LSD, afterglow and hangover: Increased episodic memory and verbal fluency, decreased cognitive flexibility

Psychedelics such as LSD, psilocybin and ayahuasca have been investigated increasingly for psychiatric indications including mood, anxiety and substance use disorders, and there is growing interest in other applications. Previous research reports acute impairments in a range of cognitive functions during intoxication, yet findings are mixed: experienced users or low (micro) doses sometimes show preserved or improved performance, and animal and human studies have suggested psychedelic-induced neuroplasticity, neurogenesis and anti-inflammatory effects that could support memory and language functions. The main pharmacological target, the serotonin 2A (5-HT2A) receptor, is implicated in executive functions, learning and memory, but the literature leaves open questions about whether observed episodic memory changes reflect encoding, consolidation or retrieval processes and whether sub-acute (post-acute) effects might differ from acute impairments. Wießner and colleagues designed an exploratory human study to examine whether a low sub-psychedelic dose of LSD (50 μg) produces measurable sub-acute changes in diverse cognitive domains 24 hours after administration. The study aimed to test multiple aspects of episodic memory, verbal and design fluency, cognitive flexibility, attention, inhibitory control and perceptual reasoning using a randomised, double-blind, placebo-controlled, crossover design in healthy volunteers, with particular interest in whether LSD might improve specific memory or language-related processes despite acute cognitive disruptions reported elsewhere.

The investigators used a randomized, double-blind, placebo-controlled, crossover design with two treatments (50 μg oral LSD vs inactive placebo) and a 14-day washout between sessions. Treatment order was randomised and balanced across participants, and where tasks had alternate versions these were counterbalanced across treatment conditions. The study received university ethics approval and followed safety guidelines for human psychedelic research. A convenience sample of 25 healthy adults was recruited; one participant withdrew after the first session leaving a final sample of 24 (8 women; mean age 35 ± 11 years, range 25–61). Inclusion criteria included age ≥22 years, at least one prior experience with LSD, and specified abstinence periods from psychedelics (≥2 weeks until study end) and from alcohol/other drugs (≥3 days before each session); tobacco and caffeine use were restricted on study days. Exclusion criteria included psychiatric symptoms, personal or first-degree family history of psychosis, psychiatric medication, severe adverse history with psychedelics, substance use disorder, significant medical conditions and non-native Portuguese speakers. The text notes that participants consumed on average three cups of coffee daily and that about one third used tobacco regularly, which the authors state may confound stimulant-like effects. Cognitive testing focused on multiple domains assessed at baseline, during the dosing day and in a session the next morning (~24 h after administration). Key instruments were: Rey–Osterrieth Complex Figure (ROCF) for spontaneous visuospatial memory (copy, immediate and delayed recall); a 2D Object-Location Memory Task (OLMT) and Rey Auditory-Verbal Learning Test (RAVLT) for overnight consolidation of visuospatial and auditory-verbal lists respectively; Verbal Fluency Task (phonological, semantic, switching) and Design Fluency Task as nonverbal fluency analogues; Wisconsin Card Sorting Test (WCST) for cognitive flexibility and perseveration; Trail Making Test (TMT) for sustained and switching attention; Stroop Task for inhibitory control; and Block Design Test (BDT) for perceptual organisation. Tasks were administered in standard or adapted paper versions as described, with learning runs for OLMT and RAVLT performed before dosing and final recall the following morning. Data were analysed using repeated-measures General Linear Models (GLMs) in IBM SPSS. Each outcome was modelled with treatment (within-subjects) and treatment order (between-subjects) factors; subsequent models included BMI and abstinence from psychedelics as covariates to account for dose relative to body mass and washout effects. Partial eta squared (ηp2) was used to estimate effect sizes, significance was set at α = 0.05 two-tailed, and p-values were not corrected for multiple comparisons given the exploratory nature. The authors note that one to two subjects' data were lost for some tasks (OLMT, RAVLT, VFT, Stroop) due to application errors.

