Speech Graphs Provide a Quantitative Measure of Thought Disorder in Psychosis

Abstract

BackgroundPsychosis has various causes, including mania and schizophrenia. Since the differential diagnosis of psychosis is exclusively based on subjective assessments of oral interviews with patients, an objective quantification of the speech disturbances that characterize mania and schizophrenia is in order. In principle, such quantification could be achieved by the analysis of speech graphs. A graph represents a network with nodes connected by edges; in speech graphs, nodes correspond to words and edges correspond to semantic and grammatical relationships.Methodology/Principal FindingsTo quantify speech differences related to psychosis, interviews with schizophrenics, manics and normal subjects were recorded and represented as graphs. Manics scored significantly higher than schizophrenics in ten graph measures. Psychopathological symptoms such as logorrhea, poor speech, and flight of thoughts were grasped by the analysis even when verbosity differences were discounted. Binary classifiers based on speech graph measures sorted schizophrenics from manics with up to 93.8% of sensitivity and 93.7% of specificity. In contrast, sorting based on the scores of two standard psychiatric scales (BPRS and PANSS) reached only 62.5% of sensitivity and specificity.Conclusions/SignificanceThe results demonstrate that alterations of the thought process manifested in the speech of psychotic patients can be objectively measured using graph-theoretical tools, developed to capture specific features of the normal and dysfunctional flow of thought, such as divergence and recurrence. The quantitative analysis of speech graphs is not redundant with standard psychometric scales but rather complementary, as it yields a very accurate sorting of schizophrenics and manics. Overall, the results point to automated psychiatric diagnosis based not on what is said, but on how it is said.

Physiology and assessment as low-hanging fruit for education overhaul

“…Regarding such computational approaches, we have developed speech analyses that are successful on cognitive deficits associated with pathological conditions such as dementia (Bertola et al 2014) and psychosis (Mota et al 2012, and can even predict psychotic breaks more than 2 years in advance during the prodromal phase, i.e. during the initial stages of the disease when symptoms are not very apparent (Bedi, Carrillo, Cecchi, Slezak, Sigman, Mota, Ribeiro, Javitt, Copelli, and Corcoran 2015).…”

“…during the initial stages of the disease when symptoms are not very apparent (Bedi, Carrillo, Cecchi, Slezak, Sigman, Mota, Ribeiro, Javitt, Copelli, and Corcoran 2015). These approaches use structural and semantic features measured on free speech recorded naturalistically, and were successful in low-SES environments in Latin American countries (Mota et al 2012Mota, Copelli, and Ribeiro 2016). Cognitive deficits related to temporal abilities impaired by attention-deficit/hyperactive disorder (ADHD) could be correctly measured by gamelike software, and the discrimination function classified 82.4% of the cases (Méndez et al 2015).…”

Characterization of the relationship between semantic and structural language features in psychiatric diagnosis

“…On the other hand, the representation of word trajectories as directed graphs has revealed that subjects with chronic psychosis speak with significantly less connectedness between words than healthy subjects, and this allows for the automated diagnosis of Schizophrenia [4,7]. Importantly, connectedness attributes were negatively correlated with the severity of negative symptoms measured during standard psychiatric evaluations [3].…”

“…However, even for very well trained psychiatrists, the aberrant thought organization identified through language is hard to measure with precision and without subjective biases. The development of natural language processing tools now enable us to quantify aberrant word trajectories analyzing structural [2][3][4] as well semantic features on patient reports [5,6].…”

Tempo de cérebro

“…Finalmente é possível testar a hipótese de que diferentes estados mentais produzem distintas malhas de relações semânticas. Agora é viável, por exemplo, utilizar análise de grafos matemáticos para quantificar sintomas psicopatológicos anteriormente acessíveis apenas por meio do exame qualitativo e subjetivo realizado por especialistas (Mota et al, 2012). Esse método é útil para realizar o diagnóstico diferencial de pacientes psicóticos, separando pacientes esquizofrênicos de pacientes bipolares em fase maníaca com mais de 93% de especificidade e sensibilidade.…”