Quid pro quo: The effect of individuals' exchange orientation on prosocial behavior and the moderating role of mortality salience

Abstract

Individuals high in exchange orientation expect immediate and comparable rewards in order to establish exchange equality after they have provided rewards for others. Therefore, such individuals should be less likely than individuals low in exchange orientation to behave prosocially because doing such usually leads to exchange inequality (i.e., postponement of reciprocal expectations). However, research on terror management theory has indicated that an adherence to prosocial norms increases after mortality salience, especially in situations where those norms are prescribed and cognitively focused. Based on this, we predicted and found evidence that when participants who were high (vs. low) in exchange orientation were directly asked in a face-to-face interaction to donate their participation payment to a charity, they were less likely to donate unless they had first been reminded of their own death.

Keywords::

• Exchange orientation
• Prosocial behavior
• Terror management theory
• Mortality salience

Research on exchange orientation suggests that individuals who are high in exchange orientation maintain exchange equality (e.g., Murstein, Cerreto, & MacDonald, 1977) and, further, are negatively affected when inequality is experienced (e.g., Buunk, Doosje, Jans, & Hopstaken, 1993; Buunk & Schaufeli, 1999). With this study, we investigated the effect of dispositional exchange orientation on prosocial behavior because acting prosocially usually implies exchange inequality (i.e., postponement of reciprocal expectations). In this case, we assumed that individuals who are high in exchange orientation act less prosocially due to their motivation to maintain exchange equality. Additionally, we examined whether being confronted with one's own death would moderate the assumed effect of exchange orientation on prosocial behavior.

Following the rules of social exchange plays an important role in social relationships (for an overview, see Cropanzano & Mitchell, 2005). Reciprocity is probably the best-known rule of social exchange. Defined as a norm, it is assumed to be “one of the universal ‘principal components’ of moral codes” (Gouldner, 1960, p. 161) that exists in all known societies. It prescribes that people should support, and not injure, those who previously supported them (Gouldner, 1960; Uehara, 1995). Accordingly, a great deal of research in social psychology has supported the idea that performing a favor will lead to higher compliance toward a future request by the favor-doer (e.g., Cialdini, Green, & Rusch, 1992; Edlund, Sagarin, & Johnson, 2007; Regan, 1971; Whatley, Webster, Smith, & Rhodes, 1999). Furthermore, social exchange approaches (e.g., equity theory) basically assume that relationships are, in general, more satisfying and stable when rewards for each partner are perceived to be reciprocal and equal (Adams, 1965; LaGaipa, 1977). For many different types of relationships, including helping relationships, exchange inequality (i.e., being underbenefited as well as being overbenefited) was shown to lead to negative feelings (e.g., Buunk & Schaufeli, 1999; Hatfield & Sprecher, 1983; Smets, Visser, & Oort, 2004), such as feelings of obligation, indebtedness, fear, and uncertainty about being unable to repay the debt (Greenberg & Westcott, 1983). Rook (1987) suggested that the perception of giving more support than one receives leads to feelings of unfairness and resentment.

Though reciprocity is assumed to be a universal norm in general, research has shown that individuals vary in the degree to which they endorse the norm of reciprocity (e.g., Clark & Mills, 1979; Eisenberger, Lynch, Aselage, & Rohdieck, 2004; Murstein et al., 1977; Perugini, Gallucci, Presaghi, & Ercolani, 2003). Analogous to this, Murstein et al. (1977) introduced the concept of dispositional exchange orientation, suggesting that exchange-oriented individuals strongly care about direct reciprocity, expect immediate and comparable rewards when they have provided rewards for others, and feel uncomfortable when they receive favors that they cannot immediately reciprocate. This concept was shown to play a crucial role for satisfaction in relationships, as well as in health and well-being (Buunk & Prins, 1998; Buunk & Van Yperen, 1991; Buunk et al., 1993; Milardo & Murstein, 1979; Murstein, Wadlin, & Bond, 1987; Murstein et al., 1977). Buunk and Van Yperen (1991), for instance, found that for individuals high in exchange orientation, perceived equality was an important factor for marital satisfaction. In contrast, individuals low in exchange orientation (i.e., for whom the input–output ratio did not matter) overall were more satisfied with their relationship. Moreover, Buunk et al. (1993) found that individuals high (vs. low) in exchange orientation were more negatively affected (e.g., irritated, depressed, confused, nervous) when they experienced a lack of perceived reciprocity, that is, when they felt underbenefited (“You owe me something!”) or overbenefited (“I owe you something!”).

