We present the MultiplEYE Text Corpus, a large-scale, document-level, multi-parallel resource designed to advance cross-linguistic research on reading and language processing. The corpus provides paragraph-level alignment for texts in 39 languages spanning seven language families and seven scripts. Unlike many existing multilingual corpora, a substantial number of documents were originally written in languages other than English, reducing English-centric bias and supporting more typologically diverse investigations. The texts are carefully selected to balance linguistic richness with experimental feasibility, particularly for eye-tracking-while-reading studies. Developed within a multi-lab initiative, the MultiplEYE Text Corpus follows unified translation, alignment, and experimental design guidelines to ensure cross-linguistic comparability. Its inclusion of texts varying in type and difficulty enables research on discourse- level processing, genre effects, and individual differences across a wide range of languages. The text corpus and accompanying metadata provide a robust foundation for multilingual psycholinguistic and computational modeling research. Data and materials are publicly available at https://doi.org/10.23668/psycharchives.22294.

This study presents a survey of research integrating eye-tracking (gaze) data into Language Models (LMs) as a means of cognitively grounding NLP models and applications in human reading behavior. Although contemporary LMs excel at learning statistical patterns from text, they fundamentally lack human-like reading and comprehension capabilities. Incorporating gaze data may offer a window into cognitive processing, yet its impact on LMs remains underexplored. Addressing a persistent bottleneck, namely, the high cost and limited scale of laboratory eye-tracking, we propose a roadmap consisting of three streams of research for advancing this novel research domain: (1) developing cognitive multimodal corpora, (2) leveraging generative models for gaze synthesis to overcome the data bottleneck caused by the high costs of human eye-tracking, and (3) training LMs with gaze-guided attention mechanisms and input augmentation. Furthermore, we illustrate practical applications in readability assessment, educational analytics, and assistive communication, demonstrating how gaze-informed models can enable adaptive technologies. Finally, we critically examine ongoing challenges, including the lack of data standardization, the misalignment between human and machine language processing, and the urgent ethical imperative for privacy-preserving architectures to protect sensitive biometric gaze data, motivating privacy-aware data practices and model designs for scalable deployment.

This eye-tracking study links language-model surprisal and contextual entropy to how 23 non-expert adults read French health texts. Participants read seven texts (clinical case, medical, general), each available in an Original and Simplified version. Surprisal and entropy were computed with eight autoregressive models (82M–8B parameters), and four complementary eye-tracking measures were analyzed. Surprisal correlates positively with early reading measures, peaking in the smallest GPT-2 models (r ≈ 0.26) and weakening with model size. Entropy shows the opposite pattern, with negative correlations strongest in the 7B-8B models (r ≈ −0.13), consistent with a skim-when-uncertain strategy. Surprisal effects are largest in Clinical Original passages and drop by ∼20% after simplification, whereas entropy effects are stable across domain and version. These findings expose a scaling paradox – where different model sizes are optimal for different cognitive signals – and suggest that French plain-language editing should focus on rewriting high-surprisal passages to reduce processing difficulty, and on avoiding high-entropy contexts for critical information.