Sham tDCS controls: blinding, reliability, and a specification-grade checklist.

Abstract

Sham transcranial direct current stimulation (tDCS) is essential for causal inference but is commonly implemented as ramp-only protocols that may not ensure physiological inertness and may fail to maintain mid-session blinding. We synthesize evidence on sham blinding integrity and test-retest reliability from January 1, 2010, to August 31, 2024, with a targeted post-search update (September 2024-September 2025) for key methodological advances directly informing sham design, blinding assessment, or physiological inertness. Evidence across conventional fade-in/short-stimulation/fade-out (FSF) approaches and emerging HD and device-locked alternatives suggests that end-of-study "guess" measures frequently overestimate blinding, whereas time-resolved probes detect perceptual divergence during stimulation. Test-retest reliability under sham conditions is generally adequate across common behavioral tasks, although practice effects and procedural heterogeneity can reduce stability and inflate variance. Based on recurrent methodological failure factors identified in the mapped literature, we propose five specification-grade recommendations: (i) use time-resolved (not end-of-study) blinding assessment with pre-specified success criteria; (ii) constrain intracranial E-fields below a conservative engineering ceiling using accessible modeling tools; (iii) standardize ramp profiles and electrode preparation and impedance protocols; (iv) implement device-locked randomization with audit trails; and (v) preregister blinding thresholds and analysis plans. These practices operationalize sham quality as measurable outcomes that can reduce trial variability and strengthen cross-study synthesis in mechanistic and clinical research.