Experiences of Awe Mediate Ketamine's Antidepressant Effects: Findings From a Randomized Controlled Trial in Treatment-Resistant Depression

Depression is a major global health burden and many patients do not respond to first-line treatments. Ketamine, an NMDA receptor antagonist, produces rapid antidepressant effects but these are typically transient and questions remain about how to prolong or amplify clinical benefit. Prior research has emphasised ketamine's molecular and circuit-level actions, whereas acute psychological effects—often labelled dissociative—have been treated as side effects and measured with instruments such as the Clinician-Administered Dissociative States Scale (CADSS). The authors note that CADSS may not sensitively capture the full range of ketamine's subjective effects and that more specific psychological constructs might help explain clinical outcomes. Aepfelbacher and colleagues therefore tested whether the acute subjective experience of awe, measured with the Awe Experience Scale (AWE-S), occurs during a ketamine infusion and whether it relates to antidepressant response. Using data from a randomised controlled trial of a single 0.5 mg/kg ketamine infusion versus saline, the study asked three primary questions: does ketamine induce awe relative to saline, are AWE-S scores associated with clinician-rated depression improvement up to 30 days postinfusion, and do AWE-S scores statistically mediate ketamine's antidepressant effects? For comparison, the investigators ran parallel analyses using CADSS scores to index general dissociative symptoms.

The study used data from a parent trial (ClinicalTrials.gov: NCT03237286) that randomised adults with moderate-to-severe depression (baseline MADRS ≥25) to receive a single intravenous infusion of ketamine (0.5 mg/kg over 40 minutes) or saline, with a 2:1 allocation ratio stratified by sex and level of prior treatment resistance. Participants had to have failed at least one prior antidepressant (45% had failed three or more) and to have stable background treatments for at least 4 weeks before and during the 30-day follow-up. Of 154 participants randomised in the parent trial, the AWE-S was added to the assessment battery midway through recruitment; 116 participants (77 ketamine, 39 saline) completed the AWE-S at approximately 40 minutes after infusion start and constitute the analytic sample reported here. Immediately prior to the AWE-S at that same timepoint, a trained rater administered the CADSS. The primary clinical outcome was the Montgomery-Åsberg Depression Rating Scale (MADRS), clinician-rated at baseline and at five postinfusion time points (24 hours and days 5, 12, 21, and 30). For interpretability the investigators expressed outcomes as percentage improvement from baseline MADRS at each time point. The AWE-S is a validated 30-item measure capturing six subdomains of awe: connectedness, time, vastness, accommodation, self-diminishment, and physiological changes. Statistical analyses used JMP and MATLAB. Because some scale distributions were non-normal, nonparametric tests were employed: Wilcoxon rank-sum tests compared AWE-S (total and subscales) between treatment groups and Spearman correlations tested associations between AWE-S and MADRS % improvement. Mediation testing used a standard three-variable path model implemented in a Multilevel Mediation and Moderation Toolbox with bootstrap testing of the indirect (a*b) pathway; false discovery rate correction adjusted for multiple comparisons across the five MADRS time points. Parallel analyses replaced AWE-S with CADSS and Fisher r-to-z tests compared correlation strengths within the ketamine group. Sensitivity analyses and further details, including the adjunct digital therapy (automated self-association training, ASAT) and CONSORT information, are reported in the Supplement.

