生成AI利用と顔表情由来の中央実行ネットワーク指標との関連：パイロットスタディ

Background/Objectives: The rapid diffusion of generative AI has raised concerns about its potential influence on human cognition, particularly during creative work. This pilot study explored task-related associations between generative AI use and facial expression-derived indices that have previously been shown to correlate with gray matter volume in the default mode network (DMN) and central executive network (CEN). Methods: Thirty-three business professionals completed three AI-supported writing tasks involving concept generation, concept combination, and a mixed task. Results: The results showed a statistically robust reduction in the CEN-related facial expression index during the concept combination task, whereas no corrected changes were observed during concept generation or the mixed task. In addition, higher creative self-efficacy was associated with smaller reductions in the CEN-related index. Conclusions: Given the indirect nature of the facial expression measures, the absence of a control condition, and the exploratory design, these findings should be interpreted cautiously and primarily as hypothesis-generating. Future research using controlled designs and direct neuroimaging methods is needed to clarify the cognitive and neural mechanisms underlying AI-assisted creativity.