While research has investigated the idea that perceiving exchange inequality can lead to negative effect, the behavioral consequences of this effect in a prosocial context, for example, have barely been considered. To our knowledge, there is only one study by Bell, Abrahams, Clark, and Schlatter (1996) that has indicated that for individuals high in exchange orientation, the door-in-the-face strategy—a persuasion strategy based on the norm of reciprocity (Cialdini et al., 1975)—increased the compliance rate toward a request. They also found that participants in the control condition were less likely to donate when they were high (vs. low) in exchange orientation. However, this effect was not discussed theoretically. The case of prosocial behavior seems particularly interesting because it usually involves the temporary postponement of reciprocal expectations. Based on the assumption that individuals who are high in exchange orientation expect immediate and comparable rewards when they have provided rewards for others, and furthermore are negatively affected when they perceive exchange inequality, we stress the idea that those individuals are less likely to behave prosocially if it implies abstaining from their rewards (i.e., exchange inequality).

Following the rules of social exchange can also serve an anxiety-buffering function when confronted with one's own death (Schindler, Reinhard, & Stahlberg, 2012, 2013). According to terror management theory (TMT; Greenberg, Pyszczynski, & Solomon, 1986), cultural worldviews function as an anxiety buffer against the ever-present terror of potential death, by providing a meaningful, orderly conception of reality that contains a set of standards and values. By living up to those cultural standards, people believe that they are valuable beings in a meaningful reality. Based on this idea, the mortality salience (MS) hypothesis states that reminding people of their mortality increases their motivation to defend and bolster their own cultural worldview, resulting in derogating those who violate important cultural standards and supporting those who uphold them (for an overview, see Greenberg, Solomon, & Arndt, 2008).

Referring to norms of social exchange, research has recently found that MS increased adherence to the norm of reciprocity (Schindler et al., 2012, 2013). For example, Schindler et al. (2013) indicated that after having received a favor (i.e., espresso on the house), participants gave a higher tip to the server under MS. Furthermore, research has indicated that MS increases prosocial attitudes and behavior (Blackie & Cozzolino, 2011; Hirschberger, Ein-Dor, & Almakias, 2008; Joireman & Duell, 2005, 2007; Jonas, Schimel, Greenberg, & Pyszczynski, 2002). For example, Jonas et al. (2002) showed that participants reported more favorability toward charities when they were interviewed in close proximity to a funeral home (vs. several blocks away). Moreover, Joireman and Duell (2005) found that MS led proself individuals to endorse self-transcendent (i.e., prosocial) values, such as protecting the environment or helpfulness. However, this effect did not occur for prosocial individuals because these individuals already appreciated and practiced those values and behaviors. To explain the opposite effects of MS, such as increased aggression against worldview-threatening others (McGregor et al., 1998), or greedy acquisition (e.g., Arndt, Solomon, Kasser, & Sheldon, 2004; Kasser & Sheldon, 2000), research on TMT has suggested that situational salience is a crucial factor. In line with this idea, several studies have shown that priming prosocial norms increases prosocial behavior after MS (Galliot, Stillman, Schmeichel, Maner, & Plant, 2008; Jonas, Sullivan, & Greenberg, 2013; Jonas et al., 2008). Jonas et al. (2013), for example, recently found a fairness norm prime to increase generous behavior toward an anonymous person.

