A conflicted and contentious literature has emerged from the proposal that visuospatial information from static images can automatically trigger associated motor representations. Curiously, investigations into this visual-motor relationship have predominantly focused on images of manipulable objects, while relatively little work has made use of images of body parts- whose referents are represented directly in the motor system. Limited work has made use of hand images as task-irrelevant primes, in an effort to determine whether a hand image automatically evokes a motor representation of the viewer’s corresponding limb. The results of these studies have provided diverging evidence and have resulted in competing theoretical accounts. Here, I present results from a series of stimulus-response compatibility experiments that were designed to probe the nature of representations generated by static hand images, while also addressing potential methodological weaknesses of the previous works. The results show that both stimulus properties and task demands influence the way in which an image of a hand is coded. Notably, I provide clear evidence that motor representations can be evoked automatically by depictions of particular hand postures, but that these representations are not an automatic, ineluctable component of the general processing of any hand image. These results not only contribute to a more unified account of hand representations, but also have wider implications for our understanding of the conditions under which static displays can engage motor representations. / Graduate

TMS can be used to generate representational maps by delivering pulses at throughout a grid, centered over the most sensitive spot to elicit a resulting MEP called the motor hotspot. The areas of these maps are modulated by muscle contraction and have been shown to increase in area with increasing contraction intensity. Both intracortical inhibition (SICI), and intracortical facilitation (ICF) are paired pulse paradigms in which contraction causes a reduction in magnitude. The present study aimed to categorize changes in the above circuits and representational maps as well as expose a possible relationship between both metrics in the context of graded contraction. To study these questions 15 healthy, right-handed volunteers participated in a study measuring SICI, ICF and cortical maps under conditions of REST, 10, 20 and 30% of MVC of the right FDI muscle. SICI and ICF showed significant reduction between REST and no differences amongst contraction levels. However, SICI displayed a graded reduction through contraction levels when analyzed on a trial-by-trial basis sorted by actual contraction level. Cortical representational area increased from REST to all contraction states and between 10 and 30% MVC confirming the graded growth observed in previous studies. Further, analysis shows that SICI, ICF and area all exhibited the majority of their modulation within the first 10% of contraction. Both SICI and ICF were not significantly correlated to the growth in representational area. This may be in part due to participants’ variability in the level of contraction sustained during measures, which also made it unfeasible to conduct a correlation of trial-by-trial data between map area and circuit magnitudes. We present evidence to corroborate previous findings for the effects of contraction on intracortical circuits and representational area during graded contraction as well as contribute to the methodology of such investigations concerning the control of varied contraction. / Thesis / Master of Science (MSc)

Il est maintenant bien établi que le cerveau humain est doté d’un système de neurones qui s’active tant à la perception qu’à l’exécution d’une action. Les neurones miroirs, ainsi que le système qu’ils forment avec des structures adjacentes appelées système neurones miroirs (SNM), ont été relié à la compréhension d’action et pourrait être impliqué dans les fonctions sociales de haut niveau tel que l’empathie et l’imitation. Dans la foulée spéculative reliant le SNM à la sphère sociale, le dysfonctionnement de ce système a rapidement gagné intérêt dans la genèse des anomalies du domaine social chez les personnes présentant le Trouble du spectre de l’autisme (TSA). Néanmoins, l’hypothèse voulant que le dysfonctionnement social des TSA repose sur une atteinte du SNM est controversée. En effet, les études soutenant cette hypothèse nécessitent des fonctions cognitives et sociales qui peuvent contribuer à l’obtention de résultats atypiques, telles que la compréhension des consignes, l’attention sur des stimuli sociaux ou la réalisation d’acte moteur. Récemment, un protocole auditif de négativité de discordance (MMN) utilisant des stimuli reliés à l’action humaine a été utilisé pour mesurer l’activité du SNM. Cette technique semble prometteuse dans la mesure où elle ne nécessite pas de capacités attentionnelles ou