Search papers, labs, and topics across Lattice.
This study investigates the performance of fourteen foundation models, ranging from 135M to 14B parameters, trained on a large dataset of human behavioral choices (Psych-101) to understand their cognitive processing capabilities. The results reveal that while smaller models (0.6B to 1B parameters) can match the performance of a much larger 70B model in familiar contexts, larger models significantly outperform smaller ones when generalizing to novel tasks. Additionally, diagnostic tests show that essential information from task stimuli and feedback is crucial for model performance, indicating that smaller models can serve as effective noise ceiling estimators in psychological experiments, albeit limited by their training paradigms.
Smaller foundation models can rival larger counterparts in familiar tasks but struggle with generalization, revealing a critical trade-off in cognitive modeling.
Large language models fine-tuned on human behavioural data have emerged as general-purpose cognitive proxies, but the scale this requires, and whether these models process task structure or exploit statistical shortcuts, remain open questions. We train fourteen models from 135M to 14B parameters across four architecture families on Psych-101, a dataset of 10.7 million trial-level choices from 160 experiments. In-distribution, scale barely matters. The models fall within a narrow band, as though against a ceiling, and 0.6B to 1B parameters suffice to match a 70B baseline on held-out participants. Out-of-distribution, that band opens into a markedly steeper scaling gradient, with larger models clearly advantaged in generalisation to novel task structure. To determine what information these models use, we run two diagnostics. We progressively strip four prompt channels -- task instructions, experimental stimuli, outcome feedback, and choice history -- across 27 experiments, and permute trial order. Masking the content of stimuli and feedback destroys 75.7% of learned information and pushes models below chance, demonstrating that choice history alone does not account for performance. Permutation reveals invariance on tasks with independent trials but sensitivity where trial order is determined by prior responses. Small cognitively fine-tuned models therefore show promise as noise ceiling estimators for psychological experiments, though their scope remains bounded by the paradigms seen in training.
