BackgroundMinor hallucinations (MH) affect 30-60% of patients with Parkinson's disease (PD), and are considered precursors to structured visual hallucinations and cognitive decline. While the link between structured visual hallucinations and dementia is well established, the neuropsychological correlates of MH in PD remain unclear; most studies finding no significant cognitive differences between patients with MH and those without any hallucinations.ObjectivesPresence hallucinations (PH) being among the most prevalent MH in PD, we used a robotic procedure delivering somatomotor conflicts inducing PH experimentally to investigate whether sensitivity to such robot-induced PH aids in detecting cognitive differences between patients with MH and without hallucinations.Methods31 PD patients with MH (PD-MH) and 37 without hallucinations (PD-nH) underwent neuropsychological assessment and the robotic procedure inducing PH. The sensitivity to report robot-induced PH was analyzed in relation to cognitive performance in neuropsychological tests.ResultsPD-MH patients reported more robot-induced PH than PD-nH patients, supporting previous findings. While both groups showed comparable performance in neuropsychological testing, we found a significant association between increased sensitivity to the PH-induction and poorer performance in frontal subcortical cognitive functions, in PD-MH patients, but not in PD-nH patients.ConclusionsThese findings demonstrate that sensitivity to robot-induced PH reveals a previously undetected link between MH and frontal subcortical cognitive deficits in PD, pointing to shared underlying mechanisms between executive dysfunction and somatomotor processes involved in MH. This approach offers a novel and clinically valuable means of identifying early cognitive vulnerability that assessments relying only on standard testing may overlook.

Abstract Dopamine replacement therapy (DRT) alleviates motor symptoms in Parkinson’s disease (PD) but can trigger hallucinations in a subset of patients, yet the neural basis of this selective vulnerability is unknown. Hallucinations are among the most disabling non-motor symptoms of PD, linked to social isolation, dementia and institutionalization. Using a validated robotic paradigm to induce and quantify hallucinations in real-time, combined with resting-state fMRI in a crossover On/Off DRT design, we studied patients with PD with (PD-H) and without (PD-nH) hallucinations. DRT selectively amplified sensitivity to robot-induced hallucinations in patients with pre-existing hallucinatory phenotype (PD-H, but not PD-nH) and was accompanied by cortico-striatal and large-scale network hyperconnectivity. Rather than supporting a uniform hallucinogenic effect of dopamine in PD, these findings indicate that DRT interacts with an intrinsic neural vulnerability that varies in patients. Prospective studies will establish whether this pharmacological–behavioural signature identifies patients at risk before clinical hallucinations emerge.

Why do some experiences remain vivid while others fade? One factor may be the sense of agency (SoA), which shares self-referential mechanisms with episodic memory. We investigated whether SoA over action outcomes modulates item and spatial long-term memory. During intentional encoding, participants’ keypresses moved an object either congruently or incongruently with the chosen direction. One hour later, item and spatial memory were tested. Participants reported higher SoA in congruent trials. Congruency and higher SoA ratings were associated with faster item-recognition responses without changes in sensitivity, suggesting enhanced decision efficiency rather than improved item-memory quality. In contrast, both predicted higher spatial memory accuracy, independent of encoding duration, spatial bias, or guessing. These findings indicate that perceived control during encoding selectively strengthens memory for the spatial consequences of one’s actions, supporting a role for SoA in binding actions and outcomes into episodic representations.

Minor hallucinations (MHs) affect up to 40% of patients with Parkinson's disease (PD) and are early indicators of cognitive decline. Unlike previous group-level studies, connectome-based brain fingerprinting enables the capture of individualized neural features. Leveraging this framework in a cross-sectional cohort of patients with Parkinson's disease with and without MHs using resting-state functional magnetic resonance imaging, here we show that each patient exhibits a unique brain fingerprint, irrespective of MH status. The most individualized features were associated with clinically relevant measures such as motor-symptom severity and cognitive performance. Patients with MHs showed a loss of individual-specific features in brain networks linked to cognitive health, whereas the cerebellar-to-cortex connectivity-typically less distinctive-became more prominent, emphasizing the role of somatosensory and motor integration in MH pathogenesis. Moreover, MH-related fingerprint regions mapped onto cortical territories with lower densities of neurotransmitter systems implicated in hallucinations. Overall, our fully data-driven findings reveal a distinct, patient-specific signature that differentiates patients with Parkinson's disease and MHs.

