“Paradoxical wakefulness” induced by psychedelic 5-methoxy-N,N-dimethyltryptamine in mice

Mammalian vigilance is conventionally divided into wakefulness, slow-wave sleep (SWS, or NREM) and paradoxical sleep (PS, or REM), each defined by characteristic behavioural and electrophysiological markers. Serotonin (5-HT) is a major neuromodulator implicated in sleep–wake regulation and in mood disorders; selective serotonin reuptake inhibitors are known to suppress PS. Psychedelic compounds acting on 5-HT receptors have renewed therapeutic interest, but their immediate and delayed effects on canonical sleep–wake signatures are not well characterised. Bréant and colleagues set out to examine how the fast-acting psychedelic 5-methoxy-N,N-dimethyltryptamine (5-MeO-DMT) affects vigilance states and cortical activity in freely moving mice. Using chronic electrophysiological recordings combined with behavioural assays, the study aimed to document acute changes in EEG, local field potentials (LFP) and multi-unit activity (MUA), to quantify effects on sleep architecture, and to test whether psychedelics can induce an altered, hybrid vigilance state during overt wakefulness.

Adult male C57BL/6J mice (n = 15, aged 9–15 weeks) were singly housed under a 12:12 light–dark cycle with ad libitum food and water. A subset of animals remained unimplanted for behavioural assays (running wheel, n = 4; feeding, n = 4). All procedures complied with UK Home Office regulations. Surgical implantation under isoflurane anaesthesia positioned EEG screws over right frontal and right occipital cortex with a left cerebellar reference; EMG wires were placed in nuchal muscles. In six animals laminar probes were additionally implanted in primary visual cortex (V1) to record LFP and MUA. Analgesia was provided peri‑operatively and animals were allowed to recover and habituate prior to recordings. The principal electrophysiology experiment used a within-subject crossover design (n = 10 implanted mice). After one baseline day, each mouse received intraperitoneal injections of either vehicle (saline) or 5 mg/kg 5-MeO-DMT (1 mg/ml) at light onset (ZT0) on day 1, two recovery days, then the alternate injection on day 4. Continuous EEG, LFP and MUA recording was performed, and animals were left undisturbed after injections. Two additional crossover experiments assessed behaviour at dark onset (ZT12): feeding with a bowl of sugar pellets (n = 4) and running-wheel activity (n = 4); in both, animals received saline or 5 mg/kg 5-MeO-DMT in the same schedule and were video recorded for one hour after injection. Vigilance scoring combined EEG spectral composition, EMG levels and behavioural observation. Consolidated SWS episodes were defined as >1 minute and consolidated PS episodes as >30 seconds. Slow-wave detection on LFP used a 0.5–4 Hz bandpass filter and retained waves with peak amplitude > mean + 1 SD; waves were aligned to positive peaks to compute average MUA profiles. Data acquisition used a multichannel Tucker‑Davis system and Synapse software. Blinding was not possible because 5-MeO-DMT produced conspicuous EEG changes. Statistical analysis was performed in Matlab. Normality was tested with Shapiro–Wilk. Parametric paired t-tests (with Cohen’s d reported) or non-parametric Wilcoxon tests were applied as appropriate, using p < 0.05 as the significance threshold. For spectral analyses the authors followed convention and did not apply corrections for multiple comparisons; detailed results were reported in supplementary tables.

