What it is
Social connection is the felt sense of closeness, belonging, and support in a person's relationships. Researchers split it into three overlapping layers. The structural layer counts how many people someone sees, how large the network is, and whether they live alone. The functional layer asks what relationships provide, such as practical help or someone to talk to. The subjective layer asks how the person feels about all of it, and this is where loneliness and belonging live. Someone can have a full calendar and feel alone.
The layers predict different things. A meta-analysis of 148 studies and 308,849 participants found that people with stronger social relationships had a 50 percent higher likelihood of survival over follow-up (odds ratio 1.50, 95% confidence interval, or CI, 1.42 to 1.59). The association was weakest for a bare yes or no on living alone (OR 1.19, 95% CI 0.99 to 1.44, an interval that includes no effect) [1]. These were observational studies, so the result shows that connection and survival travel together and does not show that raising someone's connection would extend their life. The review did not report a separate analysis for healthy adults.
This page keeps to a boundary. The separate Empathy page covers understanding and sharing another person's emotional state, and this page covers the relationship itself: closeness, belonging, and contact.
How it is measured
Most of the research below stands or falls on its instruments, so it is worth knowing what they can and cannot do.
The UCLA Loneliness Scale, version 3, is the standard tool for loneliness. Its internal consistency ran from .89 to .94 [2], a figure from the abstract only. It reads in the direction of loneliness, so a well-connected adult sits near the bottom of the scale with little room to move. We did not find a published test of the floor effect for this scale in healthy adults, though our search was partial because PsycINFO was not accessible to us. A community survey of 301 healthy adults aged 65 and over averaged 31.5 on a scale that runs from 20 to 80 [23], which is descriptive and does not settle it. A 2025 review adds that loneliness tools capture frequency and not severity [26].
The Watts Connectedness Scale was built in the psychedelic field, with 19 items and three subscales for connection to self, others, and the world, from interviews with 20 people in a depression trial [3]. Its model fit was borderline, its "others" subscale correlated .78 with the Social Connectedness Scale and so is not clearly distinct from an existing measure of closeness, the samples were self-selected and mostly white and well educated, and one of its two surveys is the retreat sample reported in [4], so those papers are not independent. We found no study that tests the scale after non-psychedelic experiences.
A related group-experience tool, the Communitas Scale, has 8 items on felt togetherness and shared humanity during a group session, and its authors doubt that it translates across cultures [4]. Laboratory tasks such as Cyberball, a virtual ball-tossing game in which one player is left out, capture a few minutes of social processing and not how relationships go a month later.
One caution applies to every measure. After an intense shared experience, a higher score could mean deeper relationships, a warm afterglow, a wish to report a good result, or some mix of these.
What is known
The evidence comes in three kinds that should not be blended: controlled laboratory studies of the acute state, uncontrolled follow-up of weeks to months, and animal work.
In the hours after ingestion: controlled laboratory studies
A double-blind, placebo-controlled crossover study in 21 healthy volunteers used Cyberball inside a brain scanner, after 8 of 29 recruits were excluded for technical, familiarity, or movement reasons. Under psilocybin, participants rated feeling less excluded when the game left them out, on a single rating item (3.14 against 5.12 under placebo, P = 0.01). Activity fell in the right anterior midcingulate cortex, tied to social pain, and in the left middle frontal gyrus, and the drop in the cingulate region correlated with a questionnaire score for felt unity (r = 0.53), an association and not proof of cause. Other items on the same questionnaire did not change, including belonging, self-esteem, control, and meaningful existence [6].
Two pooled LSD crossover studies with 40 healthy volunteers aged 25 to 60 found higher emotional empathy, but only at the higher of the two doses and mostly for positive emotions. Ratings of feeling close to others, open, trusting, and wanting to be with others also rose, on analogue scales that were not validated. At the same time, cognitive empathy fell and recognition of fearful and sad faces got worse, so people felt closer while reading others less accurately. A behavioral measure of prosociality, the social value orientation task, rose only in the pooled sample and not in either study alone [7], the weakest link in an otherwise consistent picture.
A 24-person LSD study with a placebo condition and a receptor-blocker condition used a picture-rating task in which people rated pictures, saw a computer-generated group rating, and rated them again. Under LSD they moved toward the group, but only when its rating was already close to their own. With the blocker ketanserin given first, the brain response in the medial prefrontal cortex while receiving the group rating looked like placebo, which points to the serotonin 2A receptor, but the paper reports no significant behavioral difference between LSD and blocker plus LSD [8]. The task measured agreement about pictures, which is closer to social influence than to warmth, and it had no non-social control task.
