How to test for phasic modulation of neural and behavioural responses

Benedikt Zoefel,Matthew H Davis,Giancarlo Valente,Lars Riecke

doi:10.1016/j.neuroimage.2019.116175

Benedikt Zoefel, Matthew H Davis + Show 2 more

Open Access

https://doi.org/10.1016/j.neuroimage.2019.116175

Copy DOI

Abstract

Research on whether perception or other processes depend on the phase of neural oscillations is rapidly gaining popularity. However, it is unknown which methods are optimally suited to evaluate the hypothesized phase effect. Using a simulation approach, we here test the ability of different methods to detect such an effect on dichotomous (e.g., “hit” vs “miss”) and continuous (e.g., scalp potentials) response variables. We manipulated parameters that characterise the phase effect or define the experimental approach to test for this effect. For each parameter combination and response variable, we identified an optimal method. We found that methods regressing single-trial responses on circular (sine and cosine) predictors perform best for all of the simulated parameters, regardless of the nature of the response variable (dichotomous or continuous). In sum, our study lays a foundation for optimized experimental designs and analyses in future studies investigating the role of phase for neural and behavioural responses. We provide MATLAB code for the statistical methods tested.

Highlights

Neural oscillations are cyclic variations in the excitability of neuronal ensembles
Stimuli are often presented at an a-priori unknown, random neural phase in each trial; this leads to the possibility of testing how neural or behavioural responses relate to the phase of spontaneous oscillations extracted post-hoc from concurrent EEG/MEG recordings
Most methods in the other categories do not divide data into phase bins. These methods were not affected by the number of possible phases in each trial, reflected by Nbins for phases imposed externally. These findings show that the highest possible sensitivity to detect phase effects can be achieved by using (1) regression- or permutation-based methods and (2) a high Ntrials

Summary

Introduction

Neural oscillations are cyclic variations in the excitability of neuronal ensembles. Oscillatory phase indexes the instantaneous state of excitability (Buzsaki and Draguhn, 2004) and correlates with neuronal firing in intracranial recordings (Kayser et al, 2015). In both the visual and somatosensory systems, the detection of a near-threshold stimulus and the ability to distinguish two rapidly presented stimuli correlate with EEG/MEG phase (Ai and Ro, 2014; Baumgarten et al, 2015; Busch et al, 2009; Mathewson et al, 2009; Milton and Pleydell-Pearce, 2016; Ronconi et al, 2017; but see Ruzzoli et al, 2019). Imposing phase externally in an experimentally controlled manner allows testing for causal effects of phase, but often limits the number of discrete phase values that can be tested

Methods

Results

Conclusion