Computational linguistic analysis applied to a semantic fluency task to measure derailment and tangentiality in schizophrenia

Abstract

Although rating scales to assess formal thought disorder exist, there are no objective, high-reliability instruments that can quantify and track it. This proof-of-concept study shows that CoVec, a new automated tool, is able to differentiate between controls and patients with schizophrenia with derailment and tangentiality. According to ratings from the derailment and tangentiality items of the Scale for the Assessment of Positive Symptoms, we divided the sample into three groups: controls, patients without formal thought disorder, and patients with derailment/tangentiality. Their lists of animals produced during a one-minute semantic fluency task were processed using CoVec, a newly developed software that measures the semantic similarity of words based on vector semantic analysis. CoVec outputs were Mean Similarity, Coherence, Coherence-5, and Coherence-10. Patients with schizophrenia produced fewer words than controls. Patients with derailment had a significantly lower mean number of words and lower Coherence-5 than controls and patients without derailment. Patients with tangentiality had significantly lower Coherence-5 and Coherence-10 than controls and patients without tangentiality. Despite the small samples of patients with clinically apparent thought disorder, CoVec was able to detect subtle differences between controls and patients with either or both of the two forms of disorganization.

Semantic Search in Psychosis: Modeling Local Exploitation and Global Exploration

“…This contrasts with our prediction based on findings of less coherent semantic VFT output in schizophrenia. 20 , 32 , 33 It may be important to test this hypothesis with longer periods of word generation, as differences in local cue salience may be subtle and difficult to detect within a 1-minute task.…”

“…A common computational metric utilized in the present study is semantic coherence, defined as the average similarity between each word and all other words in a particular text or word list. Prior VFT research has reported decreased semantic coherence in individuals with schizophrenia compared to healthy individuals 20 and in patients with thought disorder symptoms compared to patients without these symptoms, 20 , 21 while accounting for the number of words produced. On the less clinically severe end of the psychotic spectrum, a VFT study found comparable levels of semantic coherence between individuals with and without schizotypal traits, 23 with fewer total responses among schizotypal individuals only under conditions with an affective load.…”

“…We also predicted lower local search cue salience in schizophrenia compared to controls resulting in less semantically similar words within a cluster, as studies have found less semantically coherent VFT output in psychosis. 20 , 32 , 33 We tested whether semantic search cues were intact in individuals with schizotypal traits or intermediate between control and schizophrenia groups. Given that vocabulary ability and processing speed have positively correlated with VFT performance, 4 we examined the degree to which these cognitive abilities contributed to types of search cue salience that differed by diagnosis.…”

Identifying emerging mental illness utilizing search engine activity: A feasibility study

“…Prior research in linguistic analysis has identified significant differences at the word level in the use of certain word categories, as well as at the sentence level in terms of semantic density, coherence, and/or content, both in individuals at risk for developing psychotic disorders as well as those with established SSD and MD [43][44][45][46][47][48][49][50][51][52][53][54][55][56][57][58]. Language extracted from online search activity is distinct due to its short sentence structure and the unique nature of the online search platform.…”

Lowering costs for large-scale screening in psychosis: a systematic review and meta-analysis of performance and value of information for speech-based psychiatric evaluation

“…Generally, open interviews were conducted in scenarios close to clinical settings, 45 while task-based assessment took place in artificial environments. 41 …”