A study of the same participants tested joint attention in the scanner. LSD made joint attention with a virtual partner slower to establish by about 0.17 seconds, a difference the authors call subtle, and errors and eye contact did not differ [9]. One study measured reported closeness and another measured joint-attention efficiency, so nobody has tested whether the two come apart in the same people.
In the Ultimatum Game, a lab game in which one player proposes how to split money and the other accepts or rejects, psilocybin (an open-label infusion study of 19 men) and MDMA (a placebo-controlled study of 20 men) both reduced rejection of unfair offers, with odds ratios of 0.48 and 0.57. All participants were men with earlier experience of the drug, and the psilocybin study had no placebo and no counterbalancing, which its authors name as a major limitation. Psilocybin also reduced rejection of offers that a computer made at random (OR 0.30), so its effect was not specific to dealing with people, while the MDMA effect did not extend to computer offers and a social value orientation questionnaire did not change under MDMA [10]. In 16 healthy volunteers LSD raised plasma oxytocin only at the highest dose, and blocking the serotonin 2A receptor prevented that rise without significantly reducing the empathy increase [11].
For contrast, MDMA is not a classic psychedelic and is not a natural product. A narrative review found that it consistently raises feelings of closeness and sociability, with thinner evidence for actual social approach, which rests on a single human study at one of two doses, and mixed evidence for blunted responses to rejection, where the authors describe conflicting findings [12]. The classic psychedelics above shift social processing in a similar direction, in samples of 16 to 40 healthy volunteers with mean ages of about 25 to 29 where reported, over hours, and none of these studies asked whether anyone's relationships changed afterward.
Weeks to months: follow-up without a control group
A small blinded sample was followed for 14 months. Eighteen healthy volunteers received psilocybin across several sessions, and fourteen months later volunteers, community observers who knew them, and the study monitors all rated persisting positive change. Volunteers named better social relationships most often, and monitors rated the quality of social relationships as improved (mean 1.64 on a scale from -3 to +3) [13]. The limits are real. There were 18 people, the 14-month questionnaire told volunteers which sessions had been the strongest, the observers were chosen by the volunteers and knew about the psilocybin study, and the monitors rated change they believed came from the experience, which reflects attribution. Most volunteers also took part in weekly spiritual or religious practice.
A naturalistic online study followed people who planned to take a psychedelic on their own initiative. The first survey drew 654 respondents, and the final sample was 148 people who completed the main set of three surveys. Seventy-one percent were men, 32 percent reported a past psychiatric diagnosis, and 91 percent had used psychedelics before. Scores on the 6-point Social Connectedness Scale, whose original validation paper we did not read, rose by about 0.18 to 0.19 points at two and four weeks. The standardized effect was small, around 0.1, and the 99 percent confidence interval on the raw difference sat barely above zero (0.01 to 0.37), against the study's strict threshold of p < 0.005. A second closeness measure rose at two weeks and did not persist at four. Emotional empathy, compassion, and warmth did not change, which the authors put down to restriction of range, meaning that scores were already high at the start, an explanation and not a test. Only 22 percent of early participants finished every survey, people who did not benefit may have dropped out, and there was no control group [5].
A larger web survey followed 886 people around psychedelic ceremonies and retreats, and 399 remained at the four-week follow-up. On the 8-item Social Connectedness Scale, social connectedness rose from 31.5 to 36.3 on a maximum of 48 (Cohen's d = 0.46), with no sign of a ceiling, though that is a descriptive number and not a test. How much "communitas" a person felt during the ceremony correlated weakly with the gain (r = 0.25, about 6 percent of the variance), and in the path analysis of the 213 people who completed every time point, communitas felt about the retreat as a whole carried the link. The authors list their own limits: no experimental control, heavy attrition, a self-selected and mostly white sample, doubts about how well the scale translates across cultures, and travel to remote places that can lift mood by itself. Every outcome was self-report [4].
