Neural mechanisms underlying external distraction by unexpected environmental stimuli or by a secondary task: an intracranial EEG investigation

In our everyday lives, efficiency and ease crucially depend on our ability to focus on what we are doing at any given time. From a neuropsychological point of view, it means that we are constantly "on‐task", and that our brain always engages resources that are both necessarily and sufficient for the task at hand: at the neural level, this is characterized by a fine balance between activations and inhibitions to ensure that only perceptual, cognitive and motor processes relevant for the task are active, at the exclusion of any other. Yet, practically, such ideal patterns almost never occur because we are almost constantly distracted, most obviously by events in the environment around us, but also by spontaneous thoughts and fruitless attempts to perform several tasks at the same time. This thesis tries to reveal the deep reasons why distraction has such a profound, detrimental effect on attention and performance. Why is it so hard to stay on task in a noisy environment? Why does performance drop when we try to perform two difficult attention‐demanding tasks at the same time? Those questions have already been debated for years, and modelled in many ways, but no study so far had conducted the investigation at the deepest level of the fine neural dynamics that supports our ability to focus. We used the most precise recordings of a living human brain ‐ intracranial EEG recordings with millisecond and milimetric resolution ‐ to examine in two experiments how external distractions and multi‐tasking interfere with the optimal dynamics of a demanding, continuous attention task, and we addressed those questions in naturalistic settings, in the context of a direct social interaction to ensure our conclusions extend to real‐life situations. We found that the tight balance between excitation and inhibition is disrupted in key regions supporting attention and executive control for short periods of time, at the subsecond level, but sufficiently to knock the network off‐balance and impair performance. Altogether, our results provide explanations for many of the failures of attention of our modern lives and pave the way for new techniques to avoid them