al., 1989), the desire to make oneself happy (Batson et al., 1991), and bolstering of self-concept (Batson et al., 1988).

Challenges to the Empathy-Altruism Hypothesis

Although some theorists argue that the experience of empathy toward others leads to altruistic actions (Batson et al., 1997), other theorists believe that taking the viewpoint of the other leads to a greater sense of self-other overlap (Cialdini et al., 1997). Helping others under these conditions would not be selfless, but instead would be directed towards the self, because it leads to a more favorable mental state. Batson et al. (1997) challenged this hypothesis with two experiments discounting the influence of identification of the self with the other. Female participants in one study listened to a tape in which they heard the story of a female college student whose parents were killed in a car accident. The student was described as being in a state of financial need because her parents did not have life insurance, requiring her to take care of two younger siblings. The authors manipulated whether the character was said to be attending the same university or an “arch rival university” (Batson et al., 1997, p. 499). The authors measured self-other overlap with Aron et al.’s (1992) IOS scale, similarity on 16 attribute ratings, and a one-item judgment of similarity at the end of the experiment. Participants indicated the degree to which they had experienced feelings related to empathy by using bipolar scales to rate the adjectives sympathetic, softhearted, warm, compassionate, tender, and moved. The experimenters instructed some participants to take an objective perspective and not to get caught up in how the target feels, others were instructed to imagine how the target feels and how these events have affected her life.

Batson et al. (1997) reported a significant effect for the level of perspective taking on empathy, although this effect was weaker than previous findings. In one study, the authors reported no significant effect for group membership on empathy, however this may have been an artifact of low statistical power due to the small sample size (N = 40). The level of empathy experienced was found to have a significant effect on the number of hours volunteered to help the target, but it only approached significance for whether the participant volunteered to help. In a second study with a similar format, empathy did have a significant effect on whether participants volunteered to help. The authors report that path modeling of data in each of these studies indicated that the empathy-helping model could not be explained by similarity and that freeing the path between the IOS scale and helping did not significantly improve the model's goodness of fit (the three measures of self-other overlap did not correlate very highly, rs ranged from -.03 to .20, so they were analyzed separately). However, the small sample sizes used for these models limits confidence in the authors conclusions.

An alternative viewpoint is offered by Cialdini et al. (1997), who advocate for the primacy of self-other overlap, or “oneness,” an indicator of people’s sense of interpersonal interconnectedness in intentions for helping. Aron, Aron, & Smollan (1992) point to McAdams’s (1988) summary of the intimacy literature, which argued that most definitions of intimacy “converge on the central idea of sharing that which is inmost with others” (p. 18). Aron et al. (1992) state that the experience of overlap between oneself and another in a close relationship has been described by William James, Carl Jung, Abraham Maslow, and also by contemporary social cognition researchers.

Cialdini et al. (1997) present the results of three studies, including a path model indicating that empathic concern affected helping only through its relation to perceived oneness. The authors suggest that empathic concern is an emotional signal of oneness and that empathy per se at best leads to superficial helping. In one study, participants were told to focus on a target character that was just evicted from his or her apartment. The authors manipulated relationship closeness by instructing participants to think of either; a near-stranger, an acquaintance, a good friend, or a family member. Participants rated the amount of sadness, personal distress, empathic concern, and oneness they were feeling toward the target individual. Relationship closeness had a significant effect on empathic concern, oneness, and willingness to help. However, Cialdini et al.’s (1997) categories confound social proximity with kinship. Although kin may benefit from kin selecting effects, participants may not necessarily feel as socially close to their family members as they do to a close friend.

In a hierarchical regression, oneness was a strong predictor of helping (accounting for 30% of the variance) and reduced the contribution of empathic concern to nonsignificance. Cialdini et al. (1997) used a similar scenario in a second study, in this case, the target character died and left two children without a home. Participants were given a list of helping options, ranging from doing nothing to raising the children as your own. This second study replicated the findings of the first study.

A third study used a 4 x 3 factorial design to combine the relationship closeness levels from the first two studies with three levels of severity of need (Cialdini et al., 1997). The lowest level of need was requiring aid in making a phone call, the middle level of need was the eviction scenario from the first study, and the highest level of need was the orphan scenario from the second study. This third study also included measures of empathic concern, sadness, personal distress, and oneness as mediators of helping intentions. Main effects for relationship closeness and severity of need were found, along with an interaction between relationship closeness and severity of need. The effect of relationship closeness was stronger for higher need situations.

Figure 2. Cialdini et al.’s path model of helping. From ``Reinterpretinting the empathy-altruism relationship: When one into one equals oneness.'' By R. Cialdini, S. Brown, B. Lewis, C. Luce, & S. Neuberg. In The Journal of Personality and Social Psychology (Volume 73, p.489) Copyright 1997 by The American Psychological Association. Reprinted with permission of the author.

Using hierarchical regression, the authors found that personal distress, sadness, and situational egoistic motives reduced the impact of empathy, and the introduction of oneness further reduced the effect of empathy to nonsignificance for all three levels of need. The authors reported the best-fitting path model (see Figure 2), which indicated that oneness had a mediational role in helping and reduced the mediational role of empathic concern to nonsignificance. There is some concern over the appropriateness of this analysis. The psychological mediators are both influenced by exogenous factors and allowed to covary with each other, which is not allowed in a path model of measured variables (see Kline, 1999).