Motor awareness (MA) describes the conscious access we have to the details of our movements. Although sensorimotor control relies on effector-specific transformations, many studies report similar MA thresholds across effectors, suggesting a shared, abstract representation. We tested this by having participants perform the same goal-directed reaching task once with a hand-held joystick and once by leaning with the whole body (center-of-mass displacement), using veridical or randomly deviated visual feedback. We measured corrective motor responses and fitted psychometric functions to derive MA thresholds. MA did not differ between effectors in non-deviated control trials or in “converging” trials where compensation aided the reach, and thresholds correlated strongly between effectors. By contrast, in “diverging” trials with increased kinematic demand, MA relied more on effector-specific sensorimotor information. These findings support a largely effector-independent MA mechanism, consistent with motor-equivalence principles, that is flexibly reweighted toward effector-specific transformations under uncertainty, thereby linking low-level sensorimotor corrections with abstract motor representations and consolidating a gap in conceptual frameworks of the sense of agency.

Hallucinations are significant symptoms in psychiatric and neurodegenerative diseases, that may indicate advanced disease progression or worse disease forms. They are also frequent in healthy individuals, especially elderly or bereaved. Despite their relevance, studying hallucinations in controlled laboratory conditions remains challenging given the limited procedures inducing clinically relevant hallucinations. We have previously developed a robotics-based protocol capable of inducing a specific clinically relevant hallucination, presence hallucination (PH), in both healthy individuals and patients. Using this approach, we have systematically investigated the sensitivity and intensity of PH-induction, as well as its effects on various sensory, behavioral and cognitive aspects. In the present study, we pooled individual-participant data from 25 in-house experiments (totaling 561 individuals) and conducted a Bayesian analysis to estimate effects and moderators of PH-induction. PH-induction was reliably induced with a medium effect-size, and individuals with schizotypal traits were more sensitive. Furthermore, we identified that PH-induction may not strongly depend on altered agency, but found a synergistic relationship between passivity and induced PH. Collectively these results elucidate the role of somatomotor processes in aberrant own body perceptions, advance understanding of psychosis, and provide powerful statistical priors for future studies.

Hallucinations are common in Parkinson’s disease (PD), with presence hallucinations (PH), reported as the sensation that someone is close to you, being the most frequent. Some PD patients report PH involving deceased loved ones, known outside the context of PD as grief or bereavement hallucinations, and which we refer to as “after death experiences” (ADEs). This study explores the phenomenology of ADEs in six PD patients through in-depth interviews, focusing on how these experiences appear and are emotionally and cognitively processed. While sharing features with bereavement hallucinations in the general population, ADEs in PD are distinct in their frequency and persistence. This recurrence offers a unique window into how individuals interpret repeated ADEs. We found that patients used both rational and reflective language to make sense of these experiences, indicating that these ADEs may be a low-level perceptual experience meeting a high-level meaning-making scaffold. These findings have implications not only for understanding ADEs in PD, but also for broader cognitive studies, offering insights into the social-cognitive mechanisms behind hallucinatory phenomena.

Isolated REM sleep behaviour disorder (iRBD) represents a prodromal stage of alpha-synucleinopathies, frequently preceding Parkinson’s disease and dementia with Lewy bodies. Beyond the characteristic sleep disturbances, alterations in brain-body coupling, reflecting neural peripheral physiological interactions, have yet to be determined. Using whole-night polysomnography, we quantified brain-body coupling via heartbeat evoked potentials (HEPs) across wakefulness, NREM, and REM sleep. Individuals with iRBD (n = 13) were compared with healthy controls (n = 23), alongside analyses of ECG-derived features. During wakefulness, iRBD patients showed altered HEPs over frontal regions between 305 and 445 ms after the R-peak. No significant group differences were observed during NREM or REM sleep, and HEP amplitude alterations were not explained by ECG differences. These diurnal HEP changes may represent a novel quantitative biomarker for iRBD during wakefulness and, in the future, could serve as a marker for phenoconversion to alpha-synucleinopathy.

