Recently it has been proposed that humans possess an innate, domain-specific moral faculty, and that this faculty might be fruitfully understood by drawing a close analogy with nativist theories in linguistics. This Linguistic Analogy (LA) hypothesizes that humans share a universal moral grammar. In this paper I argue that this conception is deeply flawed. After profiling a recent and appealing account of universal moral grammar, I suggest that recent empirical findings reveal a significant flaw, which takes the form of a dilemma: either there is something wrong with the moral grammar model because we do not actually possess the innate contents (rules, principles, and concepts) it says we have, or the moral grammar model is simply the wrong model of moral cognition. In light of this dilemma, I conclude we ought to be skeptical that the Linguistic Analogy can adequately serve as a general account of moral cognition.

In the past, experiments on human morality have predominantly utilised theoretical moral dilemmas to shed light on the nature of moral judgment. However, little attention has been given to determining how these judgments might translate into moral actions. In this thesis, I utilised novel and state-of-the-art Virtual Reality environments and combined approaches from social psychology, experimental philosophy, computer science, robotics, and speculative design. Over the course of six experiments with more than 200 participants, simulated moral actions made in Virtual Reality were found to be dissociated from moral judgments made in conventional paradigms. The results suggest that moral judgment and action may be driven by distinct mechanisms. The association between personality traits and moral judgments versus actions, was also investigated. In two experiments, psychopathic and associated traits predicted moral actions and the power with which these were simulated, but failed to predict moral judgments. With research suggesting a mediating role for empathy in this relationship, two further experiments examined empathic and affective processing in moral judgment versus action. In the first of these, alcohol consumption successfully lowered affective empathy and arousal in virtual dilemmas, but moral judgment and action remained unaffected. In the second, an investigation of professionally trained paramedics and fire service incident commanders, revealed distinct differences in empathic and related personality traits, reduced emotional arousal, and less regret following moral action. Taken together, this research suggests that novel virtual technologies can provide insights into self-referent actions, which sit in contrast to judgments motivated by social norms. Ethically, incorporating Virtual Reality in investigations of morality of harm offers a balanced approach; protecting participant wellbeing while increasing the ecological validity of moral investigations. The roles of personality traits and associated emotional processes in moral judgment and action remain multifaceted and as such, I outline the necessity of considering both the characteristics of the decision-maker and the context in which the decision is undertaken, within an interactionist model of morality.

The essay aims to provide an overview of existing research on moral cognition and its neural correlates, as well as to highlight aspects in which the findings from the field of cognitive neuroscience can add to the understanding of the challenges that arise from having artificial systems as moral decision-makers. Morality is an important characteristic of the human way of life and a central mechanism in the modern society. Research suggests that morality has evolved as a function to promote cooperation and prosocial behaviour. Morality has been a topic of investigation in several interdisciplinary fields and often includes the exploration of reason and emotion and their respective roles. Contemporary technological advances include the increased presence of artificial intelligence (AI) that aims to simulate the human mind. Empirical evidence suggests that deontological responses are more closely related to emotional components of moral cognition and linked to increased activity in the ventromedial prefrontal cortex. In contrast, utilitarian responses appear to be more reliant on reason and a more complex neural network that is linked to increased activity in primarily dorsolateral prefrontal cortex. AI has successfully accomplished certain elements of simulating human cognition. However, key findings suggest that the integration of reason and emotion is essential which is possibly challenging for AI to fully replicate.

