Search

People’s ability to question and resist immoral orders is a fundamental aspect of individual autonomy and of democratic societies. To wit: the history of nations is plagued by horrendous acts of obedience that have caused wars and the loss of countless lives. But today, experimental research on neuro-cognitive processes that prevent an individual from being coerced into causing pain to others is almost completely inexistent. Since Milgram’s studies which mostly investigated psychological and contextual components which make an individual to obey or disobey immoral orders (Haslam et al., 2014), some studies have tried to understand how obedience to authority impacts several neuro-cognitive processes (Caspar et al., 2016, 2021; Caspar, Ioumpa, et al., 2020; Caspar, Lo Bue, et al., 2020). Yet, the neuro-cognitive processes supporting resistance to immoral orders have almost not been investigated. The major reasons to explain this lack of research were the ethical issues associated with Milgram’s studies, the difficulty to adapt Milgram’s protocol to imaging studies, and the difficulty to develop an experimental set-up to make participants disobey. By using a new protocol developed by Caspar (Caspar, 2021) addressing these issues, in the present fMRI study we will explore the neural signature of disobedience to immoral orders received by an experimenter in civilians and a military population. Based on previous studies (Caspar et al., 2021; Caspar, Ioumpa, et al., 2020), we will particularly focus on two key neuro-cognitive processes which have been proposed to be involved in (prosocial) disobedience: the sense of agency (SoA) and empathy for pain. In an exploratory fashion, we will also target other neuro-cognitive processes such as the interpersonal feeling of guilt and cognitive conflict.

Prosocial behavior can be defined as behaviors intending to help others (Wu & Hong, 2022). However, such type of behavior is reduced when the ‘others’ are perceived as part of another group (i.e., outgroup members; De Dreu, 2022; Louis et al., 2019). We referred to this diminution of prosocial behavior toward outgroup members as the “intergroup prosociality bias”. Importantly, prosociality appears to be a key mechanism in the prevention of conflicts (Fehr & Fischbacher, 2004; Ostrom, 2000) and to achieve reconciliation in war-torn societies (Mironova & Whitt, 2016; Nadler & Schnabel, 2015). Our main aim is thus to understand how this intergroup prosociality bias evolves in war-thorn societies, where feelings of in- and out-group membership can be enhanced. In the present study, we aim to study how intended prosociality toward other groups is altered over 2 generations after a conflict, for both victims and perpetrators. Cambodia represents a unique opportunity as the entire Cambodian population suffered from the genocide orchestrated by the Khmer Rouge regime between 1975 and 1979, that resulted from about 1,700,000 of deaths. Victims and Khmer Rouge cadres had a similar ethnicity, religion and culture for the majority and are still living together in Cambodia, thus allowing comparisons with restricted number of confounds. We will use the Intended Prosociality Task (Pech et al., Under review; Pech & Caspar, 2022) to cover different aspects of prosociality. Three prosocial scenarios will be used based on the three types of prosocial behaviors described in the review of Wu and Hong (2022), which are comforting, sharing and targeted helping (i.e., helping others for a specific goal, for instance helping to move pieces of furniture). Three individuals (i.e., a former victim, a former perpetrator, a neutral individual) will be presented to the participants with their stories. At the beginning of each block one of the three prosocial scenarios will be presented, participants will be asked to imagine themselves in these different situations. For each trial, participants will be presented with a pair of the individuals and will have to select between the two, as fast as possible, who benefit from their intention to act prosocially considering the actual scenario. In one condition the two individuals will be the same (i.e., forced-choice condition). In the other condition the two individuals will be different (i.e., free-choice condition). This task will allow us to measure participants’ explicit preference, measured by the percentage of choice of an individual in the free-choice condition, and their implicit preference measured by their reaction times (RT) and their mid-frontal theta power (FMθ). Previous studies indeed described that a higher FMθ and longer RT before a choice indicates a higher cognitive conflict (Cavanagh & Frank, 2014; Cooper et al., 2019).

An extensive amount of literature indicates that post-traumatic stress disorder (PTSD) is associated with both affective and cognitive dysfunctions (Karl et al., 2006). One of these dysfunctions involves a hyperactivity of the amygdala together with a hyporesponsivity of the medial prefrontal cortex, which involve a higher emotional reactivity, notably to potentially threatening stimuli (DiGangi et al., 2017). In electroencephalography (EEG) studies, this hypersensitivity has been characterized by lower frontal theta power (DeLaRosa et al, 2014) and higher amplitude of the Late Positive Potential (LPP, DiGangi et al., 2017), indicating a disruption of cortico-limbic communication. The hypersensitivity to threatening stimuli has been observed both in military veterans (e.g., DiGangi et al., 2017) and victims (e.g., Brunetti et al., 2010). However, how comparable are these effects between “victims” and “perpetrators” of a dramatic event and how the following generations also display such hypersensitivity to threatening stimuli is still largely understudied. In the present study, we aim to study emotional reactivity to threatening (directed towards oneself or directed towards others) vs non-threatening stimuli in survivors (including “victims” and “perpetrators”) of the Cambodian Genocide and their children over two generations. Cambodia represents a unique opportunity as the entire Cambodian population suffered from the genocide orchestrated by the Khmer Rouge regime between 1975 and 1979, that resulted from about 1,700,000 of deaths. Victims and Khmer Rouge cadres had a similar ethnicity, religion and culture for the majority and are still living together in Cambodia, thus allowing comparisons with a limited number of confounds. The hypersensitivity to threatening stimuli has been observed both in military veterans (e.g., DiGangi et al., 2017) and in victims (e.g., Brunetti et al., 2010). However, how comparable are these effects between “victims” and “perpetrators” of a dramatic event and how the following generations also display such hypersensitivity to threatening stimuli is still largely understudied. In the present study, we aim to study emotional reactivity to threatening vs non-threatening stimuli in survivors (including “victims” and “perpetrators”) of the Cambodian Genocide and their children over two generations. Cambodia represents a unique opportunity as the entire Cambodian population suffered from the genocide orchestrated by the Khmer Rouge regime between 1975 and 1979, that resulted from about 1,700,000 of deaths. Victims and Khmer Rouge cadres had a similar ethnicity, religion and culture for the majority and are still living together in Cambodia, thus allowing comparisons with restricted number of confounds.

