Closed‐Loop Targeted Memory Reactivation During Sleep Spindles Enhances Motor Memory Consolidation

Published October 23, 2026. A sound replayed when EEG detected sleep spindles was followed by greater overnight improvement for its paired movement sequence than for an uncued sequence. This was a small healthy-young-adult experiment, not evidence for treating memory impairment or improving broad cognition.

Gupta P, Mutreja V, Lungu O, Lazzouni L, Gabitov E, Benali H, Jourde HR, Beltrame G, Coffey EBJ, Lina JM, Albouy G, King BR, Boutin A, Carrier J, Doyon J.
Journal of Sleep Research · Publication date: 2026-10-23
Added to site: 2026-10-25
DOI: 10.1111/jsr.70468

Published October 23, 2026. Added to Neurofeedback.com October 25, 2026.

Modalities

EEG, Closed-loop stimulation, Targeted memory reactivation

Clinical focus

Sleep spindles, Motor memory consolidation, Brain-state timing

Population

Young adults learning two explicit finger-movement sequences

Sample

26 participants

Intervention / method

Two eight-element finger sequences were paired with distinct sounds; during subsequent NREM stage 2 and 3 sleep, a portable deep-learning EEG system detected spindles and replayed the cue for one sequence while the other sequence remained uncued

Comparator

Within-participant cued sequence versus a non-reactivated sequence

Clinician summary

Spindle-locked cueing produced greater overnight motor-sequence improvement than the within-person non-TMR condition. Benefit correlated positively with the proportion of cues delivered during true spindles and spindle trains, while a higher proportion during isolated spindles was associated with smaller gains. The abstract does not provide the between-condition effect estimate or confidence interval.

How to interpret the findings

Only 26 young adults participated, and the checked abstract does not report randomization of cue assignment, participant or assessor masking, sleep disruption, adverse effects, exact behavioral effect size, confidence intervals or retention beyond the next assessment. The outcome was a specific trained finger sequence, so general memory, clinical and real-world enhancement claims are unsupported.

Equipment and disclosures

The study used the named portable Portiloop deep-learning EEG device for online spindle detection. The source does not attribute equipment to BrainMaster or establish device effectiveness beyond this experimental task.

The retained Crossref record lists the Canadian Institutes of Health Research as a funder. Competing-interest details were not available in the checked PubMed abstract. No indexed correction, retraction or expression-of-concern relation appeared in the retained PubMed or Crossref records; this is a bounded check. AI-assisted source review.

Related evidence

Compare findings from the same clinical area, including mixed results.