Aging Brains Work Harder to Answer the Question: Who Am I?

Who we are is not a fixed thing. Psychologists have long argued that identity is a lifelong project, continuously revised as experiences accumulate, relationships shift, and priorities change. Yet while the neuroscience of self-knowledge has flourished in recent decades, almost all of that work has been conducted on young adults, typically university students in their twenties. What happens to the brain’s machinery of self-reflection as we age into healthy older adulthood has remained largely uncharted territory. A new electroencephalography study published in GeroScience by Marta Paź, Maciej Kamiński, and Anna Nowicka offers one of the most detailed looks yet at how the aging brain handles the deceptively simple task of deciding whether a personality trait describes you.

The researchers recruited 85 right-handed-dominant participants, split into 42 younger adults aged 20 to 35 and 43 older adults aged 60 to 79, all community-dwelling and free of neurological or psychiatric disorders. Each participant sat in a dark, acoustically shielded room while 150 trait adjectives flashed on a screen, one every few seconds. The words, drawn from a standardized list of personality descriptors and carefully balanced for positive, negative, and neutral valence, had to be judged as suitable or unsuitable for describing three different targets: the self, a personally chosen close other such as a partner or child, and a favorite famous person selected from a list of Polish celebrities. Sixty-two scalp electrodes recorded the brain’s millisecond-by-millisecond response throughout, allowing the team to track both the timing of neural events and the direction of information flow between brain regions.

The study’s design was deliberately built around two well-characterized electrical signatures. The first is the P1, a positive deflection peaking roughly 100 milliseconds after a stimulus appears, generated in extrastriate visual cortex and famously amplified by attention. The second is the late positive potential, or LPP, a broad frontal-central-parietal wave emerging around 400 to 500 milliseconds that indexes a global, temporary boost of attentional resources devoted to salient or motivationally significant stimuli. Decades of work show that self-relevant material, from one’s own face to self-descriptive words, reliably inflates the LPP. By measuring both components across the three reflection targets, the team could separate early perceptual gating from later evaluative processing and ask whether aging alters one, the other, or both.

The behavioral results already hinted at something interesting. Younger participants endorsed more traits as self-descriptive than they did for the close other or the famous person, a classic self-prioritization effect, while older adults distributed their yes answers more evenly across targets. Reaction times told a familiar story of generalized slowing: older adults were slower overall, and the gap was widest precisely in the self condition, where median responses stretched to nearly 1.2 seconds compared with about 940 milliseconds in the young group. Given that slower responses in the self condition appeared in both age groups relative to the famous-person condition, the data suggest that judging one’s own character is inherently the most demanding version of the task, and that aging amplifies that demand.

The electrophysiology sharpened the picture considerably. At the occipital electrode O1, P1 amplitudes were significantly reduced in older adults across all three conditions, with no condition-specific modulation in either group. This replicates a large literature showing that early visual attentional processing is among the first casualties of normal aging, and it indicates that older brains engage less early attentional amplification when processing socially relevant words, regardless of who those words describe. In other words, the age-related P1 reduction was a blanket effect, not a selective weakening of the self-bias at the perceptual stage.

The LPP results, by contrast, revealed a striking reversal. When the researchers pooled frontal electrodes F1, F3, and F5 and analyzed mean amplitudes in the 400 to 800 millisecond window, they found that self-referential traits produced significantly larger LPPs in older adults than in younger ones, while the close-other and famous-person conditions showed no age differences at all. The authors offer two compatible interpretations. The enhanced LPP may reflect stronger subjective emotional responses to self-related traits in older adults, consistent with developmental theories holding that aging shifts emotional priorities toward internally meaningful information and the self. Alternatively, or additionally, it may signal increased cognitive effort: older adults may need to recruit extra attentional and evaluative resources to accomplish self-judgments that younger brains perform more cheaply, a pattern consistent with compensation models of cognitive aging such as the CRUNCH hypothesis.

The most technically ambitious part of the study went beyond surface potentials entirely. Using the full-frequency directed transfer function, or ffDTF, a Granger-causality measure fitted through multivariate autoregressive modeling, the team estimated the direction and strength of information flow between three scalp-defined regions of interest: a visual input region, a posterior parietal region, and an anterior frontal region. Unlike fMRI connectivity, which is inherently correlational, this approach can in principle distinguish a signal traveling from sensory cortex toward frontal evaluative systems from the reverse, top-down feedback. The electrodes were grouped based on converging neuroimaging evidence linking these territories to default mode network activity and to the distributed networks underlying the LPP, though the authors caution that these are scalp-level approximations rather than direct measurements of anatomical network nodes.

The connectivity findings were unambiguous in their overall direction. In all three analyzed time windows, from stimulus onset out to 1200 milliseconds, younger adults showed stronger directed connectivity than older adults. The most robust, multiple-comparison-corrected effects emerged in the intermediate 400 to 800 millisecond window, where younger participants showed stronger feed-forward signaling from visual to anterior regions, from posterior to anterior regions, and from posterior back to visual regions, along with stronger local connectivity within the posterior region itself. This mid-latency concentration suggests a specific processing stage, associated with higher-order perceptual integration, where age-related communication deficits bite hardest. The pattern echoes prior fMRI work showing attenuated default mode network connectivity during self-referential judgments in older adults, but adds the crucial element of directionality that correlational methods cannot provide.

What biological forces might underlie this weakening of neural dialogue? The authors point to a convergence of mechanisms documented in the aging literature: declining white matter integrity that slows conduction between distant regions, loss of dendritic spines and synaptic density, impaired synaptic plasticity, and increased neural noise that degrades the signal-to-noise ratio of cortical communication. Together these support the cortical disconnection hypothesis of aging, which holds that distributed neural systems become slower, less precise, and less coordinated with advancing age. The full picture that emerges is one of paradox: an aging brain that transmits information less efficiently across its networks, yet invests more attentional and emotional resources when the subject of evaluation is the self. Older adults, the study suggests, do not lose their capacity for self-reflection; they pay more for it, both in time and in neural effort, while their sense of who they are may grow only more emotionally charged.

Subject of Research: Age-related changes in the neural correlates of self- and other-referential trait processing in healthy aging

Article Title: Age-related differences in neural activity and functional connectivity during self- and other-referential processing: An EEG study

Article References: Paź, M., Kamiński, M., & Nowicka, A. (2026). Age-related differences in neural activity and functional connectivity during self- and other-referential processing: An EEG study. GeroScience. https://doi.org/10.1007/s11357-026-02571-7

Image Credits: AI Generated

DOI: 10.1007/s11357-026-02571-7

Keywords: aging brain, self-referential processing, EEG, late positive potential, P1 component, functional connectivity, directed transfer function, default mode network, GeroScience, cognitive aging, event-related potentials, self-knowledge