Effects of Anodal Transcranial Direct Current Stimulation on Working Memory: A Systematic Review and Meta-Analysis of Findings From Healthy and Neuropsychiatric Populations

- We performed a meta-analysis investigating working memory (WM) enhancement with anodal tDCS (a-tDCS) in healthy and neuropsychiatric cohorts.
- We examined both online and offline effects of stimulation.
- We explored the role of current density and stimulation duration on WM performance.
- Our results demonstrate mixed effects of a-tDCS on WM performance.
- A-tDCS enhanced offline WM reaction times in healthy populations, with a trend towards improvement for accuracy, while online WM accuracy in neuropsychiatric populations was improved. No other significant results were obtained.
- We provide some limited evidence that higher current densities and longer stimulation durations might be more effective at modulating WM.

BackgroundSeveral studies have trialled anodal transcranial direct current stimulation (a-tDCS) for the enhancement of working memory (WM) in both healthy and neuropsychiatric populations. However, the efficacy of this technique remains to be clearly established.ObjectiveThis review provides a quantitative synthesis of the published literature investigating the effects of a-tDCS, compared to sham, on WM, as assessed using the n-back, Sternberg and digit-span tasks. We also separated results from tasks performed 'online' (during stimulation) and 'offline' (following stimulation). The secondary aim was to assess for any additional effects of current density and stimulation duration.MethodsComprehensive literature searches were performed using MEDLINE, Embase, PsychINFO, CENTRAL and Scopus from July 1998 to June 2014.ResultsIn healthy cohorts, a-tDCS produced a trend towards improvement for offline WM accuracy (p = 0.05) and a small, but significant improvement in reaction time (p = 0.04); however, no significant effects were observed for online tasks (accuracy [p = 0.29], reaction time [p = 0.42]). In the neuropsychiatric cohort, a-tDCS significantly improved accuracy for online (p = 0.003), but not offline (p = 0.87) tasks, and no effect was seen for either online (p = 0.20) or offline (p = 0.49) reaction times. Secondary analyses controlling for current density and stimulation duration provided limited support for the role of these factors in influencing a-tDCS efficacy.ConclusionsThis review provides some evidence of a beneficial effect of a-tDCS on WM performance. However, the small effect sizes obtained, coupled with non-significant effects on several analyses require cautious interpretation and highlight the need for future research aimed at investigating more optimised stimulation approaches.