Abstract
A formal correction was issued for a major meta-analysis examining the pharmacology, neuroimaging, and phenomenology of psychedelics. The correction addresses the accidental inclusion of duplicated data from three studies in the 11-dimensional Altered States of Consciousness (11D-ASC) meta-analysis. Re-analysis confirms that the core findings—demonstrating broadly similar subjective profiles for LSD and psilocybin—remain robust, with only minor statistical shifts in two specific dimensions at medium doses.
The Challenge of Data Redundancy
In large-scale meta-analyses, "double-counting" participants is a significant risk that can artificially inflate sample sizes and bias results. The original study aimed to provide a "synergistic, multi-level understanding" of substances like LSD and psilocybin. However, it was discovered that three included papers—Duerler et al. (2021), Lewis et al. (2017), and Preller et al. (2017)—utilized overlapping datasets. This correction ensures the statistical integrity of the field's understanding of psychedelic phenomenology.
Methodological Re-evaluation
The researchers focused on the 11-dimensional Altered States of Consciousness (11D-ASC) scale, a standardized tool for quantifying subjective experiences. The correction involved:
- Data Filtering: Identifying and removing redundant data points from the 11D-ASC dataset to prevent participant double-counting.
- Statistical Recalculation: Re-calculating 33 specific comparisons (11 dimensions across 3 dose levels: low, medium, and high) between LSD and psilocybin.
- Verification: Confirming that the 5D-ASC meta-analysis results remained unaffected, as no duplicates were present in that specific subset.
Stability of Phenomenological Profiles
The most significant takeaway from the re-analysis is the resilience of the original findings. Despite the removal of three studies, 31 out of 33 comparisons remained consistent. This suggests that the reported similarities between LSD and psilocybin are representative of broader trends rather than artifacts of specific datasets.
The correction did reveal subtle nuances at medium doses. Specifically, the "insightfulness" dimension became less distinct between the two drugs (shifting from significant to insignificant), while "disembodiment" became a more statistically significant differentiator in the corrected model.
Conclusion and Future Directions
This correction highlights the self-correcting nature of psychedelic science. By acknowledging previously overlooked meta-analyses (such as those by Hirschfeld and colleagues) and fixing data redundancies, the authors have strengthened the credibility of the "multi-level understanding" of these substances. Future research will likely move toward even more granular comparisons, utilizing natural language processing (NLP) of trip reports and real-time neuroimaging to map the subtle neural signatures of these phenomenological states.
Summary updated