A computational model of action resonance and its modulation by emotional stimulation

Emotional resonance deficits in autistic children

According to some theories imitation, defined as an action resonance mechanism, is deficient in autism. In contrast, other theories (e.g., the "top down control of imitation" hypothesis) state that the problem is not in imitation per se but in the way social cues modulate imitative responses. In this study, 15 high-functioning children with autism and 15 matched controls were tested for their ability to imitate finger movements preceded by neutral and emotional facial expressions (primes) in a stimulus-response compatibility task. Hand movements performed after neutral expressions did not differ between the two groups (i.e., they both showed a normal imitative tendency). However, hand movements performed after emotional expressions significantly differed between the two populations, with controls, but not autistic spectrum disorder (ASD), showing enhanced imitation in the emotional condition. This study supports the view that, in ASD, imitation abilities are spared but they are not modulated according to the emotional and social context.

Vicarious function in the motor cortex. A computational investigation

This paper presents a computational investigation of the vicarious function in the motor cortex (c/o the ability to reorganise its functioning by virtue of a shift of the lost function in the surrounding cortex which becomes able to vicariate). Several experimental studies in animals and humans have shown that motor recovery after partial destruction of the motor cortex is based on adjacent motor reorganisation. This study provides phenomenological evidence of this vicarious function. We tested the hypothesis according to which the vicarious function is possible because there is a synaptic rearrangement of the weights (which are comparable to the synapses of the brain) of the lesioned layer (unmasking of previous silent synapses hypothesis), and our results confirm this hypothesis. We argue that functional recovery is possible only when having bidirectional connections and that it is facilitated when non-M1 areas can guide the layer to relearn the lost movement.