Graph analyses of the neural structures during multipl-CUE probability learning task

Learning is a dynamic process which is affected by attention, emotional state, environment and social factors. Reward and emotion adjust the value of the information to be learned, skill or behavior for the organism. Studies showed significant functional activation changes in striatum and nucleus accumbens activations during processing information with reward and amygdala, prefrontal and orbitofrontal cortex during emotional information processing. Findings of the combined effect of emotion and reward on information processing are limited. The aim of this study is to investigate the connections between different neural structures, are activated in response to the interaction between the reward system and different emotional facial expressions, during probability learning process. 45-healthy controls are included in this study.Following psychiatric assessment, Beck Depression Inventory, Physical Anhedonia Scale, Social Anhedonia Scale, Behavioural Inhibition/Activation Scale were performed and structural, functional MRI were obtained with 3T-MRI scanner. For functional images, participants have been asked to make selections by learning the characteristic of the different hints presented after certain facial expressions in each trial. The connectivity between the active brain regions has been examined by graph theory according to the effect of emotion in the probabilistic learning process of healthy volunteers with creating fMRI task. Correct answer count, wrong answer count, reaction time and reward amount were recorded as behavioral data during task. They were compared with ANOVA. Preprocessing for fMRIs were analyzed by using Matlab-SPM12. Network structures for each emotion were formed by time series. The formed graph was analyzed with BRAPH. There has not been found statistically significant difference between the emotions in participants' behavioral data. But, with graph analysis, participants use different networks according to different emotions during the probabilistic learning task. Considering the influence of emotions on cognitive functions, different emotions network with different brain regions during learning. The absence of behavioral difference overlaps with the hypothesis of this paper. Emotions-specific network patterns have not reflected on behaviors but resulting in different connections in brain networks. Keywords: functional magnetic resonance imaging, graph theory, probability learning