The induction of dissociative states: A meta-analysis

Abstract

Dissociative states, characterised by discontinuities in awareness and perception, occur in a diverse array of psychiatric disorders and contexts. Dissociative states have been modeled in the laboratory through various induction methods but relatively little is known about the efficacy and comparability of different experimental methods. This meta-analysis quantified dissociative states, as indexed by a standardised instrument (Clinician Administered Dissociative States Scale), at baseline in varied diagnostic groups and in response to different experimental induction methods (psychological techniques and pharmacological agents) in both clinical and non-clinical samples. Primary outcomes were state dissociation effect sizes (Hedges's g) (PROSPERO registration CRD42022384886). 2,214 papers were screened, yielding 150 eligible articles and 251 effect sizes comprising 7,190 individuals. High levels of baseline state dissociation were observed in multiple diagnostic groups relative to controls, with the largest effects found in post-traumatic stress disorder (PTSD). In controlled experiments, induced state dissociation was most pronounced in response to mirror-gazing, ayahuasca, ketamine, cannabis, MDMA, and nitrous oxide relative, with effects comparable to or exceeding baseline state dissociation in PTSD. The effect sizes were characterised by pronounced heterogeneity but were not reliably associated with methodological features of the original studies. Elevated state dissociation is present in multiple diagnostic groups and comparable or higher levels can be reliably induced in controlled experiments using psychological techniques and pharmacological agents. These results demonstrate the efficacy of several methods for experimentally modelling dissociation and have implications for measuring adverse events and predicting outcomes in clinical interventions involving pharmacological agents.