The first-person perspective (1PP) refers to the spatial position and orientation from which a subject perceives the world. It constitutes a fundamental aspect of bodily self-consciousness. While typically anchored to and aligned with the physical body, 1PP can be altered (e.g., rotated of 180 deg) during out-of-body experiences (OBE) or experimentally induced bodily illusions. However, objective measures of changes in experienced 1PP remain scarce. In the present study we tested whether the extent of Peripersonal Space (PPS), which is a multisensory interface surrounding the body, can index the changes in experienced direction of 1PP, building on previous findings that PPS is dynamically remapped according to gravitational direction and experienced self-location.In a two-session experiment, we first showed that back-PPS, measured in an audio-tactile interaction task, extended farther in prone compared to supine position, consistent with the direction of the gravitational force. In Session 2, we measured PPS while exposing participants to visuo-vestibular cues via a robotic-virtual reality platform. This stimulation was designed to induce an OBE-like state, characterized by an illusory inversion of body orientation and during which participants perceived themselves as prone with a down-looking 1PP while physically lying supine. In line with our previous findings, in visual field-dependent individuals, congruent visuo-vestibular stimulation led to an extension of the back-PPS similar to that observed in prone-body orientation observed in Session 1. These results demonstrate that visuo-vestibular OBE-like manipulations can modulate PPS in a manner similar to changes in physical body posture, providing evidence for a novel, implicit measure of the 1PP.

Episodic autobiographical memory (EAM) relies on reactivations of cortical regions, originally engaged during event encoding. While reinstatement of auditory and visual regions during retrieval of external signals is well documented, reinstatement of motor regions, reflecting internal self-related signals from the person’s body, remains unexplored. Here, we investigated motor reinstatement at both cortical and peripheral levels. Using mixed-reality, participants encoded lifelike episodes involving specific motor actions. The next day, they freely retrieved these episodes while undergoing fMRI (N=30) or EMG (N=23). fMRI results revealed lateralized, effector-specific reactivation in primary and supplementary motor cortices. Moreover, hippocampal–motor connectivity increased during retrieval, and activity in premotor and supplementary motor areas scaled with subjective re-experiencing ratings. EMG recordings further showed, during retrieval, sub-threshold activation of the muscle active during encoding. Together, these results provide evidence for cortical and peripheral motor reinstatement during the re-experiencing of past events, emphasizing the embodied nature of EAM.

