The association between childhood language development and adolescent psychotic experiences in a general population sample

Sylvia and Christine Wastall Mental Health Research UK 3/4-Year PhD Studentship in Mental Health

We are looking to recruit an enthusiastic and talented PhD student to work under the supervision of Dr Sarah Sullivan, Research Fellow at the Centres for Academic Primary Care and Mental Health, University of Bristol, Professor Jeff Bowers, School of Experimental Psychology, University of Bristol and Professor Rosemary Varley, Centre for Psychology and Language, University College London. This 4-year studentship provides cross-disciplinary training in state-of-the-art epidemiological approaches to address important questions about the aetiology of psychotic experiences and psychosis. The student will combine novel analyses of a cohort dataset (Avon Longitudinal Study of Parents and Children - ALSPAC) with software which uses automated processes to measure language parameters.
Background
Language is a core human capacity, relying on dedicated mechanisms for grammar and vocabulary but also closely integrated with other cognitive systems such as episodic memory, perception, executive function, and social cognition. Disruption of any of these components may lead to changes in an individual’s communication and in grammatical structure and lexical selection. Language may therefore serve as a barometer of cognitive and mental health across the lifespan.

Psychosis is best viewed as a continuum, ranging from occasional psychotic experiences (PEs) to disorder with serious disability. PEs have a prevalence of 3-30% in community samples. PEs in adolescence are strongly associated with later psychotic disorder, share some risk factors with psychotic disorder and are associated with dysfunction and other illnesses e.g. depression.
There is rich cross-sectional evidence of atypical speech and language in psychosis and longitudinal evidence that some speech/language abnormalities precede psychosis and PEs and may be a marker of underlying processes that increase psychosis risk.

The proposed project assumes causal links between neurological development and linguistic and mental health traits. ALSPAC data offer a unique opportunity. Narrative audio data were collected from nearly 7500 children at age 8. Data on PEs at 12, 18 and 24 years are also available. Identification of childhood language patterns associated with PEs, and therefore later psychosis, would be of great applied, clinical and aetiological value.

Previous longitudinal studies investigating the association between child language and psychosis have used parental-report of their child’s ability via a checklist or questionnaire. We are only aware of one prospective study (also using parent-report data) of adolescent PEs, which found an association with socio-pragmatic aspects of language, but not intelligibility and fluency or total language ability score. Parental-report is affected by random measurement error, making the findings difficult to interpret. More importantly, it does not permit investigation of specific aspects of language. We will use primary language data in the form of transcribed audio recordings, which have not previously been used. This will be the first study to analyse the association between word use and grammatical structure and PEs in a general population sample. We will employ newly-developed methods of automated language analysis to extract quantitative variables for linguistic characteristics, specifically formulaic word combinations and semantic coherence.

Formulaicity captures the degree to which the speaker relies on common words or phrases, and how much she employs her language system to generate novel utterances. Formulas such as I don’t know, I wanna or it’s alright are pre-fabricated units stored in the lexicon and not produced by combinatorial mechanisms. They have very specific socio-pragmatic functions. How much speakers rely on formulas is a sensitive marker for aphasia and dementia (where they increase) and in right-hemisphere lesions and schizophrenia (where they decrease), although this is based on very small cross-sectional samples. Semantic coherence indicates how related units are and reflects the ability to maintain and change topic in an interactionally appropriate manner. Semantic coherence declines in disordered speech and reflects tangentiality and derailment in schizophrenic discourse.
The Aim and Approach
The overall aim of this PhD is to investigate whether specified parameters of childhood language predict later psychotic experiences.
The specific direction this project takes depends on the candidate's research interests. There would be opportunities to investigate the specificity of any association for other mental health conditions and to investigate other language parameters in association with psychosis. This is an expanding and important research area, that so far has been little investigated.

The successful candidate will be trained in psychiatric epidemiology, population health research and language analysis. They should have a willingness to embrace this multi-disciplinary project and a strong interest in risk factors and bio-markers for mental health disorders.
Candidate requirements
Applicants should have, or expect to achieve, at least an upper second class honours degree (or equivalent) and / or with a research Masters degree in a relevant discipline (e.g. psychology, epidemiology).

Interested and suitably qualified candidates should make informal contact with Dr Sarah Sullivan ([Email Address Removed] or 0117 3310074) before 27 June 2019. The intended start date for the project is September 2019, but there is flexibility.
How to apply
Please make an online application for this project at http://www.bris.ac.uk/pg-howtoapply. Please select Faculty of Health Sciences and Population Health Science PhD on the Programme Choice page. You will be prompted to enter details of the studentship in the Funding and Research Details sections of the form. For general enquiries linked to the online application process, please email [Email Address Removed]
Closing date: 5pm, Monday 25th February 2019. Interviews will be held in July 2019.