As reasoned above, in a situation of exchange inequality, individuals who are high in exchange orientation are expected to act less prosocially if doing so leads to exchange inequality. However, according to research on TMT, being confronted with their own death might increase their prosocial behavior, especially when prosocial norms are cognitively focused. Thus, although individuals high in exchange orientation, in general, are motivated to maintain exchange equality, we assume MS to increase motivation to act according to the salient norm. Accordingly, we expect that individuals high (vs. low) in exchange orientation are less likely to behave prosocially if doing so implies that they must abstain from their own rewards (i.e., exchange inequality)—but only until they are reminded of their own death. That is, after MS, exchange orientation is not expected to affect prosocial behavior, especially in a situation where prosocial norms are prescribed. For the current study, we assumed a donation scenario to constitute a situation where prosocial norms are of high cognitive focus, especially when the charity organization is explicitly introduced as being in support of disadvantaged children. Beyond that, research on cooperation in social dilemmas (e.g., Axelrod, 1984; Fox & Guyer, 1978) has suggested that cooperation decreases when decisions are made anonymously. Therefore, we assumed a direct face-to-face interaction to further increase prosocial norm salience by inducing normative pressure and social desirability to follow the prosocial request. Based on this, we hypothesized an interaction effect of participants' dispositional exchange orientation and experimental condition (mortality vs. control) on the donation of participation payment to a charity (i.e., prosocial behavior).

Method

Participants in this experiment included 67 students (39 female, 28 male; age 19–29 years, M = 21.64, SD = 2.00) recruited on the campus of a German university. Before having agreed to participate, participants were told that participation would take about 30 min and that they would be paid 5 Euros (approximately $6 US), assuming a balanced and justified relation between experiment duration (costs) and participation payment (benefit). Once participants came into the lab, the experimenter assigned them to a computer. The cover story was as follows: participants read that the experiment was about personality traits and information processing. After having filled out the demographic measures, the revised Exchange-Orientation Scale (Murstein et al., 1987) was used to assess participants' dispositional exchange orientation (e.g., “I don't like people who don't fulfill their obligations to me”; “I usually do not forget if I owe someone a favor or if someone owes me a favor”; “If I take a friend out to dinner, I expect him or her to do the same for me sometime”).1 Participants responded to all 19 items (α = .68) on a seven-point continuous scale ranging from 1 (not true for me) to 7 (very true for me).2 To avoid demand characteristics or priming effects of the measurement, participants were then asked to do several analogy tasks from the Cognitive Ability Test 4–12+R (Heller & Perleth, 2000), a German version of the CogAT by Thorndike and Hagen (1993), assessing the mental ability of students from Grade 4 to 12. Participants were not given any feedback on their performance. Afterwards, participants were randomly assigned to the MS or control condition. As is common in TMT experiments, participants in the MS condition answered two open-ended questions about death, whereas participants in the control condition answered two open-ended questions about dental pain (DP; see Rosenblatt, Greenberg, Solomon, Pyszczynski, & Lyon, 1989): “Please briefly describe the emotions that the thought of your own death arouses in you” and “Jot down, as specifically as you can, what you think will happen to you physically as you die and once you are physically dead.” The DP control treatment consisted of parallel questions with respect to the experience of DP. When using such explicit death primes, a distractor is necessary to elicit effects of distal defense (e.g., Arndt, Greenberg, & Cook, 2002). Therefore, as in most studies on TMT (see Burke, Martens, & Faucher, 2010), participants filled out 20 items of the Positive and Negative Affect Schedule (PANAS; Watson, Clark, & Tellegen, 1988). Then, the experiment ended and participants were told to enter the room next door to receive their participation payment. There, they received a 5 Euro bill by a confederate. Subsequently, the confederate introduced herself as a student assistant who was privately supporting the charity organization Big Brothers, Big Sisters. It was further pronounced that the organization involves a mentoring program for disadvantaged children. Then, participants were asked the following:

Big Brothers, Big Sisters regularly organizes events for children and their mentors, such as “Cook ‘n’ Save” classes, where children learn about healthy cooking and adequate nutrition. Since Big Brothers, Big Sisters is a volunteer-based program, donations are sorely needed. You could support Big Brothers, Big Sisters by sponsoring such a class for a child. Would you like to donate your 5 Euro to enable a child to participate in a “Cook ‘n’ Save” class?