A prospective cohort of 62 adults aged 60 and over and 62 matched younger adults attended guided group sessions. Well-being, measured on the Warwick-Edinburgh scale and not on a connection scale, rose in the older group by about 4 points at two weeks and 3 points at four weeks, while the 1.7-point rise at six months was not statistically significant (p = 0.22, 23 older adults answering). The sense of togetherness felt across the whole retreat tracked well-being gains at four weeks (r = .37, about 27 people), while the intensity of the drug experience did not, and the age-group difference in that link did not reach significance (p = 0.08), so it is an exploratory finding. Sixteen of the 62 older adults (26 percent) reported a past psychiatric diagnosis, and the study was observational. The authors note that the psychedelic session itself could be a non-essential part of the benefit [14].
What about very small repeated amounts? A 2026 systematic review of 24 microdosing studies with 3,681 participants searched for effects on psychological outcomes in healthy adults, but only six studies fed the meta-analyses. The randomized-trial analysis rested on two parallel trials (117 participants) and found no difference from placebo on depression, anxiety, or stress, and the authors call the evidence inconclusive. The abstract does not report a social connection outcome, and the full text was closed to us [15]. We found no placebo-controlled microdosing study with connection as its primary outcome.
In animals
In male mice, several psychedelics reopened a developmental window for learning that social contact is rewarding, for under a week with ketamine to at least four weeks with ibogaine, a duration the authors call proportional to each drug's acute effects in humans, though with no statistical test. A lower dose did not open it, and the oxytocin response in the nucleus accumbens changed alongside the reopening without oxytocin signaling being blocked in living animals, so the mechanism is not established [16]. A later paper cites a preprint reporting a failure to replicate psilocybin's effect on social reward learning, which we have not read [17]. A 2026 mouse study found sex-specific and time-specific effects of psilocybin. Males preferred a cage-mate more at 24 hours in a barrier-climbing test, but the three-chamber test showed no change in either sex at 24 hours or one week, and the authors stress there is no uniform prosocial effect. It used a single dose, with many significant results near p = 0.03 to 0.05 and no correction for multiple comparisons in the text we read [17]. None of this shows that anything similar happens in people, and human studies are small and often pool men and women, so they cannot check sex-specific or time-specific effects.
What is not known
No controlled trial in healthy adults has yet reported lasting change in social connection. Searches of PubMed and ClinicalTrials.gov on 30 September 2026 found none, and a second check on 2 October 2026 changed nothing, though our search was partial because PsycINFO was not accessible to us. The nearest is a University of Zurich trial (NCT06081179) of 122 healthy adults aged 18 to 40 comparing psilocybin, MDMA, and methylphenidate. It lists ten primary outcomes with no stated hierarchy, measured about four weeks after the session. The registry, checked again on 2 October 2026, still shows the trial as completed in September 2025 with no results posted, and a PubMed search that day returned no publication [22]. When it reports, it will be the best evidence on this question, and we will update this page.
Who drops out. In the two large surveys above, 55 and 78 percent of participants were lost before the key follow-up, and people who had a hard time or noticed no change may be the ones who stop answering, which would inflate every uncontrolled number.
Blinding and expectancy. In high-dose studies, participants guess correctly that they received the drug at rates approaching 95 percent, according to a 2026 position paper [20]. The same paper describes the one formal expectancy analysis it found, from a trial in depression and not in healthy adults, in which expectancy did not predict response to psilocybin but did predict response to the comparison drug. Its authors call this preliminary and say it needs replication [20]. We found no equivalent analysis for social outcomes.
The substance or the group. In the retreat study, the statistical link ran through the togetherness felt across the whole retreat [4], and the authors of the older-adult cohort raise the same possibility [14]. A group virtual-reality experience offered during lockdowns, in which 58 of the 109 people who signed up answered the after-survey, produced scores on ego-dissolution and inclusion scales similar to values published in some psychedelic studies, and its communitas score was higher than the retreat sample's (44.1 against 39.6, p = 0.002). On a mystical-experience questionnaire the results ranged from lower to higher than different psychedelic studies, and 29 percent reported a complete mystical experience [18]. The authors compared averages across studies with no control group and say no suitable placebo could be defined, and here "similar" means no significant difference, with no test of equivalence, so the comparison does not show that virtual reality matches a psychedelic. How much of the closeness reported after group psychedelic sessions comes from the substance and how much from the room full of people is unknown. The international consensus on reporting the setting of psychedelic clinical trials lists the people present and the therapeutic alliance among the items to record [21], and its authors say the guidelines may allow future studies to separate these factors, though they were written for clinical trials and a Delphi consensus is expert opinion, so using them for healthy adults would extend their intended use.
Harms of togetherness. An anonymous online survey was started by 2,151 people, and 1,221 past psychedelic users were analysed. Of these, 51.6 percent reported feeling lonely or alienated during a psychedelic experience at least once, most of them rarely (28.3 percent) or sometimes (18.8 percent), and 3.8 percent often or more. Separately, 7.9 percent said that they, or someone they know, had experienced inappropriate sexual contact from a sitter, guide, or therapist, most often an underground sitter or guide (46.3 percent) or a friend or acquaintance (21.1 percent), with clinical-trial therapists accounting for 4.2 percent. The sample was recruited largely through an event and social media in Michigan and was self-selected, and the "someone they know" wording makes a rate impossible to estimate [19].
What Metanoia seeks to learn
Metanoia is a research and education charity, and it has no data of its own on this dimension yet. These are the questions it wants answered, through published research and through observational records kept under full legal authorization, without supplying any substance:
- Does a change in connection on a validated scale last at six and twelve months, and does it appear in reports from people who know the participant, not only in self-report?
- How much of any change tracks setting variables such as group size, who was present, and preparation, recorded with the international reporting guidelines for psychedelic trials as a model [21]?
- Can we build measures that detect gains in adults who already score well, and that pair a rating with concrete behavior such as actual contact with other people?
- How often does connection get worse, and in what circumstances?
Null and negative results will be reported alongside positive ones.
How it compares with other paths
Every path in this section is measured on similar outcomes, but the populations differ. Most comparison trials enrolled lonely older adults and measured how much loneliness fell, while the psychedelic studies above mostly enrolled self-selected, non-clinical people and measured a rise in connectedness, so the numbers below cannot be ranked head to head.
Design changes the answer. A meta-analysis of loneliness interventions found a mean effect of -0.37 in 12 single-group before-and-after studies, -0.46 in 18 non-randomized comparisons, and -0.20 in 20 randomized trials (95% CI -0.32 to -0.08), and 14 of the 20 randomized trials had confidence intervals that included zero. Four randomized trials that targeted distorted social thinking averaged -0.60, against -0.16 for social support, but the test for differences between these types was borderline (p = 0.05), all four used psychological approaches, and 11 of the 20 randomized trials were in adults aged 60 and over [23]. Designs without randomization produced effects roughly twice as large as randomized ones, which bears on how the uncontrolled psychedelic figures above should be read.
Group programs. In community-living older adults, 11 randomized trials of group-based treatment gave a standardized mean difference (SMD) of -0.27 (95% CI -0.48 to -0.08), with moderate certainty (the review's main text reports -0.25 for the same analysis). Group-based treatment there included psychological content such as cognitive-behavioral therapy, and 12 of the 16 group-based studies recruited people at raised risk of loneliness. Group exercise had one randomized trial (SMD -0.52, interval including zero) and low certainty. The review was funded by Humana, whose representatives took part in setting its scope and interpreting results [25]. A wider review of 70 randomized trials in adults aged 65 and over found estimates from near zero to very large, mostly with wide intervals that included zero, and rated the quality of evidence very low [24]. A rapid review of 95 studies found the largest before-and-after effect for psychological interventions (-0.79), but the authors point to regression to the mean, and over half the studies had no control group. Against control groups the psychological effect fell to -0.64 and was not significant (95% CI -1.37 to 0.09) [26].
Contemplative training outside clinical samples. A meta-analysis of eight mindfulness trials found a 6.33-point drop on the UCLA scale in an analysis of three small trials (94 participants) of people with no known mental health condition, all against a waiting list. The three groups were older adults, women with a history of breast cancer, and medical students, so "no known condition" is not the same as healthy. Certainty was low, and seven of the eight trials were at high risk of bias. Against an active comparison without mindfulness content the drop was 3.74 points and not significant, the one low-risk trial found no effect, and in older adults the pooled difference was not significant (SMD -0.12) [27]. In a nine-month open trial, 362 healthy adults were assigned to training or a control group and 226 were analysed, and paired practices raised social closeness, though the outcome was tracked within the training arms and not against an untrained group [28]. In an online trial of 809 adults, loving-kindness meditation and light exercise both raised well-being with no difference between them, 82 percent dropped out, and a difference in a small donation task (relative risk 3.57) was not statistically significant (95% CI 0.82 to 15.50) [29].
Volunteering. A systematic review of 40 papers, searched to January 2013, found that volunteers in cohort studies did better on depression, life satisfaction, and well-being, and lived longer (mortality risk ratio 0.78, 95% CI 0.66 to 0.90 across five cohorts), although three of seven mortality studies found no association. The experimental studies did not confirm the mental health results: none found a consistent benefit, three found no difference in depression, and the one trial that measured loneliness found no significant effect [30].
Where psychedelics fit. The controlled evidence supports a real shift in social processing during the acute state, usually less social pain and more felt closeness, measured in small samples over hours. The evidence for lasting change in relationships or belonging comes from uncontrolled follow-up with heavy dropout, and the size of the effect varies fourfold between studies. Group programs, contemplative training, and cognitive approaches have randomized evidence, mostly in lonely older adults, with effects from near zero to moderate that are often of low certainty or weaker against active comparison conditions. On present evidence, lasting benefit to social connection in healthy adults is unproven for psychedelics, and the comparison paths have more randomized evidence, though much of it is modest. Whether the psychedelic state adds something distinct is not settled, and the group setting may do the work.
History and tradition
Graded 4: traditional and ethnographic description. This section describes practices. It is not evidence of benefit or safety.
Many traditional uses of psychedelic plants and fungi are communal. One scholarly account concerns Santo Daime, a Brazilian religion that uses ayahuasca inside a highly structured group ritual. Members tend to attend the same congregation for years, and the author suggests that some may keep attending mainly to see friends after their own wish to drink the sacrament has faded. He notes that the anthropological literature draws on Victor Turner's concept of "communitas" to describe the solidarity these rituals build, and cites the anthropologist MacRae, who argued that Santo Daime rituals push social life toward order and structure, close to the opposite of the anti-structure that Turner's concept describes [31].
The same account is not romantic. The ritual language draws on military images, and the author describes traditional gender roles and a strict dress code. Overregulation can leave people feeling guilty or inadequate, and congregational conflicts can last for years and sometimes break communities apart [31].
A 2021 hypothesis paper proposes that psilocybin mushrooms were part of early human diets and later of rituals, and that this shaped human sociality. Its authors say psychedelics are not a missing link in human evolution, present the idea as one factor among several, and offer predictions that could disprove it [32]. We cite it as an idea, not as evidence.
The word "communitas" now appears in a scale built for group psychedelic sessions [4]. Whether the closeness that ceremonial participants describe is what the scale measures, and whether either predicts a lasting change in relationships, has not been tested.
Sources
Grades: 1 primary peer-reviewed report or official record. 2 reliable secondary (review, meta-analysis, perspective). 3 grey or community (none cited on this page). 4 traditional or ethnographic description. "Read" states how much of the source was read, as of 2 October 2026. "Full text" means the main article from start to end, without online supplements. "In part" means selected sections only and is not counted as a full read. "Abstract only" means the full text was not accessible. Bibliographic details were retrieved through PubMed, Europe PMC, and ClinicalTrials.gov.
- Holt-Lunstad J, Smith TB, Layton JB. Social relationships and mortality risk: a meta-analytic review. PLoS Med 2010. DOI 10.1371/journal.pmed.1000316 PMID 20668659. Grade 2, meta-analysis of observational studies, 148 studies. Read in part: methods, results, discussion, and limitations. Note: the association was strongest for measures that captured many aspects of integration at once (OR 1.91).
- Russell DW. UCLA Loneliness Scale (Version 3): reliability, validity, and factor structure. J Pers Assess 1996. DOI 10.1207/s15327752jpa6601_2 PMID 8576833. Grade 1, psychometric study. Read: abstract only. Full text not accessible. Note: the internal consistency of .89 to .94 was across samples of students, nurses, teachers and older adults, and scores correlated .73 across a year.
- Watts R et al. The Watts Connectedness Scale: a new scale for measuring a sense of connectedness to self, others, and world. Psychopharmacology 2022. DOI 10.1007/s00213-022-06187-5 PMID 35939083. Grade 1, scale development, survey n=1,226 plus a small trial in depression. Read in part: abstract, sample, methods, main results, discussion, and limitations. References, supplement, and long interpretive passages were not read. Note: the scale was tested in a survey sample of 1,226 completers and in a small psilocybin-versus-antidepressant trial in depression. The comparative fit index was 0.902, where values near 0.90 are conventionally the minimum acceptable, and the internal consistency of the three subscales was .84, .87 and .90. In the retreat sample scores stayed above baseline at two, four and six months, with some decline between two weeks and six months and heavy attrition.
- Kettner H et al. Psychedelic communitas: intersubjective experience during psychedelic group sessions predicts enduring changes in psychological wellbeing and social connectedness. Front Pharmacol 2021. DOI 10.3389/fphar.2021.623985 PMID 33995022. Grade 1, observational web survey, n=886. Read: full text. Supplementary material not read. Note: the Communitas Scale has an internal consistency of .92. Cohen's d is an effect-size measure in which about 0.2 counts as small, 0.5 as medium and 0.8 as large. In the path analysis more personal sharing and a good relationship with facilitators predicted more communitas.
- Weiss B et al. Examining psychedelic-induced changes in social functioning and connectedness in a naturalistic online sample using the five-factor model of personality. Front Psychol 2021. DOI 10.3389/fpsyg.2021.749788 PMID 34899488. Grade 1, prospective observational, 654 respondents at the first survey, final sample n=148. Read: full text. Supplementary tables not read. Note: about half of the sample took LSD and about a quarter took psilocybin. The graphical relatedness scale is the second closeness measure.
- Preller KH et al. Effects of serotonin 2A/1A receptor stimulation on social exclusion processing. Proc Natl Acad Sci USA 2016. DOI 10.1073/pnas.1524187113 PMID 27091970. Grade 1, RCT, double-blind crossover, n=21. Read: full text, including supplementary methods and results (PMC author manuscript, PMC4983864). Note: the sample was 12 men and 9 women, and the result was t(20) = 2.71. The authors call the right anterior midcingulate cortex the dorsal anterior cingulate cortex.
- Dolder PC et al. LSD acutely impairs fear recognition and enhances emotional empathy and sociality. Neuropsychopharmacology 2016. DOI 10.1038/npp.2016.82 PMID 27249781. Grade 1, two pooled double-blind crossover studies, n=40. Read: full text. Supplement not read. Note: the 40 volunteers were 20 men and 20 women, most with no earlier psychedelic experience. The closeness ratings were visual rating scales taken around the peak of the experience. The social value orientation task has people choose how to split resources between themselves and another person.
- Duerler P et al. LSD-induced increases in social adaptation to opinions similar to one's own are associated with stimulation of serotonin receptors. Sci Rep 2020. DOI 10.1038/s41598-020-68899-y PMID 32699231. Grade 1, RCT, double-blind crossover, n=24. Read: full text. Supplementary material not read. Note: the picture-rating task used street-art pictures. The result also shows the other side of openness to others, greater susceptibility to their views, and the participants probably overlap with those of [9].
- Preller KH et al. Role of the 5-HT2A receptor in self- and other-initiated social interaction in LSD-induced states. J Neurosci 2018. DOI 10.1523/JNEUROSCI.1939-17.2018 PMID 29555857. Grade 1, RCT, double-blind crossover, n=24, same research group and apparently the same participants as [8]. Read: full text. Note: 24 people took part, 18 of them men. Joint attention means two people directing their attention to the same thing, and the delay was 1.82 seconds against 1.65 under placebo. Activity fell in brain regions for self-processing and social cognition.
- Gabay AS et al. Psilocybin and MDMA reduce costly punishment in the Ultimatum Game. Sci Rep 2018. DOI 10.1038/s41598-018-26656-2 PMID 29844496. Grade 1, one open-label study of intravenous psilocybin (n=19) and one placebo-controlled study of MDMA (n=20), all participants men with earlier experience of the drug. Read: full text of the main article. Supplementary material not read. Note: the psilocybin was given by infusion.
- Holze F et al. Role of the 5-HT2A receptor in acute effects of LSD on empathy and circulating oxytocin. Front Pharmacol 2021. DOI 10.3389/fphar.2021.711255 PMID 34326773. Grade 1, RCT, double-blind crossover, n=16. Read: full text. Note: the empathy test took place six hours after dosing, and oxytocin was measured only at the highest dose.
- Kamilar-Britt P, Bedi G. The prosocial effects of MDMA: controlled studies in humans and laboratory animals. Neurosci Biobehav Rev 2015. DOI 10.1016/j.neubiorev.2015.08.016 PMID 26408071. Grade 2, narrative review. MDMA is not a classic or natural psychedelic and is cited as contrast only. Read: full text. The tables of individual studies were scanned, not read row by row.
