1
|
Fradkin I, Nour MM, Dolan RJ. Theory-Driven Analysis of Natural Language Processing Measures of Thought Disorder Using Generative Language Modeling. BIOLOGICAL PSYCHIATRY. COGNITIVE NEUROSCIENCE AND NEUROIMAGING 2023; 8:1013-1023. [PMID: 37257754 DOI: 10.1016/j.bpsc.2023.05.005] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 11/30/2022] [Revised: 05/19/2023] [Accepted: 05/19/2023] [Indexed: 06/02/2023]
Abstract
BACKGROUND Natural language processing (NLP) holds promise to transform psychiatric research and practice. A pertinent example is the success of NLP in the automatic detection of speech disorganization in formal thought disorder (FTD). However, we lack an understanding of precisely what common NLP metrics measure and how they relate to theoretical accounts of FTD. We propose tackling these questions by using deep generative language models to simulate FTD-like narratives by perturbing computational parameters instantiating theory-based mechanisms of FTD. METHODS We simulated FTD-like narratives using Generative-Pretrained-Transformer-2 by either increasing word selection stochasticity or limiting the model's memory span. We then examined the sensitivity of common NLP measures of derailment (semantic distance between consecutive words or sentences) and tangentiality (how quickly meaning drifts away from the topic) in detecting and dissociating the 2 underlying impairments. RESULTS Both parameters led to narratives characterized by greater semantic distance between consecutive sentences. Conversely, semantic distance between words was increased by increasing stochasticity, but decreased by limiting memory span. An NLP measure of tangentiality was uniquely predicted by limited memory span. The effects of limited memory span were nonmonotonic in that forgetting the global context resulted in sentences that were semantically closer to their local, intermediate context. Finally, different methods for encoding the meaning of sentences varied dramatically in performance. CONCLUSIONS This work validates a simulation-based approach as a valuable tool for hypothesis generation and mechanistic analysis of NLP markers in psychiatry. To facilitate dissemination of this approach, we accompany the paper with a hands-on Python tutorial.
Collapse
Affiliation(s)
- Isaac Fradkin
- Max Planck University College London Centre for Computational Psychiatry and Ageing Research, London, United Kingdom.
| | - Matthew M Nour
- Max Planck University College London Centre for Computational Psychiatry and Ageing Research, London, United Kingdom; Department of Psychiatry, University of Oxford, Oxford, United Kingdom
| | - Raymond J Dolan
- Max Planck University College London Centre for Computational Psychiatry and Ageing Research, London, United Kingdom; Wellcome Trust Centre for Human Neuroimaging, University College London, London, United Kingdom; State Key Laboratory of Cognitive Neuroscience and Learning, IDG/McGovern Institute for Brain Research, Beijing Normal University, Beijing, China
| |
Collapse
|
2
|
The altered state of consciousness induced by Δ9-THC. Conscious Cogn 2022; 102:103357. [DOI: 10.1016/j.concog.2022.103357] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/29/2021] [Revised: 03/07/2022] [Accepted: 05/19/2022] [Indexed: 11/22/2022]
|
3
|
Tagliazucchi E. Language as a Window Into the Altered State of Consciousness Elicited by Psychedelic Drugs. Front Pharmacol 2022; 13:812227. [PMID: 35392561 PMCID: PMC8980225 DOI: 10.3389/fphar.2022.812227] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/09/2021] [Accepted: 03/01/2022] [Indexed: 11/22/2022] Open
Abstract
Psychedelics are drugs capable of eliciting profound alterations in the subjective experience of the users, sometimes with long-lasting consequences. Because of this, psychedelic research tends to focus on human subjects, given their capacity to construct detailed narratives about the contents of their consciousness experiences. In spite of its relevance, the interaction between serotonergic psychedelics and language production is comparatively understudied in the recent literature. This review is focused on two aspects of this interaction: how the acute effects of psychedelic drugs impact on speech organization regardless of its semantic content, and how to characterize the subjective effects of psychedelic drugs by analyzing the semantic content of written retrospective reports. We show that the computational characterization of language production is capable of partially predicting the therapeutic outcome of individual experiences, relate the effects elicited by psychedelics with those associated with other altered states of consciousness, draw comparisons between the psychedelic state and the symptomatology of certain psychiatric disorders, and investigate the neurochemical profile and mechanism of action of different psychedelic drugs. We conclude that researchers studying psychedelics can considerably expand the range of their potential scientific conclusions by analyzing brief interviews obtained before, during and after the acute effects. Finally, we list a series of questions and open problems that should be addressed to further consolidate this approach.
