Neural processes for action observation

Abstract

Through actions we explore the world around us, we express ourselves, achieve goals and interact with others. We are thinking about actions, planning, executing, imitating, observing, and understanding them. In other words, our life is filled with motor cognition. Yet our knowledge of the brain processes underlying these tasks is limited. Based on extensive literature we know that action observation and production share common neural mechanisms and a common neural network. Throughout the studies in this thesis, I use this connection to explore neural processes related to motor cognition.In the first chapter, I examine current theories on action observation and identify key concepts investigated in the later experiments. I review a wide range of literature from the fields of neural disorders and healthy participants tested on various tasks, such as visual illusions, motor expertise, and conscious and unconscious visual processing. Focusing on the bidirectional information flow between motor and perceptual areas, I examine how well the common coding theory, the direct matching hypothesis, and predictive coding models fit the current experimental results. I argue that, while predictive coding theories are best to explain the wide variety of results related to motor cognition, there are still important questions unanswered in the literature. There is a relative lack of studies investigating motor cognition in close-tonatural settings, and attentional modulation is often ignored. Furthermore, more research is needed to explore the spatial and temporal dynamics of how observed and executed actions present on the neural level.In the first experiment, I investigate how the brain processes actions when they are not consciously attended. I recorded brain activity related to action observation under an attentionally demanding visual task. Data indicate that even when our attention is directed away from actions both the motor and perceptual systems -as part of the action observation network- show a systematic change depending on the novelty of action properties. The results of this experiment also suggest that action related information is prioritised even when attentional resources are limited. In the second experiment, I examined how attention on specific action representations, such as kinematics, goals and agency, influences brain response during the processing of actions. The results of this experiment suggest that, even though novelty-related changes are very strong throughout in the action observation network, attention can specifically modulate neural activity to enhance the processing of task-specific information.Finally, in the third experiment, I investigated how a common system can deal with the parallel processing of action execution and observation. I aimed to give a comprehensive picture of neural activity related to motor cognition, and thus analysed event-related magnetic field activity and power changes in theta, alpha and beta frequency bands. Data indicated that neural processes are sensitive to conflict between observed and executed actions as early as 100ms after stimulus presentation. Furthermore, theta and beta frequency bands were found to be the most sensitive to the concurrent effects of action execution and observation, while the alpha frequency range showed neural processes related more to attentional mechanisms than to motor preparation.During these studies, I have attempted to test how the brain deals with motor cognition in everyday situations. I have described neural activation with fMRI and MEG and shown the involvement of a widespread network in motor cognition. I focused my investigation on attentional processes, and found that actions are processed even when attention is diverted away to a different demanding task. However, when actions are task-relevant and explicitly attended, task-specific areas are selectively enhanced to help discriminate effectively between sensory information. Indeed, a conflict between planned and observed actions is registered in the brain as rapidly as within 100ms. Furthermore, with a detailed description of brain oscillations and event related field changes, this thesis provides a comprehensive picture of how neuronal changes manifest on a temporal level during the early processing of an observed action. While this research clarified several important questions, further research is needed to explore one of the fundamental brain processes, motor cognition.