# Psychosis involves a deficit of information compression in connected speech

arXiv: https://arxiv.org/abs/2609.15522 · HTML: https://mlsanity.com/paper/2609.15522/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 13 of its 36 references that are in the map.
- First posted on arXiv 14 Sep 2026.

## Builds on (most cited first, 13 of 13)

- [Estimating the intrinsic dimension of datasets by a minimal neighborhood information](https://mlsanity.com/paper/1803.06992/) (2018)
- [Training Trajectories of Language Models Across Scales](https://mlsanity.com/paper/2212.09803/) (2022)
- [Scikit-dimension: a Python package for intrinsic dimension estimation](https://mlsanity.com/paper/2109.02596/) (2021)
- [Interpreting neural computations by examining intrinsic and embedding dimensionality of neural activity](https://mlsanity.com/paper/2107.04084/) (2021)
- [DADApy: Distance-based Analysis of DAta-manifolds in Python](https://mlsanity.com/paper/2205.03373/) (2022)
- [The optimality of syntactic dependency distances](https://mlsanity.com/paper/2007.15342/) (2020)
- [Entropy and type-token ratio in gigaword corpora](https://mlsanity.com/paper/2411.10227/) (2024)
- [Distributional Results for Model-Based Intrinsic Dimension Estimators](https://mlsanity.com/paper/2104.13832/) (2021)
- [Scale adaptive and robust intrinsic dimension estimation via optimal neighbourhood identification](https://mlsanity.com/paper/2405.15132/) (2024)
- [A general framework for adaptive nonparametric dimensionality reduction](https://mlsanity.com/paper/2511.09486/) (2025)
- [Unveiling Intrinsic Dimension of Texts: from Academic Abstract to Creative Story](https://mlsanity.com/paper/2511.15210/) (2025)
- [Lyapunov Spectral Analysis of Speech Embedding Trajectories in Psychosis](https://mlsanity.com/paper/2602.16273/) (2026)
- [The grip of grammar on meaning uncertainty: cross-linguistic evidence, neural correlates, and clinical relevance](https://mlsanity.com/paper/2605.01537/) (2026)

## Abstract (arXiv)

Large language models (LLMs) with human-like performance on linguistic tasks have transformed the study of language in neurodiverse conditions. LLMs provide representations of linguistic input in the form of high-dimensional vectors (embeddings), and next-token predictions computed from these embeddings. Previous crosslinguistic evidence suggests a complexity reduction in the form of both lower intrinsic dimensionality (ID) of LLM representations and higher mean surprisal (prediction error) in psychosis. We hypothesized that these metrics reflect a general deficit of information compression in psychosis, linked to grammatical organization as what enables predictions in language.We operationalized surprisal difference as the difference between surprisal as estimated from word frequency and surprisal as based on a contextual LM, which is sensitive to grammatical organization over and above lexical concepts. Using a dataset of 144 Turkish speakers, including 106 patients with schizophrenia-spectrum disorders (SSD) - 56 with chronic schizophrenia (SZH), 33 with first-episode psychosis (FEP), and 17 with schizoaffective disorder (SZA) - and 38 healthy controls. We report: (1) Surprisal difference is attenuated in all clinical groups relative to controls, independently of word count; (2) Compressibility (intrinsic dimension) is reduced in SZH and FEP; (3) Syntactic complexity and compressibility both predict surprisal difference. These results, further refining an alteration in the geometry of the semantic space in psychosis as previously attested, suggest a broader deficit in information compression in this disorder, with a mechanistic underpinning in the operations of grammar.