The pathophysiology of functional neurological disorders (FND) has been discussed to include dysfunctions in interoception, the modality about perceiving and processing internal bodily signals. However, findings on abnormal interoception in FND have been inconsistent and mainly limited to measures of accuracy and self-report. Interoceptive neuronal markers have only been investigated in specific symptoms, and interoceptive attentional modulation has been completely overlooked. In a cohort of patients with mixed FND (N = 44) and sex- and age-matched healthy controls (N = 48), we set out to assess first, interoceptive accuracy with an adapted version of the heartbeat counting task; secondly, interoceptive self-report with two different questionnaires (Multidimensional Assessment of Interoceptive Awareness and Interoceptive Accuracy Scale) and thirdly, neuronal trait markers under attention modulation by measuring heartbeat-evoked potentials, the neurophysiological signal related to the heartbeat. We searched for group differences (FND versus controls) across two attentional conditions, by asking participants to either focus on their heartbeat (i.e. interoceptive condition) or an external sound (i.e. exteroceptive control condition). Cardiac covariates (heart rate and heartrate variability or normalized electrocardiac amplitude) were included in the analysis as control. Patients with FND scored lower in both interoceptive self-report questionnaires (P < 0.020), reported higher difficulty concerning the focus towards their heartbeat (P = 0.004), while no significant difference was found in interoceptive accuracy using the heartbeat counting task (P = 0.060). Global field analyses revealed short intervals of a group-by-condition interaction in global field power (285-298 ms) and topographical differences (310-321 ms) confirming that patients with FND have lower overall activity and frontal deactivation during the interoceptive condition. Preselected electrodes for a targeted analysis of the heartbeat-evoked potential based on earlier work revealed a medium effect size attenuation at the frontal-lateralized F8 electrode at 250-595 ms following R-peak for patients with FND (P = 0.028), surviving correction for cardiac covariates. Exploratory analyses further identified an earlier difference at F1 (185-210 ms post-R-peak) in FND patients for interoceptive attention (P = 0.001), also surviving covariate control. While behavioural interoceptive accuracy was marginally preserved, these findings indicate overall altered interoceptive processing in FND, characterized by reduced self-report and difficulty to focus on cardiac signals, along with attenuated neural processes, especially in frontal-lateralized regions that further depend on attentional mechanisms. By identifying objective neural markers of interoceptive dysfunction in FND, this study highlights the involvement of interoception in a multidimensional assessment including the relevance of attention.

Background: Hallucinations, ranging from minor (MH) to structured, are a common non-motor symptom in Parkinson's disease (PD). Structured hallucinations have been associated with altered functional connectivity (FC) between dorsal/ventral attention (DAN, VAN) and default mode (DMN) networks. As structured hallucinations are linked to rapid cognitive decline and MH are often viewed as their precursor, it is imperative to understand the neural basis of MH, and its relationship with cognitive alterations. Objectives: We aimed to identify a whole-brain FC pattern associated with MH and alterations in attention-executive functioning in PD, leveraging a robotic procedure inducing presence hallucinations (riPH) experimentally, to which patients with hallucinations previously showed increased sensitivity. Methods: Non-demented PD patients (N = 53) were categorized into three subgroups based on their hallucination symptoms: no hallucinations (nH; n = 19), MH (n = 18), and structured hallucinations, with or without MH (SMH; n = 16). We combined results from the riPH procedure and neuropsychological tests and applied multivariate methods capturing their shared variance in resting-state fMRI data across the three subgroups. Results: We identified a distributed FC pattern more strongly expressed in patients with hallucinations (MH, SMH), and equally so across both groups, significantly associated with alterations in attention-executive functions and differences in riPH sensitivity. The pattern was primarily driven by FC between subcortical areas and visual network, DAN and DMN, and within-cerebellar and within-subcortical FC. Conclusions: Our results highlight the role of subcortical-cortical connectivity in PD hallucinations, associated with cognitive alterations and already present in less advanced MH patients. ### Competing Interest Statement OB is one of the inventors on patent US 10,286,555 B2 (Title: Robot-controlled induction of the feeling of a presence) held by the Swiss Federal Institute of Technology (EPFL) that covers the robot-induction of the presence hallucinations (riPH). O.B. is one of the inventors on patent US 10,349,899 B2 (Title: System and method for predicting hallucinations) held by the Swiss Federal Institute of Technology (EPFL) that covers a robotic system for the prediction of hallucinations for diagnostic purposes. OB, JP and FB are inventors on patent n. EP4407627A1 (Title: Numerosity estimation impairment measurement system) held by the Swiss Federal Institute of Technology (EPFL) that covers the implicit measure of presence hallucinations. ### Author Declarations I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained. Yes The details of the IRB/oversight body that provided approval or exemption for the research described are given below: The Cantonal Ethics Committee of Geneva gave ethical approval for this work (protocol reference 2019-02275), in accordance with the Declaration of Helsinki. I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals. Yes I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance). Yes I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable. Yes Data sharing will require a formal data use agreement. Requests will be evaluated based on institutional and departmental policies to determine whether the data requested is subject to intellectual property or patient privacy obligations. Swiss National Science Foundation, https://ror.org/00yjd3n13 Fondazione Teofilo Rossi di Montelera e di Premuda Fondation Bertarelli, https://ror.org/04h7fev25 Parkinson Schweiz Fondation Leenaards, https://ror.org/004h88r69 Empiris foundation Synapsis Foundation, https://ror.org/03jrr8f73

