The Caveats of observing Inter-Trial Phase-Coherence in Cognitive Neuroscience

Abstract

Many studies have now consistently reported that the phase angle of ongoing oscillatory activity (measured using EEG/MEG), at time of stimulus presentation influences detection when stimuli are near-threshold. However, studies examining whether the adjustment of the phase angle of oscillations is under top-down attentional control have thus far yielded conflicting results. A possible source for the discrepancy could be that the estimation of the phase of ongoing oscillations as well as its uniformity across trials could be affected by task induced changes in the power of oscillations or concurrent evoked responses. One measure, Inter-Trial Phase-Locking (ITPC), or the uniformity of phase angles across trials, is particularly vulnerable to these factors. Here, using various simulations modelling the common task induced changes in the EEG reported in the literature, we demonstrate that apparent changes in Inter-Trial Phase-Locking of oscillatory activity can occur independent of any actual change in the phase of the ongoing activity.

Stimulus-driven brain rhythms within the alpha band: The attentional-modulation conundrum

“…Note that both formulas only differ in the order in which weighted sums and absolute values are computed. Also note that formula [2] is highly similar to the calculation of intertrial phase coherence (ITC), a popular measure of phase locking (Cohen, 2014; Gross, 2014; van Diepen and Mazaheri, 2018). ITC calculation additionally includes a trial-by-trial amplitude normalisation.…”

Sleepless and Desynchronized: Impaired Inter Trial Phase Coherence of Steady-State Potentials Following Sleep Deprivation

“…Our analysis revealed that both the task-related amplitude and the phase coherence of the ssVEP response were decreased under sleep deprivation, and that this decrease was strongly related among the two measures. This finding raises the possibility that the observed reduction in ITPC results from the weaker task-evoked signal measured in sleep deprivation compared to sleep rested (van Diepen and Mazaheri, 2018). Although the possibility that the ITPC differences arise from amplitude differences cannot be completely ruled out, our findings point towards decreased ITPC as a reliable indicator of sleep deprivation, as demonstrated by its correlation with well established markers for sleep deprivation (PVT and theta power), even when statistically controlling for evoked amplitude levels.…”

“…Fluctuations between high and low cortical excitability states are represented by the phase of neural oscillations, which is the time-varying angle of the oscillatory signal (Buzsáki, 2010;Buzsáki and Watson, 2012;Klimesch et al, 2007;Lakatos et al, 2007Lakatos et al, , 2008Thut et al, 2012). In humans, inter trial phase locking of neural oscillations, also referred to as inter trial phase coherence (ITPC), is a measure of phase consistency of the neural response over experimental trials (van Diepen and Mazaheri, 2018).…”

“…The ssVEPs are typically recorded in occipital electrodes and are measured as oscillatory waveforms, stable in phase and amplitude (Regan, 1966), with the same frequency peak as the delivered stimulus. Thus, ssVEPs are relatively narrow band oscillations and are advantageous in their excellent signal to noise ratio, which is essential for a reliable phase coherence estimation (Tallon-Baudry et al, 1996;van Diepen and Mazaheri, 2018). In addition, ssVEPs can be detected on a single trial level (Vialatte et al, 2010), which empowers their relevance for behavioral performance.…”

Inter-trial phase coherence of visually evoked postural responses in virtual reality

“…Phase coherence, in general, constitutes a measure of temporal synchronicity between a set of oscillations. In EEGrelated studies, a phase-lock value to determine synchronization between two neuroelectric signals was introduced by Lachaux et al (1999) and subsequently led to intertrial phase coherence (ITPC) being established as a method to investigate phase synchronicity between trials (Cohen 2014;van Diepen and Mazaheri 2018). Analogously, we used the data achieved through the wavelet analysis to investigate phase relationships of all frequency bands across trials for each subject and condition.…”