AUTHOR=Fischer Florian U. , Kollmann Bianca , Wolf Dominik , Sebastian Alexandra , Knaepen Kristel , Riedel David , Mierau Andreas , Ruffini Nicolas , Endres Kristina , Winter Jennifer , Strüder Heiko K. , Bischof Gerard N. , Faraza Sofia , Baier Bernhard , Binder Harald , Drzezga Alexander , Teipel Stefan , Fellgiebel Andreas , Tüscher Oliver 

TITLE=Cognitive training gain transfer in cognitively healthy aging: per protocol results of the German AgeGain study

JOURNAL=Frontiers in Aging Neuroscience

VOLUME=Volume 17 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/aging-neuroscience/articles/10.3389/fnagi.2025.1587395

DOI=10.3389/fnagi.2025.1587395

ISSN=1663-4365

ABSTRACT=IntroductionCognitive decline is part of the normal aging process, but also a major risk factor for dementia. Cognitive training interventions aim to attenuate cognitive decline, but training gains need to be transferable to untrained cognitive abilities to influence everyday function. Furthermore, the neurobiological basis of cognitive training gain transfer remains elusive. A possible candidate is increased bilateral hemisphere usage enabled by efficient structural connectivity, especially of prefrontal regions. Therefore, the present multicentric study used a cognitive training intervention to demonstrate training transfer and identify neurobiological modulators of successful transfer.MethodsIn total 235 subjects were enrolled in AgeGain; 180 underwent a broad 4-week cognitive training intervention at three study sites. Pre- and post-training neuropsychological testing was conducted and successful transferers were identified according to preregistered definitions. Pre-training, subjects underwent diffusion and functional MRI to assess interhemispheric connectivity, measured as microstructural integrity of the corpus callosum and lateralization of functional activation patterns during a cognitive control task. Logistic regression models were estimated to predict successful transfer based on structural connectivity and bilateralization of activation patterns.ResultsOut of 180 subjects, 74 showed short-term training gain transfer that was maintained over 3 months in 19 subjects. Neither microstructural integrity of the corpus callosum, nor bilateralized activation predicted training gain transfer alone. However, their interaction was associated with long-term transfer over 3 months: subjects with higher mean diffusivity of the corpus callosum and more bilateral functional activity or conversely with lower diffusivity of the corpus callosum and more lateral functional activity were more likely successful long-term transferers.DiscussionWe demonstrated successful training gain transfer in 41.1% of subjects, among whom 25.7% maintained the transfer over 3 months. Successful long-term transfer of training gains may depend on divergent mechanisms of structural and functional connectivity, which may explain previous heterogeneous results in the literature.Trial registerGerman Clinical Trials Register (DRKS), ID: DRKS00013077. Registered on November 19th 2017.