Studying the neural mechanisms of encoding new memories is crucial for understanding and restoring cognitive function. In this study, we utilized functional connectivity metrics to identify hubs responsible for successful or unsuccessful encoding in a free recall task among eleven epileptic patients undergoing stereo EEG presurgical evaluation. Firstly, several types of functional connectivity metrics were calculated across standard EEG bands, ripples, and fast ripples for each patient, feature, encoded word and time window across all possible pairs of contacts. Whole-brain connectivity maps were calculated for recalled and forgotten words using the median feature values, identifying connections exhibiting differential connectivity during encoding of recalled and forgotten words. Using significance threshold, we calculated the number of significantly different connections for each contact. Global hubs were identified using a three-sigma limit, and their corresponding brain structures were determined with the MNI-AAL3 atlas. Finally, for each feature, frequency band, and time window the percentage of patients for whom a given structure served as a global connectivity hub was calculated. Middle temporal gyrus and hippocampus emerged as prominent global connectivity hubs for distinguishing between recalled and forgotten words. Local connectivity hubs were also quantified. The percentage of significantly different connections for each pair of structures in whole-brain connectivity maps was calculated for each patient, feature, frequency band and time window. Finally, the interpatient mean was enumerated to determine local hubs for distinguishing between recalled and forgotten words. These findings enhance understanding of memory encoding and provide potential targets for neuromodulation therapies in cognitive deficit disorders.