From moral to legal judgment: the influence of normative context in lawyers and other academics Stephan Schleim,1,2 Tade M. Spranger,3 Susanne Erk,2,4 and Henrik Walter2,4,5 1Department of Psychology, University of Groningen, Grote Kruisstraat 2/1, 9712 TS Groningen, The Netherlands, 2Division of Medical Psychology, University of Bonn, Sigmund-Freud-Strasse 25, 53105 Bonn, 3Institute of Science and Ethics, University of Bonn, Bonner Talweg 57, 53113 Bonn, 4Charite, �� Department of Psychiatry and Psychotherapy, Division for Mind and Brain Research, Campus Mitte, Universitaetsmedizin Berlin, Germany, and 5Department of Psychiatry and Psychotherapy, University of Bonn, Sigmund-Freud-Strasse 25, 53105 Bonn, Germany Various kinds of normative judgments are an integral part of everyday life. We extended the scrutiny of social cognitive neuroscience into the domain of legal decisions, investigating two groups, lawyers and other academics, during moral and legal decision-making. While we found activation of brain areas comprising the so-called ���moral brain��� in both conditions, there was stronger activation in the left dorsolateral prefrontal cortex and middle temporal gyrus particularly when subjects made legal decisions, suggesting that these were made in respect to more explicit rules and demanded more complex semantic processing. Comparing both groups, our data show that behaviorally lawyers conceived themselves as emotionally less involved during normative decision-making in general. A group condition interaction in the dorsal anterior cingulate cortex suggests a modulation of normative decision-making by attention based on subjects��� normative expertise. Keywords: legal decision-making moral decision-making neurolaw fMRI prefrontal cortex neuroethics INTRODUCTION Normative judgments are ubiquitous in everyday life. For example, judging people as tall or small, or as beautiful or unsightly, refers to respective norms. Besides these examples of normativity in a wide sense, there is one particularly strong understanding related to norms of right or wrong human conduct. One kind of such norms, namely moral norms, has previously been subject to experimental psychol- ogy and social, cognitive and affective neuroscience. Some researchers speak of a ���moral brain��� comprising areas in the frontal, temporal and parietal lobes as well as limbic structures (see, e.g. Greene and Haidt, 2002 Moll and de Oliveira-Souza, 2007), involving brain areas associated with a variety of tasks of social cognition. But not only moral norms are related to right or wrong human conduct. The domain of law, as it is formulated and applied, poses another example that is of high relevance to our social life. What happens on the neural level if subjects are engaged in legal reasoning and judgment? Is there an overlap between brain activation during legal and moral decision-making or are different regions involved in the legal condition? And does such neuroscientific knowledge imply anything for our understanding of normative decision-making? The tension between moral and legal norms is illustrated by debates in the scholarly literature, where the application of legal rules is contrasted with the reliance on moral intu- itions (Goodenough, 2001 Goodenough and Prehn, 2004). The predominant view conceives law ideally as purely ratio- nal, free from emotion and passion (Gewirtz, 1996). In the light of recent scientific evidence emphasizing the role of emotion and intuition in moral perception and judgment (Haidt, 2001 Greene and Haidt, 2002 Heekeren et al., 2003 Greene et al., 2004 Moll et al., 2005 Koenigs et al., 2007 Ciaramelli et al., 2007), it is pertinent to know whether legal judgments are also subject to people���s emotions and intuitions. Although we do not think that neuroimaging data can prove or disprove stances in philosophy of law, we are convinced that such empirical investigations can shed new light on these rather theoretical debates. Considering the recent discussion about the neuroscienti- fic implications for the legal system (Goodenough and Prehn, 2004 Greene and Cohen, 2004 Garland, 2005 Zeki and Goodenough, 2006 Mobbs et al., 2007 Tovino, 2007 Gazzaniga, 2008), sometimes even referred to as ���neurolaw��� (Wolf, 2008 Schleim et al., 2009), it is apparent that neu- roimaging research so far has concentrated on investigating criminals and psychopaths (Blair, 2008 Yang et al., 2008) or developing forensic applications such as lie detection (Sip et al., 2008 Spence et al., 2004). By contrast, we were also interested in investigating what impact legal expertise would have on the neural mechanisms of normative cogni- tion and thus investigated two groups, experienced lawyers and other academics. Received 18 May 2009 Accepted 13 January 2010 Advance Access publication 1 March 2010 This work was supported by grants to H.W. from the Volkswagen Foundation, Germany (AZ: II/80 777) and the BMBF (German Ministery of Education and Research, AZ 01GP0804). We would like to thank Hauke Heekeren, Stephanie Melzig, Christiane Rieke, Knut Schnell, Markus Staudinger, two anonymous reviewers and the technical staff of the Life and Brain Center, Bonn for their kind support. Correspondence should be addressed to Stephan Schleim, Department of Psychology, University of Groningen, Grote Kruisstraat 2/1, 9712 TS Groningen, The Netherlands. E-mail: s.schleim@rug.nl. doi:10.1093/scan/nsq010 SCAN (2011) 6, 48^57 �� The Author (2010). Published by Oxford University Press. For Permissions, please email: journals.permissions@oup.com by Fran?ois Bogacz on June 24, 2011 scan.oxfordjournals.org Downloaded from

