Prefrontal theta-gamma transcranial alternating current stimulation improves non-declarative visuomotor learning in older adults

Abstract

The rise in the global population of older adults underscores the significance to investigate age-related cognitive disorders and develop early treatment modalities. Previous research suggests that non-invasive transcranial Alternating Current Stimulation (tACS) can moderately improve cognitive decline in older adults. However, non-declarative cognition has received relatively less attention. This study investigates whether repeated (16-day) bilateral theta-gamma cross-frequency tACS targeting the Dorsolateral Prefrontal Cortex (DLPFC) enhances non-declarative memory. Computerized cognitive training was applied alongside stimulation to control for the state-of-the-brain. The Alternating Serial Reaction Time (ASRT) task was employed to assess non-declarative functions such as visuomotor skill and probabilistic sequence learning. Results from 35 participants aged 55-82 indicated that active tACS led to more substantial improvements in visuomotor skills immediately after treatment, which persisted 3 months later, compared to sham tACS. Treatment benefit was more pronounced in older adults of younger age and those with pre-existing cognitive decline. However, neither intervention group exhibited modulation of probabilistic sequence learning. These results suggest that repeated theta-gamma tACS can selectively improve distinct non-declarative cognitive aspects when targeting the DLPFC. Our findings highlight the therapeutic potential of tACS in addressing deficits in learning and retaining general skills, which could have a positive impact on the quality of life for cognitively impaired older individuals by preserving independence in daily activities.

Keywords: Aging; Cognition; Cross-frequency coupling; Elderly; Implicit learning; Non-declarative memory; Transcranial alternating current stimulation (tACS).

Figure 1

Study design. In this triple-blinded, placebo-controlled, parallel-group research study, participants came to the clinic for 19 appointments. The first appointment included history taking, implementation of the Montreal Cognitive Assessment (MoCA), and an initial administration of the Alternating Serial Reaction Time (ASRT) task (outlined in red). At each of the subsequent 16 appointments, participants received 20 min of active or sham tACS treatment bilaterally targeting the dorsolateral prefrontal cortex (F3/F4, according to the international 10–20 system) while completing Computerized Cognitive Training (CCT) based on the n-back paradigm on a tablet (outlined in blue). After completion of treatment, two further appointments were held to examine direct treatment effects and long-term effects (outlined in green), separated by a 3-month break (outlined in yellow).

Figure 2

The Alternating Serial Reaction Time (ASRT) task. (a) Participants were placed in front of a screen and provided with a keyboard containing only four keys. Each key corresponded to one of four circles displayed horizontally on the screen. When a stimulus (a dog's head) appeared in any of the circles, the participant's task was to press the corresponding key as fast as possible. Once the correct key was pressed, a new stimulus appeared. The position of the stimulus alternated between predetermined pattern (P, yellow background) and random (R, red background) elements. (b) The predetermined pattern elements were part of an eight-element sequence, in which they alternated with random elements (e.g. 3-R-2-R-4-R-1R). The numbers represented the corresponding position on the screen in ascending order from left to right. Every trial was characterized as the third element of the underlying triplet (three consecutive trials). Due to the probabilistic sequence structure, this results in so-called high-probability triplets, which occur with a probability of 62.5%, and low-probability triplets, which occur with a probability of 37.5%. (b) One ASRT session (green background) consisted of 20 blocks (blue background), each containing 80 trials (yellow and red background), which was ten times the repetition of the eight-element sequence. Between blocks, participants could take self-paced breaks (purple background).

Figure 3

Active theta—gamma transcranial Alternating Current Stimulation (tACS) boosts offline visuomotor skill learning. (a) Participants from the active tACS group (blue line, n = 17) exhibited greater decrease in reaction time from baseline to the post-treatment and to the follow-up (3 months later) assessments, compared to participants from the sham tACS group (orange line, n = 18). (b) This resulted in greater direct (measured as the RT difference between baseline and post-treatment) as well as long-term (measured as the RT difference between baseline and follow-up) offline visuomotor learning in the active tACS group (blue bars), compared to the sham tACS group (orange bars). The error bars represent the standard error of the mean (SEM).

Figure 4

Active tACS treatment is more effective in younger and cognitively weaker participants. (a) In the active tACS group (upper right plot), older adults of younger age (≤ 68 years, light blue line, n = 18) revealed higher offline visuomotor learning than those of older age (> 68 years, purple line, n = 17) 3 months after treatment completion, whereas in the sham tACS group (upper left plot), both age groups showed the same level of offline visuomotor learning 3 months after treatment. (b) In the active tACS group (lower right plot), participants with MoCA (i.e. Montreal Cognitive Assessment) scores in the range of Mild Cognitive Impairment (MCI) (< 26, orange line, n = 17) revealed higher offline visuomotor learning than participants with MoCA scores in the healthy range (≥ 26, dark green line, n = 18) 3 months after treatment completion, whereas in the sham tACS group (lower left plot), both cognitively healthy and cognitively impaired participants exhibited the same level of offline visuomotor learning 3 months after treatment. The error bars represent the standard error of the mean (SEM).