A key aim of researchers in the clinical psychology field is to identify risk factors for psychological disorder, with the ultimate goal of informing applied research through generating novel intervention targets. However, many challenges exist in this context, including high levels of complexity in terms of contributing factors, often coupled with sample sizes/datasets that are limited in size to support such complexity. This, in turn, has correspondingly limited progress in the field. In particular, while it is clear on a conceptual level that multiple risk factors are likely to interact in the development of psychological disorders, studies are frequently underpowered to investigate statistical interactions, and these often prove unreliable. Machine learning based approaches offer a potential solution to this issue, as they tackle complexity in a fundamentally different way to standard statistical tests and may provide a new means to identifying novel sets of interacting risk factors, even in relatively small samples. However, to date they have had limited application in the field, and there are some major challenges to be overcome in adopting such an approach in a way that is responsible and reliable.
The proposed project will apply machine learning techniques specifically to the study of child trauma and associated psychological disorder. Children who are exposed to trauma are at risk of developing posttraumatic stress disorder in particular, and mental health problems more broadly. Overall, it is estimated that childhood exposure to adverse experiences explains around 40% of the variance in common mental health problems (e.g., depression, anxiety). Nonetheless, the factors that contribute to psychological disorder in some trauma exposed children, versus resilient outcomes in others, are still poorly understood. One problem that has limited progress in the developmental trauma field is the need to study complex interactions, since it is clear that individual risk and resilience factors combine with adverse environmental exposures in predicting disorder. Machine learning based data mining techniques have significant potential to address this complexity, but significant questions remain with respect to their appropriate application. Consistent with this, the aims of the proposed project are twofold.
First, using a variety of existing datasets the successful applicant will apply machine learning and data mining based approaches to construct models of predictive risk factors for the development of trauma-related psychopathology in young people.
Second, through this process the applicant will simultaneously explore key ethical and practical issues in applying machine learning approaches in a field where strong links between research outputs and clinical interventions mean we have a responsibility to generate reliable findings. Specifically, the proposed PhD will examine how we can ensure responsible application of machine learning approaches in the field, and how their robustness (or lack thereof) can be communicated effectively to reviewers, researchers and clinicians who are non-experts in this area.
This project is associated with the UKRI Centre for Doctoral Training in Accountable, Responsible and Transparent AI (ART-AI), which is looking for its second cohort of at least 10 students to start in September 2020. Further details can be found at: http://www.bath.ac.uk/centres-for-doctoral-training/ukri-centre-for-doctoral-training-in-accountable-responsible-and-transparent-ai/
Applicants should hold, or expect to receive, a First or Upper Second Class Honours degree. A master’s level qualification would also be advantageous. Students with a background in psychology or computer science may apply. A willingness to engage with interdisciplinary research is essential.
Informal enquiries about the project should be directed to Prof Sarah Halligan on email address [email protected]
Enquiries about the application process should be sent to [email protected]
Formal applications should be made via the University of Bath’s online application form: https://samis.bath.ac.uk/urd/sits.urd/run/siw_ipp_lgn.login?process=siw_ipp_app&code1=RDUCM-FP02&code2=0002
Start date: 28 September 2020.