Grammatical errors could aid in the early detection of schizophrenia

By Dr. Priyom Bose, Ph.D.Reviewed by Lauren HardakerJun 18 2025

New research uncovers the hidden language clues in speech that could help diagnose schizophrenia earlier and shape more effective therapies.

Study: Relationship between grammar and schizophrenia: a systematic review and meta-analysis. Image credit: Andrey_Popov/Shutterstock.com

Schizophrenic individuals experience significant difficulties with grammar, particularly in sentence comprehension and production. A Communications Medicine study performed a comprehensive analysis to evaluate the extent to which schizophrenia impacts language comprehension and creation. 

Grammatical impairment in schizophrenia

Schizophrenia is a severe mental health disorder that affects how an individual thinks, feels, and behaves. These patients often experience difficulties communicating with others, which stem from syntactic deficits.

Syntax production involves creating grammatically correct and contextually appropriate sentences, whereas syntax comprehension requires deciphering and interpreting syntactic structures in real time. Syntax-level deficits are irregularities in the way words are composed in an order, creating difficulties in communication for patients with schizophrenia.

Impaired syntax production and comprehension lead to conversational incoherence, persecutory delusions, uncooperativeness, and a lack of insight. Evaluating the relative deficiencies in syntax production and comprehension is imperative, as this sheds light on distinct cognitive mechanisms.

To date, few systematic reviews and meta-analyses have explained the impact of grammatical impairment in sentence comprehension and creation in patients with schizophrenia. Quantifying the scope of grammatical impairment in schizophrenia is crucial because the use of various linguistic markers in speech helps predict clinically significant outcomes. Furthermore, previous studies have shown that social interactions positively affect functional recovery. Therefore, scientists across the world have focused on developing interventions that improve communication deficits in schizophrenia.

About the study

The systematic review and meta-analysis obtained all relevant literature, published up to May 1, 2024, from multiple electronic databases, including Web of Science, PubMed (MEDLINE), and Scopus. The search also included PsycINFO, ensuring a comprehensive review of available studies.

Research published in English and linked to grammar/syntax and schizophrenia was included. Studies associated with adults diagnosed with schizophrenia spectrum disorders, along with a control group of healthy adults, were considered.

This systematic review primarily focused on empirical studies with quantitative measures. Cohen’s d and log coefficient of variation ratio were extracted, and a Bayesian meta-analysis (BMA) was performed across six domains: two in comprehension and four in production in patient-control comparisons. A modified Newcastle–Ottawa Scale was used to assess data quality, with moderators, such as age, sex, language, and study quality, tested using meta-regression.

Study findings

Out of 820 studies in the initial database search, 45 studies fulfilled the eligibility criteria. All studies included in the systematic review and meta-analysis were published between 1982 and 2024. They included 1,679 people with schizophrenia and 1,281 healthy participants (control). The mean age of the participants was 32.31 years. Among the selected studies, 29.2% of participants were women. It must be noted that five studies recruited only men, and eight studies’ cohorts comprised more than 40% of women.

Since a standardized method for estimating grammatical impairment in mental health conditions is lacking, selected studies employed diverse methods in their analysis. This lack of overlap in how language variables are measured between studies has made standardizing predictive models for clinical use difficult.

BMA results provided robust evidence that patients with schizophrenia experience reduced syntactic comprehension, production length, error detection, production integrity, and phrasal complexity. Similarly, random effects analysis across the six domain-specific effects also provided evidence for an overall grammatical impairment in schizophrenia.

Production length, global complexity, and integrity showed strong between-study heterogeneity (tau). The meta-regression analysis indicated that age was a significant moderator of global complexity. The moderator/effect-size bivariate correlations were not significant for the total symptom severity index or antipsychotic dose across all domains.

The meta-analysis found that people with schizophrenia differed in their ability to understand syntax, produce longer sentences, and use complex phrases. This means that not all individuals with schizophrenia have the same type or degree of grammatical impairment, which has important implications for how language clues are used in a clinical setting.

Robust BMA did not indicate any significant publication bias within the individual meta-analyses. Borrowing of Strength analysis suggested that the adjustment for correlations among the domains supported all precision estimates.

These findings suggest that these individuals understand simpler sentences more effectively and may not possess a complete syntactic structure. People with schizophrenia often speak in less sophisticated and shorter sentences and ignore syntactical errors.

Notably, the degree of grammatical impairment was not uniform. The analysis suggested possible subgroups of patients with different levels of deficit.

The study also found that syntactic deficits were not strongly related to overall symptom severity or medication dosage, suggesting that grammatical impairments are a distinctive feature rather than merely a reflection of general illness severity.

The authors emphasized that syntactic impairment is just one dimension of a broader language dysfunction in schizophrenia. Deficits in grammar can interact with or exacerbate problems in other areas, such as semantic (meaning) and pragmatics (contextual language use). For example, reduced production length and phrasal complexity might help mask semantic incoherence in some cases. While improvement in syntax could benefit overall communication, some language differences might serve as compensatory or adaptive mechanisms.

The current literature has limitations, including the diversity of methods used to assess syntax, the underrepresentation of women and non-English speakers, and the likelihood that real-world impairments may be greater than estimates from controlled research samples.

The paper also points out that most available studies are cross-sectional, so the true extent of language impairment over time may be underestimated.

Conclusions

The meta-analysis provided statistically significant evidence linking grammatical impairment with schizophrenia. People diagnosed with schizophrenia experience challenges in comprehension of complex syntax and identification of errors in sentences.

In the future, researchers must investigate whether grammatical impairment in people with schizophrenia occurs independently of lexico-semantic abnormalities or as a part of broader linguistic impairment. A targeted intervention focused on linguistic differences could be developed to improve severe symptoms.

The findings also highlight the potential for speech-based language analysis as a tool for early detection and personalized intervention in schizophrenia. Future research should include more diverse populations, examine the relationship between different language domains, and address the practical implications for therapy and daily communication support.

Download your PDF copy now!