Double-blind comparison of the two hallucinogens psilocybin and dextromethorphan: effects on cognition

Abstract

Objectives: Classic psychedelics (serotonin 2A receptor agonists) and dissociative hallucinogens (NMDA receptor antagonists), though differing in pharmacology, may share neuropsychological effects. These drugs, however, have undergone limited direct comparison. This report presents data from a double-blind, placebo-controlled within-subjects study comparing the neuropsychological effects of multiple doses of the classic psychedelic psilocybin with the effects of a single high dose of the dissociative hallucinogen dextromethorphan (DXM).

Methods: Twenty hallucinogen users (11 females) completed neurocognitive assessments during five blinded drug administration sessions (10, 20, and 30 mg/70 kg psilocybin; 400 mg/70 kg DXM; and placebo) in which participants and study staff were informed that a large range of possible drug conditions may have been administered.

Results: Global cognitive impairment, assessed using the Mini-Mental State Examination during peak drug effects, was not observed with psilocybin or DXM. Orderly and dose-dependent effects of psilocybin were observed on psychomotor performance, working memory, episodic memory, associative learning, and visual perception. Effects of DXM on psychomotor performance, visual perception, and associative learning were in the range of effects of a moderate to high dose (20 to 30 mg/70 kg) of psilocybin.

Conclusions: This was the first study of the dose effects of psilocybin on a large battery of neurocognitive assessments. Evidence of delirium or global cognitive impairment was not observed with either psilocybin or DXM. Psilocybin had greater effects than DXM on working memory. DXM had greater effects than all psilocybin doses on balance, episodic memory, response inhibition, and executive control.

Keywords: Cognition; Dextromethorphan; Hallucinogen; Psilocybin; Psychedelic drug.

Figure 1.

Timeline of drug effects and neurocognitive assessments. The abscissa indicates time relative to capsule administration. The top half of the figure displays the average time course of participant-rated strength of drug effects, with error bars denoting SEM, and filled symbols indicating a significant difference from placebo (p < 0.05) at the respective time point. Each “X” in the bottom half of the figure designates the time, relative to capsule administration, at which each neurocognitive assessment was administered in each drug administration session. BL: Baseline measurements, before capsule administration; DSST: Digit Symbol Substitution Task; MMSE: Mini-Mental Status Examination; PLOT: Penn Line Orientation Test.

Figure 2.

Time Course of Effects of Drug Condition on Measures of Gross Motor Function. The outcome measure (on the ordinate) for (A) the circular lights task, and (B) the balance task are plotted against the time point during the experimental drug sessions at which assessments of gross motor function were made (on the abscissa): 0 (before capsule administration) and 2, 4, and 6 hours after capsule administration. Brackets indicate ±1 standard error. Filled markers indicate significant difference from placebo at that time point (p < 0.05, using Tukey’s correction for multiple comparisons).

Figure 3.

Effects of Drug Condition on Rated Strength of Drug Effects, Psychomotor Performance, and Executive Function. Outcome measures in each panel (on the ordinate) are plotted against the drug conditions of this experiment (on the abscissa): 0 mg (placebo), 10 mg/70 kg psilocybin, 20 mg/70 kg psilocybin, and 30 mg/70 kg psilocybin, and 400 mg/70 kg dextromethorphan (DXM). The thick horizontal lines in each panel indicate the medians, the outer boxes indicate the 1^st and 3^rd quartiles (25% and 75%), and vertical lines (or “whiskers”) indicate 1.5 times the inter-quartile range. (A) Observed strength of drug effects was rated by study monitors at regular intervals during each drug administration session. The average of the peak of observed strength of drug effects for each session are plotted. (B) Effects of drug condition on psychomotor performance were assessed using average response time during the timed portion of the motor praxis task. (C) Effects of drug condition on cognitive impairment were assessed using the total score from the Mini-Mental Status Examination (MMSE). A dashed line at 24 on the ordinate indicates the cutoff for a “cognitively normal” score on the MMSE, where scores at or above this score are considered to not be impaired. (D-F) Effects of drug condition on executive function as assessed using the Digit Symbol Substitution Task (DSST) were analyzed using (D) number of trials attempted, (E) accuracy (percent correct) of responses in trials attempted, and (F) accuracy (percent correct) of substitution recall trials at the end of the DSST. DSST responses from baseline as well as 2, 4, and 6 hours post-drug-administration were included in the analysis. Because peak effects generally occurred at 2 hours, data at that time point are displayed graphically. A more detailed figure including data from 4 and 6 hours post-drug-administration is included in the online Supplemental Materials and Methods (Figure S1-S3). * = p < 0.05, ** = p < 0.01, *** = p < 0.001, **** = p < 0.0001, using Tukey’s correction for multiple comparisons.

Figure 4.

Effects of Drug Condition on Memory Outcome Measures. Outcome measures in each panel (on the ordinate) are plotted against the drug conditions of this experiment (on the abscissa): 0 mg (placebo), 10 mg/70 kg psilocybin, 20 mg/70 kg psilocybin, and 30 mg/70 kg psilocybin, and 400 mg/70 kg dextromethorphan (DXM). (A-C) Effects of drug condition on working memory performance in the letter n-back task were assessed using (A) discriminability, (B) response bias, and (C) median response time (in ms) to correct responses. 0-back, 1-back, and 2-back conditions were contained within the analysis, but no significant differences were observed between 0-back and 1-back conditions. Therefore, plotting is restricted to 0-back (red bars) and 2-back (green bars) conditions for brevity. Full plots including the 1-back condition are available in Figure S4-S6, in the Supplemental Materials and Methods. (D) Word recall performance is expressed as the total number of encoded words that were freely recalled. (E-H) Effects of drug condition on outcome measures for word recognition were assessed using (E) Receiver Operating Characteristic (ROC) curve analysis and ROC area under the curve, (F) the non-parametric sensitivity index A’, and the (G) familiarity index and (H) recollection index derived from the dual-process ROC curve analysis model. The horizontal lines of each panel indicate the median, the outer boxes indicate the 1^st and 3^rd quartiles (25% and 75%), and vertical lines (or “whiskers”) indicate 1.5 times the interquartile range. * = p < 0.05, ** = p < 0.01, *** = p < 0.001, **** = p < 0.0001, using Tukey’s correction for multiple comparisons.

Figure 5.

Effects of Drug Condition on Visual Perception. Outcome measures in each panel (on the ordinate) are plotted against the drug conditions of this experiment (on the abscissa): 0 mg (placebo), 10 mg/70 kg psilocybin, 20 mg/70 kg psilocybin, and 30 mg/70 kg psilocybin, and 400 mg/70 kg dextromethorphan (DXM). The horizontal lines of each panel indicate the median, the outer boxes indicate the 1^st and 3^rd quartiles (25% and 75%), and vertical lines (or “whiskers”) indicate 1.5 times the inter-quartile range. Effects of drug condition on visual perceptual abilities as assessed using the Penn Line Orientation Test (PLOT) were demonstrated on (A) mean excess clicks for correct trials, and (B) median response time (in ms) for errors. * = p < 0.05, ** = p < 0.01, *** = p < 0.001, **** = p < 0.0001, using Tukey’s correction for multiple comparisons.