Neuronal networks exhibit oscillatory activity at various frequencies, which are thought to coordinate local neuronal processing and cortico-cortical communication through alternating periods of excitation and inhibition. In the motor cortex, single neuron spiking rates (Haegens et al., 2011) and neuronal population activity (Miller et al., 2012) are highest at mu and beta oscillatory troughs (maximum surface negativity) and lowest at peaks (maximum surface positivity). This relationship is strongest when subjects are at rest, when power in these frequency bands is known to be high (Pfurtscheller and Da Silva, 1999). Yet, how sensorimotor oscillatory phase and power interact to influence corticospinal excitability in intact humans is not known. Here, we explored this issue in an open-loop TMS design (i.e., timing of TMS delivery was determined by the passage of time, but not brain oscillatory state), focusing on mu and beta oscillatory activity due to their relevance to human motor control. We collected 600 MEPs for each of 20 healthy subjects using single-pulse TMS (Magstim, UK) at 120% or resting motor threshold delivered to the scalp hotspot for the right first dorsal interosseous during simultaneous EEG recordings (BrainProducts GmbH, Germany). Subsequently, data from the C4 sensor were Hjorth-transformed (central: C4; surround: FC2, FC6, CP2, CP6) and the instantaneous oscillatory phase and pre-stimulus power within 150 ms of the TMS pulse in the mu and beta bands was calculated. Peak and trough trials were defined as trials where TMS occurred at maximum surface positivity (45–135°) and maximum surface negativity (225–315°) of the oscillatory cycle, respectively. The influence of sensorimotor oscillatory phase (categorical variable, peak vs. trough), power (continuous variable) and their interaction on MEP amplitudes was statistically evaluated using separate trial-by-trial linear mixed-effects models for each frequency band. For mu activity, there was a significant PHASE x POWER interaction (p = 0.001), evident as a significantly more positive relationship between pre-stimulus mu power and MEP amplitudes for troughs than peaks (mu peak slope = −0.001 [95% CI = −0.029 to 0.027]; mu trough slope = 0.063 [95% CI = 0.034 to 0.090]). Notably, when mu power was high, MEPs were larger at troughs versus peaks, while the opposite was true when mu power was low. For beta activity, there was a significant main effect of POWER (p = 0.007), revealing a positive relationship between pre-stimulus beta power and MEP amplitudes independent of beta phase. Corticospinal excitability was highest at mu troughs during periods of high mu power, and during periods of high beta power regardless of beta phase. We conclude that sensorimotor oscillatory power and phase interact to shape human motor function, supporting the view that sensorimotor rhythms dynamically gate corticospinal excitability.
Read full abstract