This essay aims to provide an overview of some key theories and frameworks regarding moral cognition and its neural correlates, in order to examine the possibilities of enhancement of moral cognition. Moral cognition arises from the functional integration of several distinct brain regions and networks. These neural systems correspond to different socioaffective abilities, such as empathy and compassion, as well as sociocognitive abilities, such as theory of mind. Due to this neural distinction, these moral abilities, behaviors, and emotions can be targeted and trained separately. Recent research suggests that training sociocognitive and socioaffective abilities increases cortical thickness in corresponding brain regions and networks, hence providing support for adult neural plasticity in relation to moral cognition. Increased cortical thickness also corresponds to enhanced performance in socioaffective and sociocognitive abilities. Training compassion and empathy induce enhanced abilities to pick up emotional cues, as well as strengthen the motivation to alleviate others’ distress. Practicing theory of mind allows for a better understanding of the perspective of others, which has been indicated to reduce biases between individuals or societal groups. Thus, enhanced moral cognition can contribute to an increase in consideration for those affected by our choices and behavior, which may yield more compassionate, just, and safe societies.

This dissertation is about moral cognition and the production of spiritual experiences in the Chinese version of the massively multiplayer online game, World of Warcraft (CWoW). Chapter 1 introduces the game, field sites, data collection methods, and the questions and paradigms that structure the dissertation's narrative. This chapter explains that during fieldwork CWoW players frequently described gameplay as yielding positive moral and spiritual experiences. Chapter 2 outlines the cognitive mechanisms – Haidt and Joseph's 'moral foundations' (2004; 2007) – that are hypothesised to inform the in-game experiences to which players attributed moral significance. Chapter 3 argues that some aspects of the WoW gameworld are high in cognitive 'relevance' (following Sperber & Wilson, 1986) for Haidt and Joseph's moral foundations. This relevance yields the game a capacity for cross-cultural appeal. Chapter 4 argues that social agencies operant in the lives of CWoW players motivated players to utilise these morally relevant aspects of gameplay as a 'religion-like tool for group cohesion' (following Sosis & Alcorta, 2008). These extra-game agencies help explain why these aspects of gameplay were cultivated in lieu of, or as superordinate to, aspects that may be 'relevant' to other cognitive mechanisms. Chapter 5 argues that reports of soul merger experiences obtained in CWoW resemble Durkheim’s 'collective effervescence' (Durkheim, 1995 [1912]) and may be understood at the cognitive level as 'flow' (Csikszentmihalyi, 1990) experiences conjoined with affective outputs from Haidt and Joseph's moral foundations. Chapter 6 summarises the key points made in the preceding chapters and concludes that in the contexts sampled here, high-contingency aspects of CWoW gameplay may be viewed as Durkheimian religious rituals.

Humans are a social species. Automatic affective responses generated by neural systems wired into our brains create a moral intuition or “gut-feeling” of wrong and right that guides our moral judgments. Humans are also an intelligent, problem solving and planning species with neural structures that enable cognitive control and the ability to reason about the costs and benefits of decisions, and moral judgments, not the least. Previous research suggests that moral intuition and moral reasoning operates on different neural networks - a dual process of moral cognition, that sometimes gives rise to an inner conflict in moral judgments. Early lesion studies found correlations between damage to the ventromedial prefrontal cortex (VMPFC) and changes in moral behaviour. This has been further established through brain imaging studies and the suggestion is that VMPFC mediates affective signals from the amygdala in moral decision making and is highly involved in generating the gut-feeling of right and wrong. However, some moral issues are complex and demand higher level processing than intuition, and the dorsolateral prefrontal cortex (DLPFC) seems to be responsible for the rational, cost-benefit reasoning during moral judgments. Further, recent research suggests that during moral judgments, the brain employ neural systems that generates the representation of value, perspective and cognitive control as well as the representation of the mental and emotional states of others. The present thesis aims to investigate prominent and up to date research on the neural correlates of necessary components in moral cognition, and to examine the function of moral intuition versus reason in relation to current complex moral issues. Moral intuition is supposedly an adaption to favour “us” before “them”, not to be concerned with large scale cooperation, which may explain why we treat many moral issues with ignorance. Understanding how the moral brain works involve understanding what sort of tasks the neural mechanisms in moral cognition evolved to handle, which may explain why some modern issues are so difficult to solve.