Sleep architecture and timing: 5-MeO-DMT produced a significant delay in PS onset, with an average postponement of 44.1 ± 10.3 minutes (t(6) = 3.81, p = 0.008, d = -1.44). SWS onset was not significantly changed (t(6) = 2.07, p = 0.08, d = -0.78). Total time spent in each vigilance state over the 24 hours following injection showed no significant differences between drug and vehicle conditions, and measures of state incidence and interruptions were likewise unaltered. EEG, LFP and MUA during wakefulness: The dominant and most striking effect was the acute appearance of high-amplitude slow-wave activity (SWA, 0.5–4 Hz) in cortical EEG and LFP while animals remained behaviourally awake with elevated EMG. Visual observation and video confirmed normal waking behaviours (exploration, grooming, walking) during these episodes. Quantitatively, frontal EEG showed significant increases in SWA and in the 12–14 Hz band after 5-MeO-DMT; occipital derivations showed increased SWA and a significant decrease in theta power (6–10 Hz) during wake. Spectrograms indicated these cortical effects lasted approximately 45 minutes and generally dissipated within an hour, although sparse frequency-bin differences persisted in some animals and derivations up to a few hours. Local neuronal activity accompanying slow waves: In the subset with laminar probes, LFP slow waves observed during wake after 5-MeO-DMT were accompanied by neuronal OFF-periods—periods of suppressed MUA—resembling OFF-periods seen during physiological SWS. Detection and averaging of individual slow waves demonstrated that the average LFP slow wave during SWS and during wake after 5-MeO-DMT was associated with strong suppression of spiking activity. Behavioural assays and adverse signs: Classic acute psychedelic-associated behaviours were observed, notably head-twitch responses. The investigators did not observe overt features of serotonin syndrome such as hind-limb abduction or flattened posture. In the feeding assay, vehicle-injected mice retrieved the first sugar pellet rapidly (mean latency 0.20 ± 0.08 minutes), whereas 5-MeO-DMT-injected mice showed markedly longer latencies (mean 21.16 ± 7.2 minutes), with individual latencies ranging from ~4.8 to 37.4 minutes. Running-wheel activity over a 30-minute observation period was reduced after 5-MeO-DMT: vehicle mice completed 310.7 ± 179.8 wheel revolutions (mean speed 23.5 ± 8.3 rpm), while drug-treated mice completed 67.0 ± 37.6 revolutions (mean speed 11.03 ± 4.4 rpm). The authors note substantial inter-individual variability and small sample sizes for these behavioural assays. Temporal profile and spectral specifics: The drug-induced cortical slowing was transient; one hour post-injection most differences versus vehicle had resolved except for isolated frequency bins (for example 3.25 Hz). No notable changes were found above 30 Hz immediately or an hour after injection, and changes in SWS spectral content were confined to the first 2 hours post-injection.

Bréant and colleagues interpret their findings as evidence that acute systemic 5-MeO-DMT induces a hybrid vigilance state they term "paradoxical wakefulness" (PW): animals are behaviourally awake with high muscle tone yet exhibit cortical slow waves and neuronal OFF-periods that are typical markers of SWS. This state contrasts with paradoxical sleep (PS), which pairs wake-like cortical activation and muscle atonia; PW is therefore a dissociated state in which electrophysiological sleep signatures co-occur with overt wake behaviour. The authors suggest that PW may underlie, or at least contribute to, the delay in PS onset observed after psychedelics such as LSD, psilocybin or psilocin. The suppression of hippocampal/occipital theta during PW is highlighted as potentially important because theta is linked to spatial and temporal processing; the authors hypothesise that theta suppression might reflect disrupted processing of space and time during the psychedelic state. They also discuss that SWA during PW raises questions about the exclusivity of slow-wave activity as a sleep hallmark and whether SWA during PW shares the same cellular and network mechanisms as SWS. The paper notes that neuromodulatory changes beyond 5-HT (for example acetylcholine or noradrenaline) could contribute to the observed reorganisation of cortical dynamics, but these were not measured. Key limitations acknowledged by the investigators include the short-lasting pharmacokinetics of 5-MeO-DMT, which may explain differences from studies of longer-acting psychedelics, and the possibility that intraperitoneal injection itself is a stressor influencing immediate and delayed effects. They also note the pronounced inter-individual variability and the small sample sizes in behavioural assays. The authors refrain from asserting mechanistic conclusions about homeostatic regulation of SWA during PW, stating that it remains to be determined whether PW is homeostatically regulated or whether it contributes to dissipation of sleep pressure. Implications and future directions discussed by the authors include investigating whether PW occurs spontaneously, exploring the cellular mechanisms that generate slow oscillations during PW, assessing longer-term effects on sleep and plasticity following psychedelic exposure, and examining the role of other neuromodulators and routes of administration.

The authors conclude that acute intraperitoneal injection of 5-MeO-DMT in mice produces a short-lasting hybrid vigilance state characterised by active behaviour and high muscle tone accompanied by cortical slow-wave activity and neuronal OFF-periods. They term this state "paradoxical wakefulness" and propose it as a potential rodent analogue of the altered conscious state induced by psychedelics in humans, emphasising that wake and sleep phenotypes can be dissociated rather than strictly mutually exclusive.