Three findings emerge: in the restless task, neural models trained without rewards predict held-out choices better than four simple training-fitted behavioral baselines, and at some reward weights, neural networks predict better than a pooled reinforcement-learning model but have less faithful changes in choice probabilities.
Abstract
Using predictive models to explain cognition requires more than accurate behavioral predictions. Input ablations offer an appealing route: remove information from a model and interpret the resulting performance change as evidence of its importance for behavior. Yet this inference assumes that the model's dependence on information reflects the dependence of the process generating the behavior. We test it in two synthetic sequential bandit tasks with known generating policies, where past choices can remain informative when feedback is unavailable to a predictor. We compare GRUs and Transformers trained from scratch, a fine-tuned LLaMA model, and cognitive models across systematically varied reward contributions. Our analyses distinguish prediction after training without reward observations from the response of a fixed predictor to donor-reward replacement. Three findings emerge. First, in the restless task, neural models trained without rewards predict held-out choices better than four simple training-fitted behavioral baselines. Second, under matched donor replacement, accurate predictors can respond much less than the known generator. Third, at some reward weights, neural networks predict better than a pooled reinforcement-learning model but have less faithful changes in choice probabilities; the model ordering differs between the two tasks. These independent-test results separate information sufficient for prediction from response fidelity under a specified ablation in sequential choice. They motivate validating model-ablation responses independently of predictive performance before using them to infer how the observed behavior was generated.
Behavioral Foundation Models (BFMs) aim to solve a wide range of downstream tasks without test-time policy learning by inferring a task vector from the reward function. While efficient, the retrieved policies are often suboptimal because of how this task vector is inferred, typically with ordinary least squares (OLS)....
Kuan-Hsun Tu, Chi-Hsun Chiang, Hsin-Wei Chen et al.· 0 citations
BBT outperforms traditional approaches at predicting human behavior while also revealing psychological insights, resulting in models that can both mimic Bayesian priors and capture heuristics and biases that violate simple modeling assumptions.
Brenden M. Lake, A. Jagadish, Guang-Yuan Jiang· 0 citations
This work provides the first comprehensive empirical investigation of humans' internal models play a mediating role in feedback behaviour through a randomized controlled trial and shows that this relationship is invariant across visual contexts and is robust to three common feedback types.
Taha Shaheen, S. West, Yu Zhang· Proceedings of the Thirty-Fi...· 0 citations
The component that dominates a latent world model's prediction error need not be the one whose repair most improves action selection. We show this by comparing action sequences from identical physical starts and separating endpoint error into a candidate-pool center and action-relative responses. Across four model fami...
This work proposes RewardExplainer, a framework that obtains feedback from the target reward model through counterfactual rewriting and uses this feedback to further optimize the explainer, enabling the explainer to more faithfully capture the target RM's scoring preferences and sensitive behaviors than single-pass gen...
Jing-Yi He, Ni-Er Wu, Shuang Liu et al.· arXiv.org· 0 citations
A reward can identify a correct answer while leaving the computation needed for new inputs undetermined. We study how pretraining and midtraining supply the information and computation that make reward adaptation effective. In sequential state computation and contextual memory, we characterize mechanisms that agree on...