Virtual reality-based retrieval–extinction training reduces cue reactivity in methamphetamine use disorder: a randomized controlled trial with 6-month follow-up

Methamphetamine (MA) use disorder (MUD) is characterized by a high relapse risk driven by persistent drug-associated memories. Retrieval–extinction (R–E) training, grounded in memory reconsolidation theory, may attenuate cue reactivity more effectively than standard extinction training, and virtual reality (VR) may enhance memory reactivation during retrieval. The efficacy, durability, and generalizability of VR-based R–E training in MUD remain unclear. Fifty-three male individuals with MUD were randomly assigned to one of three groups: R–10 min–E (MA-related VR retrieval followed by extinction 10 min later, within the reconsolidation window), NR–10 min–E (neutral VR retrieval followed by extinction 10 min later), or R–6 h–E (MA-related VR retrieval followed by extinction 6 h later, outside the reconsolidation window), yielding a final analytic sample of 48 participants. The intervention was delivered over two consecutive days. Cue-elicited self-reported craving and physiological reactivity, including skin conductance responses (SCR) and heart rate (HR), were assessed at pre-test and at 1-day, 1-week, 1-month, and 6-month follow-ups in response to both original and novel MA-related VR scenes. Groups were compared with baseline-adjusted ANCOVAs and false discovery rate (FDR) correction. MA Cue reactivity decreased in all groups. Compared with NR–10 min–E, the R–10 min–E group showed lower craving at 1 day and lower SCR and HR at 1 week, and lower HR than R–6 h–E at 1 day ( q = 0.08). In the novel scene, SCR was lower in R–10 min–E at 1 day and in both retrieval groups at 1 week ( q < 0.05), and HR was lower in R–10 min–E at 1 week ( q = 0.08); craving did not differ. No group differences remained at 1 or 6 months. Exploratory correlations between retrieval-phase craving and 6-month change did not survive correction. Extinction after MA-related VR retrieval was followed by short-lived reductions in some indices of cue reactivity, with partial short-term generalization of physiological effects. As these effects did not depend on the timing of extinction and may partly reflect additional MA-cue exposure, reconsolidation-specific effects were not demonstrated; larger trials with equated cue exposure and no-treatment controls are needed. Chinese Clinical Trial Registry, ChiCTR1800018899, registered on October 16, 2018. http//www.chictr.org.cn/showproj.aspx? proj=30​854.