Encoding EEG Signals to Examine Human-Like Next-Word Prediction Behaviour in Language Models

• URL: http://arxiv.org/abs/2607.16549v1
• Date: Fri, 17 Jul 2026 23:11:16 GMT
• Title: Encoding EEG Signals to Examine Human-Like Next-Word Prediction Behaviour in Language Models
• Abstract summary: Next-word predictability influences brain response, as recorded at millisecond resolution using electroencephalography (EEG)<n>We generate regressors for both humans and LMs based on two information measures, including top-1 prediction and surprisal, to predict event-related potential (ERP)<n>Our results indicate that only surprisal potentially correlates with language-processing ERPs, especially for open-class words with high semantic content.
• Score: 3.1487162256026573
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Language models (LMs) are trained to excel at predicting the next word in the sequence given prior context, and humans also share this predictability in reading comprehension. Neuroscience research reveals that next-word predictability influences brain response, as recorded at millisecond resolution using electroencephalography (EEG). While our evidence indicates that advanced LMs achieve accuracies closely aligned with human performance at the next-word prediction task, this raises the question: Does higher prediction accuracy necessarily mean that these models adequately capture the cognitive signals associated with human reading comprehension? Here, we generate regressors for both humans and LMs based on two information measures, including top-1 prediction and surprisal, to predict event-related potential (ERP) elicited from EEG recordings which reflect different stages of cognitive processing during reading. We argue that modelling ERP patterns offers fine-grained analysis of the cognitive plausibility of various LMs during reading. Our results indicate that only surprisal potentially correlates with language-processing ERPs, especially for open-class words with high semantic content. Moreover, our findings challenge the assumption that scaling LMs with increased parameters and computational budgets will consistently lead to improved convergence with human-like linguistic processing.

Related papers

• Probing for Reading Times [60.30314214800421]
  We probe language model representations for human reading times in English, Greek, Hebrew, Russian, and Turkish.<n>We find that the representations from early layers outperform surprisal in predicting early-pass measures such as first fixation and gaze duration.
  arXiv  Detail & Related papers  (2026-04-20T18:12:59Z)
• E^2-LLM: Bridging Neural Signals and Interpretable Affective Analysis [54.763420895859035]
  We present ELLM2-EEG-to-Emotion Large Language Model, first MLLM framework for interpretable emotion analysis from EEG.<n>ELLM integrates a pretrained EEG encoder with Q-based LLMs through learnable projection layers, employing a multi-stage training pipeline.<n>Experiments on the dataset across seven emotion categories demonstrate that ELLM2-EEG-to-Emotion Large Language Model achieves excellent performance on emotion classification.
  arXiv  Detail & Related papers  (2026-01-11T13:21:20Z)
• Spontaneous Speech Variables for Evaluating LLMs Cognitive Plausibility [0.7061230262755125]
  We propose using spontaneous speech corpora to derive production variables (speech reductions, prosodic prominences) and applying them in a similar fashion.<n>We then test models trained with a standard procedure on different pretraining datasets for their ability to predict these two variables.<n>Our results show that, after some fine-tuning, the models can predict these production variables well above baselines.
  arXiv  Detail & Related papers  (2025-05-22T06:23:02Z)
• Interpretable Next-token Prediction via the Generalized Induction Head [59.500195503897764]
  Generalized Induction-Head Model (GIM) is an interpretable model for next-token prediction.<n>In language modeling, GIM improves next-token prediction by up to 25%p over interpretable baselines.<n>In an fMRI setting, GIM improves neural response prediction by 20%.
  arXiv  Detail & Related papers  (2024-10-31T12:33:26Z)
• Decomposition of surprisal: Unified computational model of ERP components in language processing [7.760815504640362]
  We advance an information-theoretic model of human language processing in the brain in which incoming linguistic input is processed at first shallowly and later with more depth. We show that the information content (surprisal) of a word in context can be decomposed into two quantities: (A) shallow surprisal, which signals shallow processing difficulty for a word, and corresponds with the N400 signal; and (B) deep surprisal, which reflects the discrepancy between shallow and deep representations, and corresponds to the P600 signal.
  arXiv  Detail & Related papers  (2024-09-10T18:14:02Z)
• Language models emulate certain cognitive profiles: An investigation of how predictability measures interact with individual differences [1.942809872918085]
  We revisit the predictive power of surprisal and entropy measures estimated from a range of language models (LMs) on data of human reading times. We investigate if modulating surprisal and entropy relative to cognitive scores increases prediction accuracy of reading times. Our study finds that in most cases, incorporating cognitive capacities increases predictive power of surprisal and entropy on reading times.
  arXiv  Detail & Related papers  (2024-06-07T14:54:56Z)
• Humans and language models diverge when predicting repeating text [52.03471802608112]
  We present a scenario in which the performance of humans and LMs diverges. Human and GPT-2 LM predictions are strongly aligned in the first presentation of a text span, but their performance quickly diverges when memory begins to play a role. We hope that this scenario will spur future work in bringing LMs closer to human behavior.
  arXiv  Detail & Related papers  (2023-10-10T08:24:28Z)
• Better Language Model with Hypernym Class Prediction [101.8517004687825]
  Class-based language models (LMs) have been long devised to address context sparsity in $n$-gram LMs. In this study, we revisit this approach in the context of neural LMs.
  arXiv  Detail & Related papers  (2022-03-21T01:16:44Z)
• Long-range and hierarchical language predictions in brains and algorithms [82.81964713263483]
  We show that while deep language algorithms are optimized to predict adjacent words, the human brain would be tuned to make long-range and hierarchical predictions. This study strengthens predictive coding theory and suggests a critical role of long-range and hierarchical predictions in natural language processing.
  arXiv  Detail & Related papers  (2021-11-28T20:26:07Z)
• On the Predictive Power of Neural Language Models for Human Real-Time Comprehension Behavior [29.260666424382446]
  We test over two dozen models on how well their next-word expectations predict human reading time on naturalistic text corpora. We evaluate how features of these models determine their psychometric predictive power, or ability to predict human reading behavior. For any given perplexity, deep Transformer models and n-gram models show superior psychometric predictive power over LSTM or structurally supervised neural models.
  arXiv  Detail & Related papers  (2020-06-02T19:47:01Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.