Perplexity of utterances in untreated first-episode psychosis: an ultra–high field MRI dynamic causal modelling study of the semantic network
Bibliographic record
Abstract
BACKGROUND: Psychosis involves a distortion of thought content, which is partly reflected in anomalous ways in which words are semantically connected into utterances in speech. We sought to explore how these linguistic anomalies are realized through putative circuit-level abnormalities in the brain's semantic network. METHODS: Using a computational large-language model, Bidirectional Encoder Representations from Transformers (BERT), we quantified the contextual expectedness of a given word sequence (perplexity) across 180 samples obtained from descriptions of 3 pictures by patients with first-episode schizophrenia (FES) and controls matched for age, parental social status, and sex, scanned with 7 T ultra-high field functional magnetic resonance imaging (fMRI). Subsequently, perplexity was used to parametrize a spectral dynamic causal model (DCM) of the effective connectivity within (intrinsic) and between (extrinsic) 4 key regions of the semantic network at rest, namely the anterior temporal lobe, the inferior frontal gyrus (IFG), the posterior middle temporal gyrus (MTG), and the angular gyrus. RESULTS: We included 60 participants, including 30 patients with FES and 30 controls. We observed higher perplexity in the FES group, indicating that speech was less predictable by the preceding context among patients. Results of Bayesian model comparisons showed that a DCM including the group by perplexity interaction best explained the underlying patterns of neural activity. We observed an increase of self-inhibitory effective connectivity within the IFG, as well as reduced self-inhibitory tone within the pMTG, in the FES group. An increase in self-inhibitory tone in the IFG correlated strongly and positively with inter-regional excitation between the IFG and posterior MTG, while self-inhibition of the posterior MTG was negatively correlated with this interregional excitation. LIMITATION: Our design did not address connectivity in the semantic network during tasks that selectively activated the semantic network, which could corroborate findings from this resting-state fMRI study. Furthermore, we do not present a replication study, which would ideally use speech in a different language. CONCLUSION: As an explanation for peculiar speech in psychosis, these results index a shift in the excitatory-inhibitory balance regulating information flow across the semantic network, confined to 2 regions that were previously linked specifically to the executive control of meaning. Based on our approach of combining a large language model with causal connectivity estimates, we propose loss in semantic control as a potential neurocognitive mechanism contributing to disorganization in psychosis.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.004 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".