langagières, elle est brève et demande un montage minimal d’électrodes. Le premier article avait comme objectif principal de mesurer la validité de convergence du protocole MMN relié à l’action avec celui du rythme mu, le protocole le plus utilisé pour enregistrer l’activité miroir à l’aide de l’électroencéphalographie (EEG). Les modes de stimulation ont été délivrées en bloc successif à un groupe de 12 adultes en santé. Alors que les deux techniques ont modulé efficacement les régions fronto-centrales et centrales respectivement, mais ne sont pas corrélées, nous avons conclu qu’il est possible 2 qu’elles mesurent des aspects différents du SNM. Le deuxième article avait comme objectif principal de mesurer l’activité du SNM à l’aide du protocole MMN relié à l’action chez 10 enfants présentant un TSA ainsi que chez 12 enfants neurotypiques dans la même tranche d’âge (5-7ans). Chez les enfants TSA, nous avons montré un patron de latence inversée, comparativement aux enfants du groupe contrôle; ils traitaient plus rapidement les sons contrôles que les sons reliés à l’action humaine, alors que la tendance inverse était observée chez les contrôles. De plus, bien que les deux groupes différaient quant aux sons d’action, ils ne différaient pas quant aux sons contrôles. Quant à l’amplitude, les enfants TSA se distinguaient du groupe contrôle par une amplitude restreinte du son d’action provenant de la bouche. Par ailleurs, les mesures neurophysiologiques et neuropsychologiques n’étaient pas corrélées. En sommes, basé sur la prémisse que ce protocole MMN pourrait mesurer l’activité du SNM, cette thèse a comme but d’améliorer les connaissances quant à son utilisation chez l’adulte et l’enfant neurotypique ainsi que chez l’enfant TSA. Celui-ci pourrait ultimement être utilisé comme un biomarqueur potentiel du TSA. / Mirror-neuron system (NMS) has been suggested to underlie action understanding, which is believed to be involved in higher social functions such as empathy and imitation. Numerous studies have also provided indirect evidence supporting the existence of a MNS in the human brain using functional magnetic resonance imaging, transcranial magnetic stimulation, magnetoenceophalography and electroencephalography (EEG). As evidence relating MNS function with social cognition accumulated, its dysfunction was proposed to underlie social impairments in Autism Spectrum Disorder (ASD). Although some studies have reported structural and/or functional data supporting this hypothesis, it remains controversial. Indeed, the techniques used to probe MNS activity often require the participant to understand the task, display sustained attention and present spared visual cognitive functions, which could be altered in ASD population. This, in turn, could partly explain differences in MNS activity found between invididuals with ASD and neurotypical participants. The auditory mismatch negativity (MMN) protocol has recently been proposed as a viable tool to assess MNS activity using action-related sounds. This relatively short oddball paradigm does not require visual, attentional or language skills and can be performed with minimal electrode installation (minimal use of 4 electrodes). This action-related MMN may therefore be well-suited to investigate MNS function in very young children or populations with mental disabilities, such as ASD. The first article of this thesis assessed the convergent validity of the action-related MMN protocol with a mu rhythm suppression paradigm, which is the most commonly used EEG measure of MNS activity. To do so, both protocols were consecutively administered in healthy adults in the same experimental session. While both techniques successfully 4 modulated the fronto-central and central brain regions using action-related stimuli, they did not correlate significantly. This suggests that the two techniques may not probe the same aspects of MNS function. The second article aimed at investigating MNS activity in children with ASD using the action-related MMN protocol. Data were compared to age-matched typically developing children. We show that children with ASD present an opposite pattern to that of control children : whereas MMN latencies are shorter for action-related sounds compared to control sounds in neurotypical children, the opposite pattern is observed in children with ASD. Furthermore, although the two groups present different response patterns with regards to action-related sounds, they repond similarly to control sounds. For amplitude measurements, children with ASD display reduced amplitude to action-related sounds produced by the mouth compared to neurotypical children. In summary, this thesis aimed at determining whether and action-related MMN protocol could be used to measure MNS activity in neurotypical adults and children, as well as in ASD children. The action-related MMN could potentially be used as a biological biomarker of ASD pending further studies.