The meta-analyses shown on this website are based on the results in Hirschfeld et al. (in peer-review)
Data has been retrieved from the ASDB repository on Open Science Framework version ASDB_v.2022-12-31 which contains data from MEDLINE-listed studies on altered states of consciousness published from 1975-01-01 until 2022-12-31. The ASDB stems from a systematic literature review according to PRISMA standards as described in Prugger et al. (2022)
This meta-analysis included psychometric data from commonly applied questionnaires that assess the phenomenology of altered states of consciousness, namely from two versions of the Altered States of Consciousness Rating Scale (the 5D-ASC and the 11-ASC), and from the Mystical Experience Questionnaire (MEQ30).
The Altered States of Consciousness Rating Scale (Dittrich 1998) is a self report questionnaire with 94 items rated on a visual analog scale. Two different analysis schemata are in use:
In the 5D-ASC version (5 Dimensional Altered States of Consciousness Rating Scale) items are assigned to five core dimensions: (1) Auditory Alterations, (2) Dread of Ego Dissolution, (3) Oceanic Boundlessness, (4) Visionary Restructuralization, and (5) Vigilance Reduction.
In the more recent 11-ASC version (11 factor Altered States of Consciousness Rating Scale) only 42 of the 94 questionnaires items are used in the analysis, where item scores are summarized along 11 factors: (1) Experience of Unity, (2) Spiritual Experience, (3) Blissful State, (4) Insightfulness, (5) Disembodiment, (6) Impaired Control and Cognition, (7) Anxiety, (8) Complex Imagery, (9) Elementary Imagery, (10) Audio Visual Synesthesia, and (11) Changed Meaning of Percepts.
Both analysis schemes have been validated and demonstrate good reliability (5D-ASC: Hoyt 0.88–0.95 (Dittrich et al. 2006)); 11-ASC: mean Cronbach’s alpha of 0.83 (Studerus et al. 2010)
We performed meta-regression analyses using restricted cubic splines. A restricted cubic spline model is non-linear and consists of a series of piecewise cubic polynomials that are connected at the position of the knots, with a linear curve before the first and after the last knot (Stone & Koo 1985). It was suggested that four or five knots are usually adequate to fully capture the underlying shape, provided the amount of included studies is sufficiently large.(Harrell et al. 1988, Harrell 2006). Given the limited number of available studies for the 5D-ASC and MEQ30, we chose three knots for those questionnaire-data and 4 knots for the 11-ASC. To account for repeated measures and potential heteroscedasticity of error-terms resulting from repeated measurements on the same or similar set of participants, we clustered standard errors at the study level and estimated the cluster-robust variance with the small sample adjustment proposed by Pustejovsky & Tipton (2018). We assumed a within-study correlation p=0.8 for the variance-covariance matrix. The analyses were performed in R version 4.2.2 with the package metafor using the restricted cubic spline model from the package drc. Cluster-robust variance was estimated with the package clubSandwich in the R robust function. The R-syntax of this analysis is available on Github.
The authors received no financial support - neither for the published article nor for the development of the app. The authors declare no competing interests.
The App was developed by Tim Hirschfeld in R with Shiny and Echarts4r. For further information contact me: tim.hirschfeld@charite.de