Trauma-focused interventions are effective for PTSD; a key active ingredient being the memory reprocessing/updating. Despite this model being introduced almost 25 years ago, the neural representations of trauma memories remain poorly characterised, and it remains to be tested whether reprocessing impacts these representations, particularly in people with co-occurring PTSD and psychosis symptoms.
Aims: To examine whether trauma-focused CBT for psychosis (TF-CBTp) impacts trauma representations, using functional MRI to drive at three potential neurobiological mechanisms of change:
1) Retrieval: In a dual-process model, trauma memories arise from excessive threat item memory, encoded in amygdala, alongside weak hippocampal representations of spatial/temporal context (Bisby & Burgess, 2017). We will ascertain whether TF-CBTp normalises amygdala-hippocampal connectivity, with Dynamic causal modelling (DCM) revealing change in network structure and information flow.
2) Reactivation: Recent advances implicate hippocampal reactivation of memory traces during (re)consolidation, and this “replay” of trauma stimuli predicts intrusion symptoms days after encoding (Hørlyck, 2018). We have developed a method that captures real-time trauma reconsolidation using support vector machine learning and representational similarity analysis, and will ascertain whether these multivariate signatures are impacted by TF-CBTp.
3) Reappraisal of social threat: Empirical evidence suggests that CBTp normalises prefronto-amygdalar connections underpinning social threat, predicting long-term remission (Mason et al 2016; 2017). We will ascertain whether this mechanism mediates real-life improvements in paranoia (smartphone-based Ecological Momentary Assessment; ESM).
Dataset: Secondary analysis STAR (Study of Trauma And Recovery) trial data; participants with psychosis+PTSD symptoms receiving TF-CBTp or Treatment-As-Usual. Approximately n=75 completed fMRI+EMA, pre- and post-therapy. The student would spearhead an fMRI study in trauma-resilient controls (experienced trauma, did not develop PTSD).
Techniques & Skills: Multilevel modelling, fMRI (DCM, MVPA). The student will work alongside a world-leading, multisite team.
Y1: Training in analyses, literature review, finalising research questions.
Y2: Conduct fMRI study in traumas-exposed controls,
Y3: Analyses of study hypotheses, start write up papers.
Y4: Complete write up papers and thesis.