File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Conference Paper: REINFORCEMENT LEARNING IMPAIRMENT AND PRIMARY NEGATIVE SYMPTOMS IN INDIVIDUALS AT CLINICAL HIGH-RISK FOR PSYCHOSIS

TitleREINFORCEMENT LEARNING IMPAIRMENT AND PRIMARY NEGATIVE SYMPTOMS IN INDIVIDUALS AT CLINICAL HIGH-RISK FOR PSYCHOSIS
Authors
Issue Date2019
PublisherOxford University Press. The Journal's web site is located at http://schizophreniabulletin.oxfordjournals.org/
Citation
2019 Congress of the Schizophrenia International Research Society (SIRS), Orlando, Florida, 10-14 April 2019. In Schizophrenia Bulletin, 2019, v. 45 n. Suppl. 2, p. S222-S222 How to Cite?
AbstractBackground: Reinforcement learning (RL) impairment has been observed in chronic schizophrenia and first-episode psychosis. Yet, there is a paucity of research of RL impairment in individuals at clinical high-risk (CHR) for psychosis. The current study aimed to examine RL performance in a representative cohort of Chinese CHR subjects, with particular focus on its relationship with primary negative symptoms (PNS). Methods: Ninety-seven Chinese individuals with CHR for psychosis, aged 15–40 years, were recruited from a territory-wide specialized early intervention service for psychosis in Hong Kong. CHR status was verified using Comprehensive Assessment of At-Risk Mental State (CAARMS). Thirty-four demographically-matched healthy controls were recruited as a comparison group. Each subject completed two computerized RL tasks (Go-No-Go, GNG and Gains-Loss Avoidance GLA tasks) which have been studied in chronic and first-episode schizophrenia patients. In both tasks, rapid / gradual and positive / negative RL measures were derived for analyses. CHR subjects were categorized as having PNS if they had (1) global score ≥ 3 on at least two of the following SANS subscales: Affective flattening, Alogia, Avolition-apathy or Anhednoia-asociality; and (2) no or clinically non-significant depression with total score < 16 in Montgomery-Asberg Depression Rating Scale (MADRS). Results: On GNG task, three-group comparison (PNS, non-PNS, control groups) revealed significant difference in RL accuracy, with post-hoc contrasts showing that controls performed better than both PNS and non-PNS groups in gradual positive RL. No between-group difference in Go-response bias or rapid learning was observed. On GLA task, main effect of group was noted in three-group comparison analysis on RL accuracy. Post-hoc tests indicated that PNS group displayed significantly lower accuracy than both non-PNS group and controls in gradual positive and negative RL. Additionally, PNS group exhibited significantly lower overall and block-1 win-stay rates than controls. Discussion: Our results indicate RL impairment in CHR sample. In particular, such RL impairment was more evident in CHR subjects presenting with PNS relative to those without PNS. Further investigation is required to verify and confirm our findings on the relationship between negative symptoms and RL deficits. In addition, a prospective follow-up of our CHR cohort will help clarify the potential utility of baseline RL impairment in enhancing prediction of psychosis transition at follow-up.
DescriptionPoster Session I - no. T47
Persistent Identifierhttp://hdl.handle.net/10722/279539
ISSN
2019 Impact Factor: 7.958
2015 SCImago Journal Rankings: 4.051

 

