Memory encoding can be facilitated by different memorization techniques. Amongst them, the Mind Palace is a very efficient encoding strategy that implies spatial processing and the use of emotional valency. In this study, we tested this method's efficacy by comparing it to a control group using a visual imagery memorisation technique. Our goal was to investigate the EEG correlates that underlie the efficacy of the method during encoding and retrieval and to test the modulation of the memorization correlated to the valency of the words. To that end, we performed an EEG encoding, recognition and recall experiment with two groups of participants (one using the Mind Palace and the other as a control) and three conditions, with words encoded as either neutral, positive or negative in valency. We found that participants who used the Mind Palace had higher recall scores and that in recognition, the negative valency of words could affect memorization. ERP analysis in the encoding tasks revealed a late effect for the Mind Palace group with an increase in amplitude for the frontal and temporal regions compared to the control group. At the same time, time-frequency analysis showed that the Mind Palace group seemed to have stronger decrease in theta power in the right fronto-temporal and bilateral temporal regions. In the time-frequency analysis of recognition, we observed stronger theta power for the Mind Palace group in a network of regions including frontal, temporal and occipital regions and stronger decrease in alpha/beta power in the overall scalp for the control group.