Anhedonia is the core feature of negative symptoms of schizophrenia, and mainly manifests as the deficits of anticipatory pleasure, i.e., reduced pleasurable experiences when anticipating future events. The reduced ability to activate and maintain the representation of value and emotion is associated with working memory (WM) deficit. On the other hand, empirical findings suggest the neural mechanisms of hedonic processing and WM share recruitment of the dorsolateral prefrontal cortex, the anterior cingulate cortex and the striatum.The present dissertation aimed specifically to examine whether WM training would improve anhedonia observed in patients with schizophrenia and individuals with social anhedonia.
Study 1 adopted the activation likelihood estimation (ALE) meta-analysis to examine whether there would be neuroplastic effect of WM training in healthy volunteers and patients with schizophrenia.The results showed that there was a widespread distribution of activation changes with WM training in the prefrontal, the parietal, the temporal regions, as well as the insula and the putamen in healthy individuals. WM was also accompanied with brain activation changes in patients with schizophrenia, mainly in the dorsolateral prefrontal cortex, the precuneus and the fusiform gyrus. The analysis of neural effects of WM training in healthy volunteers at the dorsolateral prefrontal and the sub-cortical regions provides the basis for the understanding of neuroplastic changes in patients with schizophrenia and sheds light on the possible neural transfer effects to hedonic processing.
Study 2 aimed to examine whether WM training would improve anhedonia observed in individuals with social anhedonia. We recruited 17 individuals with high social anhedonia to undertake a 20-sessions training on the dual n-back task. Another 17 individuals with high socialanhedonia who did not undertake the dual n-back raining were recruited as control group. The results showed that individuals who received the WM training not only showed significant improvement in WM performance compared to controls, but also showed significant improvement in approach sensitivity to rewards.
Study 3 recruited an independent sample of high social anhedonia group to examine whether the observed behavioural improvement shown in study 2 could be extended to the underlying mechanisms. Fifteen individuals with high social anhedonia were recruited to undertake the said dual n- back training described in study 2. Moreover, they were required to undertake two functional imaging tasks on the Monetary Incentive Delay (MID) task and the Affective Incentive Delay (AID) task both before and after the training. The results indicated that WM training could improve the behavioural performance of WM tasks and reduce the severity of self-reported anhedonia. Enhanced brain activations related to pleasure anticipation were observed at the anterior cingulate cortex, the left dorsal striatum and the left precuneus on the AID task, and at the dorsolateral prefrontal cortex and the supramarginal gyrus on the MID task. Reduced brain activations were observed in the consummation of both affective and monetary incentives across several frontal and parietal regions, as well as some subcortical regions.
In study 4, we examined whether the observed neural improvement shown in study 3 could be extended to schizophrenia patients with prominent negative symptoms. We recruited 15 patients with schizophrenia to undertake the 20-sessions of the dual n-back task for 4 weeks. Moreover, we also recruited seven patients with schizophrenia who did not take part the training as the clinical controls. The results showed that there was reduced brain activation at the middle frontal gyurs when performing the WM task, and the reduction of activation correlated significantly with behavorial improvement on the dual n-back task. Moreover, increased brain activation at the superior temporal gyrus in the anticipating of affective rewards was observed, and the increased brain activation at the superior temporal gyrus negatively correlated with reduction of negative symptoms on affective flattening and avolition-apathy in the training group.
Taken together, the present findings indicated the neuroplasticity of WM training in both individuals with social anhedonia and patients with schizophrenia. The results demonstrated positive transfers to hedonic processing abilities with WM training, manifesting as enhanced motivation towards affective rewards and increased brain activations in the anticipating of rewards. These findings highlight the importance of a non-pharmacological intervention on alleviating anhedonia in both clinical and subclincal samples. It is hoped that these findings can shed light onto the early intervention and rehabilitation of patients with schizophrenia and individuals at-risk for psychosis in mainland China.