Computational Psycholinguistics

Overview

Computational psycholinguistics builds models of how people learn, produce, and understand language, and uses them to explain measurable behavior. Approaches include Bayesian noisy-channel and rational-speaker models, surprisal-based prediction, neural language models, and explicit accounts of syntactic parsing and memory. Researchers test these models against reading times, eye movements, cloze judgments, speech, and EEG or brain-imaging data, across languages and developmental settings. A central tension is that predictive fit—especially from language-model surprisal—does not by itself establish a human-like mechanism, and models often miss integration costs or vary across tasks, languages, and measures.

Latest papers

CardsList

93 papers8 papers in the last four weeks, down 33% on the four weeks before. 0.1% of all new papers.

  1. LLM surprisal is necessary but not sufficient to capture English garden-path effects: Evidence from joint latent modeling of reading paradigms

    Feb 4, 2026Dario Paape, Tal Linzen, Shravan VasishthLatent Variable ModelsCognitive Modeling

  2. From Associations to Activations: Comparing Behavioral and Hidden-State Semantic Geometry in LLMs

    Jan 31, 2026Louis Schiekiera, Max Zimmer, Christophe Roux +2Representation Geometry in Language ModelsComputational Psycholinguistics

  3. Graded strength of comparative illusions is explained by Bayesian inference

    Nov 18, 2025Yuhan Zhang, Erxiao Wang, Cory ShainBayesian InferenceComputational Psycholinguistics