Autonoetic consciousness (ANC) enables individuals to relive past events by recalling sensory and emotional details tied to a specific place and time related to a past event. Next to ample evidence linking ANC to memory-related mechanisms, recent behavioral evidence suggests that ANC is modulated by sensorimotor and multisensory bodily mechanisms that are fundamental for bodily self-consciousness (BSC). However, the neural mechanisms of how BSC modulates ANC remain unclear. Based on previous work showing that activity in dorsal premotor cortex (dPMC) and hippocampus indexes BSC effects and its related sense of agency (SoA) in episodic memory, we here explored whether functional connectivity between dPMC and hippocampus at encoding is related to later autonoetic reliving. Using virtual reality (VR), during the encoding of virtual scenes, we manipulated BSC by varying the level of visuomotor (feedback of participants' right-hand movements) and perspectival congruency (first-person versus third-person perspective), while recording brain activity, and collected ANC 1 week after encoding by established questionnaire procedures. Our results show that changes in functional connectivity strength between left hippocampus and left dPMC (contralateral to hand movements) were associated with different levels of visuomotor and perspectival congruency. Moreover, this connectivity modulated the relationship of SoA during encoding with later subjective ANC during reliving. Thus, higher connectivity was associated with a stronger SoA-ANC association, while lower connectivity led to a weaker association. Critically, this modulatory effect was absent for dPMC-hippocampus connectivity in the right hemisphere (i.e., ipsilateral to hand movements). These findings support the role of dPMC, a key SoA region, in ANC, linking subjective bodily experience during encoding to the subsequent subjective re-experiencing of past events.

Introduction: The preservation of the human self—a fundamental yet underexplored aspect of neurosurgical practice—has gained increasing attention in recent years. Research question: How can neural correlates of self-consciousness be identified, monitored, and protected during brain tumor surgery, and how might this reshape the concept of “onco-functional balance”? Material and methods: This review synthesizes emerging evidence from neuroimaging, neuropsychology, and intraoperative neurophysiology to build a framework for integrating the concept of self into modern neurosurgical practice. Results: We describe the anatomical and functional basis of bodily and cognitive self-awareness, highlighting the roles of interoception, multisensory integration, and higher-order cortical networks such as the medial prefrontal cortex, insula and temporoparietal junction. We outline perioperative tools for clinical assessment, including validated scales for anosognosia and disownership, as well as the Self-Other Voice Discrimination (SOVD) paradigm and Heartbeat-Evoked Potentials (HEPs), which offer quantifiable markers of self-processing. Discussion and conclusion: We argue for a reconceptualization of “eloquent” cortex to include regions critical for the preservation of self. As neurosurgery advances toward precision-guided, patient-centered care, protecting the self must become an explicit goal alongside motor, sensory, and language preservation. Future directions include real-time intraoperative monitoring of HEPs, development of functional risk maps for self-related structures, and broader implementation of personalized, neurocognitive surgical planning.Ultimately, this work proposes a shift from an “onco-functional” to an “onco-functional-identity” paradigm—where the integrity of the patient's personality, agency, and awareness becomes a measurable, preservable endpoint of neurosurgical care.

