Analytical bias accounts for some of the reported effects of tACS on auditory perception

Background: Transcranial alternating current stimulation (tACS) has been shown to modulate auditory, visual, cognitive and motor function. However, tACS effects can often be small and difficult to reproduce. Thus, the establishment of robust experimental and analysis procedures is of high importance. We reviewed the analysis used in six studies that investigated if tACS can phase-modulate auditory perception. All studies used analytical methods that introduce bias and could produce false positive results. Four studies corrected for this bias but two did not. Objective: Our objectives were two-fold: 1) Use simulated null hypothesis datasets, where no tACS effect is present, to determine if uncorrected analytical bias could account for some of the reported effects on auditory perception. 2) Help establish best practices to correct for bias when analyzing tACS phase effects on perception. Methods: We simulated null hypothesis datasets (i.e. no tACS effect) by drawing samples for all tACS and sham conditions from the same normal distribution. We then applied the reported analyses to the null hypothesis datasets. Results: Reported results from studies that did not correct for analytical bias could be reproduced from the null hypothesis datasets. However, results for studies that did correct for analytical bias could not be reproduced from the null hypothesis datasets. Conclusion: True effects of tACS on auditory perception can be detected if analytical bias is accounted for by using correction procedures. However, to fully establish the effects of tACS on auditory perception a reanalysis of the data for the studies that used biased analysis without correction procedures is needed. (C) 2019 Elsevier Inc. All rights reserved.