Order Effects in Moral Judgment

Order Effects in Moral Judgment Alex Wiegmann (Alex.Wiegmann@psych.uni-goettingen.de) Jonas Nagel (jnagel1@uni-goettingen.com) Stefan Mangold (smangol@gwdg.de) Department of Psychology, University of Gottingen, Germany Yasmina Okan(yokan@ugr.es) Department of Experimental Psychology, University of Granada Campus Universitario de la Cartuja s/n ,18071, Granada, Spain Abstract Explaining moral intuitions is one of the hot topics of recent cognitive sciences. In the present article we focus on a factor that attracted surprisingly little attention so far, namely the temporal order in which moral scenarios are presented. We argue that previous research points to a systematic pattern of order effects that has been overlooked until now: Only judgments of actions that are normally regarded as morally acceptable are affected by the order of presentation. Additionally, this is only the case for dilemmas immediately preceded by a scenario where the proposed action was judged as morally unacceptable. We conducted an experiment that confirmed this pattern and allowed us to analyze the individual level responses it was generated by. We argue that investigating order effects is necessary for approaching a complete descriptive moral theory. Keywords: moral intuitions; trolley dilemmas; order effects Introduction In the past decades, trolley dilemmas have been used extensively for testing philosophical and psychological theories of moral judgments. In the standard description of the trolley dilemma introduced by Philippa Foot (1967), an out-of-control train threatens to kill five people standing on its track. The only way to prevent this is to pull a switch that redirects the train onto a different track where it will kill only one person. In a modification of this scenario (Thomson, 1976), the only possibility to prevent the five people from being killed is to push a heavy person from a footbridge onto the track. This would stop the train but kill the heavy person. Numerous studies (e.g., Hauser, Cushman, Young, Jin, & Mikhail, 2007) have shown that given the same number of people being killed vs. saved, participants approve of acting in the first but not in the second scenario. Several competing descriptive theories explicate psychological principles supposed to underlie this pattern of moral intuitions (e.g. Greene et al., 2001; Hauser, 2006). However, surprisingly little is known about potential effects of the order in which several consecutive scenarios are presented. It is plausible to assume that consecutive scenarios will not be judged independently of each other: A principle or mechanism that is activated when a particular scenario is represented or evaluated might later be applied to a series of subsequent scenarios. However, only few studies have dealt with this issue so far (Petrinovich & O’Neill, 1996; Lanteri, Chelini, & Rizzello, 2008). Their results suggest that under certain circumstances moral judgments can indeed be transferred from one situation to another. If such order effects could be replicated systematically, this would have important implications for psychological theories aiming to explain patterns of moral reasoning at a descriptive level. Furthermore, relevant practical implications would arise both for methodological considerations inside the research laboratory (in terms of controlling for order effects when designing experiments) and for everyday judgments outside the lab. The present work has three main goals. First, we will provide the first comprehensive review of previous empirical research on order effects in moral judgments, and we will demonstrate that a systematic pattern of results has been overlooked so far. Second, we will empirically test the existence, extent and direction of order effects in reasoning about moral dilemmas. Finally, we will discuss the theoretical and practical implications of our findings, focusing on psychological theories of moral reasoning. Order Effects in Previous Research Speaking of order effects in moral judgment, there are at least two possible interpretations that could be labelled “within-scenario order effect” and “between-scenarios order effect”, respectively. The first type of effect results if the order in which information concerning one particular situation is presented affects judgment. If, for example, the task is to judge the permissibility of an action, and the results solely differ as a function of the particular sequence in which positive and negative consequences are presented, a “within-scenario order effect” occurs. Second, a judgment regarding an action in a particular scenario might be influenced by a judgment that had previously been made about a different scenario. To illustrate, consider two conditions in which a given scenario C is preceded by one of two different scenarios (A vs. B). Differences in judgments of the action scenario C between the two conditions would – all other things being equal – instantiate a “between-scenarios order effect”. The present research will focus on this second category of order effects.