Aroma touch therapy is widely used in clinical fields for alleviating pain-related symptoms; however, few studies have reported the pain-relief mechanisms. The present study aimed to elucidate the analgesic effects of aroma touch therapy with Citrus junos oil based on the quantitative evaluation of deep brain network (DBN) activity using electroencephalogram (EEG) occipital alpha-2 rhythm (10-13 Hz) powers. Experimental investigations were performed with 13 healthy volunteers using the cold pressor task for simulating chronic pain in three different sessions: a baseline session with no therapies, a control session with a touch therapy, and an aroma touch therapy. We have found for the first time that the interviewed pain ratings represented by Numeric Rating Scale (NRS) scores were strongly correlated with a DBN activity index, which was derived from the slow fluctuation components of occipital EEG alpha-2 rhythm powers. The correlation was characterized by a V-shaped curve in the DBN activity index versus the pain rating, i.e., the NRS score, which provided the complete analgesic states (NRS = 0) for some subjects under aroma touch therapy at an appropriate DBN activity index. Such analgesic states were not so strongly correlated with emotional valence. In conclusion, aroma touch therapy may directly modulate DBN activity so that pain-induced outcomes are minimized.