PSYTECH
News
Essay12 min read

The age a model assigns you

When software estimates how fast a brain is ageing, a private process becomes a score. Identity and threat follow the number — not, automatically, the tissue.

identitystressperceptionexecutive function

Written by

Nou Heravi

Editor, PSYTECH News

The scene

A person sits after a scan that did not hurt. On a screen a number appears that was not there yesterday: a brain-age estimate, a few years older than the birthday they already know. Nothing in the body has announced itself. The number is not a symptom. It is information with the posture of a verdict.

The mind does not receive a statistic. It receives a story about the self. From that moment, a forgotten name or a slow afternoon can be recruited as evidence. The scan is over. The interpretation has only begun.

The psychological stake

Identity is a running inference, updated by salient cues. A model that outputs “brain age” inserts a high-status, apparently objective cue into that inference. Health-anxiety research has long shown that ambiguous bodily information is read through threat. A number with the authority of imaging is not ambiguous in the same way a headache is. It feels like a fact about the future.

This piece does not claim that brain-age models are fraudulent, nor that they are ready to tell a person who they are. It asks what happens to self-concept when a research biomarker is presented as a personal score.

What the named evidence actually holds

Cole and Franke (2017), writing in Trends in Cognitive Sciences, treated neuroimaging-based brain age as an innovative biomarker of brain health: the gap between predicted and chronological age can track group differences in later outcomes. That is a research claim about cohorts. It is not a licence to narrate an individual life.

Later deep-learning systems have tightened error on clinical-looking scans. Kim and colleagues (2025) report a framework for routine 2D MRI in cognitively unimpaired adults. Wang and colleagues (2025), training across large multi-cohort T1 data, report a mean absolute error near 3.3 years on an ADNI cognitively normal test split, with the gap widening as cognition declines in other tasks. Papouli and Staff (2025) review the same family of methods in neurodegenerative disease and stress the usual limits: scanner, population, and the jump from group signal to the person in the chair.

None of these papers, read fairly, say: this number is your fate. Several imply the opposite. Error of a few years is excellent for a biomarker and crude as a identity. Confounders — depression, sleep debt, medication, which scanner you lay down in — are not trivia. They are the difference between a useful research tool and a story that will follow someone home.

Mechanisms (hypotheses, not verdicts)

Three psychological pathways are offered here as hypotheses, grounded in health psychology and labelling research, not as findings of the imaging papers themselves.

First, self-concept. A number that implies accelerated ageing is a direct hit on working models of competence and future self. Second, interoceptive amplification: once decline is expected, ordinary noise in memory is recoded as signal. Third, loss of narrative autonomy: ageing used to be interpreted slowly, with biography. A score arrives without it.

These pathways would matter even if the model were perfect. They matter more because it is not.

Second-order effects, held lightly

If such scores become ordinary, people may perform youth for the model, and organisations may be tempted to treat the gap as a proxy for reliability. That is foresight, not data. It is also why disclosure design is an ethical problem, not only a statistical one.

For living and designing

If you receive such a number, keep its unit: a noisy estimate, at a moment, in a population you may not represent. Ask what decision it supports. If there is no decision — only a mood — it is entertainment with extra authority.

Researchers: distinguish trial utility from personal dashboard. Designers: show uncertainty; refuse leaderboards of ageing. A mind is not a ranking.

Close

The mind has always aged. What is new is the invitation to watch it age as a number, on a schedule the person did not choose, with a precision the individual-level evidence does not yet earn. A model can estimate a trajectory. Only a life — and a careful reading of the paper — can decide what that estimate is allowed to mean.

PSYTECH does not diagnose, treat, or advise clinically. Interpretations above are editorial.

If this continues

What this may shift in identity

If the dynamics described here continue to scale, expect second-order effects on identity, stress, perception. The desk treats these as working hypotheses — grounded enough to watch, not certain enough to declare.

identity · stress · perception · executive function

Sources & further reading

Original editorial. Empirical claims are tied to the named papers below. We paraphrase; we do not reproduce others' sentences. Mixed findings are left mixed.

  1. Cole JH, Franke K (2017). Predicting age using neuroimaging: innovative brain ageing biomarkers — Trends in Cognitive Sciences — Primary review of brain-predicted age as a research biomarker; group associations, not a personal destiny score.
  2. Kim H, Park S, Seo SW, et al. (2025). A novel deep learning-based brain age prediction framework for routine clinical MRI scans — npj Aging
  3. Papouli A, Staff RT (2025). Brain age prediction from MRI scans in neurodegenerative disease — a review — PMC / review literature
  4. Wang C, et al. (2025). Deep-learning based multi-modal models for brain age, cognition, and amyloid — Alzheimer's Research & Therapy — MAE about 3.3 years on an ADNI cognitively normal test split — useful precision in a cohort, not a private oracle.
  5. PSYTECH News editorial constitution — PSYTECH — House method: scene → claim → named evidence → implication. Editorial, not clinical.

Editorial note. PSYTECH News is not a medical service. Nothing here is diagnosis, therapy, or clinical advice — only careful editorial sense-making at the edge of mind and machine.

Continue exploring

Nearby pieces in the same intellectual neighbourhood.