Schizophrenia symptom profiles are reflected in patients’ written language

Recent research published in the Journal of Writing Research suggests that the written language of people with schizophrenia provides evidence about their specific symptom profiles. The study indicates that analyzing how these individuals summarize a story can reveal distinct patterns in their writing, depending on whether they experience predominantly positive or negative symptoms. These findings highlight the potential of using writing tasks to monitor clinical changes and tailor therapy interventions for people with schizophrenia.

Schizophrenia is a severe mental health condition that affects thinking, emotion, and behavior. The condition tends to be divided into two main categories of symptoms. Positive symptoms involve an excess of abnormal experiences, such as hallucinations or delusions, which are fixed false beliefs. Negative symptoms involve a decrease in typical behaviors, such as a lack of motivation, emotional flatness, and reduced speech.

Scientists recognize that schizophrenia often disrupts how a person uses and understands language. This disruption typically affects the social and practical use of language, making effective communication difficult. While speech difficulties are well documented, writing is an area that has received less attention from researchers. The new study sought to better understand how the specific language deficits seen in verbal communication might translate into written form.

“Mental health has been a prominent part of my family context, and from the beginning of my training I knew I wanted to orient my career toward this field. As I delved into the linguistic characteristics of schizophrenia, I noticed that most of the literature focuses on spoken language,” said study author Alfonso Martínez Cano, a doctor of medical Sciences and faculty member at the University of Castilla–La Mancha.

“However, it is reasonable to expect that linguistic alterations are also expressed in reading and writing, which can directly affect academic performance, employment, everyday bureaucratic demands, and overall functional outcomes. We therefore designed a study to examine whether the deficits described in speech are also present in writing and, if so, whether they mirror oral-language profiles or show a distinct pattern.”

Free daily newsletter

To explore this, the researchers designed a study involving 41 adults diagnosed with schizophrenia. The participants ranged in age from 20 to 79 years old, with an average age of about 53. Of this group, 24 individuals primarily exhibited positive symptoms, while 17 primarily exhibited negative symptoms.

All participants were native Spanish speakers and completed the same reading and writing task. The researchers asked each participant to read a short, two-page story called The Tale of Landolfo Rufolo, which contains 530 words. Participants were allowed to read the story twice and ask for definitions of any unfamiliar words.

After reading, they were instructed to write a summary of the story by hand. They were given no time or space limits and were encouraged to include all the details they could remember. Four researchers then evaluated the handwritten summaries across three different levels of language production.

The first level looked at the overall structure and coherence of the text. This involved checking if the participants included the main ideas and maintained the chronological order of the story. The second level examined cohesion at the sentence and word level.

This second level of analysis included examining vocabulary variety and counting the use of specific verb tenses. The scientists also tallied the use of function words, such as articles and conjunctions, as well as content words, such as nouns and verbs. The third level focused on basic writing and spelling skills.

For this third level, the researchers looked for punctuation use, capital letters, spelling mistakes, and illegible words. They then compared the scores across all these areas between the participants with positive symptoms and those with negative symptoms. The analysis revealed several distinct patterns in how the participants wrote.

Overall, the summaries tended to maintain the basic timeline of the story, showing that the participants understood the core plot. The participants also demonstrated a correct use of past tense verbs, indicating that basic grammar rules remained intact.

However, the researchers noticed that participants frequently recalled minor details better than the central themes of the narrative. For example, they were more likely to remember that the main character became a pirate than to remember the main plot point of a woman helping him.

“One striking finding was that, in some cases, participants recalled secondary ideas from a text better than the main ideas, particularly among those with a greater predominance of positive symptoms,” Martínez Cano told PsyPost. “We were also surprised that, despite distortions in content or narrative, many texts preserved central elements and a degree of chronological coherence, suggesting partial preservation of discourse-level functioning.”

The written texts also showed low vocabulary variation, a reliance on simple sentence structures, and a high number of spelling mistakes.

“Linguistically, we observed relatively preserved morphology (e.g., verb variety and functional use of tense) alongside a tendency toward less complex syntax (simpler sentence structures),” Martínez Cano said. “At present, we cannot determine whether this reflects a primary linguistic impairment or a broader cognitive contribution. We also observed reduced lexical flexibility, with increased repetition across both shorter and longer texts.”

When comparing the two symptom groups, the researchers found significant differences. Individuals with predominantly positive symptoms tended to write longer summaries that contained more ideas. These texts were often less connected and contained more distorted information. Their writing also showed a higher overall word count, more repetitions, and a greater use of words related to paranoia.

Individuals with predominantly negative symptoms demonstrated a different writing profile. They produced shorter summaries that contained fewer ideas, but these ideas tended to be more concrete and central to the story. Their writing showed less vocabulary variation, indicating a more rigid use of language.

Additionally, this negative symptom group exhibited more basic writing difficulties. For instance, they had a tendency to combine separate words together incorrectly and used fewer punctuation marks. A somewhat unexpected finding was that delusional thinking did not strongly interfere with the basic functionality of the writing.

Most of the written summaries did not contain obvious delusional content. The scientists suggest that the structured nature of the summarizing task might help individuals focus and inhibit irrelevant or disorganized thoughts.

Martínez Cano highlighted three key takeaways from the study:

“Writing can provide clinically meaningful information. Certain indicators in written production appear to relate to symptom profiles (e.g., predominance of positive vs. negative symptoms), suggesting potential value for monitoring clinical course and change over time.”

“Language difficulties are not confined to speech. The linguistic deficits described in oral language tend to be reflected in writing as well, which may contribute to academic, occupational, and broader functional difficulties.”

“More applied evidence is needed. Further research—especially well-designed intervention trials—is required to determine whether targeted language rehabilitation improves these skills and whether such improvements translate into meaningful gains in real-world functioning.”

While the study provides evidence linking writing patterns to schizophrenia symptoms, there are a few potential misinterpretations to keep in mind. One limitation is the lack of a healthy control group for direct comparison.

The researchers also point out that standard cognitive assessments, which measure attention and memory, were not included in the study. Without these measures, it is difficult to know if the writing differences stem directly from language impairments or from broader cognitive challenges.

Medication is another factor that could influence the results. Different symptom profiles often require different medication types and dosages, which can affect a person’s cognitive and physical functioning.

“Accordingly, the findings should be interpreted cautiously, and future studies should more systematically control for these variables,” Martínez Cano said.

The scientists plan to explore whether these linguistic variables can act as early warning signs for individuals at high risk for developing psychosis. They also hope to test specific language therapy interventions to see if improving writing skills can translate into better social integration and daily functioning for people with schizophrenia.

“Overall, this work suggests that written language may constitute a valuable source of clinical and functional information,” Martínez Cano told PsyPost. “The immediate priority, however, is to replicate these findings in larger samples with tighter control of key variables (cognition, medication, educational level) and with longitudinal designs, in order to evaluate clinical utility with greater rigor.”

The study, “Macrotextual, microtextual and writing analysis of texts written by people with schizophrenia differentiated by their symptoms,” was authored by Alfonso Martínez-Cano, Alberto Martínez-Lorca, Juan José Criado Álvarez, and Manuela Martínez-Lorca.