bioRxiv
July 22, 2025
Alice Caulfield, Lorraine Li, Famia Askari et al.
preprint
Psychedelics show potential for enhancing associative learning in animals, according to a systematic review of 31 studies (29 animal, 2 human). The review found that psychedelic administration improved learning across classical and operant conditioning paradigms, including fear extinction and reversal learning, though effects varied by dose, timing, training intensity, and sex. Possible mechanisms include increased prediction error sensitivity, serotonin receptor agonism, and structural plasticity. Learning enhancements may persist into the post-acute phase and depend on active environmental engagement. These findings have not yet been confirmed in humans, but suggest a window of enhanced learning that could inform psychedelic-assisted psychotherapy.
bioRxiv
April 21, 2025
Joanna Kuć, Rosalind G. McAlpine, Amelia Sellers et al.
preprint
After a single 12 mg dose of the psychedelic 5-MeO-DMT, people's speech patterns shifted toward more cognitive language and fewer social words, and their voices showed increased jitter and shimmer, indicating altered vocal quality. Baseline speech features predicted how psychologically prepared participants were, their anxiety about ego dissolution, emotional breakthrough, and later well-being. The findings suggest that analyzing voice journals can track a shift from external focus to introspection and serve as a tool for monitoring psychological transformation after psychedelic use.
bioRxiv
December 17, 2024
Dian Lyu, Ram Adapa, Robin L. Carhart-Harris et al.
preprint
The default mode network (DMN) and frontoparietal control network (FPCN), typically anticorrelated at rest in healthy brains, show continuous rather than absolute anatomical boundaries in the posterior precuneus. Connectivity differences along the dorsal-ventral axis follow linear slopes, forming functional gradients that exist only within each network's territory. These gradients flatten in altered states of consciousness (ASC), with the gradient magnitude similarly impaired across different ASC types, while spatial entropy differs between psychedelic and sedative states. The findings suggest the DMN and FPCN, though appearing distinct, may originate from a single integrated mechanism, and the loss of functional differentiation between them characterizes altered conscious states.
bioRxiv
October 26, 2024
Hang Yang, Bruno Herbelin, Chuong Ngo et al.
preprint
Practicing meditation from a third-person perspective in virtual reality, compared with a first-person perspective, produces stronger feelings of detachment and disconnection, a less distinct sense of one's body boundary, and reduced identification with the body. These subjective changes are accompanied by a more negative heartbeat evoked potential amplitude and activation of the posterior cingulate cortex and medial prefrontal cortex. The findings link altered states of self during meditation to neural and behavioral measures of the bodily self, demonstrating that manipulating perspective in VR can modulate the sense of self during meditation.
bioRxiv
January 29, 2024
Juan Ignacio Piccinini, Yonatan Sanz Perl, Carla Pallavicini et al.
preprint
The brain state induced by psychedelic drugs is often studied as a static snapshot, but this work focuses on the transition itself. Using a time-dependent whole-brain model and fMRI data from 15 volunteers given intravenous DMT, the authors show that the drug briefly pushes the brain near a critical point where it becomes maximally responsive to perturbations. This heightened reactivity is concentrated in fronto-parietal regions and visual cortices and correlates with serotonin 5HT2a receptor density. The findings suggest that even a short psychedelic episode can have a lasting influence because minimal perturbations during this transient achieve maximal effect, with the temporal evolution aligning with the drug's pharmacokinetics.
bioRxiv
December 7, 2023
Joseph Starkey, Robin L. Carhart-Harris, Andrea Pigorini et al.
preprint
A measure of 'statistical complexity' that calculates the diversity of statistical interactions in neural activity, removing randomness, was tested on human brain data from different sleep stages and from people under ketamine, LSD, and psilocybin. Statistical complexity differentiated sleep stages similarly to the standard Lempel-Ziv complexity measure and increased relative to placebo for all three psychedelic substances. The measure is a useful alternative for investigating neural activity complexity across states of consciousness.
bioRxiv
June 29, 2026
This theoretical paper proposes a new analytical framework for studying consciousness by examining how brain networks balance integration and segregation. The authors introduce an equal and cross-frequency connectivity analysis that partitions neural activity into distinct frequency bands to assess how information is integrated across brain regions versus segregated within specialized networks. The approach aims to provide a more nuanced understanding of consciousness by moving beyond simple measures of overall connectivity to capture the dynamic interplay between global integration and local specialization. The paper argues that consciousness depends on a critical balance between these two processes, and that disruptions in this balance may underlie altered states of consciousness.
