Randomized–controlled pilot study of tDCS and prism adaptation in adult ADHD: effects on working memory and executive functions

INTRODUCTION

Adult ADHD is characterized by persistent deficits in executive functions and working memory, associated with dysfunctions of fronto-parietal and cerebellar circuits involved in attention, inhibitory control, and visuospatial integration. Transcranial direct current stimulation (tDCS) over the dorsolateral prefrontal cortex (DLPFC) has been proposed to modulate cortical excitability and enhance executive control, with heterogeneous results. Prism adaptation (PA) is a bottom-up intervention based on visuo-motor conflict induced by prism displacement. In this protocol, prism deviation was directed to the right, inducing compensatory sensorimotor adaptation involving parietal and cerebellar circuits, with recalibration of visuospatial coordinates and transient increases in fronto-parietal network plasticity. The present pilot study evaluated combined tDCS over the left DLPFC and prism adaptation on executive functions and working memory in adults with ADHD, exploring effects as a function of cognitive load.

MATERIALS AND METHODS

This was an open-label, randomized, controlled pilot study conducted on 6 patients with ADHD, aged 17–35 years and with 8–18 years of education, with no major neurological comorbidities. Participants were randomly assigned to prism adaptation (PA) alone or combined tDCS + PA. The protocol consisted of 10 sessions. During the first 10 minutes, tDCS (2 mA) was applied with the anode over the left DLPFC (F3; Brodmann areas 46 and 9) and the cathode over the contralateral homologous region (F4), with electrode placement determined according to the international 10–20 EEG system.. Subsequently, participants performed prism adaptation (MindLenses) with visuo-motor reaching tasks under rightward prism displacement, followed by post-adaptation training. Cognitive assessment was conducted pre- and post-intervention across working memory and executive function domains. Analyses included descriptive statistics and between-group comparisons of change scores (Δ = post–pre) using the Mann–Whitney U test. Non-parametric effect sizes were calculated as r = |Z|/√N, with significance set at p < 0.05.

RESULTS

Between-group comparisons of change scores (Δ) showed differences between groups. Working memory (digit span) improved significantly in the tDCS + PA group compared to PA (Δ = 1.33 ± 0.57 vs −1 ± 1.73; p = 0.043; r = 0.83) whereas visuospatial working memory (backward Corsi blocks) was significantly worse in the tDCS + PA group (Δ = −1 ± 0 vs 0.6 ± 1.15; p = 0.034; r = 0.87). Additional measures showed a coherent trend, although not significant: phonological fluency showed lower improvement in the tDCS + PA group compared to PA (Δ = 4.67 ± 6.42 vs 7.3 ± 11.93; p = 1), while backward digit span showed a trend favoring PA (Δ = 0 ± 1 vs 0.66 ± 0.57; p = 0.346).

CONCLUSIONS

The combination of tDCS + PA appears to facilitate some aspects of working memory but to be less effective in high-load executive tasks. The data suggest load-dependent modulation consistent with non-linear effects. This pattern aligns with the inverted U-shaped model of neuromodulation, according to which tDCS effects depend on an optimal level of cortical excitability: both insufficient and excessive excitability may impair performance, particularly in cognitively demanding tasks. These findings are preliminary and require confirmation in larger samples.

References

No references