DC FieldValueLanguage
dc.contributor.authorChang, WC-
dc.contributor.authorWo, SF-
dc.contributor.authorWong, CF-
dc.contributor.authorLee, HC-
dc.contributor.authorWaltz, J-
dc.contributor.authorGold, J-
dc.contributor.authorChan, SI-
dc.contributor.authorChiu, S-
dc.contributor.authorLee, HME-
dc.contributor.authorChan, KW-
dc.contributor.authorHui, CLM-
dc.contributor.authorSuen, YN-
dc.contributor.authorChen, EYH-
dc.date.accessioned2019-11-01T07:19:17Z-
dc.date.available2019-11-01T07:19:17Z-
dc.date.issued2019-
dc.identifier.citation2019 Congress of the Schizophrenia International Research Society (SIRS), Orlando, Florida, 10-14 April 2019. In Schizophrenia Bulletin, 2019, v. 45 n. Suppl. 2, p. S222-S222-
dc.identifier.issn0586-7614-
dc.identifier.urihttp://hdl.handle.net/10722/279539-
dc.descriptionPoster Session I - no. T47-
dc.description.abstractBackground: Reinforcement learning (RL) impairment has been observed in chronic schizophrenia and first-episode psychosis. Yet, there is a paucity of research of RL impairment in individuals at clinical high-risk (CHR) for psychosis. The current study aimed to examine RL performance in a representative cohort of Chinese CHR subjects, with particular focus on its relationship with primary negative symptoms (PNS). Methods: Ninety-seven Chinese individuals with CHR for psychosis, aged 15–40 years, were recruited from a territory-wide specialized early intervention service for psychosis in Hong Kong. CHR status was verified using Comprehensive Assessment of At-Risk Mental State (CAARMS). Thirty-four demographically-matched healthy controls were recruited as a comparison group. Each subject completed two computerized RL tasks (Go-No-Go, GNG and Gains-Loss Avoidance GLA tasks) which have been studied in chronic and first-episode schizophrenia patients. In both tasks, rapid / gradual and positive / negative RL measures were derived for analyses. CHR subjects were categorized as having PNS if they had (1) global score ≥ 3 on at least two of the following SANS subscales: Affective flattening, Alogia, Avolition-apathy or Anhednoia-asociality; and (2) no or clinically non-significant depression with total score < 16 in Montgomery-Asberg Depression Rating Scale (MADRS). Results: On GNG task, three-group comparison (PNS, non-PNS, control groups) revealed significant difference in RL accuracy, with post-hoc contrasts showing that controls performed better than both PNS and non-PNS groups in gradual positive RL. No between-group difference in Go-response bias or rapid learning was observed. On GLA task, main effect of group was noted in three-group comparison analysis on RL accuracy. Post-hoc tests indicated that PNS group displayed significantly lower accuracy than both non-PNS group and controls in gradual positive and negative RL. Additionally, PNS group exhibited significantly lower overall and block-1 win-stay rates than controls. Discussion: Our results indicate RL impairment in CHR sample. In particular, such RL impairment was more evident in CHR subjects presenting with PNS relative to those without PNS. Further investigation is required to verify and confirm our findings on the relationship between negative symptoms and RL deficits. In addition, a prospective follow-up of our CHR cohort will help clarify the potential utility of baseline RL impairment in enhancing prediction of psychosis transition at follow-up.-
dc.languageeng-
dc.publisherOxford University Press. The Journal's web site is located at http://schizophreniabulletin.oxfordjournals.org/-
dc.relation.ispartofSchizophrenia Bulletin-
dc.relation.ispartof2019 Congress of the Schizophrenia International Research Society (SIRS)-
dc.titleREINFORCEMENT LEARNING IMPAIRMENT AND PRIMARY NEGATIVE SYMPTOMS IN INDIVIDUALS AT CLINICAL HIGH-RISK FOR PSYCHOSIS-
dc.typeConference_Paper-
dc.identifier.emailChang, WC: changwc@hku.hk-
dc.identifier.emailWong, CF: scfwong@hku.hk-
dc.identifier.emailChan, SI: sherinac@hku.hk-
dc.identifier.emailLee, HME: edwinlhm@hku.hk-
dc.identifier.emailChan, KW: kwsherry@hku.hk-
dc.identifier.emailHui, CLM: christyh@hku.hk-
dc.identifier.emailSuen, YN: suenyn@hku.hk-
dc.identifier.emailChen, EYH: eyhchen@hku.hk-
dc.identifier.authorityChang, WC=rp01465-
dc.identifier.authorityLee, HME=rp01575-
dc.identifier.authorityChan, KW=rp00539-
dc.identifier.authorityHui, CLM=rp01993-
dc.identifier.authoritySuen, YN=rp02481-
dc.identifier.authorityChen, EYH=rp00392-
dc.identifier.doi10.1093/schbul/sbz019.327-
dc.identifier.hkuros308313-
dc.identifier.volume45-
dc.identifier.issueSuppl. 2-
dc.identifier.spageS222-
dc.identifier.epageS222-
dc.publisher.placeUnited Kingdom-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats