Haas, Shalaila S. (2019): Elucidating the efficacy and response to social cognitive training in recent-onset psychosis. Dissertation, LMU München: Faculty of Medicine |
Preview |
PDF
Haas_Shalaila_S.pdf 5MB |
Abstract
Neurocognitive deficits are one of the core features of psychosis spectrum disorders (PSD), and they are predictive of poor functional outcome and negative symptoms many years later (Green, Kern, Braff, & Mintz, 2000). Neurocognitive interventions (NCIs) have emerged in the last two decades as a strong potential supplementary treatment option to improve cognitive deficits and drop in functioning affecting patients with PSD. Social cognitive training (SCT) involving e.g., facial stimuli, has gained considerably more attention in recent studies than computerized NCIs, that use basic visual or auditory stimuli. This is due to the complex character of social cognition (SC) that draws on multiple brain structures involved in behaviors and perception beyond default cognitive function. SC is also tightly interlinked with psychosocial functioning. Although they are cost-effective and quite independent of clinical staff, such technological approaches as SCT are currently not integrated into routine clinical practice. Recent studies have mapped the effects of SCT in task-based studies on multiple brain regions such as the amygdala, putamen, medial prefrontal cortex, and postcentral gyrus (Ramsay & MacDonald III, 2015). Yet, the degree to which alterations in brain function are associated with response to such interventions is still poorly understood. Importantly, resting-state functional connectivity (rsFC) may be a viable neuromarker as it has shown greater sensitivity in distinguishing patients from healthy controls (HC) across neuroimaging studies, and is relatively easy to administer especially in patients with acute symptoms (Kambeitz et al., 2015). In this dissertation, we employed 1) a univariate statistical approach to elucidate the efficacy of a 10-hour SCT in improving cognition, symptoms, functioning and the restoration of rsFC in patients undergoing SCT as compared to the treatment as usual (TAU) group, and 2) multivariate methods. In particular, we used a Support Vector Machine (SVM) approach to neuromonitor the recovery of rsFC in the SCT group compared to TAU. We also investigated the potential utility of rsFC as a baseline (T0) neuromarker viable of predicting role functioning approximately 2 months later. First, current findings suggest a 10-hour SCT has the capability of improving role functioning in recent-onset psychosis (ROP) patients. Second, we have shown intervention-specific rsFC changes within parts of default mode and social cognitive network. Moreover, patients with worse SC performance at T0 showed greater rsFC changes following the intervention, suggestive of a greater degree of rsFC restoration potential in patients with worse social cognitive deficits. Third, when referring to neuromonitoring results, it is important to state that only greater transition from ROP to “HC-like” SVM decision scores, based on the resting-state modality, was paralleled by intervention specific significantly greater improvement in global cognition and attention. Finally, we were able to show the early prediction of good versus poor role functioning is feasible at the individual subject level using a rsFC-based linear SVM classifier with a Balanced Accuracy (BAC) of 74 %. This dissertation sheds light on the effects and feasibility of a relatively short computerized SCT, and the potential utility of multivariate pattern analysis (MVPA) for better clinical stratification of predicted treatment response based on rsFC neuromarkers.
Item Type: | Theses (Dissertation, LMU Munich) |
---|---|
Subjects: | 600 Technology, Medicine 600 Technology, Medicine > 610 Medical sciences and medicine |
Faculties: | Faculty of Medicine |
Language: | English |
Date of oral examination: | 22. July 2019 |
1. Referee: | Koutsouleris, Nikolaos |
MD5 Checksum of the PDF-file: | e1a6e0f83c17f73facf932055ff0639b |
Signature of the printed copy: | 0700/UMD 18614 |
ID Code: | 24519 |
Deposited On: | 06. Aug 2019 14:08 |
Last Modified: | 23. Oct 2020 15:17 |