Alertness fluctuations when performing a task modulate cortical evoked responses to transcranial magnetic stimulation

Abstract

Transcranial magnetic stimulation (TMS) has been widely used in human cognitive neuroscience to examine the causal role of distinct cortical areas in perceptual, cognitive and motor functions. However, it is widely acknowledged that the effects of focal cortical stimulation can vary substantially between participants and even from trial to trial within individuals. Recent work from resting state functional magnetic resonance imaging (fMRI) studies has suggested that spontaneous fluctuations in alertness over a testing session can modulate the neural dynamics of cortical processing, even when participants remain awake and responsive to the task at hand. Here we investigated the extent to which spontaneous fluctuations in alertness during wake-to-sleep transition can account for the variability in neurophysiological responses to TMS. We combined single-pulse TMS with neural recording via electroencephalography (EEG) to quantify changes in motor and cortical reactivity with fluctuating levels of alertness defined objectively on the basis of ongoing brain activity. We observed rapid, non-linear changes in TMS-evoked responses with decreasing levels of alertness, even while participants remained responsive in the behavioural task. Specifically, we found that the amplitude of motor evoked potentials peaked during periods of EEG flattening, whereas TMS-evoked potentials increased and remained stable during EEG flattening and the subsequent occurrence of theta ripples that indicate the onset of NREM stage 1 sleep. Our findings suggest a rapid and complex reorganization of active neural networks in response to spontaneous fluctuations of alertness over relatively short periods of behavioural testing during wake-to-sleep transition.

Different underlying mechanisms for high and low arousal in probabilistic learning in humans

“…We fitted a mixed-effects model separately for RT variability and omissions as dependent variables. As predicted from other cognitive tasks, 17,18,30,33…”

“…Indeed, we have previously shown that decreased levels of arousal can fragment or reconfigure specific aspects of cognition while preserving crucial executive control processes such as the capacity to detect and react to incongruity, 18 the efficiency in perceptual decision making, 31 and the precision of conscious access. 17 Upon further examining the microstructure of learning derived from trial-by-trial performance of the PRL task, we uncovered the existence of prolonged periods of time-on-task in which participants did not seem to apply any particular high-order behavioural strategy.…”

“…12,13 These arousal fluctuations play a crucial role in modulating cognition, facilitating or hindering certain cognitive processes and performance to internal and external stimuli. 14,15,16,17,18…”

“…31 However, in drowsiness, and even during highly reduced arousal states, pre-attentive and early bottom-up attentive processing can still be accomplished with and without conscious awareness. 17,32,33…”

“…Based on the premises that (1) drowsiness hinders the extraction of task-relevant information from external stimuli and its integration, fragmenting specific aspects of cognition while preserving crucial executive control processes; 18,31,33,41 (2) drowsiness has been associated with more liberal decision-making; 17,30,31 (3) moderate-to-high intensity endurance exercise leads to a selective enhancement of executive control processes while lower and higher intensities result in an impairment or minimal effect; 40,42,43 and (4) high arousal promotes habitual responding and reduced engagement of complex cognitive strategies; 44,45,46 we predicted that behavioural performance would be enhanced in moderate-intensity physical exercise, while drowsiness and high-intensity exercise would lead to diminished performance in light of the inverted U-shaped Yerkes-Dodson Law. Specifically, we hypothesized that reduced arousal states would be associated with an impairment of performance (compared to baseline), which would be attributed to a tendency to apply a simple strategy (win-stay/lose-shift) instead of using an integrated history of choices and outcomes to drive performance (probabilistic switching behaviour).…”