Twenty-four participants completed both sessions and were analysed. A small amount of task data was lost for one to two participants in OLMT, RAVLT, VFT and Stroop because of application errors. Analyses reported are from GLMs with BMI and abstinence included as covariates; the authors additionally comment that models without these covariates yielded fewer treatment effects and more period effects. Episodic memory: Compared with placebo, LSD increased ROCF immediate recall points (p = 0.044) and ROCF immediate recall percentage (p = 0.018), interpreted as enhanced spontaneous visuospatial encoding and immediate recall. LSD also increased OLMT consolidation percentage (p = 0.022), indicating better overnight consolidation of visuospatial locations relative to the immediate pre-drug recall. No treatment effects were found for ROCF copy, ROCF delayed recall, OLMT consolidation points or any RAVLT variables measuring auditory-verbal consolidation and recognition. Fluency: LSD increased phonological verbal fluency total responses (VFT total; p = 0.036) and correct responses (p = 0.027), and produced a marginal reduction in switch errors on the VFT (p = 0.058). There were no significant LSD effects on semantic fluency or on any Design Fluency Task (basic, filter, switch) measures. Cognitive flexibility and other executive tests: On the WCST, LSD was associated with impairments in cognitive flexibility as reported by the authors — fewer categories achieved, increased total errors and increased perseverative responses and errors — indicating reduced ability to shift conceptual rules and heightened perseveration. The authors detail that perseverative error type 3 (three consecutive errors within any prior or current perseveration principle) was most pronounced, followed by type 1 and marginally type 2, and they interpret these patterns as reflecting a mixture of perseveration and other processes such as short-term memory or attention rather than a pure perseverative deficit. No treatment effects were observed for sustained and shifting attention (TMT), inhibitory control (Stroop) or perceptual organisation (BDT). Order and period effects: Several tasks showed practice (period) effects with improved performance in session 2. Significant period effects included ROCF immediate recall points (p = 0.006), ROCF immediate percentage (p = 0.002) and ROCF delayed recall points (p = 0.008). Fluency and design fluency measures also showed session effects (e.g. VFT phonological total p = 0.010; DFT basic total p = 0.001). An order effect appeared for VFT switch total (p = 0.047) and correct responses (p = 0.016), with higher switching performance when placebo preceded LSD. For TMT and Stroop, improvements across sessions suggested practice rather than treatment effects. The authors reiterate that p-values were not adjusted for multiple comparisons and that effect sizes (ηp2) were estimated though specific values are not reported in the extracted text.

Wießner and colleagues interpret the findings as evidence that a single low dose of LSD (50 μg) produces selective sub-acute effects on cognition up to 24 hours post-administration, comprising both improvements and impairments. The main positive findings were enhanced visuospatial memory — immediate ROCF recall (post-dose learning) and overnight consolidation of pre-dose visuospatial material (OLMT) — and improved phonological verbal fluency. The authors link visuospatial improvements to hippocampal 5-HT2A-rich circuitry and suggest that psychedelic-induced serotonergic action could support consolidation and short-term maintenance; they also note concordance with animal studies showing improved consolidation when psychedelics are administered between encoding and retrieval. For verbal fluency, the selective benefit for phonological (frontal-dependent, strategy-based) rather than semantic (temporal-dependent, conceptual) fluency led the authors to propose that LSD may facilitate frontal-based strategic retrieval processes. Conversely, LSD impaired aspects of cognitive flexibility on the WCST, including fewer categories achieved and increased errors and perseverative responses. The investigators consider these deficits to be mixed with other cognitive processes (attention, working memory or short-term memory) rather than reflecting a pure perseverative syndrome. They acknowledge that such sub-acute impairments are unlikely to be permanent given reports of unchanged or improved flexibility in regular ritual users and following therapeutic courses, but they emphasise that dose, substance, sample characteristics and setting likely modulate outcomes and must be disentangled in future work. The authors discuss practical and clinical implications: improved memory consolidation and phonological retrieval suggest potential utility for conditions involving memory and language impairments (for example, mild cognitive impairment, dementia, stroke or brain injury), while the observed sub-acute reduction in cognitive flexibility advises caution because it could impair performance monitoring or the ability to carry out demanding tasks the day after consumption. They therefore recommend considering a resting day after psychedelic administration in therapeutic, research or recreational contexts and highlight possible risks for psychotic populations. The investigators state several limitations: the exploratory design with a modest sample and no correction for multiple comparisons; practice (period) effects that complicate interpretation and underscore the need for well-matched parallel task versions; the relatively low LSD dose which may have limited effect sizes; and the healthy volunteer sample, which restricts generalisability to clinical populations. They further note that analyses were sensitive to covariates (BMI and abstinence), implying that dose and washout intervals materially influence sub-acute cognitive effects and should be addressed in future studies. Overall, the authors call for replication in larger samples, more careful control of practice effects, dose-ranging work and investigation in clinical populations before therapeutic recommendations can be made.

The authors conclude that cognitive effects of a single 50 μg dose of LSD measured 24 hours later are mixed: there is evidence of an 'afterglow' with improved visuospatial memory consolidation and phonological fluency, alongside a 'hangover' in the form of reduced cognitive flexibility. Practically, sub-acute impairments in flexibility should be considered when planning therapy, research or daily activities after LSD use, yet the observed improvements point to the potential for psychedelics to enhance memory- and language-related functions and motivate further research in conditions such as brain injury, stroke and dementia.