- Griffiths RR et al. Psilocybin occasioned mystical-type experiences: immediate and persisting dose-related effects. Psychopharmacology 2011. DOI 10.1007/s00213-011-2358-5 PMID 21674151. Grade 1, double-blind comparison of conditions, n=18, 14-month follow-up. Read in part: abstract, methods, 14-month follow-up results, monitor ratings, discussion, and limitations. The acute-session results were scanned, not read in full. Note: the volunteers received psilocybin in a blinded comparison of conditions, and the better relationships were with family and others.
- Kettner H et al. Effects of psychedelics in older adults: a prospective cohort study. Am J Geriatr Psychiatry 2024. DOI 10.1016/j.jagp.2024.05.007 PMID 38849218. Grade 1, observational, 62 older and 62 matched younger adults, includes participants with psychiatric histories (26 percent of the older group). Read: full text. Supplementary tables not read. Note: the change in well-being did not differ significantly between older and younger adults.
- Meshkat S et al. Efficacy and safety of psychedelic microdosing on psychological outcomes in healthy adults: a systematic review and meta-analysis. CNS Drugs 2026. DOI 10.1007/s40263-026-01325-5 PMID 42593636. Grade 2, 24 studies, 3,681 participants in all, six contributing to the meta-analyses. Read: abstract only. Full text not accessible. Note: the narrative summary describes minimal effects on cognition, creativity or sustained mood in randomized trials. The randomized-trial analysis had three comparisons.
- Nardou R et al. Psychedelics reopen the social reward learning critical period. Nature 2023. DOI 10.1038/s41586-023-06204-3 PMID 37316665. Grade 1, mouse study, male mice only. Read: full text. Extended Data were read as figure legends only.
- Shadani S et al. Psilocybin modulates social behaviour in male and female mice in a time-dependent manner. Neuropsychopharmacology 2026. DOI 10.1038/s41386-026-02450-x PMID 42185639. Grade 1, mouse study. Read: full text. Supplementary material not read. Note: in the 2026 mouse study females huddled more in the first hour, and a week after the dose females shifted their preference toward familiar animals over novel ones.
- Glowacki DR et al. Group VR experiences can produce ego attenuation and connectedness comparable to psychedelics. Sci Rep 2022. DOI 10.1038/s41598-022-12637-z PMID 35637199. Grade 1, uncontrolled, 58 of 109 registered participants answered the after-survey, compared with published values. Read: full text. Part of the supplementary material not read.
- Kruger DJ et al. Psychedelic therapist sexual misconduct and other adverse experiences among a sample of naturalistic psychedelic users. Psychedelic Med 2025. DOI 10.1089/psymed.2024.0011 PMID 40342400. Grade 1, anonymous online survey, 1,221 analysed of 2,151 who started, convenience sample. Read: full text.
- Marseille E et al. Well-being as a primary endpoint in clinical trials: a call for consensus. npj Ment Health Res 2026. DOI 10.1038/s44184-026-00234-1 PMID 42581324. Grade 2, position paper that reports no new data. The expectancy finding it cites comes from a trial in depression, and its authors call the finding preliminary. Read: full text of the main article. Supplementary information not read.
- Pronovost-Morgan C, Greenway KT, Roseman L et al. An international Delphi consensus for reporting of setting in psychedelic clinical trials (ReSPCT). Nat Med 2025. DOI 10.1038/s41591-025-03685-9 PMID 40461819. Grade 1, Delphi consensus (expert opinion, not experimental evidence), 89 experts registered, 62 completed the last round, 17 countries, 30 items. Read: full text. Supplementary material not read.
- ClinicalTrials.gov NCT06081179. Does serotonin system stimulation increase pro-social behavior? University of Zurich. https://clinicaltrials.gov/study/NCT06081179. Grade 1, official registry record. Read: full record, 30 September 2026, and verified again on 2 October 2026. No results posted at that date, and no publication found in PubMed. Note: the ten primary outcomes include prosocial games and tasks, the Compassion Scale, an inclusion-of-others-in-the-self measure, and a social network questionnaire on emotional support and strain.
- Masi CM, Chen HY, Hawkley LC, Cacioppo JT. A meta-analysis of interventions to reduce loneliness. Pers Soc Psychol Rev 2011. DOI 10.1177/1088868310377394 PMID 20716644. Grade 2, meta-analysis. Read: full text. The long tables of individual studies were scanned, not read row by row. Note: the randomized trials averaged -0.16 for social support, 0.02 for social skills and -0.06 for more chances to meet people, against -0.60 for distorted social thinking.
- Hoang P et al. Interventions associated with reduced loneliness and social isolation in older adults: a systematic review and meta-analysis. JAMA Netw Open 2022. DOI 10.1001/jamanetworkopen.2022.36676 PMID 36251294. Grade 2, 70 randomized studies in adults aged 65 and over, 8,259 participants. Read: full text. Supplementary eTables and eFigures not read. Note: the review covered 8,259 participants.
- Shekelle PG et al. Interventions to reduce loneliness in community-living older adults: a systematic review and meta-analysis. J Gen Intern Med 2024. DOI 10.1007/s11606-023-08517-5 PMID 38200279. Grade 2, 60 studies, funded by Humana. Read: full text. Supplementary material not read. Note: internet training gave -0.22 in the same review.
- Blodgett JM et al. What works to reduce loneliness: a rapid systematic review of 101 interventions. J Public Health Policy 2025. DOI 10.1057/s41271-025-00561-1 PMID 40050496. Grade 2, 95 studies, includes before-and-after comparisons and studies without control groups. Read: full text. Supplementary material not read. Note: the rapid review covered 101 interventions, with smaller before-and-after effects for social interaction (-0.50) and social support (-0.34).
- Teoh SL et al. Can mindfulness help to alleviate loneliness? A systematic review and meta-analysis. Front Psychol 2021. DOI 10.3389/fpsyg.2021.633319 PMID 33716901. Grade 2, 8 trials, 815 participants, with the main result drawn from 3 trials (n=94). Read: full text. Note: the meta-analysis covered 815 participants, and the 95% confidence interval of the 6.33-point drop was -9.39 to -3.26.
- Kok BE, Singer T. Effects of contemplative dyads on engagement and perceived social connectedness over 9 months of mental training: a randomized clinical trial. JAMA Psychiatry 2017. DOI 10.1001/jamapsychiatry.2016.3360 PMID 28030741. Grade 1, open nine-month trial, 362 assigned and 226 analysed. Read: abstract only. Full text not accessible. Note: 226 were analysed because the control group did not complete the main measures. Paired practices are ones in which two people take turns sharing and listening.
- Galante J et al. Loving-kindness meditation effects on well-being and altruism: a mixed-methods online RCT. Appl Psychol Health Well Being 2016. DOI 10.1111/aphw.12074 PMID 27333950. Grade 1, RCT, n=809. Read: abstract only. Full text not accessible.
- Jenkinson CE et al. Is volunteering a public health intervention? A systematic review and meta-analysis of the health and survival of volunteers. BMC Public Health 2013. DOI 10.1186/1471-2458-13-773 PMID 23968220. Grade 2, 40 papers, searched to January 2013. Read: full text. Supplementary tables not read.
- Hartogsohn I. Set and setting in the Santo Daime. Front Pharmacol 2021. DOI 10.3389/fphar.2021.651037 PMID 34017252. Grade 4, scholarly description of one tradition's ritual and community life. Read in part: a targeted search of the full text for the passages cited here (seating, uniforms, military language, conflict, attendance, guilt, and the communitas claim). Not read from start to end. Note: participants have assigned places, sing hymns together in coordination and often wear uniforms. Some of the tradition's own foundational hymns tell members to bicker less. Ceremonies elsewhere differ in size, leadership and rules, and much of the ethnographic record was written by outsiders.
- Rodriguez Arce JM, Winkelman MJ. Psychedelics, sociality, and human evolution. Front Psychol 2021. DOI 10.3389/fpsyg.2021.729425 PMID 34659037. Grade 4, hypothesis paper, not evidence. Read in part: abstract, the relevant discussion and conclusions, and a targeted search for the passages cited here. Not read from start to end. Note: the authors say plainly that psychedelics are not a missing link in human evolution.
Search method: PubMed (NCBI E-utilities and the PubMed connector) and ClinicalTrials.gov, 30 September 2026, with the registry record and PubMed checked again on 2 October 2026. Europe PMC and PubMed Central were used to open full texts. Search results were used only to find papers. Every claim above rests on a paper or record that was opened, and where only an abstract could be read, the Read field says so and the claim is limited to what the abstract states. PsycINFO, where much of the loneliness and connection literature sits, was not accessible to us, so the map of the field is partial. The bioRxiv preprint server was not searched by keyword because its interface allows only date and category filters.
The book on all twelve dimensions Back to the twelve dimensions