Cognitive and emotional processes both seem to contribute in the production of moral judgments, but how they interact is still under investigation. Greene’s dual-process model suggests that these processes constitute dissociable systems in the brain, which are hypothesized to give rise to two qualitatively different ways of moral thinking characterized by two normative moral theories, consequentialism and deontology. Greene indicates that this research undermine deontology as a normative theory. The empirical investigation of moral judgments implies that the dual-process model only seems to accurately predict and explain moral judgments in moral dilemmas involving physical harmful intentions. Regardless of the models empirical support, the empirical findings in the study of moral judgments could have normative and metaethical implications.

Perception of moral violations and people’s misbehaviours are crucial elements that allow people to live in the society. The literature suggested that both explicit and implicit processes are involved in the moral cognition (Cameron et al., 2012). However still nowadays, the majority of the literature focused only on the explicit aspects of morality, while scarce attention was given to the study of implicit moral attitudes (Cameron et al., 2018). Furthermore, researchers on morality are currently attracting by the investigation of moral cognition in people with impairments in social cognition (e.g., people with autism spectrum disorder; ASD) who may process the moral violations differently compared neurotypical people (see e.g., Gallese, 2006), However, this line of research is at the beginning stage, and the use of implicit measures to understand the mechanisms underpinning the morality in autism (or in people with high autistic traits) have not yet been considered in previous studies. The present dissertation aimed to investigate important aspects of moral cognition that currently are still under-investigated in literature, offering a wider view on implicit aspects of morality in autism. Through six experiments, this work wanted to provide new empirical findings concerning the implicit mechanisms underpinning moral cognition in both neurotypical and autistic people without intellectual disability. Experiment 1 and 2 According to Moral Foundation Theory (Graham et al., 2013) specific emotions are associated to different types of moral violations. For instance, when people perceive that a victim is physically or emotionally harmed by someone else, they feel anger against the perpetrator. By contrast, people feel disgust when they perceive that someone ate inedible food or made blasphemous practices (Haidt et al., 1993). However, empirical finding that assessed the link between different types of moral violations and specific emotions showed mixed results. Indeed, it was suggested that other important factors might also modulate the relations between different moral violations and emotions (Atari et al., 2020; Kemper & Newheiser, 2018). Specifically, the benign violation hypothesis (McGraw & Warren, 2010) claimed that sometimes people could perceive moral violations as amusing, and this specific emotion seems elicited by specific contextual (e.g., psychological distance) as well as interpersonal factors (e.g., humor traits). However, previous studies never tested whether people perceive amusement in response to all types of moral violation, or whether amusement emerged only in response to specific norm violations (e.g., purity violations). Furthermore recently, Dempsey and colleagues (2020a) emphasized the importance to test the assumptions of Moral Foundation Theory in autistic spectrum disorder (ASD). Interviewing six autistic participants, the authors found that the participants endorsed all moral domains as morally relevant. However, the relations between emotional reaction and moral domains in autism have never been tested, as well as whether autistic people perceive moral violations as benign and amusing. In the first two experiments of this thesis, it was assessed the amusement reaction in response to different types of moral violations (purity vs. harm) and extending such investigation involving a group of autistic participants. Furthermore, the studies conducted so far on moral cognition has prevalently used explicit methodology. By contrast, in these experiments it was assessed the emotional reaction in response to different types of moral violations analysing the spontaneous facial expressions in neurotypical people (experiment 1), and the cardiac response of both typical and autistic individuals (experiment 2). The aim of these experiments was to investigate whether neurotypical people perceive specific types of moral violations as benign and amusing, and then whether autistic people expressed different emotions compared neurotypical people in response to different types of moral violations. Experiment 3, 4 and 5 As opposite of Moral Foundation Theory, the Dyadic Model claimed that a sine qua non for moral perception is the implicit activation of harm (Gray et al., 2012). In other words, always immorality automatically activates perceptions of harm