bioRxiv
April 22, 2026
Conscious experience requires both adequate cerebral metabolism and preserved large-scale neural integration, not just one alone. A cross-study synthesis of published FDG-PET and TMS-EEG data mapped metabolic values and perturbational complexity index (PCI) across states including disorders of consciousness, sleep, and anesthesia. A metabolic lower bound of about 42-46% of normal cortical metabolism and a PCI threshold of approximately 0.31 consistently separated unconscious from conscious conditions. All nine conscious states met both thresholds, while all six unconscious states fell below at least one. NREM sleep, with preserved metabolism but low complexity, was unconscious, indicating metabolic support is necessary but not sufficient without integrative dynamics.
bioRxiv
November 11, 2025
preprint
A reanalysis of sleep-stage transitions challenges claims that consciousness changes follow two distinct patterns—gradual versus discrete. Using electroencephalography data from 12 healthy adults, the authors tested 622 transitions between sleep stages. When they selected analysis metrics tailored to each transition (e.g., alpha power for wake-to-sleep), the transition from light sleep (N1) to deeper sleep (N2) appeared uniquely steep, consistent with a discrete threshold. However, when they applied the same metrics uniformly across all transitions, no transition showed meaningful changes; the apparent steepness vanished. The authors conclude that the earlier finding resulted from circular reasoning—choosing metrics that already matched known neurophysiological changes—and serves as a methodological caution for consciousness research.
bioRxiv
April 7, 2025
preprint
A family of functional connectivity measures based on the barycenter—tracking the 'center of mass' between two neural signals—best decoded conscious vision from MEG data, outperforming other measures across brain regions central to both Integrated Information Theory and Global Neuronal Workspace Theory. Neural mass models showed that GNWT-based dynamics, featuring delayed ignition, better matched observed connectivity patterns than IIT-based highly synchronized sensory dynamics. The work introduces a generalizable approach for identifying and testing neural correlates of consciousness in an unbiased, data-driven way.
bioRxiv
January 18, 2023
Cristina Delgado-Sallent, Thomas Gener, Pau Nebot et al.
preprint
In mice with NMDAR hypofunction induced by subchronic phencyclidine (PCP), memory and auditory perception impairments are linked to disrupted connectivity between the medial prefrontal cortex and the dorsal hippocampus. Short-term memory relies on high gamma connectivity from the prefrontal cortex to the hippocampus, while long-term memory retrieval depends on theta connectivity from the hippocampus to the prefrontal cortex. Subchronic PCP impairs both short-term and long-term memory, increases prefrontal activity, decreases hippocampal activity, and disrupts this connectivity. Two weeks of daily risperidone rescues memory deficits, attenuates hippocampal desynchronization, and also ameliorates auditory perception impairments and their neural correlates.
bioRxiv
October 12, 2021
Sophie Waldron, Rachel Pass, Simonas Griesius et al.
Genetic evidence links disruptions in the DLG2 gene to increased risk for schizophrenia, autism, and intellectual disability. To explore psychiatric traits associated with reduced DLG2 function, a rat model with one functional copy of the Dlg2 gene (Dlg2+/-) was tested for anxiety, sensorimotor gating, reward responses, social behavior, and movement after phencyclidine administration. The rats showed reduced PSD-93 mRNA and protein without compensation from related genes. They exhibited an exaggerated locomotor response to phencyclidine, a hallmark of psychosis models, but no deficits in other behaviors. These findings suggest that DLG2 haploinsufficiency produces subtle, psychosis-related effects, unlike the more severe impairments seen in models with complete gene loss.
bioRxiv
August 31, 2021
O. Alitalo, S. Kohtala, Marko Rosenholm et al.
preprint
Both pharmacological and non-pharmacological treatments for depression activate TrkB receptors—a known antidepressant target—by inducing a physiological response linked to sedation. Rapid-acting antidepressants trigger TrkB signaling by evoking a state characterized by electroencephalographic slow-wave activity, behavioral immobility, reduced cerebral glucose utilization, and lowered body temperature. This signaling was not impaired in animals with reduced activity-dependent BDNF release but was diminished by maintaining animals in a warm ambient temperature. Preventing the hypothermic response attenuated the behavioral effects of the rapid-acting antidepressant nitrous oxide. The findings suggest that changes in energy expenditure and thermoregulation are essential, but not sufficient, for antidepressant responses, challenging pharmacology-centric hypotheses and highlighting the role of bioenergetics and thermoregulation.