Predicting Working Memory Training Benefits From Transcranial Direct Current Stimulation Using Resting-State fMRI

The effects of transcranial direct current stimulation (tDCS) on working memory (WM) performance are promising but variable and contested. In particular, designs involving one session of tDCS are prone to variable outcomes with notable effects of individual differences. Some participants benefit, whereas others are impaired by the same tDCS protocol. In contrast, protocols including multiple sessions of tDCS more consistently report WM improvement across participants. The objective of the current project was to test whether differences in resting-state connectivity between stimulation site and two WM-relevant networks [default mode network (DMN) and central executive network (CEN)] could account for initial and longitudinal responses to tDCS. Healthy young adults completed 5 days of visual WM training during sham or anodal right frontal tDCS. The behavioral data showed that only the active tDCS group significantly improved over the visual WM training period. There were no significant correlations between initial response to tDCS and resting-state activity. DMN activity in the anterior cingulate cortex significantly correlated with WM training slope. These data underscore the importance of sampling in studies applying tDCS; homogeneity (e.g., of gender, special population, and WM capacity) may produce more consistent data in a single experiment with limited power, whereas heterogeneity is important in determining the mechanism(s) and potential for tDCS-linked protocols. This issue is a limitation in tDCS findings that continues to hamper its optimization and translational value.