regardless of someone is harmed (acts which involve physical and emotional damage; harm violations) or not (i.e., acts which involve impure and degradation behaviour; purity violations). Indeed, empirical evidence suggested that, during the reading of a scenario depicting a harmless moral violation in the domain of purity, people nevertheless implicitly infer that harm was involved. Across three experiments, we tested whether ostensibly harmless moral violations activated implicitly the harm concept. In all these three experiments the participants’ implicit attitude in response to moral violations was tested by using the affective misattribution paradigm (AMP; Payne et al., 2014), which is one of the most reliable methods to assess individual unconscious mental processes in psychology (Znanewitz et al., 2018). Specifically, Experiment 3 aimed to replicate previous findings which highlighted the implicit role of harm in response to harmless moral violations (Gray et al., 2014). Experiment 4 extended previous results with an investigation aimed to verify whether people with high autistic traits showed less implicit activation of harm in response to harmless scenarios than people with low autistic traits. Finally, Experiment 5 aimed to investigate whether autistic people (without intellectual disability) implicitly activated the harm concept in response to harmless scenarios as well as neurotypicals. Experiment 6 The majority of the studies which investigated the moral cognition in autism started to the premise that autistic people have a selective impairment in Theory of Mind (ToM; Moran et al., 2011), which is also a fundamental skill to understand the moral behaviours (e.g., Hamlin, 2015). However, the literature is far to show the role of Theory of Mind in the moral processing. Recently, it was suggested that ToM is a cognitive skill that involves both explicit and implicit processes (e.g., Southgate et al., 2007). The born of new implicit measures to assess the ToM brought an initial enthusiasm in this area of study. For instance, studies on clinical psychology, using both explicit and implicit ToM measures in order to understand selective ToM difficulties by autistic individuals without intellectual disability showed that, although this population presented high performance to explicit ToM tasks, they showed lower performance in the newer implicit ToM tasks compared to neurotypical participants (e.g., Schuwerk et al., 2015; Senju et al., 2009). Late unfortunately, these interesting results were put into question by other studies, which argued against the reliability of these new implicit measures (Kulke et al., 2018b). Moreover, another important gap in implicit ToM literature was that, while previous models in developmental psychology and psychopathology have suggested a link between attachment and the development of ToM, the relationship between attachment and implicit ToM has been so far neglected. In this experiment, we assessed both implicit and explicit mental state attribution, by means of a new implicit ToM task, as well as some attachment dimensions, comparing a group of autistic participants, without intellectual disability, and a control group. Specifically, in this experiment it was investigated participants’ spontaneous anticipatory look in response to false beliefs tasks using the eye-tracking methodology (one of the most used implicit ToM paradigms in literature). This study aimed to investigate the Theory of Mind (ToM) at both explicit and implicit level (Southgate et al., 2007), investigating possible differences between neurotypical individuals and autistic people. Furthermore, in the same study it was also investigated the link between the participants’ attachment and the explicit and implicit ToM skills. Moral cognition is a complex object of study that should be investigated with a multi-dimensional approach. In the experiments described in this thesis it was proposed a combined methodology that allowed to gather both explicit and implicit cognitive and emotional mechanisms that stay at the root of the moral judgment. Thanks to this approach it was found that deliberate and involuntary processes did not present convergent responses in the moral cognition of participants. Indeed, self-reported emotions seem to show divergent responses compared to the spontaneous facial expressions and the cardiac response to moral violations. These results support extant literature that raises concerns about the intuitionist assumptions of the moral cognition based on the assessment of only explicit responses of participants (e.g., Gutierrez & Giner-Sorolla, 2007; Nosek, 2007). Furthermore, this thesis highlighted the importance of studying the moral cognition in people affected by the autism spectrum disorder. Despite the social impairments that this neurodevelopmental disorder presents, we found that autistic people with cognitive level in average with the typical population, did not show any differences compared to the control group in both explicit and implicit mechanisms of moral cognition. These interesting findings may bring important theoretical reflections in moral and clinical psychology.

