Previous studies investigating the differences in olfactory processing and judgments between trained sommeliers and control participants have shown increased activations in brain regions involving higher-level cognitive processes in sommeliers [1, 2]. Despite these findings, there is little or no information about the influence of expertise on causal connectivity and topological properties of the connectivity networks between these regions. Therefore, the current study focuses on addressing these questions in an fMRI study of olfactory perception in trained master sommeliers.FMRI data were acquired from thirteen master sommeliers and thirteen control participants during different olfactory and non-olfactory tasks. A single odorant was presented during the olfactory tasks and the participants were informed as to which olfactory task to perform. For one of the olfactory tasks, participants reported whether odorant was wine or non-wine, and for the other they reported if the odorant presented was red or white wine. One of the non-olfactory task was a visual discrimination task during which participants were randomly presented a pixelated image of a fingerprint or zebra skin and asked to identify the image correctly. The other non-olfactory task was a motor task, where the participants were delivered a stream of air without any odorant and asked to respond with a button press. After standard pre-processing using SPM12, mean time series were extracted from 76 different ROIs (based on AAL [3] and Jülich atlas [4]) and underlying neuronal variables were extracted using Cubature Kalman filter based blind hemodynamic deconvolution [5]. The resultant neuronal variables were then input into a dynamic multivariate autoregressive [6, 7] model to obtain connectivity between every pair of ROIs as a function of time. The absolute causal connectivity values during task of interest (for wine/non wine task and fingerprint or zebra task) were then populated into different samples separately for the sommelier and control groups. One-sided two-sample t-tests were performed between these samples and paths that were significantly greater during the wine/non-wine task were identified. Among these, paths that were significantly different between the sommeliers and controls were obtained. The obtained connectivity matrices were further studied using graph theory by labeling effective connectivity (EC) as an edge and a brain region as a node. Using Brain Connectivity Toolbox [8] the small-world parameters were obtained at different sparsity thresholds (0.1 < S < 0.98, ∆S = 0.01) for each participant and compared between the two groups.There were a total of 80 EC paths that were significantly different (p<0.01 nonparametric testing [9]) between the two groups (54 paths were significantly greater in sommeliers compared to controls and 26 paths were significantly greater in controls compared to sommeliers). For simplicity, fig. 1 shows only those EC paths that were extremely significant (p<<0.01, t-value > 9) in the between-group comparison. It can be seen that there are two different connectivity networks that exist in the two groups. The right paracentral lobule (PCL) is a major hub in the controls, while the left thalamus (THA) and the left inferior occipital gyrus (IOG) are hubs in the sommeliers. Fig. 2 shows the differences in overall network characteristics between the two groups. The connectivity networks of both the groups represented a small-world organization for a defined range of sparsity. Subsequent evaluation of the integrated area under the curve (AUC) values over the range of sparsity showed significantly higher small-world index, normalized clustering coefficient and normalized path length in master sommeliers when compared to untrained participants (p<0.05 corrected, effect size > 1).The current study revealed differences in causal connectivity and topological properties of networks involved in olfactory perception among master sommeliers and untrained control participants. Specifically connections from the left THA and the left IOG were significantly greater in sommeliers and connections from right PCL were greater in controls. Furthermore, both groups showed a small-world organization, but the master sommeliers exhibited significantly increased small-world topology as compared to the controls. These observations support the view that specialized expertise and training might result in enhancements in the brain well into adulthood. These findings are mainly important, given the fact that, learning about neuroplasticity in adulthood in these regions may, then, have added clinical importance in diseases like Alzheimer’s and Parkinson’s where early neurodegeneration is isolated to regions important in smell.