Post-Traumatic Stress Disorder (PTSD) results from exposure to a traumatic event and is associated with perceptual and cognitive dysfunctions (Vieweg et al., 2006). Individuals with PTSD show abnormalities in higher-order emotional processes, including emotion regulation and recognition (Passardi et al. 2019), likely due to an abnormal functioning of the amygdala (e.g., Bremmer et al., 1999). The literature on facial emotion recognition in PTSD showed contrasting results. While some research showed deficits in the recognition of negative facial emotions such as fear or sadness in PTSD patients compared to non-PTSD individuals (Poljac, Montagne & de Haan, 2011), others have found a deficit in the recognition of positive facial emotions (Passardi et al., 2018), and a bias for neutral facial expressions (Pfaltz, Passardi et al., 2019). Another recent study (Castro-Vale, Severo & Carvalho, 2020) indicated that lifetime PTSD was associated with an impairment in the recognition of all emotions, an effect also observable in the offspring. It thus appears that, even though the results are spare in the literature, PTSD is associated with a deficit in emotion recognition. In the present study, we aim to study how emotion recognition is altered over 2 generations after a trauma, for both victims and perpetrators. Cambodia represents a unique opportunity as the entire Cambodian population suffered from the genocide orchestrated by the Khmer Rouge regime between 1975 and 1979, that resulted from about 1,700,000 of deaths. Victims and Khmer Rouge cadres had a similar ethnicity, religion and culture for the majority and are still living together in Cambodia, thus allowing comparisons with restricted number of confounds.

Sensory gating (SG) is a mechanism that allows to filter out repeating irrelevant information (Alder et al., 1982; Gjini et al. 2008). Classical tasks using electroencephalography (EEG) have documented this phenomenon, also called ‘repetition suppression’, by presenting twice the same stimulus (Gjini et al., 2013; Boutros et al., 2004). The amplitude of the Event-related potentials (ERPs) associated with the second stimulus (S2) is classically reduced compared to the amplitude of the ERP response to the first stimulus (S1). Post-Traumatic Stress Disorder (PTSD) results from exposure to a traumatic event and is associated with perceptual and cognitive dysfunction (Vieweg et al., 2006). A study showed that stimulus encoding (when S1 is presented) and SG (when S2 is presented) is altered in patients suffering from PTSD for both visual and auditory inputs (Gjini et al., 2013). In the present study, we aim to study how stimulus encoding (S1) and SG (S2) are altered over generations after a trauma, for both victims and perpetrators. Cambodia represents a unique opportunity as the entire Cambodian population suffered from the genocide orchestrated by the Khmer Rouge regime between 1975 and 1979, that resulted from about 5,000,000 of deaths. Victims and Khmer Rouge cadres had a similar ethnicity, religion and culture for the majority and are still living together in Cambodia, thus allowing comparisons with restricted number of confounds. We will investigate to what extent stimulus encoding and sensory gating could be altered over 3 generations.

Post-Traumatic Stress Disorder (PTSD) results from exposure to a traumatic event and is associated with perceptual and cognitive dysfunction (Vieweg et al., 2006). Previous studies showed that due to hyperarousal and impaired concentration, pre-attentive processing could be altered in individuals suffering from PTSD (e.g., Morgan & Grillon, 1999). The mismatch negativity (MMN) is a component of the event-related potential (ERP) evoked by rare changes (i.e., deviant) after repetitive sensory stimuli, and has been used to study pre-attentive processing. Using an auditory oddball paradigm, previous studies showed that the mismatch negativity (MMN) was greater following deviant stimulus in individuals suffering from PTSD compared to individuals without PTSD (e.g., Morgan & Grillon, 1999; Ge, Wu et al., 2011). But to what extent this effect extends to the following generations, and results from the trauma itself or from the development of PTSD symptoms is still unknown. In the present study, we aim to study how the MMN is altered over generations after a trauma, for both victims and perpetrators. Cambodia represents a unique opportunity as the entire Cambodian population suffered from the genocide orchestrated by the Khmer Rouge regime between 1975 and 1979, that resulted from about 1,700,000 of deaths. Victims and Khmer Rouge cadres had a similar ethnicity, religion and culture for the majority and are still living together in Cambodia, thus allowing comparisons with restricted number of confounds.

In the present study that will be conducted in Rwanda, we will investigate how the neural empathic response varies between former perpetrators and victims of the genocide and their children when they visualized former perpetrators, victims or their children, receiving painful or non-painful stimulations.