Item-specific overlap between hallucinatory experiences and cognition in the general population: A three-step multivariate analysis of international multi-site data

Chinchani, Abhijit M., Menon, Mahesh, Roes, Meighen, Hwang, Heungsun, Allen, Paul, Bell, Vaughan, Bless, Josef, Bortolon, Catherine, Cella, Matteo, Fernyhough, Charles, Garrison, Jane, Kozáková, Eva, Laroi, Frank, Moffatt, Jamie, Say, Nicolas, Suzuki, Mimi, Toh, Wei Lin, Zaytseva, Yuliya, Rossell, Susan L., Moseley, Peter and Woodward, Todd S. (2021) Item-specific overlap between hallucinatory experiences and cognition in the general population: A three-step multivariate analysis of international multi-site data. Cortex, 145. pp. 131-144. ISSN 0010-9452

Multisite_paper_2021_01_31_PsyArXiv.pdf - Accepted Version
Available under License Creative Commons Attribution Non-commercial No Derivatives 4.0.

Download (674kB) | Preview
Official URL:


Hallucinatory experiences (HEs) can be pronounced in psychosis, but similar experiences also occur in nonclinical populations. Cognitive mechanisms hypothesized to underpin HEs include dysfunctional source monitoring, heightened signal detection, and impaired attentional processes. Using data from an international multisite study on non-clinical participants (N = 419), we described the overlap between two sets of variables - one measuring cognition and the other HEs - at the level of individual items. We used a three-step method to extract and examine item-specific signal, which is typically obscured when summary scores are analyzed using traditional methodologies. The three-step method involved: (1) constraining variance in cognition variables to that which is predictable from HE variables, followed by dimension reduction, (2) determining reliable HE items using split-halves and permutation tests, and (3) selecting cognition items for interpretation using a leave-one-out procedure followed by repetition of Steps 1 and 2. The results showed that the overlap between HEs and cognition variables can be conceptualized as bi-dimensional, with two distinct mechanisms emerging as candidates for separate pathways to the development of HEs: HEs involving perceptual distortions on one hand (including voices), underpinned by a low threshold for signal detection in cognition, and HEs involving sensory overload on the other hand, underpinned by reduced laterality in cognition. We propose that these two dimensions of HEs involving distortions/liberal signal detection, and sensation overload/reduced laterality may map onto psychosis-spectrum and dissociation-spectrum anomalous experiences, respectively.

Item Type: Article
Additional Information: Funding information: This research was supported by the Wellcome Trust grant WT108720 awarded to CF. SLR is supported by a NHMRC senior research fellowship (GNT1154651), and WLT is supported by a NHMRC New Investigator project grant (GNT1161609). AMC is supported by Brain Canada, in partnership with Health Canada, for the Canadian Open Neuroscience Platform (CONP) initiative. EK and YZ are supported by the Internal Grant Agency of the Ministry of Health of the Czech Republic [grant number AZV 17-32957A] and by the project “Sustainability for the National Institute of Mental Health”, under grant number LO1611. JB is supported by grants from ERC Advanced Grant (ERCAdG #693124), from the Helse-Vest Samarbeidsorganet (#912045), and from the Research Council of Norway (NORMENT#213363).
Subjects: C800 Psychology
Department: Faculties > Health and Life Sciences > Psychology
Depositing User: John Coen
Date Deposited: 06 Oct 2021 12:48
Last Modified: 01 Oct 2022 08:00

Actions (login required)

View Item View Item


Downloads per month over past year

View more statistics