A Self-regulation Perspective on Hidden-profile Problems: If–Then Planning to Review Information Improves Group Decisions

In hidden-profile (HP) problems, groups squander their potential to make superior decisions because members fail to capitalize on each other’s unique knowledge (unshared information). A new self-regulation perspective suggests that hindrances in goal striving (e.g., failing to seize action opportunities) contribute to this problem. Implementation intentions (if–then plans) are known to help deal with hindrances in goal striving; therefore, supporting decision goals with if–then plans should improve the impact of unshared information on group decisions. Indeed, in line with past research, control participants in two experiments rarely identified the best alternative despite monetary incentives and setting decision goals. In contrast, simply adding if–then plans to review advantages of the non-preferred alternatives before making the final decision significantly increased solution rates. Process manipulations (Experiment 1) and measures (Experiment 2) indicate that conceptualizing HP problems as a self-regulation challenge provides explanatory power beyond existing accounts. Copyright © 2014 John Wiley & Sons, Ltd. key words decision making; self-regulation; implementation intentions; hidden-profile paradigm; group performance Suboptimal decisions are costly as they squander performance potential. Improving decision making is thus a key interest to researchers and practitioners (Milkman, Chugh, & Bazerman, 2009). Groups can make more informed decisions than individuals when group members bring their unique knowledge to the table and consider it jointly. However, research from the past 25 years suggests that groups frequently fail to realize this potential (review and meta-analysis by Lu, Yuan, & McLeod, 2012; Stasser & Titus, 2003). Even when groups are committed to making an informed decision, they consistently focus on what all the group members know (shared information) rather than capitalizing on individual, unique knowledge (unshared information). When unshared information is necessary to identify the best decision alternative (i.e., in hidden-profile [HP] problems, Stasser & Titus, 1985), groups thus squander their potential to make decisions superior to those made by each group member individually (Stasser & Titus, 2003; Wittenbaum & Park, 2001). In the present paper, we focus on the challenges of considering mentioned information and suggest a new selfregulation perspective on HP problems. Despite their commitment to the goal of making an informed decision, group members might fail to act on this goal because of hindrances during goal striving (e.g., missing the opportunity to review relevant information before making the final decision). Forming implementation intentions (Gollwitzer, 1999) that plan out in advance when, where, and how to act in an if–then format (e.g., “If situation y occurs, then I will show the goal-directed response z!”) is known to facilitate goal attainment (Gollwitzer & Sheeran, 2006), even in the presence of hindrances. Accordingly, we asked whether forming if–then plans to review the advantages of the non-preferred alternatives in support of the goal to make an informed decision helps groups consider unshared information that has already been mentioned and to make more informed decisions. A self-regulation perspective: challenges to considering unshared information Groups commonly possess more information than single individuals, and thus, group decisions are highly informed when group members capitalize on their unique knowledge (unshared information; Lu et al., 2012; Mesmer-Magnus & DeChurch, 2009). Unfortunately, research suggests that groups neglect unshared information even when it comes up during the discussion (Gigone & Hastie, 1993, 1997; Mojzisch & Schulz-Hardt, 2010) and instead focus on what everybody already knows (i.e., shared information; Wittenbaum & Park, 2001). This leads to suboptimal decisions when unshared information is key to finding the best alternative (i.e., in HP situations). Indeed, this problem persists even when groups are explicitly told that they have to consider all the available information to find the optimal decision alternative (Stasser, Stewart, & Wittenbaum, 1995; Stasser, Vaughan, & Stewart, 2000) and when group members receive monetary incentives for optimal group decisions and therefore should be highly motivated (Greitemeyer, SchulzHardt, Brodbeck, & Frey, 2006; Lightle, Kagel, & Arkes, 2009). From a rational-economic perspective, it seems counterintuitive that groups fail to solve HP problems even when good decisions are incentivized: Why would one fail to attain a goal one is strongly committed to? Although this question has not been addressed in the context of HP problems, Lewin’s psychology of action perspective (Lewin, Dembo, Festinger, & Sears, 1944) suggests that setting strong goals *Correspondence to: J. Lukas Thürmer, Department of Psychology, University of Konstanz, 78457 Konstanz, Germany. E-mail: [email protected] Copyright © 2014 John Wiley & Sons, Ltd. Journal of Behavioral Decision Making, J. Behav. Dec. Making, 28: 101–113 (2015) Published online 2 September 2014 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/bdm.1832 is only the first step toward their attainment. The second step is to implement goal-directed actions and responses during goal striving. Thus, humans do not attain goals they are highly committed to when the implementation of goaldirected actions and responses fails. Indeed, meta-analytic findings suggest that a medium-to-large change in the strength of a goal (d= 0.66) only leads to a small-to-medium change in the intended behavior (d = 0.36, Webb & Sheeran, 2006), and this gap is mainly due to people who have a strong intention but fail to act (Sheeran, 2002). A common hindrance to implementing goal-directed actions and responses is that people fail to get started with acting toward their goal (Gollwitzer & Sheeran, 2006). Reasons for this failure include having detrimental routines, failing to seize opportunities, and experiencing a high cognitive load. First, behavioral routines are difficult to break when they have been applied repeatedly or have strong intuitive appeal. Although routines are an advantage in stable environments, they become a problem when they cannot be applied to a new problem at hand (Aarts, Dijksterhuis, & Midden, 1999; see Betsch & Haberstroh, 2005, for routines in decision making). Second, many goal-directed behaviors have to be performed within a certain time frame, for instance, when an action opportunity presents itself only for a short period or when one has to wait for it to occur. It is easy to miss such opportunities because one may fail to act in time (Prestwich et al., 2005). Lastly, it is difficult to act when one is “preoccupied with something else” or “distracted” (i.e., one is under a high cognitive load). Such a high cognitive load makes it difficult to initiate a goal-directed action deliberately (Brandstätter, Lengfelder, & Gollwitzer, 2001; Cohen & Gollwitzer, 2008). Past research suggests that these hindrances to goal striving are present in HP situations. First, research suggests that considering unshared information deviates from groups’ decision routine. Gigone and Hastie (1993) manipulated in repeated group decisions whether each quantitative information item about a student (e.g., percentage of attendance and SAT score) was available to everyone before the discussion (shared) or just available to some (unshared) and then assessed the impact that each information item had on the group decisions (i.e., the course grade assigned to each student). Whether information was shared or unshared was highly predictive of group members’ initial preferences and the grades groups assigned; however, groups took very little time to discuss each student, and whether information was brought up during the discussion had no impact. Apparently, groups commonly base their decisions on pre-discussion information and preferences (Gigone & Hastie, 1993; Mojzisch & Schulz-Hardt, 2010). Second, unshared information is usually mentioned later in the discussion than shared information (Larson, Christensen, Abbott, & Franz, 1996; Larson, Christensen, Franz, & Abbott, 1998), and one therefore has to wait until it is on the table and then seize the opportunity to consider it before the final decision is made. Thus, merely being committed to considering relevant information might not suffice; one might actually have to plan when to review it. The finding that unshared information has a greater impact when discussions are structured rather than unstructured supports this idea (Mesmer-Magnus & DeChurch, 2009). Third, when group discussions cover a lot of information, they create a high cognitive load, which makes it difficult to pursue the goal of making an informed decision. Stasser and Titus’ (1987) observation supports this reasoning. They manipulated the information load: Groups either received 12 (low load) or 24 (high load) different information items on three decision alternatives before their discussion. After the discussion, low-load groups recalled more available information than high-load groups, suggesting that they discussed more information. In sum, our self-regulation perspective suggests that group members facing HP problems might be highly motivated and have the goal to consider all crucial information but fail to do so because of problems during goal striving: Failing to interrupt routines and to seize opportunities and experiencing a high cognitive load are all known to hinder acting on one’s goals even when one is highly committed (Gollwitzer & Sheeran, 2006). In order to improve decisions in HP situations, one therefore has to apply strategies that are known to support goal attainment when such hindrances are present. Planning to review relevant information Research from the past 20 years suggests that forming implementation intentions (Gollwitzer, 1993, 1999, in press) supports goal attainment. In an implementation intention, one plans out in advance when, where, and how one wants to act on one’s goal in an if–then format (e.g., “And if situation Y occurs, then I will show response Z!”). Implementation intentions promote goal attainment by two related processes: First, forming an implementation intention entails selecting a future action opportunity (if-part). The mental representation of this situation becomes highly activated, and one thus easily recognizes it once it presents itself (Achtziger, Bayer, & Gollwitzer, 2012; Parks-Stamm, Gollwitzer, & Oettingen, 2007, Study 1; Wieber & Sassenberg, 2006). Second, implementation intentions create a strong situation–response link: Once the situation specified in the if-part occurs, it triggers the response specified in the then-part immediately and efficiently (Brandstätter et al., 2001; Parks-Stamm et al., 2007, Study 2; Webb & Sheeran, 2007) without requiring further conscious intent (Bayer, Achtziger, Gollwitzer, & Moskowitz, 2009). Implementation intentions have been shown to help overcome the most common challenges of initiating goal-directed action. First, implementation intentions facilitate interrupting routine behavior in order to achieve one’s goal. In one study, Aarts et al. (1999) promised participants a voucher that was to be collected on the way to the cafeteria, which required taking an unusual turn. Some participants furnished this goal with an implementation intention. After a distracting filler task, participants were asked to go to the cafeteria for another study. Whereas only half of the control participants were successful in interrupting this routine walk to obtain the Implementation intentions do not impact the commitment toward a set goal (motivation) but support goal attainment by these two processes (Gollwitzer & Sheeran, 2006; Webb & Sheeran, 2008). 102 Journal of Behavioral Decision Making Copyright © 2014 John Wiley & Sons, Ltd. J. Behav. Dec. Making, 28, 101–113 (2015)