Background: The aim of this thesis was to investigate the relationship of moral cognition and emotions to the pathophysiology of major depressive disorder (MDD). Patients with MDD may experience excessive guilt or self-blaming biases despite recovery from the depressed state. Since guilt is a moral emotion thought to motivate altruistic behaviours, it has been hypothesized that elevated self-blame in MDD may result in pathological increases to altruism in some patients. The relationship of self-blame to altruistic choices in individuals with remitted MDD (rMDD), however, has not been established. Guilt has been shown to activate the subgenual cingulate and adjacent septal region (SCSR) which is of known importance to the pathophysiology of MDD. Since MDD is thought to arise from network-level dysfunctions, and moral cognition and emotions are hypothesized to emerge from network-level binding, investigating resting-state SCSR functional connectivity in rMDD patients and healthy control (HC) participants could reveal networks of potential relevance both to MDD and to moral cognition and emotions. Chapter 2: We investigated whether melancholic rMDD patients could be distinguished from non-melancholic and HC groups on the basis of resting-state functional connectivity to an SCSR seed region. Lower SCSR-amygdala connectivity distinguished the melancholic rMDD group from non-melancholic and HC groups. Chapter 3: We investigated whether patients who remained resilient to recurring depressive episodes were distinguishable from recurring episode MDD and HC groups on the basis of resting-state connectivity to an SCSR seed region. Lower interhemispheric SCSR connectivity distinguished the resilient MDD patients from the recurring episode MDD and HC groups. Chapter 4: We measured explicit and implicit preferences for social options with and without altruistic motivations relative to selfish options in the rMDD and HC groups during emotion priming to modulate feelings of guilt. The rMDD patients explicitly preferred prosocial options (i.e., social options and altruism directed towards friends or colleagues) less than HC participants. Regardless of group, guilt priming increased explicit and implicit preferences for altruism towards strangers. Chapter 5: We investigated whether explicit and/or implicit preferences for prosocial options during guilt priming were correlated with resting-state connectivity to an SCSR seed region, and whether this relationship could distinguish the rMDD and HC groups. Across all participants, implicit prosocial choice preference negatively correlated with connectivity between the SCSR and right temporoparietal junction (TPJ). The relationship of SCSR-TPJ connectivity to implicit preferences for social options and for altruism towards friends and colleagues was weaker in the rMDD group compared to the HC group, particularly for implicit altruism. Conclusions: We identified resting-state SCSR networks associated with vulnerability to melancholia and with resilience to recurring depressive episodes. Patients with rMDD explicitly preferred options entailing social withdrawal, a symptom associated with MDD vulnerability. Irrespective of group, guilt motivated altruism towards strangers but not friends and colleagues. Implicit prosociality was negatively associated with connectivity in a social agency network, and the comparatively weak relationships between connectivity and implicit choice preferences in rMDD patients may reflect a vulnerability factor for MDD.

In this dissertation, I criticize current approaches to moral cognition research and develop an embodied account of moral cognizing. Contemporary research into moral cognizing is strongly influenced by an orthodox cognitive mindset. Modern approaches to cognitive science, especially situated, i.e., embodied-enacted, approaches emphasize the role of affectivity, emotions, bodily experiences, sociality, culture, interpersonal relationship, and dynamic interactions in human cognizing. However, these non-orthodox approaches have only occasionally been applied to the study of the moral mind. The embodied account defended analyzes moral actions in terms of an agent's engagement with the world. Agents acquire a repertoire of possible interactions with the world rooted in their sensorimotor expertise. This embodied view of moral action focuses on the role culture, sociality, and interpersonal relations play in contextualizing actions and constituting an agent's identity. Enactive theory provides a broader cognitive framework to understand embodied action and moral identity, overcoming the dominant logic of doing and being done to, opening up moral agency to a sphere of joint action.