Elusive Gains of Cognitive Training: Limited Effects on Neural Activity Across Sessions

Cognitive training paradigms rely on the idea that consistent practice can drive neural plasticity, improving not only connectivity within critical brain networks, but also ultimately result in overall enhancement of trained cognitive functions, irrespective of the specific task. Here we opted to investigate the temporal dynamics of neural activity and cognitive performance during a structured cognitive training program.

A group of 20 middle-aged participants completed 20 training sessions over 10 weeks. Quantitative EEG (qEEG) parameters, including alpha and theta power, alpha/theta ratio, and fronto-parietal coherence, were analyzed at four time points to assess changes in neural activity.

Results revealed significant overall improvements in the trained task (n-back) performance, without an effect on the untrained task (OSPAN). qEEG analyses showed increased change in posterior (and a less robust in frontal) alpha power, particularly during mid-training, suggesting an improved neural efficiency in regions associated with attentional allocation and task engagement. Theta power remained stable across sessions, indicating a limited influence on neural processes underlying working memory and attentional control. The parietal alpha/theta ratio showed weak increases during mid-training, reflecting subtle shifts in the neural efficacy and cognitive engagement. There were no significant changes in functional connectivity between frontal and parietal locations.

Our findings suggest that cognitive training primarily influences localized neural activity, rather than network-level connectivity. This lack of a longer-range network-level effect might also explain the failure of cognitive training paradigms to induce performance enhancements on the untrained tasks.