In this scenario, participants were forced to choose between a donation of their full payment and no donation. Thus, the dependent measure (prosocial behavior) was binary.3 Participants who agreed with the request were asked to drop their 5 Euro bill into a donation box, which was actually given to Big Brothers, Big Sisters. After having made their decision, participants were thanked and fully debriefed.

Results

First, the data were checked for any effects of MS on the PANAS and, consistent with prior research on TMT, no effects were found (Fs <  1).4 To test our hypothesis, we used a logistic regression with dispositional exchange orientation, mortality condition, and the two-way interaction term (exchange orientation ×  mortality condition) to predict the probability of donation. We coded donation as 1 (n = 40) and no donation as 0 (n = 27). Exchange orientation as a predictor was standardized, and mortality condition was dummy coded, with MS as 1 and DP as 0. Exchange orientation was hypothesized to determine the probability of donation, depending on the mortality condition.

Mortality condition and exchange orientation were both entered simultaneously in a first analytical step (see Model 1, Table 1). The analysis revealed that exchange orientation significantly predicted the probability of donation, b = (.68, p = .017, odds ratio = .51, indicating that participants high in exchange orientation were less likely to donate. Furthermore, there was no significant effect of mortality condition on donation decision, b = .64, p = .230, odds ratio = 1.90. As can be seen in Table 1, this first model significantly predicted participants' donation decision, χ²(1, N = 67) = 8.63, p = .013, accounting for approximately 16% (Nagelkerke's R²) of the variance. In a second analytical step, the interaction term (exchange orientation ×  mortality condition) was entered into the model (see Model 2, Table 1). As hypothesized, the interaction term significantly predicted participants' donation decision, b = 1.25, p = .049, odds ratio = 3.50, increased the χ² value compared to Model 1 by 4.30 (p = .038), and resulted in a significant overall fit, χ²(1, N = 67) = 12.93, p <  .005, which accounted for approximately 24% (Nagelkerke's R²) of the variance.

Table 1 Logistic regression results of participants' donation choice as a function of exchange orientation and mortality condition (N = 67)

		Model 1^a				Model 2^b
Model χ^2 (df)		8.631 (2)				12.93 (3)
Statistical significance		.013				.005
Nagelkerke's R^2		.16				.24
Change in χ^2						4.30
Statistical significance of Δχ^2						.038
Parameter estimate results following stepwise modeling
		Parameter estimates				Model
Model	Parameter	b	SE	Wald's χ^2	p	Δχ^2	p
(1)	Exchange orientation	− 0.68	0.29	5.74	.017	8.63	.013
	Mortality condition	0.64	0.54	1.44	.230
(2)	Exchange orientation ×  mortality condition	1.25	0.63	3.89	.049	4.30	.038

a Model 1 consists of the two independent variables (Step 1).

b Model 2 consists of the two independent variables and the interaction (Step 2).

We probed the interaction by testing the effect of exchange orientation on probability of donation for the MS and DP conditions (Hayes & Matthes, 2009). The analysis revealed that exchange orientation significantly predicted the probability of donation in the DP condition, b = − 1.30, p = .007. Thus, in line with our assumption, the higher participants were in exchange orientation, the lower the probability of donating their participation payment (see Figure 1). However, exchange orientation was not a significant predictor in the MS condition, p = .92. We additionally probed the interaction by testing the effect of mortality condition on probability of donation for high (1 SD above the mean) and low (1 SD below the mean) exchange orientation. The analysis indicated that MS significantly increased the probability of donation for participants who were high in exchange orientation, b = 1.81, p = .027, odds ratio = 6.09, but not for participants who were low in exchange orientation, p = .424. In other words, participants under MS who reported a high exchange orientation were more likely to donate their participation payment than were participants in the DP condition. No effect of sex on probability of donation was found, b = .017, p = .888.

Figure 1 Donation probability of participation payment as a function of participants' exchange orientation and mortality condition.

Discussion

Results of the current study indicated that the probability of donating the participation payment decreases as exchange orientation increases. We interpret this finding as evidence for the hypothesis that individuals who are high in exchange orientation behave less prosocially compared to individuals who are low in exchange orientation because prosocial behavior usually involves exchange inequality. Specifically, in our case, acting prosocially meant abstaining from a justified reward. Although we can only speculate about the process, we assume that those who are high in exchange orientation tend to monitor the balance of benefits exchanged in order to ensure that equality is maintained (e.g., Murstein et al., 1987). Furthermore, such individuals were shown to be negatively affected when they experienced inequality (e.g., Buunk & Schaufeli, 1999; Buunk et al., 1993). Therefore, to maintain equality and to avoid negative effects, they acted in favor of exchange equality, and thus refrained from donating their participation fee. Literature on social exchange has suggested that immediate reciprocity is strongly expected among strangers and within one-time interactions rather than within close, communal relationships, such as friendships (e.g., Clark & Mills, 1979; Rook, 1987). Therefore, our findings are restricted to short-term relationships.

Regarding participants who were reminded of their own death, results indicated that dispositional exchange orientation is not a relevant factor in their donation decision: both types of participants (those who were low as well as those who were high in exchange orientation) were equally likely to donate their participation payment after MS. According to the idea that MS increases motivation to fulfill situationally prescribed norms, participants high in exchange orientation were more likely to donate after MS compared to the DP control condition. Participants who were low in exchange orientation showed no increased donation likelihood after MS. This might be due to a ceiling effect because the donation likelihood for these participants was already high in the DP control condition. According to the findings of Joireman and Duell (2005, 2007) which indicated that MS increased prosocial values of proself-oriented but not prosocial-oriented individuals, one could also speculate that participants who were low in exchange orientation were not affected by MS because they were already living up to culturally prescribed prosocial norms.

Given that MS was recently shown to increase adherence to the norm of reciprocity (Schindler et al., 2012, 2013), one could suggest that MS would cause the likelihood of individuals high in exchange orientation to donate their payment to charity to be less, due to an increased motivation to maintain exchange equality. However, although we did not manipulate norm salience in our experimental setting, we assumed the donation scenario to have constituted a situation where prosocial norms were of high cognitive focus because the charity organization was explicitly introduced as one in support of disadvantaged children.

According to research on cooperation in social dilemmas (e.g., Axelrod, 1984; Fox & Guyer, 1978) which suggested that cooperation decreases when decisions are made anonymously, we further assumed that direct face-to-face interaction is a crucial factor for our findings because it may have further increased prosocial norm salience by inducing normative pressure and social desirability to follow the prosocial request. Interestingly, Jonas et al. (2008, Study 1) found a proself prime to decrease reported willingness to support day-care facilities for children after MS. Moreover, Jonas et al. (2013, Study 1) found MS to decrease the amount of donated money toward an out-group-focused charity. In this study, participants were told to put their donation amount in an envelope and to drop the envelope in a box. So, in both studies, prosocial behavior was assessed anonymously. Thus, it remains an open question whether MS decreases prosocial behavior after a proself prime when the decision is made publicly. Following this, we believe that the case of anonymity in prosocial behavior is an important and promising, but so far disregarded, issue in TMT research.

In any case, based on our results, we speculate that although individuals who are high in exchange orientation are, in general, motivated to maintain exchange equality, being confronted with their own death increases their motivation to act in favor of the situationally prescribed prosocial norm.

Notes

¹ Murstein et al. (1987) published a revised version of their original Exchange-Orientation Scale (Murstein et al., 1977) to correct some weaknesses in construct validity.

² Although the scale does not directly capture immediate versus delayed reciprocity, individuals who were high in exchange orientation were shown to expect a more immediate reciprocity in exchanges (Milardo & Murstein, 1979; Murstein et al., 1977, 1987).

³ At our university, monetary compensation of 5 Euros is usually given in a 5 Euro banknote. Suspecting giving the payment in coins to increase participants' suspicion that the donation scenario is part of the experiment, we therefore decided to use this binary judgment as our dependent variable.

⁴ Because we used physical pain as a control group, the PANAS might not have been sensitive enough to pick up mood effects of the MS treatment. However, research on TMT has indicated that compared to neutral control groups (e.g., watching television), no effects on the PANAS occurred after MS (e.g., Greenberg, Arndt, Simon, Pyszczynski, & Solomon, 2000).