Caught in the act: The impact of audience on the neural response to morally and socially inappropriate behavior

Abstract

We examined the impact of witnesses on the neural response to moral and social transgressions using fMRI. In this study, participants (N = 16) read short vignettes describing moral and social transgressions in the presence or absence of an audience. In line with our hypothesis, ventrolateral (BA 47) and dorsomedial (BA 8) frontal cortex showed increased BOLD responses to moral transgressions regardless of audience and to social transgressions in the presence of an audience relative to neutral situations. These findings are consistent with the suggestion that these regions of prefrontal cortex modify behavioral responses in response to social cues. Greater activity was observed in left temporal-parietal junction, medial prefrontal cortex and temporal poles to moral and to a lesser extent social transgressions relative to neutral stories, regardless of audience. These regions have been implicated in the representation of the mental states of others (Theory of Mind). The presence of an audience was associated with increased left amygdala activity across all conditions.

Introduction

Several neurologic and psychiatric disorders are notable for a propensity for behaviors deemed inappropriate or unacceptable in society. Such disorders include frontotemporal dementia (FTD) (The Lund and Manchester Groups, 1994, McKhann et al., 2001, Miller et al., 1997, Neary, 1999), anti-social personality disorder (DSM-IV), psychopathy (Frick et al., 1994, Hare, 1991, Harpur et al., 1989), and traumatic brain injury and frontal lobe lesions (Benton, 1991, Damasio et al., 1994, Grafman et al., 1996, Harlow, 1968, Price et al., 1990). In many of these disorders, the affected patients are able to correctly identify moral transgressions or inappropriate behaviors as wrong, in sharp contrast to their actual behaviors and their ability to make emotionally based moral judgments (Blair, 1995, Eslinger and Damasio, 1985, Eslinger et al., 2004, Grattan and Eslinger, 1992, Lough et al., 2006, Mendez et al., 2005, Saver and Damasio, 1991).

Moral emotions of guilt, shame and embarrassment are thought to prompt attempts at restitution or appeasement actions (Ausubel, 1955, Keltner and Buswell, 1997, Lewis, 1971, Tangney et al., 1996). The failure of moral emotions to prompt reparative actions may contribute to the inappropriate behavior of patients with the neuropsychiatric disorders considered above. While the neural systems facilitating complex social behaviors such as restitution and appeasement have yet to be characterized, the mechanisms of more basic response reversal paradigms suggest several likely substrates. Several studies have reported ventrolateral prefrontal cortex and dorsomedial prefrontal cortex activation when such a change in response is required (Cools et al., 2002, Evers et al., 2005, Nagahama et al., 2001, O’Doherty et al., 2003, Remijnse et al., 2005, Rogers et al., 2000). It is argued that these regions are activated not only when changes in reinforcement contingencies initiate alternative motor responses, but also when social cues initiate a reversal of current behavior (Blair and Cipolotti, 2000). In line with this suggestion, Kringelbach and Rolls (2003) found ventrolateral and dorsomedial frontal cortex activation when angry faces cued the need for a response reversal. Given these findings, it is likely that other forms of aversive social information, such as those associated with aversive moral and social emotions including guilt and embarrassment, generate activity related to the initiation of alternative motor responses.

In line with this suggestion, some previous studies investigating the neural responses to intentional and unintentional social transgressions (Berthoz et al., 2002) and contexts inducing guilt or embarrassment (Takahashi et al., 2004) have reported ventrolateral and dorsomedial prefrontal cortex activation to these situations. We wished to extend this work by investigating whether the presence or absence of an audience during victim-based (moral) and social transgressions would have a differential impact on this response. For victim-based moral transgressions, a victim is harmed, creating a social demand that the offender should make attempts at restitution whether the act is witnessed or not. In contrast, appeasement or submissive behaviors are only socially demanded if a social transgression is witnessed (Keltner, 2000, Keltner and Buswell, 1997, Keltner et al., 1995, Miller and Leary, 1992, Tangney et al., 1996). We thus predicted that the processing of moral transgressions should lead to increased BOLD response in ventrolateral and dorsomedial prefrontal cortex in either audience condition, while this response should be seen when processing social transgressions only in the presence of an audience.

Prior studies of stories containing moral or social transgressions have identified BOLD responses in regions associated with representation of the mental states of others; i.e., Theory of Mind processing (Berthoz et al., 2002, Greene et al., 2001, Greene et al., 2004, Heekeren et al., 2003, Heekeren et al., 2005, Moll et al., 2002a, Moll et al., 2002b, Takahashi et al., 2004). These regions include the superior temporal sulcus/temporal parietal junction, the temporal poles, and medial prefrontal cortex (BA 8 and 9). Berthoz et al. (2002) found increased activity within these regions during intentional relative to unintentional social transgressions and suggested that this increase in activity was related to attempts to determine the intentions of the offender. We examined the impact of the presence or absence of an audience on Theory of Mind activity related to moral and social transgressions. We anticipated that the presence of an audience watching the participant would activate Theory of Mind regions, as the participant represented the mental state of such observers. Following Berthoz et al. (2002), we also anticipated that these regions would be particularly active during the moral transgressions (which were intentional), compared to the embarrassing social transgressions (which were unintentional). The current study tests these predictions.