Agentic education · Learning science

Actionable companion · Assessment & reasoning

Learning with agency: Large language models predict cognition and education close to or better than genomics or expert assessment

AI-assisted editorial implementation guide, separate from the original research. Proposed actions are not claims or recommendations attributed to the source authors. No independent peer review of this guide is asserted.

Read original paperSearch paper agentCC BY 4.0

Evidence and context

Title and abstract screening; stage labels describe reported settings, not verified numerical ages. Design labels require full-text appraisal. Reported stage(s): Age / stage not established. Verify the participant population and task before transferring the findings.

Original authors’ abstract, reproduced under the linked source license:

Abstract Previous research using standard social survey data has emphasized a relative lack of power when predicting educational and psychological outcomes. Leveraging a unique longitudinal dataset, we explore predictability of educational attainment, cognitive abilities, and non-cognitive traits. Integrating various measures of computational linguistics and large language model-based embeddings within a SuperLearner framework trained on short aspirational essays written at age 11, we accurately predict cognition and non-cognitive traits at the same and later age to a similar degree as teacher assessments, and better than genomic data. The same is true for predicting final educational attainment. Combining text, genetic markers, and teacher assessments into an ensemble model, we can predict cognitive ability at close to test-retest reliability of gold-standard tests (RHoldout2=0.7) and explain 38% of individual differences in attainment. A sociological model comparable to the baseline of the Fragile Family Challenge replicates the FFC’s findings regarding the level of predictability achievable with such data. These findings show that recent advances in large language models and machine learning equip behavioural scientists with tools for prediction of psycho-social features. Subject terms: Machine learning, Human behaviour, Statistics Large Language models are powerful prediction tools. Based on short aspirational essays written at age 11, these models predicted cognitive and non-cognitive traits up to the level of teacher assessments.

What kind of evidence is this?

Supporting AI learning science · Survey / observational study

Stage: Age / stage not established. Screening labels describe reported context and study-design language. They are not a risk-of-bias rating or proof of effectiveness for an age band.

Participant ages · source passage

Integrating various measures of computational linguistics and large language model-based embeddings within a SuperLearner framework trained on short aspirational essays written at age 11, we accurately predict cognition and non-cognitive traits at the same and later age to a similar degree as teacher assessments, and better than genomic data.

Study methods · source passage

The dataset used for our analyzes is the National Child Development Study (NCDS), an ongoing British birth cohort study starting with 17,415 children born in a single week of 1958. No new data were collected, and no re-identification or data linkage to individual participants was performed. According to the ethical guidelines of the DGPs/BDP and Bielefeld University’s Ethics Review Board, such research does not require formal ethical approval.

Limitations and uncertainty · source passage

Although we demonstrate the predictability of psycho-social outcomes, our study is not without limitations. First, it is unclear to what extent our results are generalizable as they are based on a particular sample of individuals born in the year 1958 in Great Britain, and their educational biographies might not be comparable to those of today’s contemporary pupils within or outside of the UK. This underscores the need for more longitudinal datasets that combine different and diverse types of non-standard data with classical social surveys and behavioral assessments. In our supplementary information (Section Additional Analyzes), we address potential concerns related to variations in sample composition due to different non-response patterns across different variables. Specifically, we reran our primary analyses on a subset of 1618 individuals with full information across all key variables. Although predictive performance showed increased variation in this reduced sample, our findings remained robust. Despite these robustness checks, we acknowledge the potential influence of systematic missingness—especially in the genetic subsample—on our results.

Design the system before the pilot

Use this study to inform a learning mechanism or implementation question. It is not direct evidence that autonomous agents improve learning. Keep the proposed agent behavior separate from what the source actually tested.

  • Use common rubrics, blind scoring where feasible and a fresh unaided reasoning task. Document permissible assistance and avoid treating AI-detector scores as proof of misconduct.

A 90-day district pilot

  1. Days 1–15 · Appraise. Have the curriculum lead and a teacher read the methods and limitations. Record the study design, sample, comparator, outcomes and evidence gaps. Define a local question that the study can help inform.
  2. Days 16–30 · Design. Treat the paper as background evidence until its population is verified in the full text. Do not assign an age-specific intervention or extend adult results to children. Name an accountable owner and set the baseline, success threshold and budget before implementation.
  3. Days 31–60 · Test. Start with a small voluntary cohort, preserve an accessible non-AI route and require teacher review. Use synthetic or de-identified examples; approve data handling and tool terms through district processes.
  4. Days 61–90 · Decide. Extract the participant ages or education stage, recruitment, task and comparator before selecting a local pilot. If these cannot be established, retain the paper for design discussion rather than age-band efficacy claims. Compare with baseline, document uncertainty and stop or revise if learning worsens, access is unequal or critical errors persist.

The board decision brief

Report the learning problem, evidence quality, participating population, total cost including review time, measured benefit, errors, access differences and remaining uncertainties. Make a dated recommendation to stop, revise or expand. A positive perception survey alone does not establish learning improvement.

Attribution and reuse

Tobias Wolfram. Large language models predict cognition and education close to or better than genomics or expert assessment. Communications Psychology (2025-07-03). https://doi.org/10.1038/s44271-025-00274-x

Companion prepared 2026-10-30. Original authorship and DOI are preserved. The abstract is an attributed source extract; the implementation plan is new AI-assisted editorial material, offered under the same CC BY 4.0 license. This guide has not undergone independent peer review.