Autobiographical memory (AM) is a core component of human cognition; it defines who we are, it helps us relate to others, and supports decision-making and future planning. The variability embedded in AMs offers a unique opportunity to examine the neural dynamics engaged during the retrieval of memories of different ages. Here, we focus on the early access (or search) phase and test whether theta-band (4-9 Hz) cortical activity carries age-sensitive signatures distinguishing recent (<1 year) from remote (>3 years) AMs. Forty-one participants performed an AM retrieval task while EEG was recorded. Theta power and time-resolved Granger causality (GC) were quantified during the first 1200 ms of the access period. Remote AM access elicited a significant increase in midline fronto-central theta power around 900 ms, consistent with enhanced internally driven, reconstructive processing. GC analyses further revealed that remote AMs exhibited denser, more widespread theta-band connectivity, including a prominent anterior-to-posterior information flow between 600-800 ms, whereas recent AMs showed more localized fronto-central interactions. These findings demonstrate that the earliest phase of AM access is shaped by memory age, with remote memories recruiting broader, long-range theta-mediated coordination. Our results identify theta oscillations and their directed network interactions as temporally precise markers of AM remoteness, providing new insights into the systems-level mechanisms that support autobiographical retrieval.
It’s About Time: The Temporal and Modal Dynamics of Copilot Usage
arXiv:2512.11879v1 Announce Type: cross Abstract: We analyze 37.5 million deidentified conversations with Microsoft’s Copilot between January and September 2025. Unlike prior analyses of AI usage,




