Inhaled N, N-dimethyltryptamine diminishes connectivity between the ventral tegmental area and the nucleus accumbens: relevance to pathologies of mesolimbic and mesocortical pathways

The paper outlines unresolved questions about neural systems of reward, described in terms of separable functions — liking (hedonic impact), wanting (incentive salience and motivation) and learning (prediction and adaptation). The mesolimbic and mesocortical dopaminergic pathways, both originating in the ventral tegmental area (VTA) and projecting respectively to the nucleus accumbens (NAc) and to prefrontal cortical regions including the medial prefrontal cortex (mPFC) and anterior cingulate cortex (ACC), are central to these functions. Dysfunctional reward processing and altered mesocorticolimbic connectivity are implicated across multiple psychiatric conditions (for example depression, addiction, OCD and chronic pain). Psychedelic tryptamines, including N,N-dimethyltryptamine (DMT), have been shown in prior studies to alter large-scale brain network dynamics and regional activity in networks relevant to reward, affect and salience. G. and colleagues set out to characterise, in healthy volunteers, how inhaled DMT acutely alters resting-state functional connectivity (rsFC) within core mesocorticolimbic reward regions. The primary aim was to test whether inhaled DMT versus a control condition changes connectivity between key nodes of the dopamine-related reward system (notably the VTA and NAc, and their cortical targets), with the hypothesis that DMT would alter connectivity within this network. The authors justify using healthy participants to measure direct neural effects without disease or medication confounds and position this as a preliminary investigation of DMT's impact on reward circuitry.

The researchers used a within-subject design in which each participant completed an active DMT session and a control session. The active session involved self-administration by inhalation of a Mimosa hostilis (root bark) sample described as approximately 50–70 mg; the sample was quantified by HPLC-DAD to contain 30.92% DMT. The extraction does not clearly report the absolute milligrams of pure DMT delivered per inhalation. Each dose was weighed immediately before administration. Washout intervals between sessions were at least one month (mean interval 86 days, range 34–215 days). All participants received the DMT session first and the control (no inhalation) session second; the authors report this fixed order was chosen to respect ritualistic set-and-setting and to reduce expectancy effects. Eleven healthy adults (mean age 37 ± 12.4 years; 7 male, 4 female), all with prior inhaled DMT and other psychedelic experience, were recruited. Screening included structured psychiatric (M.I.N.I.), cognitive (MMSE) and socioeconomic assessments, physical examination and targeted laboratory testing. Exclusion criteria included personal or family history of psychotic or bipolar spectrum disorders, current or past addictions, major medical comorbidities, certain medications, pregnancy/breastfeeding and conditions associated with altered reward function (e.g. eating disorders, obesity, chronic pain). Participants abstained from recreational psychoactive drugs for two months and from caffeine/tobacco for 24 hours prior to sessions. Procedurally, participants self-administered DMT just outside the MRI suite, handed the pipe to facilitators after exhalation, walked a short distance to the scanner and were assisted into the scanner; mean time from smoking to scan start was approximately 4 minutes, with effects peaking at about 2–3 minutes and diminishing by the end of the scan. Anatomical T1 and resting-state fMRI data were acquired (T1: 1 mm isotropic resolution; other acquisition parameters reported). Subjective acute effects were measured after scanning using the Hallucinogen Rating Scale (HRS), a 100-item questionnaire covering somaesthesia, affect, perception, cognition, volition and intensity. Imaging preprocessing and analysis used CONN and SPM with standard steps (realignment/unwarp, slice timing correction, outlier detection, segmentation and normalization to MNI space, smoothing with 8 mm FWHM). Denoising included CompCor components (white matter 10 components, CSF 5 components), motion parameters and derivatives, outlier scans and high-pass filtering >0.01 Hz. ROI-to-ROI rsFC matrices were computed among eight a priori reward-related ROIs: bilateral VTA, bilateral NAc, bilateral amygdala, ACC and mPFC. Connectivity strength was Fisher-transformed bivariate correlations from a weighted general linear model; group-level contrasts (DMT > control) used a random-effects GLM and were thresholded at p-FDR < 0.05 across all ROI pairs. HRS data were analysed with repeated-measures ANOVA (significance p<0.05) and Spearman correlations assessed associations between change in connectivity (Δr/Δz) and change in HRS scales (ΔHRS), with authors noting FDR correction for multiple comparisons. Effect sizes for connectivity were reported as Δz (difference in Fisher z-transformed correlations) and HRS effects as partial η². A post-hoc power analysis reported achieved power of 96.3% for paired t-tests given a conservative effect size (d = 1.26) from prior work and n=11.

Comparing DMT and control conditions within subjects, the primary ROI-to-ROI analysis identified significant condition-related connectivity changes involving the mPFC, ACC, bilateral NAc and left VTA (overall F(1,10)=20.99, p=0.006, Δz=0.061, FDR-corrected across ROI pairs). Specific pairwise results included a decrease in resting-state functional connectivity (rsFC) between the right NAc and left VTA (t(10) = -2.25, p = 0.048, Δz = -0.042). In contrast, several cortico-striatal connections increased under DMT: right NAc–ACC (t(10) = 2.74, p = 0.021, Δz = 0.081), mPFC–ACC (t(10) = 3.45, p = 0.006, Δz = 0.134) and left NAc–ACC (t(10) = 4.24, p = 0.002, Δz = 0.069). An apparent increase in amygdala–ACC connectivity was noted but did not survive correction for multiple comparisons. To check for non-specific effects, the authors tested connectivity between motor cerebellum and reward regions and found no significant changes (all p‑FDR > 0.05). On the Hallucinogen Rating Scale, the authors report significant increases across HRS scales in the DMT condition compared with control (the extract does not provide a complete breakdown of each subscale in the prose). Correlational analyses showed positive associations between connectivity changes and subjective effects: change in VTA–NAc connectivity correlated with increases in HRS volition (Spearman r = 0.66, p = 0.026), and change in NAc–ACC connectivity correlated with increases in HRS perception (r = 0.67, p = 0.024). Neither association survived FDR correction for multiple comparisons. No adverse events were reported during or after either session.

The authors interpret their findings as evidence that inhaled DMT acutely reorganises functional relationships within mesolimbic and mesocortical reward networks. Specifically, DMT reduced connectivity between a core mesolimbic pair (right NAc – left VTA) while increasing cortico-striatal coupling (bilateral NAc – ACC) and mPFC–ACC connectivity. They note the ACC's role as an integrative hub in the salience network that modulates limbic, reward and pain-related circuitry, and the mPFC's dual involvement in mesocortical reward processing and the default mode network (DMN). From this perspective, the pattern of reduced VTA–NAc coupling alongside enhanced NAc–ACC and mPFC–ACC coupling could reflect a transient dampening of incentive-motivational signalling together with strengthened integration of affective and executive cortical processes. The authors link the observed VTA–NAc decrease to potential reductions in incentive salience or "wanting" and discuss how this might counteract sensitised reward-cue responses implicated in addiction. They suggest the increased NAc–ACC coupling may support improved integration of motivation and emotion, aligning with prior reports that psychedelics modulate ACC function. Correlations between connectivity changes and HRS subscales (volition and perception) are presented as preliminary evidence that mesocorticolimbic modulation is related to subjective aspects of agency and perceptual alteration, although these correlations did not survive correction for multiple comparisons. G. and colleagues acknowledge that DMT's pharmacology is multi-modal (serotonergic 5‑HT1A/2A/2C agonism and interactions with glutamate, dopamine, sigma-1, TAARs and others), and therefore the connectivity changes likely reflect combined neurochemical effects; nonetheless, serotonin–dopamine interactions are emphasised as relevant to motivational processing. They situate the findings within reinforcement-learning frameworks (reward prediction errors mediated by VTA–NAc signalling) and note parallels between their acute connectivity results and clinical reports of rapid antidepressant and analgesic effects following DMT or ayahuasca. The authors highlight that the increased cortico-cortical coupling (mPFC–ACC) is consistent with prior reports of enhanced DMN–SN interactions under psychedelics and propose this may underlie increased psychological and cognitive flexibility. Key limitations acknowledged by the authors include the small sample size (n = 11), which constrains statistical power and the capacity to control for confounds such as age and sex, and the within-subject fixed-order design (all participants received DMT first), which the authors justified on set-and-setting grounds but which may nonetheless influence expectancy or order effects. They also note that the present study cannot establish causal mechanisms and that the functional significance and clinical implications of the observed rsFC modulations require further investigation in larger and clinical samples.