# Theory Trace Card: Theory-Driven Socio-Cognitive Evaluation of LLMs

arXiv: https://arxiv.org/abs/2601.01878 · HTML: https://mlsanity.com/paper/2601.01878/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 22 of its 77 references that are in the map.
- First posted on arXiv 5 Jan 2026.

## Builds on (most cited first, 22 of 22)

- [Sparks of Artificial General Intelligence: Early experiments with GPT-4](https://mlsanity.com/paper/2303.12712/) (2023)
- [Shortcut Learning in Deep Neural Networks](https://mlsanity.com/paper/2004.07780/) (2020)
- [Concrete Problems in AI Safety](https://mlsanity.com/paper/1606.06565/) (2016)
- [Aligning AI With Shared Human Values](https://mlsanity.com/paper/2008.02275/) (2020)
- [DailyDialog: A Manually Labelled Multi-turn Dialogue Dataset](https://mlsanity.com/paper/1710.03957/) (2017)
- [GoEmotions: A Dataset of Fine-Grained Emotions](https://mlsanity.com/paper/2005.00547/) (2020)
- [Measurement and Fairness](https://mlsanity.com/paper/1912.05511/) (2019)
- [Evaluating Large Language Models in Theory of Mind Tasks](https://mlsanity.com/paper/2302.02083/) (2023)
- [Evaluating the Moral Beliefs Encoded in LLMs](https://mlsanity.com/paper/2307.14324/) (2023)
- [Understanding Social Reasoning in Language Models with Language Models](https://mlsanity.com/paper/2306.15448/) (2023)
- [Measuring what Matters: Construct Validity in Large Language Model Benchmarks](https://mlsanity.com/paper/2511.04703/) (2025)
- [Troubling Trends in Machine Learning Scholarship](https://mlsanity.com/paper/1807.03341/) (2018)
- [Mapping global dynamics of benchmark creation and saturation in artificial intelligence](https://mlsanity.com/paper/2203.04592/) (2022)
- [Evaluating Theory of Mind in Question Answering](https://mlsanity.com/paper/1808.09352/) (2018)
- [Can Large Language Models be Good Emotional Supporter? Mitigating Preference Bias on Emotional Support Conversation](https://mlsanity.com/paper/2402.13211/) (2024)
- [Theory of Mind in Large Language Models: Examining Performance of 11 State-of-the-Art models vs. Children Aged 7-10 on Advanced Tests](https://mlsanity.com/paper/2310.20320/) (2023)
- [Designing AI for Trust and Collaboration in Time-Constrained Medical Decisions: A Sociotechnical Lens](https://mlsanity.com/paper/2102.00593/) (2021)
- [Are Large Language Models More Empathetic than Humans?](https://mlsanity.com/paper/2406.05063/) (2024)
- [Line Goes Up? Inherent Limitations of Benchmarks for Evaluating Large Language Models](https://mlsanity.com/paper/2502.14318/) (2025)
- [The Greatest Good Benchmark: Measuring LLMs' Alignment with Utilitarian Moral Dilemmas](https://mlsanity.com/paper/2503.19598/) (2025)
- [MORABLES: A Benchmark for Assessing Abstract Moral Reasoning in LLMs with Fables](https://mlsanity.com/paper/2509.12371/) (2025)
- [Preventing Another Tessa: Modular Safety Middleware For Health-Adjacent AI Assistants](https://mlsanity.com/paper/2509.07022/) (2025)

## Abstract (arXiv)

Socio-cognitive benchmarks for large language models (LLMs) often fail to predict real-world behavior, even when models achieve high benchmark scores. Prior work has attributed this evaluation-deployment gap to problems of measurement and validity. While these critiques are insightful, we argue that they overlook a more fundamental issue: many socio-cognitive evaluations proceed without an explicit theoretical specification of the target capability, leaving the assumptions linking task performance to competence implicit. Without this theoretical grounding, benchmarks that exercise only narrow subsets of a capability are routinely misinterpreted as evidence of broad competence: a gap that creates a systemic validity illusion by masking the failure to evaluate the capability's other essential dimensions. To address this gap, we make two contributions. First, we diagnose and formalize this theory gap as a foundational failure that undermines measurement and enables systematic overgeneralization of benchmark results. Second, we introduce the Theory Trace Card (TTC), a lightweight documentation artifact designed to accompany socio-cognitive evaluations, which explicitly outlines the theoretical basis of an evaluation, the components of the target capability it exercises, its operationalization, and its limitations. We argue that TTCs enhance the interpretability and reuse of socio-cognitive evaluations by making explicit the full validity chain, which links theory, task operationalization, scoring, and limitations, without modifying benchmarks or requiring agreement on a single theory.