Social cognition across development and pathology

Chair:   Bruno Wicker

Aix-Marseille Université & CNRS

Abstract:

Social cognition is a broad term that refers to an array of behaviours ranging from basic face perception and recognition of primary emotional expression, to complex interpretation of body movements and social contexts, empathy and decision making in highly demanding social interaction contexts. Humans, and perhaps other non human primate species, have developed incredibly efficient brain processes - automatic or controlled - specifically dedicated to social information processing. On top of it, humans exhibit an ubiquitous tendency to anthropomorphize, which extends to agents that are not biological entities.  Our ability to automatically process social signals is fascinating and social interactions is one of the major forces driving our evolution. Lacking it would transform our everyday life into a nightmare, just as it may be for even very high level functioning individuals affected by neurodevelopmental disorders such as autism.  The last decade has seen a dramatic rise of interest in the study of social neuroscience and a large number of independent studies have shown remarkable consistency in identifying the brain regions that are involved in social cognition. The contributors to this symposium will address various themes of research that aimed at deepening our understanding of the so-called ‘social brain’, its development, and its dysfunction in clinical disorders.

Talk 1:

Intact Mirroring in Autism

Geoff Bird
Social Interaction Lab, Department of Psychological Sciences, Birbeck University of London, UK

The ability of those with autism spectrum conditions (ASC) to mirror actions (imitation) and emotions (empathy) has been a focus of much research. The standard view is that ASC is associated with deficits in both processes with accompanying deficits in those neural systems subserving them (collectively referred to as 'mirror systems'). It shall be argued that in fact ASC is associated with deficits in neither process. With respect to imitation, data will be presented which demonstrates that rather than problems with imitation, those with ASC exhibit problems with imitation-inhibition, possibly due to a failure of top-down modulation in ASC. Furthermore, the data suggests that problems with imitation-inhibition are indicative of problems with theory of mind and perspective-taking, and are associated with wider social deficits.
With respect to problems with empathy, it shall be argued that the previously reported deficits in ASC, and potentially other reported emotional deficits, are instead due to comorbid Alexithymia. Alexithymia is the term used to describe an inability to identify and describe one's own emotion, and prevalence is several times greater in the ASC population than in the typical population. It shall be argued that the presence of a high proportion of individuals with alexithymia is responsible for many of the affective symptoms of ASC including the claimed empathy deficit. Furthermore, failure to account for the presence of alexithymia in those with ASC may account for conflicting results in the processing of emotion more generally in ASC.

Talk 2:

Emotion and action observation in the teenage brain

Marie Hélène Grosbras
Department of Psychology, University of Glasgow, Scotland

Emotion recognition from facial expressions continues to improve during late childhood and adolescence (review in Herba and Phillips, 2004). How the ability to perceive emotion form other social signals develops remains largely unexplored, however. We argue that body cues are at least as important as facial cues to convey affective meaning, in particular during the transition between childhood and adulthood when social interactions are remodelled.
In this symposium I will present longitudinal and cross-sectional behavioural and brain imaging data that reveal the developmental trajectory of emotion processing not only from facial expressions but also from non-communicative hand and body movements. We show that the ability to accurately recognize basic emotions follows a curvilinear development, with rapid improvement until about 10 years of age followed by a small dip, reaching adult level only in the mid-teenage years. This is delayed when point-light displays are used as stimuli, indicating that children rely more on form information. Using functional magnetic resonance imaging we observe that the activity in some brain networks (including amygdala, supramarginal gyrus and inferior frontal gyrus) during passive exposure to such emotionally laden stimuli follows a similar cubic developmental trajectory, while other brain regions (including temporal regions and premotor regions) undergo more linear changes with age.
Further, I will explore the effects personality traits -such as empathy or resistance to peer influence- or formal experience in dance or drama have on interindividual differences in the ability to recognize emotion from body cues, and on brain activity.
   

Talk 3:

Interactions between motor and emotional resonance investigated with a humanoid robot

Thierry Chaminade
Institut de Neurosciences de la Timone, CNRS & Aix-Marseille University, Marseille, France

Humanoid robots, because they have a generic anthropomorphic form but lack human fine-grained details, are useful tools to investigate the neural bases of human social cognition.  Here we describe an fMRI experiment in which behavioural and neural responses to a humanoid upper torso displaying expressions of happiness, anger and disgust or the emotionally neutral condition speech, were compared to human expressions of the same emotions while participants were required to judge either the emotional or the motion content of the videos. Increased response for robot stimuli in the occipital and posterior temporal cortex suggests increased visual processing when perceiving a mechanical anthropomorphic agent. In contrast activity in areas involved in emotional or motor resonance specific to the perceived action, as the insula for disgust or left Broca’s area for speech, is reduced when actions are depicted with the humanoid robot. Finally, activity in regions generally involved in motor resonance in the ventral premotor and inferior frontal gyrus display different effects of the agent used to display the action and of the characteristic of the stimulus being judged. While activity in the ventral premotor is not affected by the task or the agent, activity in the more anterior Brodmann areas 44 and 45 is influenced by both experimental factors, supporting complex interactions between visual information and the object of attention in motor resonance. Altogether, these results are informative as to the features influencing the different areas involved in action perception.
 

Talk 4:

When cartoon differ from real faces: Facial emotion processing in Autism Spectrum Disorders

Delphine Rosset
Institut de neurosciences de la Timone, CNRS & Aix-Marseille University, Marseille, France

Deficits in facial emotion recognition are a hallmark of autistic spectrum disorders (ASD). To better understand the origins of these deficits, we conducted a series of experiments with children with and without ASD, exploring the influence of expertise in processing cartoon versus real face displays. Children with ASD demonstrated greater interest in cartoon than human faces. Further, their responses suggested a typical configural strategy for recognizing emotions on cartoon faces, but an atypical, local strategy with human faces. This suggests that a lack of expertise in examining human faces was associated with atypical perceptual processing. Interestingly, our results showed that the lack of expertise with faces influenced other domains of competences, as well. The ASD group showed greater sensitivity for cartoon faces in a perceptual categorisation task, in visual search, and in a priming task. The current results showed disparities between cartoon and human faces processing in children with ASD, linked to the role of expertise. It is possible that cartoons may be particularly useful in interventions for ASD.

Talk 5:

Early Modulation of Perceptual Neural Activity Induced by Top-Down Social Information

Martial Mermillod
LAPSCO, CNRS UMR 6024 & Université Blaise Pascal, Clermont Ferrand, France

It is generally assumed that neural perceptual processes mainly rely on bottom-up spreading of activity from sensory receptors (e.g. retina) to high-level conceptual areas. On the basis of behavioral and electroencephalogram (EEG) experiments, we will show that both behavioral and neuronal responses to stimulation induced by emotional stimuli are modulated by the presence of social information as early as 100 to 120 ms onset. Both P1 and N170 components are significantly modulated by the positive or negative valence of social information preceding the emotional target (angry, disgusted, fearful, happy, sad or surprise facial expressions). These modulations of the P1 and N170 components are accompanied by behavioral modulation in the feeling but also in the perception of the emotional stimuli. These effects suggest that social information in an emotional target can induce early neuronal modulation at the level of perceptual extrastriate cortex (P1 component) and low-level cognitive areas (N170 component). Implications will be discussed in the context of recent theoretical models of top-down regulation (Bar, 2004; Niedenthal, Mermillod, Maringer, & Hess, 2010).