Background and aims: Despite the inclusion of Compulsive Sexual Behavior Disorder (CSBD) in the ICD11, there are many open questions on its neuronal pathogenesis, especially regarding the role of the amygdala. In this study, we aimed to further unravel this issue via a parcellation method based on Recurrence Quantification Analysis (RQA). Methods: The RQA pipeline was applied to resting-state functional magnetic resonance imaging data from 45 heterosexual males with CSBD and 26 Healthy Controls. Each amygdala was divided into two subdivisions in each group. In the CSBD group, the scores of psychological questionnaires were used as covariates in a second-level seed-to-voxel connectivity analysis with the amygdala as a region of interest. Results: Obtained parcellations revealed bilateral differences in the sizes of dorsomedial (DM) and ventrolateral (VL) amygdala between groups. Mean values of Shannon?s Entropy in the left DM and right VL amygdala correlated negatively with depression level, anxiety, and impulsivity, which might represent a vulnerability to CSBD, but only the right VL was implicated in the severity of CSBD symptoms. Multiple correlations between resting-state functional connectivity of the amygdala subdivisions and CSBD severity were observed, especially between the left VL amygdala and several default mode network nodes. Discussion and Conclusions: This is the first attempt to explore the role of the amygdala in CSBD by a parcellation method. Our results suggest the importance of the right VL amygdala in understanding the pathogenesis of the severity of CSBD symptoms, which highlights the rising need to explore the amygdala as a complex structure with diverse functions.  Kulcsszavak: Compulsive Sexual Behavior Disorder, amygdala parcellation, Recurrence Quantification Analysis