Individual Learning Trajectories During Adaptive Cognitive Training and Post-Intervention Executive Function in Middle-Aged Adults With a Family History of Alzheimer's Disease

Abstract Objectives. Cognitive training studies typically evaluate outcomes using average pre-post changes. This study examined whether individual learning trajectories during adaptive executive training were associated with post-intervention cognitive performance beyond baseline ability in cognitively unimpaired middle-aged adults with a family history of Alzheimer’s disease (AD). Method. Seventy-eight adults aged 50-65 years with a parental history of sporadic AD completed a 10-week adaptive computerized programme including task-switching and n-back training. Learning trajectories were estimated from repeated training performance using linear mixed-effects models. Individual learning slopes were evaluated in relation to post-intervention executive functioning, episodic memory, language, and global cognition after accounting for baseline performance. Post-intervention analyses used the complete-case sample (n = 68). Results. Significant incremental associations were confined to executive-function outcomes and involved task-switching learning trajectories. Task-switching level-efficiency slopes were associated with post-intervention TMT-A (ΔR² = .064, β = -.275, p = .038) and TMT-B completion time (ΔR² = .073, β = −.283, p = .022), whereas task-switching correct-response-efficiency slopes were associated with Stroop Word-Color performance (ΔR² = .043, β = .210, p = .028). No significant incremental associations were observed for episodic memory, language, or global cognition. The largest average pre-post changes occurred in episodic memory, despite the absence of significant trajectory associations with post-intervention memory performance. Conclusions. Learning trajectories obtained during training may provide information complementary to baseline assessment and average pre-post change. Their incremental associations were selective for executive-function outcomes, suggesting that repeated within-training measures may be useful behavioural indicators of individual training responsiveness in this midlife sample.