No Benefit in Memory Performance after Nocturnal Memory Reactivation Coupled with Theta-tACS

Round 1

Reviewer 1 Report

This study used transcranial alternating current stimulation (tACS) to probe the hypothesized causal role of theta oscillations for targeted memory reactivation TMR during non-rapid eye movement (NREM) sleep. Participants learned Dutch-German word pairs before sleep, then were subjected to theta-tACS and control-tACS during NREM in a randomized order, in a within-subject design. Besides, one group of participants received tACS coupled with TMR time-locked two seconds after the reminder cue (time-locked group); the other group received continuous tACS while TMR cues were presented (continuous group). They were then tested on the next day on memory for nightly cued and uncued material. Contrary to the theoretical predictions, results failed to disclose a beneficial effect on memory performance of combined tACS – TMR, whatever the condition (time-locked or continuous, theta- or control-tACS). Rather, the interventions seemed to be detrimental as they blocked the TMR-induced memory benefits during sleep observed in prior studies from this group. The authors conclude that « tACS might have an unspecific blocking effect on memory-benefits typically observed after TMR during non-REM sleep ».

I enjoyed reading this interesting, clearly and fairly reported study, that contributes the available evidence about the effect of electrical stimulation on enhancement of cognitive functions, and in this case coupled with TMR during sleep. I have only three points that I believe the author should address.

My main concern is related to the peculiar choice to administer in succession the control- and theta-tACS (even if randomised) conditions, something that the authors should justify. Indeed, it can’t be determined with this design whether a specific stimulation condition was not detrimental to the entire consolidation process, or whether the fact to alternate TMR with two different types of frequency stimulations in a same night is not responsible for the lack of effects. In my view, it would have been more appropriate to compare across two separate nights the theta- and control-tACS conditions to avoid this potential caveat.

Regarding the EEG data analyses, the authors show a difference between Hits- and Misses- related theta activity during the presentation of cues during TMR, but it was not associated with any behavioural effect. If EEG recordings were also acquired during the learning session and especially during the immediate pre-sleep retrieval, it would be interesting to check that theta-related differences between Hit and Misses at delayed retrieval were not already present at that stage; in which case, differences observed during sleep TMR might be due to initial differences in the depth of encoding at learning, carried over during sleep .

Finally, at the behavioural level, retrieval results at the end of the pre-sleep learning session indicate that subjects achieved to learn +/- 50% of the material, and that they lost a few percent (but not much actually) over the night. Do the pre-sleep values correspond to the results obtained by the authors in their prior TMR studies? Also, as far as I have understood, both correctly retrieved and incorrectly retrieved stimuli at pre-sleep were presented during TMR, would it not have been more appropriate to use only the items correctly retrieved at the pre-sleep testing ?

Author Response

Please see attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Baselgia and colleagues aimed to determine whether coupling two different stimulation paradigms (continuous or time-locked)of transcranial alternative current stimulation (theta-tACS) to targeted memory reactivation (TMR) cues during non-REM sleep increased memory relative to a control-tACS among 37 healthy adults. They predicted that 1) both stimulation paradigms would increase memory and 2) greater theta activity during the cueing of later remembered words. In contrast to their expectations, neither TMR nor the stimulation paradigms (continuous or tine-locked) increased memory. The inability of TMR to increase memory here paralleled their prior publication [39]. Thus, the rationale for coupling theta-tACS to TMR without first observing the expected effect on memory is unclear. Moreover, without an effect of TMR or either of the stimulation paradigms and the noted significant differences between the paradigms (i.e., time in bed, slow wave sleep and time awake after sleep onset), it is difficult to understand the results for the oscillatory responses etc. Last, section 4.2 in the methods indicated that subjective sleep quality was measured and a post-stimulation questionnaire was administered. This section indicated that the latter was ‘used to evaluate whether some effect of the stimulation had been felt by the participants during sleep”. Was subjective sleep quality poor (Pittsburgh Sleep Quality Index) and was the an indication that the stimulation had been felt by the participants during sleep – did either of these contribute to the lack of the effect of TMR etc. 

very minor editing needed.

Author Response

Please see the attachment. 

Author Response File: Author Response.pdf