Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
JAMA Psychiatry. Author manuscript; available in PMC 2018 January 1.
Published in final edited form as:
PMCID: PMC5337304

Distinct Relationships between Visual and Auditory Perceptual Abnormalities and Conversion to Psychosis in a Clinical High-Risk Population


Hallucinations are a ubiquitous symptom experienced across psychotic disorders. This symptom, particularly in its auditory form, is widely thought to share a common substrate with other positive symptoms1. However, only unusual thought content and thought disorganization, but not perceptual abnormalities (e.g., hyperacusis, illusions, momentary hallucinations), predict conversion to psychosis in clinical high-risk (CHR) populations2,3, despite the high incidence of perceptual abnormalities2. Importantly, although the frequencies of auditory and visual perceptual abnormalities in CHR individuals are similar4, differences in the prevalence of auditory and visual hallucinations in full psychosis5 may suggest distinct neurobiological substrates. Therefore, we hypothesized that auditory and visual perceptual abnormalities would have distinct clinical correlates in CHR individuals.


Participants were help-seeking CHR outpatients, aged 13–30, in the Center of Prevention and Evaluation (COPE), NYSPI, and part of a longitudinal study approved by the NYSPI IRB. All met criteria for the Attenuated Positive Symptom Syndrome (APSS)6 defined in the Structured Interview for Psychosis-Risk Syndromes (SIPS6; n=203; 57 converters, 144 non-converters, 2 unknown; Table 1). Symptoms were not better explained by DSM disorders or current substance use. Participants were followed up every 3 months for up to 2 years. Conversion to psychosis was defined based on standard SIPS criteria6. Most converters (81%) had follow-up DSM diagnoses of schizophrenia or related disorders.

Table 1
Demographic and Baseline Clinical Characteristics of Clinical High Risk (CHR) Individuals at Enrollmenta

Using detailed vignettes and rater comments in the SIPS collected at enrollment (baseline), two independent, SIPS-certified raters (RRG, GB) assigned perceptual-abnormality scores, originally encompassing any abnormalities across sensory domains within the past month (P4)6, separately to auditory (denoted P4a, intra-class correlation coefficient rICC=0.98) and visual experiences (denoted P4v, rICC=0.91; Figure 1). Relationships to conversion were assessed via logistic regression and Cox proportional-hazards regression. Adults provided written informed consent; minors provided assent with written consent by a parent or legal guardian.

Figure 1
Longitudinal relationship between baseline severity of auditory and visual perceptual abnormalities (P4a and P4v, respectively) and clinical outcomes at follow-up


Consistent with previous research, the original global P4 scores did not predict conversion status (β=0.01, p=0.92). However, when split into two variables, both P4a and P4v independently predicted conversion status, although in opposite directions (βa=0.25, p=0.046; βv=−0.44, p=0.001; Figure 1A). P4a scores were higher than P4v scores (p<0.001; 67% versus 57% scored above 1, p<0.001); P4a and P4v overlapped moderately (R2=0.25; 48% scored above 1 for both). Additionally, P4a, but not P4v, predicted summed scores across P1, P2, P3, and P5 (βa=0.37, p=0.037; βv=0.14, p=0.45; βav: p=0.014; see Table 1 for item descriptions) cross-sectionally, indicating that auditory abnormalities are more strongly related to other attenuated positive symptoms than are visual abnormalities. Post-conversion P4a and P4v scores remained stable relative to baseline and longitudinally correlated within, but not between, domains (Figure 1B).

Similarly, survival analyses showed that global P4 scores did not predict days to conversion (β=0.01, p=0.95) while P4a and P4v were independent predictors in opposite directions (βa=0.19, p=0.052; βv=−0.37, p<0.001). In an extended model controlling for known predictors of conversion (P scores other than P4, NT, DT, GT, GAF, and demographics), P4v was the single strongest (negative) independent predictor of days to conversion (βv=−0.38, Z=−3.36, p<0.001); P4a, P1, P5, and NT were the only additional (positive) independent predictors that were at least marginally significant (1.64<Z<2.11; p-values 0.073, 0.035, 0.045, 0.099, respectively).


Visual perceptual abnormalities strongly predicted lower risk of conversion to psychosis, while auditory abnormalities predicted higher risk, along with P1 and P5 scores. Additionally, our cross-sectional data support that core positive symptoms (including auditory, but not visual, abnormalities) may have a common substrate. Visual abnormalities in the CHR may thus have distinct underlying substrates and forecast distinct clinical outcomes (e.g., less severe or non-psychotic disorders). Further research into the clinical value of concurrently measuring visual and auditory abnormalities and their corresponding neurobiological substrates is warranted.


This project was supported by NIMH grants K23MH106746 (RRG), K23MH101637 (GH), and R01MH093398, and by the New York State Office of Mental Hygiene. The sponsors had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. Dr. Girgis receives research support from Otsuka, Forest, Bioadvantex and Genentech. Drs. Horga and Girgis had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Data analysis was conducted by Drs. Horga and Girgis, both affiliated with Columbia University, Department of Psychiatry, New York State Psychiatric Institute, New York, New York, USA.


Authors’ contributions:

Study concept and design: All authors.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the first version of the manuscript: Lehembre-Shiah and Leong.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Horga, Girgis


1. Fletcher P, Frith C. Perceiving is believing: a Bayesian approach to explaining the positive symptoms of schizophrenia. Nat Rev Neurosci. 2009;10(1):48–58. [PubMed]
2. Addington J, Liu L, Buchy L, et al. North American Prodrome Longitudinal Study (NAPLS 2): The Prodromal Symptoms. J Nerv Ment Dis. 2015 May;203(5):328–335. [PMC free article] [PubMed]
3. Poe SL, Gill KE, Brucato G, Girgis RR. Parental age and attenuated thought disorder in a cohort at clinical high risk for psychosis. Schizophrenia Research. 2015;161(2–3):511–512. [PubMed]
4. Marshall C, Denny E, Cadenhead KS, et al. The content of attenuated psychotic symptoms in those at clinical high risk for psychosis. Psychiatry Res. 2014 Nov 30;219(3):506–512. [PMC free article] [PubMed]
5. Waters F, Collerton D, Ffytche DH, et al. Visual hallucinations in the psychosis spectrum and comparative information from neurodegenerative disorders and eye disease. Schizophr Bull. 2014 Jul;40(4):S233–S245. [PMC free article] [PubMed]
6. Miller TJ, McGlashan TH, Rosen JL, et al. Prodromal assessment with the Sstructured Interview for Prodromal Syndromes: Predictive validity, interrater reliability, and training to reliability. Schizophrenia Bull. 2003;29:703–715. [PubMed]