Social Value Orientation Moderates the Effects of Intuition versus Reflection on Responses to Unfair Ultimatum Offers

We investigated whether social value orientation (SVO) moderates the effects of intuitive versus reflective information processing on responses to unfair offers. We measured SVO one week prior to an ultimatum game experiment in which participants had to accept or reject a series of 10 ultimatum offers including very low (unfair) ones. Before making these decisions, participants mentally contrasted their individual goals with the obstacle of pondering at length or acting in a hasty way; then they made the plan to adopt an intuitive or a reflective mode of processing (intuitive and reflective condition, respectively), or made no such plans (control condition). Participants with rather high (prosocial) SVO scores were more likely to accept unfair offers in the reflective than the intuitive condition. This effect also evinced for a subset of selfish individuals; however, the majority with rather low (selfish) scores made similar decisions in both conditions. This pattern of results suggests that SVO moderates the effects of intuitive versus reflective modes of processing on responses to low ultimatum offers. Copyright © 2016 John Wiley & Sons, Ltd. key words ultimatum game; social value orientation (SVO); dual-process models; fairness; intuition versus reflection Many people dislike being treated in an unfair manner, and they are willing to spend resources to punish those who treat them unfairly (Camerer & Thaler, 1995; Fehr, Fischbacher, & Gächter, 2002; Fehr & Gächter, 2000; Kahneman, Knetsch, & Thaler, 1986; Rabin, 1993). This preference has often been analyzed with the ultimatum game (Güth, Schmittberger, & Schwarze, 1982; for review see Güth & Kocher, 2014). In this two-player game, the proposer receives a certain amount of money and offers the responder an allocation of this amount. If the responder accepts the offer, the amount is allocated as proposed; otherwise both players receive no money. Assuming common knowledge of rationality and selfishness, responders will accept any positive offer and proposers will make the smallest possible offer. In contradiction to this account, however, proposers commonly make offers in the range of 40–50% of the money, while many responders reject low offers below 20% (review by Camerer, 2003). The large offers made by proposers can at least partially be explained by strategic considerations (Bolton & Zwick, 1995), but responder rejections of low offers are purely non-strategic and hence interpreted by most researchers as indicating a preference for being treated fairly (Fehr & Gächter, 2000). In recent years, it has been debated whether and how the way people process information in the ultimatum game affects the decision to accept or reject unfair offers. This question lies at the heart of dual-process models which distinguish between two modes of information processing, an intuitive and a reflective one, that govern decision making (J. St. B. T. Evans, 2008; Haidt, 2001; Kahneman, 2011; Strack & Deutsch, 2004; Weber & Johnson, 2009). The intuitive mode is assumed to concert processes that are fast, efficient, and affect-based, permitting swift and effortless responses. The reflective mode, in contrast, is thought to rely on time and cognitive resources, affording prudent and considered decisions. But how does adopting one or the other mode of processing affect responses to unfair offers as observed in the ultimatum game? Are people more likely to accept unfair offers when they reflect upon their decisions, or when they rely on their intuition? A growing body of literature addresses this question and consistently reports that adopting different modes of processing is consequential for the decision to accept or reject unfair offers. However, the nature of these consequences is still an open question. EFFECTS OF ADOPTING INTUITIVE VERSUS REFLECTIVE MODES OF PROCESSING ON RESPONSES TO UNFAIR ULTIMATUM OFFERS On the one hand, there is considerable research suggesting that low acceptance rates for unfair offers result from negative affect (e.g., anger and disgust) or scarce cognitive resources, both characteristic of an intuitive mode of processing. For instance, acceptance decisions have been observed to correlate negatively with self-reported feelings of anger (Pillutla & Murnighan, 1996) as well as with activity in brain areas that are associated with intuitive processing (e.g., *Correspondence to: Maik Bieleke, Department of Psychology, University of Konstanz, 78457Konstanz, Germany. E-mail [email protected] Copyright © 2016 John Wiley & Sons, Ltd. Journal of Behavioral Decision Making, J. Behav. Dec. Making, 30: 569–581 (2017) Published online 3 August 2016 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/bdm.1975 Sanfey, Rilling, Aronson, Nystrom, & Cohen, 2003; Tabibnia, Satpute, & Lieberman, 2008). Speaking to the causal role of negative emotions for these effects, providing people with emotion-regulation strategies increases acceptance rates (Kirk, Gollwitzer, & Carnevale, 2011; van ’t Wout, Chang, & Sanfey, 2010), whereas depleting the resources required for self-regulation reduces them (Halali, Bereby-Meyer, & Meiran, 2013, Exp. 1). Additional supportive evidence comes from studies in which the cognitive resources available to participants were strained, tantamount to promoting intuitive modes of processing which, in contrast to reflective modes, do not require such resources. For instance, forcing people to make their decisions in short time resulted in lower acceptance rates of unfair offers compared with imposing rather loose time constraints (Sutter, Kocher, & Strauß, 2003), whereas people obliged to pause before making their decision exhibit higher acceptance rates compared with situations without such a delay (Grimm & Mengel, 2011; Neo, Yu, Weber, & Gonzalez, 2013). Taken together, these studies yield profound support for the assumption that adopting a reflective mode of processing increases the likelihood of accepting unfair offers, as compared with an intuitive mode. Interestingly, there are also studies with opposite results, demonstrating that unfair offers are more likely to be accepted in an intuitive than a reflective mode of processing. For instance, Knoch, Gianotti, Baumgartner, and Fehr (2010) have shown that baseline activity in prefrontal cortex areas is negatively correlated with acceptance rates for unfair offers. As lower baseline activity in these areas is commonly associated with less reflective processing (Miller & Cohen, 2001), this finding suggests that people intuitively act in a selfish manner and accept unfair offers. If the relation is causal, derogation of the prefrontal cortex should lead to an increase in acceptance rates. This hypothesis was supported by applying repeated transcranial magnetic stimulation (rTMS) to disrupt the prefrontal cortex, causing responders to accept more unfair offers (Knoch et al., 2008; Knoch, Pascual-Leone, Meyer, Treyer, & Fehr, 2006; van ’t Wout, Kahn, Sanfey, & Aleman, 2005). Although physiological studies using rTMS are commonly considered a valuable complement in research on social decision making processes (e.g., Rilling & Sanfey, 2011), these results should be taken with a grain of salt. rTMS provides researchers with strong experimental control but in turn lacks ecological validity, compared with other processing manipulations (e.g., cognitive load). Consequently, it is not clear whether impeded prefrontal cortex activity can be directly equated with intuitive processing, or rather reflects a mere lack of reflective processing. It is thus important to note that the findings from rTMS studies have been replicated using more common manipulations; once by depleting selfregulatory resources (Achziger, Alós-Ferrer, & Wagner, 2016) and once again by instructing participants to make quick decisions rather than thoughtful ones (Hochman, Ayal, & Ariely, 2015). Both manipulations promoted an intuitive over a reflective mode of processing in a way that resembles everyday contexts (i.e., being exhausted and intending to use one processing mode over the other) and rendered participants more likely to accept unfair offers. At the bottom line, all of these studies share the observation that adopting an intuitive versus a reflective mode of processing has consequences for the decision to accept unfair offers in the ultimatum game. Research linking ultimatum rejections to intuitive processing has so far generated more support in terms of the quantity and diversity of experimental evidence (e.g., ego depletion, emotion regulation, and time pressure) than research linking them to reflective processing, especially because the latter is still primarily based on evidence from rTMS studies. Yet, none of the hypotheses can be clearly rejected at this point, leaving the nature of intuitive versus reflective processing consequences on responder decisions unclear. How can we move on and address such inconsistent findings? A promising approach is to identify moderators of intuitive versus reflective processing effects on responses to unfair offers. THE MODERATING ROLE OF SOCIAL VALUE ORIENTATION In the present research, we turned to the concept of social value orientation (SVO) (Messick & McClintock, 1968; Murphy & Ackermann, 2014; van Lange, 1999), a simple measure of prosociality that is likely related to the decision to accept or reject unfair offers in an ultimatum game (see below). SVO captures preferences for allocating resources between oneself and another person, and two main types of preferences are commonly distinguished (Au & Kwong, 2004; Bogaert, Boone, & Declerck, 2008). Individuals with cooperative (prosocial) preferences are willing to sacrifice their own resources to establish equal allocations and/or to maximize the mutual benefit. Individuals with individualistic (selfish) preferences, in contrast, solely focus on their personal benefits, trying to maximize their own resources and largely ignoring consequences for others. Besides these two main types, there are also small groups of people with either competitive preferences (trying to obtain more resources than the other person, even if this comes at an own cost) or altruistic preferences (trying to maximize the resources of the other, even when this requires to forfeit own benefits). The preferences reflected by SVO have been associated with patterns of social interactions from early childhood to old age (e.g., van Lange, Otten, De Bruin, & Joireman, 1997), and they are ubiquitous in everyday life. For instance, differences in SVO govern pro-environmental behavior (Gärling, Fujii, Gärling, & Jakobsson, 2003), political ideologies (van Lange, Bekkers, Chirumbolo, & Leone, 2012), the willingness to sacrifice in close relationships (van Lange, Agnew, Harinck, & Steemers, 1997), the readiness to help others (van Lange, Schippers, & Balliet, 2011), and the generosity of charitable donations (Bekkers, 2007). SVO is rather stable over time (Bogaert et al., 2008; Murphy, Ackermann, & Handgraaf, 2011), and has consistently been shown to predict social decisions in experiments. In particular, prosocial individuals are more likely to cooperate in social dilemmas than selfish individuals (reviews by Balliet, Parks, & Joireman, 2009; Bogaert et al., 2008; van Lange, Joireman, Parks, & Van Dijk, 2013). 570 Journal of Behavioral Decision Making Copyright © 2016 John Wiley & Sons, Ltd. J. Behav. Dec. Making, 30, 569–581 (2017)