While a recent neuroimaging study addressed punishment of legal transgressions (Buckholtz et al., 2008), we focused on a more basic step in the process of legal decision-making that precedes the assessment of punishment, namely the decision whether an action is considered as legally right or wrong. We constructed short stories that could be evaluated from a moral as well as a legal point of view and the subjects were confronted with these stories in our fMRI experiment. They had to decide whether behaviors were right or wrong from either of these points of view. As stories were rando- mized over subjects, we were able to investigate the impact of the context or the framing (moral vs legal) independent of the respective story���s concrete content. The rational of our investigation consisted in testing the following three hypotheses. First, we wanted to test whether activations within the ���moral brain��� could also be found during legal judgment. Given that the evaluation of norma- tive behaviors depends on the attribution of beliefs and intentions, as has been suggested by the so-called ���Rawlsian��� model in moral psychology (Hauser, 2006 Huebner et al., 2009), we expected an overlap in brain regions related to mentalizing and theory of mind (TOM) such as the medial prefrontal cortex (Walter et al., 2004 Amodio and Frith, 2006 Singer, 2006 Lieberman, 2007) and the temporo-parietal junction (Gallagher et al., 2000 Saxe and Kanwisher, 2003 Frith and Singer, 2008 Adolphs, 2009) when comparing the moral and legal with the neutral condition. With our second hypothesis, we wanted to investigate the differences between moral and legal judgment in the light of the traditional understanding of law separating moral intuitions from the rational appli- cation of legal rules (Gewirtz, 1996 Goodenough, 2001). We thus expected stronger brain activation in areas related to rule-based decision-making such as the dorsolateral prefron- tal cortex (Miller and Cohen, 2001 Bunge, 2004) for legal decisions. Third, we assumed that as a function of expertise, lawyers would pay more attention to normatively salient features than other academics and thus show less activation related to processing of emotions, such as the amygdala (Dalgleish, 2004), during normative decision-making. MATERIALS AND METHODS Participants Our participants were 46 healthy adults without reported history of psychiatric or neurological disorders. Four of them did not complete the experimental design and two had incidental findings that were dealt with according to our ethical guidelines (Schleim et al., 2007). Out of the 40 remaining subjects (22 male 31.05 4.02 years of age 20.31 1.91 years of education all right-handed mean s.d.), 20 were qualified lawyers having attained the German second state examination and 20 were other aca- demics matched for age (31.95 3.69 vs 30.15 4.22 years, respectively P 0.1), education (20.48 1.31 vs 20.13 1.91 years, respectively P 0.1) and gender (nine female in each group). Experimental procedures were approved by the local ethics committee and all subjects gave written informed consent. Experimental design We developed 36 target stories adapted from moral and legal issues in the media as well as the scholarly literature and 18 control stories taken from everyday life experience (see Supplementary Data), similar to Greene���s and col- leagues��� control stimuli (Greene et al., 2001, 2004). Importantly, target stories were constructed such that they were understandable from the moral as well as the legal point of view. Instructions were randomized between subjects and were used to assign the normative cases to either the moral or the legal condition as follows. Each trial began with a 2 s presentation of a cue indicating the experimental condition, ���neutral���, ���moral��� or ���legal���, followed by the story presented together with the question whether this behavior was right from either the personal, moral or legal view. Subjects had as much time as necessary to make a decision, as in previous research (Greene et al., 2001, 2004 Borg et al., 2006), and could answer either ���yes, rightly��� or ���no, not rightly���, using buttons in both hands. The decision ended the trial that was then followed by a centered crosshair for 12 s in order to allow for blood-oxygene-level dependent (BOLD) relaxation. The stimuli were presented on a computer screen using fMRI-compatible video goggles (Nordic Neurolab, Bergen, Norway) and Presentation (Neurobehavioral Systems, Albany, CA, USA). The stories were split equally into two blocks of data acquisition (i.e. 27 stories per block) and presented individually randomized for each subject. Before entering the MRI scanner, subjects received written instruc- tions and practiced the task with four additional stories, two of them as moral, the other two as legal condition. They had to acknowledge their understanding of the differ- ence between both views before proceeding with the experiment. After the fMRI experiment, we presented all 54 stories including the individual answers in random order on a PC to the subjects and asked them to rate whether the story touched them emotionally (emotion), how realistic they found it (reality), how difficult they found their decision (difficulty) and how certain they were about it (certainty). Answers were recorded using five-point Likert scales ranging from ���not at all��� to ���very much��� for each question. We calculated condition group ANOVAs for reaction time, emotion, reality, difficulty and certainty in order to test for main effects of condition (neutral, moral and legal), group (lawyers, other academics) and group condi- tion interactions. To ascertain whether normative context shaped the subjects��� decision, we calculated an endorsement score for each subject and condition by dividing the num- ber of given yes-responses by the number of possible yes-responses. These were analysed with a condition group ANOVA. Paired tests were used to find significant From moral to legal judgment SCAN (2011) 49 by Fran?ois Bogacz on June 24, 2011 scan.oxfordjournals.org Downloaded from