The AWE-S was completed by 116 participants (77 ketamine, 39 saline). Baseline characteristics were reported as comparable to the parent trial; the extracted text does not present the full table. Distribution checks showed AWE-S scores were approximately normal in the ketamine group but skewed in the saline and total samples, prompting the use of nonparametric tests. In a Wilcoxon rank-sum comparison, total AWE-S scores and all subdomain scores were significantly higher in the ketamine group than in the saline group (p < .0001). The accommodation subscale yielded the highest subscale scores in both groups. Spearman correlations within the ketamine arm revealed significant associations between total AWE-S scores and percentage MADRS improvement at all five postinfusion time points after false discovery rate correction (i.e. 24 hours and days 5, 12, 21, and 30). Exploratory analyses found that each of the six AWE-S subscales correlated with MADRS improvement at one or more time points. Mediation analyses showed that total AWE-S scores statistically mediated the relationship between ketamine treatment and MADRS % improvement at every assessed time point (reported a*b p-values: 24 hours p = .0028; day 5 p = .0154; day 12 p = .0027; day 21 p = .0018; day 30 p = .0052; all FDR-corrected p < .05). A direct (main) effect of ketamine on MADRS remained at 24 hours and day 5 (24 hours p < .0001; day 5 p < .001) but was not significant from day 12 onward. The investigators interpret this pattern to indicate that a subset of ketamine-treated patients who reported particularly strong awe showed larger and more persistent antidepressant responses. In parallel, CADSS total scores at 40 minutes were also significantly higher in the ketamine group than saline (p < .0001). Unlike the AWE-S, CADSS scores did not correlate with MADRS improvement in the ketamine arm at any time point and did not mediate outcomes (a*b mediation tests non-significant, ps > .372). From day 12 onward, correlations between AWE-S and MADRS were significantly stronger than CADSS-MADRS correlations. The extracted text notes a few modest CADSS correlations in the saline group that did not survive multiple comparison correction.

Aepfelbacher and colleagues conclude that ketamine infusions elicited strong self-reported experiences of awe and that these acute awe experiences statistically mediated clinician-rated antidepressant improvements from 24 hours up to 30 days postinfusion. The mediation effects persisted beyond the time when a significant group-level effect of ketamine on depression was detectable, suggesting that individuals who experienced pronounced awe during infusion retained benefit for longer than those who did not. By contrast, CADSS scores, commonly used to index ketamine-associated dissociation, did not predict or mediate depression outcomes and were generally unrelated to clinical improvement in the ketamine-treated sample. The investigators situate their findings within broader literature indicating ketamine can produce psychedelic or mystical-type experiences not fully captured by dissociation scales. They note parallels with studies where measures of mystical or unitive experiences predicted benefit in substance use disorders and with research showing ketamine can elicit spiritual or unitive experiences comparable to classical psychedelics on some scales. The authors posit that acute awe may reflect a psychologically meaningful component of ketamine's action that operates alongside neurobiological mechanisms such as increased neuroplasticity; awe has been linked previously to enhanced cognitive flexibility and altered neural connectivity, which could plausibly interface with ketamine's known effects. Practical implications discussed include the possibility that treatment context, mindfulness, or adjunctive psychotherapeutic techniques could be used to foster awe and thereby augment ketamine's therapeutic impact, although the investigators emphasise that randomised trials are needed to validate such adjunctive strategies. The authors acknowledge multiple limitations. Effect sizes for AWE-S–depression associations were small to medium, indicating other sources of variance in outcomes. The AWE-S was added mid-study so only a subset of the parent trial completed it; CADSS was always administered before the AWE-S, introducing a possible order effect; and no baseline measure of trait propensity for awe or expectancy was collected. Use of saline as an inert placebo leaves open the possibility of functional unblinding and expectancy influences. The sample excluded bipolar disorder and some comorbidities, and lacked racial and ethnic diversity, limiting generalisability. Finally, because no active psychoactive comparator was included, the design cannot prove that the observed relationships are specific to ketamine rather than to acute psychoactive interventions more broadly. The authors recommend further research integrating psychological and neurobiological levels of analysis and testing whether deliberate fostering of awe in the treatment context enhances clinical outcomes.

The study suggests that experiences of awe during a single ketamine infusion mediated both rapid and sustained antidepressant effects over a 1-month period. Individuals reporting very strong awe-like experiences were comparatively protected against relapse for at least 30 days after a single infusion, whereas a commonly used dissociation measure (CADSS) did not mediate clinical outcomes. These randomised controlled trial data indicate that awe may be a clinically relevant component of the subjective experience unfolding during ketamine treatment for depression and warrant further investigation.