Age-related differences in intrinsic neural timescales (INTs): an analysis of autocorrelation windows (ACWs) in resting-state EEG across young and older adults

Autocorrelation windows (ACWs) quantify the temporal continuity of neural activity and have been linked to self-related processing and conscious experiences. Research suggests that longer ACWs during the resting state reflect a brain’s capacity to maintain temporal integration over extended periods. However, the influence of aging on these fundamental intrinsic neural timescales (INTs) remains largely unexplored, despite extensive evidence of age-related changes in brain structure and function. We analyzed resting-state EEG data from 196 healthy adults (137 younger: 20–35 years; 59 older: 59–77 years) from the Leipzig Study for Mind-Body-Emotion Interactions (LEMON) dataset. Three ACW measures were calculated across multiple electrode selection strategies: ACW-0, ACW-e, and ACW-50. Statistical analyses included parametric and non-parametric group comparisons, principal component analysis, and hierarchical modeling approaches with comprehensive methodological controls. Older adults demonstrated consistently shorter ACWs than younger adults (Cohen’s d range: -0.33 to -0.48, all p < 0.01), indicating decreased temporal continuity in neural activity with age. Parametric and non-parametric tests confirmed these effects. This effect was robust across multiple electrode selection criteria and represented a global brain-wide phenomenon (the first principal component explained 55–70% of the variance). These findings provide empirical evidence for robust age group differences in intrinsic neural timescales (INTs). Results demonstrate that older adults exhibit systematically shorter neural timescales compared to younger adults, with implications for understanding age-related changes in neural network organization.