Feed

Experience-dependent modulation of the visual evoked potential (VEP) can be used as a non-invasive human correlate of invasive measurements of long-term potentiation (LTP)-type neuroplasticity in animal research. However, conventional electroencephalogram (EEG) recording is high burden and time consuming, hindering adoption. Further, reliability of participant-level VEP modulation measures, even in highly controlled conditions, has been disappointing. To increase the practicality and utility of the assessment we trialled a short (11-min) paradigm with wireless dry-EEG headset and applied multiscale frequency-domain analyses to extract more precise measures of neuroplasticity. This approach was tested in healthy participants in laboratory (n = 17) and Phase 1 clinical studies (n = 33). Typical session duration was < 30 min. Test-retest reliability of VEP waveforms were compared to literature on full-length paradigms with wet-EEG hardware. Group-level time-domain event-related potential (ERP) analyses showed post-modulation P1 amplitude increase in both groups (Cohen's d = 0.77, 0.78, respectively) and N1b amplitude decrease in clinical study group (Cohen's d = - 0.52). Timeseries-based test-retest reliability of the VEP modulation effect was low but was substantially improved to moderate levels (some intraclass correlation ≥ 0.5) with a wavelet-based analysis. This suggests that VEP modulation assessments could be employed across multiple sites to objectively measure experience-dependent plasticity in real-world clinical trials of central nervous system therapies.