Collapse
Affiliation(s)
- Enzo Tagliazucchi
- Latin American Brain Health Institute (BrainLat), Universidad Adolfo Ibanez, Santiago, Chile
- Departamento de Física, Universidad de Buenos Aires and Instituto de Física de Buenos Aires (IFIBA, CONICET), Pabellón I, Ciudad Universitaria (1428), Buenos Aires, Argentina
| |
Collapse
|
4
|
Angers K, Suhr JA, Buelow MT. Cognitive-perceptual and disorganized schizotypal traits are nonlinearly related to atypical semantic content on tasks of semantic fluency. J Psychiatr Res 2021; 136:7-13. [PMID: 33545647 DOI: 10.1016/j.jpsychires.2021.01.038] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 07/28/2020] [Revised: 01/19/2021] [Accepted: 01/22/2021] [Indexed: 11/26/2022]
Abstract
Language deficits emerge early in the course of schizophrenia, yet research findings in those at-risk for schizophrenia, such as those with schizotypy, are mixed. The purpose of the present study was to elucidate the relationship of language ability, measured via semantic fluency, to schizotypy, examining both linear and non-linear relations. Semantic fluency data from 295 individuals with varying amounts of schizotypal traits were analyzed utilizing traditional methods (i.e., counting words generated that fit a specific semantic category). The content of semantic fluency responses was also analyzed via a semantic infrequency score (i.e., how infrequent participant responses were relative to all responses generated for the category in the study sample) and a total semantic productivity score (i.e., how many unique words generated overall, including those that did not fit the semantic category). Using traditional methods of scoring, schizotypy was not related to semantic fluency. However, schizotypy was non-linearly related to semantic infrequency and productivity, reflecting atypical semantic activation and processing. In particular, cognitive-perceptual and disorganized, but not interpersonal, traits were related to semantic infrequency and productivity. Valuable content-based information is missed when only analyzing semantic fluency data via the traditional method in the schizophrenia spectrum population. Cognitive-perceptual and disorganized traits, attenuated thought disorder symptoms, evidence the strongest relationship to semantic fluency, further illustrating the link between language and schizophrenia symptoms along the schizophrenia spectrum.
Collapse
Affiliation(s)
- Kaley Angers
- Ohio University, Department of Psychology, Porter Hall, 22 Richland Ave., Athens, OH, 45701, USA.
| | - Julie A Suhr
- Ohio University, Department of Psychology, Porter Hall, 22 Richland Ave., Athens, OH, 45701, USA
| | - Melissa T Buelow
- The Ohio State University Newark, Department of Psychology, 2048 Founders Hall, 1179 University Drive, Newark, OH, 43055, USA
| |
Collapse
|
5
|
Lundin NB, Todd PM, Jones MN, Avery JE, O'Donnell BF, Hetrick WP. Semantic Search in Psychosis: Modeling Local Exploitation and Global Exploration. SCHIZOPHRENIA BULLETIN OPEN 2020; 1:sgaa011. [PMID: 32803160 PMCID: PMC7418865 DOI: 10.1093/schizbullopen/sgaa011] [Citation(s) in RCA: 9] [Impact Index Per Article: 2.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Indexed: 01/10/2023]
Abstract
Impairments in category verbal fluency task (VFT) performance have been widely documented in psychosis. These deficits may be due to disturbed “cognitive foraging” in semantic space, in terms of altered salience of cues that influence individuals to search locally within a subcategory of semantically related responses (“clustering”) or globally between subcategories (“switching”). To test this, we conducted a study in which individuals with schizophrenia (n = 21), schizotypal personality traits (n = 25), and healthy controls (n = 40) performed VFT with “animals” as the category. Distributional semantic model Word2Vec computed cosine-based similarities between words according to their statistical usage in a large text corpus. We then applied a validated foraging-based search model to these similarity values to obtain salience indices of frequency-based global search cues and similarity-based local cues. Analyses examined whether diagnosis predicted VFT performance, search strategies, cue salience, and the time taken to switch between vs search within clusters. Compared to control and schizotypal groups, individuals with schizophrenia produced fewer words, switched less, and exhibited higher global cue salience, indicating a selection of more common words when switching to new clusters. Global cue salience negatively associated with vocabulary ability in controls and processing speed in schizophrenia. Lastly, individuals with schizophrenia took a similar amount of time to switch to new clusters compared to control and schizotypal groups but took longer to transition between words within clusters. Findings of altered local exploitation and global exploration through semantic memory provide preliminary evidence of aberrant cognitive foraging in schizophrenia.
Collapse
Affiliation(s)
- Nancy B Lundin
- Department of Psychological and Brain Sciences, Indiana University, Bloomington, IN.,Program in Neuroscience, Indiana University, Bloomington, IN
| | - Peter M Todd
- Department of Psychological and Brain Sciences, Indiana University, Bloomington, IN.,Cognitive Science Program, Indiana University, Bloomington, IN
| | - Michael N Jones
- Department of Psychological and Brain Sciences, Indiana University, Bloomington, IN.,Cognitive Science Program, Indiana University, Bloomington, IN
| | - Johnathan E Avery
- Department of Psychological and Brain Sciences, Indiana University, Bloomington, IN.,Cognitive Science Program, Indiana University, Bloomington, IN
| | - Brian F O'Donnell
- Department of Psychological and Brain Sciences, Indiana University, Bloomington, IN.,Program in Neuroscience, Indiana University, Bloomington, IN.,Department of Psychiatry, Indiana University School of Medicine, Indianapolis, IN
| | - William P Hetrick
- Department of Psychological and Brain Sciences, Indiana University, Bloomington, IN.,Program in Neuroscience, Indiana University, Bloomington, IN.,Department of Psychiatry, Indiana University School of Medicine, Indianapolis, IN
| |
Collapse
|
6
|
Minor KS, Willits JA, Marggraf MP, Jones MN, Lysaker PH. Measuring disorganized speech in schizophrenia: automated analysis explains variance in cognitive deficits beyond clinician-rated scales. Psychol Med 2019; 49:440-448. [PMID: 29692287 DOI: 10.1017/s0033291718001046] [Citation(s) in RCA: 34] [Impact Index Per Article: 6.8] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 12/31/2022]
Abstract
BACKGROUND Conveying information cohesively is an essential element of communication that is disrupted in schizophrenia. These disruptions are typically expressed through disorganized symptoms, which have been linked to neurocognitive, social cognitive, and metacognitive deficits. Automated analysis can objectively assess disorganization within sentences, between sentences, and across paragraphs by comparing explicit communication to a large text corpus. METHOD Little work in schizophrenia has tested: (1) links between disorganized symptoms measured via automated analysis and neurocognition, social cognition, or metacognition; and (2) if automated analysis explains incremental variance in cognitive processes beyond clinician-rated scales. Disorganization was measured in schizophrenia (n = 81) with Coh-Metrix 3.0, an automated program that calculates basic and complex language indices. Trained staff also assessed neurocognition, social cognition, metacognition, and clinician-rated disorganization. RESULTS Findings showed that all three cognitive processes were significantly associated with at least one automated index of disorganization. When automated analysis was compared with a clinician-rated scale, it accounted for significant variance in neurocognition and metacognition beyond the clinician-rated measure. When combined, these two methods explained 28-31% of the variance in neurocognition, social cognition, and metacognition. CONCLUSIONS This study illustrated how automated analysis can highlight the specific role of disorganization in neurocognition, social cognition, and metacognition. Generally, those with poor cognition also displayed more disorganization in their speech-making it difficult for listeners to process essential information needed to tie the speaker's ideas together. Our findings showcase how implementing a mixed-methods approach in schizophrenia can explain substantial variance in cognitive processes.
Collapse
Affiliation(s)
- K S Minor
- Department of Psychology,Indiana University- Purdue University Indianapolis,Indianapolis, IN,USA
| | - J A Willits
- Department of Psychology,University of California-Riverside,Riverside, CA,USA
| | - M P Marggraf
- Department of Psychology,Indiana University- Purdue University Indianapolis,Indianapolis, IN,USA
| | - M N Jones
- Department of Psychology,Indiana University,Bloomington, IN,USA
| | - P H Lysaker
- Roudebush VA Medical Center,Indianapolis, IN,USA
| |
Collapse
|
7
|
Semantic coherence in psychometric schizotypy: An investigation using Latent Semantic Analysis. Psychiatry Res 2018; 259:63-67. [PMID: 29028526 DOI: 10.1016/j.psychres.2017.09.078] [Citation(s) in RCA: 8] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 06/26/2016] [Revised: 05/23/2017] [Accepted: 09/25/2017] [Indexed: 12/30/2022]
Abstract
Technological advancements have led to the development of automated methods for assessing semantic coherence in psychiatric populations. Latent Semantic Analysis (LSA) is an automated method that has been used to quantify semantic coherence in schizophrenia-spectrum disorders. The current study examined whether: 1) Semantic coherence reductions extended to psychometrically-defined schizotypy and 2) Greater cognitive load further reduces semantic coherence. LSA was applied to responses generated during category fluency tasks in baseline and cognitive load conditions. Significant differences between schizotypy and non-schizotypy groups were not observed. Findings suggest that semantic coherence may be relatively preserved at this point on the schizophrenia-spectrum.
Collapse
|