Individual variability in connectome organization offers a unique framework for capturing patient-specific alterations and advancing personalized models in medicine. Minor hallucinations (MH) affect up to 40% of Parkinson’s disease (PD) patients and are early indicators of cognitive decline and dementia, hence crucial for early intervention. While previous studies focused on group-level differences, connectome-based brain fingerprinting enables deeper, individualized analysis of neural change. Applying this approach to PD patients with and without MH using resting-state fMRI, we show that each patient exhibited unique brain fingerprint, revealing rich quantifiable personalized features with medical relevance. MH-patients showed a loss of subject-specific features in brain networks linked to cognitive health, while somatosensory regions – typically less distinctive – became more prominent, emphasizing their role in MH pathogenesis. These differences enabled to identify – in an entirely data driven manner – patient-specific networks linked to early subclinical cognitive alterations, as well differential spatial fingerprinting organization linked to cortical densities of neurotransmitters. These findings reveal a distinct, patient-specific connectomic signature that differentiates PD patients with MH, uncovering early neural markers for precision medicine in PD. ### Competing Interest Statement The authors have declared no competing interest. Swiss National Science Foundation, https://ror.org/00yjd3n13, 188798, 221182 CARIGEST SA (Fondazione Teofilo Rossi di Montelera e di Premuda and a second one wishing to remain anonymous), , Parkinson Suisse, , EPFL Neuro X Post-doctoral fellowship program, , Fondation Leenaards, https://ror.org/004h88r69, Synapsis Foundation, , CIBERNED (Carlos III Institute) and FIS, , PI18/01717 Instituto de Salud Carlos III, https://ror.org/00ca2c886, PERIS Generalitat de Catalunya, , SLT008/18/00088

INTRODUCTION:Parkinson's Disease (PD) may lead to cognitive symptoms, including visuo-spatial attentional deficits such as unilateral spatial neglect (USN). Although there is some evidence for USN in PD patients, especially in those with left-sided onset of motor-symptoms (LPD), previous studies revealed inconsistent and highly variable results in neglect tasks using line bisection and have not systematically compared LPD with RPD (PD patients with right-sided onset) or healthy controls (HC). METHODS:We designed a fully automatized online web-based line bisection task and tested a group of 170 PD patients (81 RPD, 66 LPD) and 45 HC. RESULTS:Our data reveals (1) a rightward bias on line bisection in LPD compared to HC and RPD, (2) right-sided pseudoneglect in HC as well as RPD. Our data also reveals that (3) PD patients, independent of the primarily affected side, are more variable in their line bisections. CONCLUSION:These data provide robust evidence for systematic changes in visuo-spatial perception in PD and present a new online diagnosis and monitoring tool for USN to evaluate patients at their home.

Episodic autobiographical memory retrieval (EAM) is characterized by the vivid sense of re-experiencing past events. While doing so, different viewpoints over the scene can be taken, often referred to as the field or observer perspectives. However, the precision with which individuals can accurately reconstruct their original viewpoint over an event, within the field perspective, has not yet been characterized. Here we developed a task and continuous implicit measure to quantify this ability (viewpoint retrieval accuracy, VRA) and study its association with explicit measures of re-experienced vividness. Twenty-five participants experienced lifelike events in mixed reality, with five controlled observer–object angles at encoding. After a 24 h delay, they retrieved each event, rated re-experiencing vividness, and indicated their original viewpoint (VRA). Mean VRA decreased with increasing encoding eccentricity (observer–object angles) and was consistently biased toward more central viewpoints. Moreover, VRA significantly explained re-experiencing vividness, with higher re-experiencing ratings being associated with smaller errors. These findings establish VRA as an implicit and continuous proxy for re-experiencing vividness, and a new tool for investigating the spatial and bodily foundations of EAM retrieval.

Episodic autobiographical memory (EAM) is a building block of self-consciousness, involving recollection and subjective re-experiencing of personal past experiences. Any life episode is originally encoded by a subject within a body. This raises the possibility that memory encoding is shaped by bodily self-consciousness (BSC), a basic form of self-consciousness arising from the multisensory and sensorimotor perceptual signals from the body. Recent studies in healthy subjects showed that embodied encoding improves EAM, with the involvement of the hippocampus. However, there are only few imaging studies to date, hippocampal data are not consistent, and the role of hippocampal damage is not understood. We investigated how different BSC states during encoding, modulate later EAM retrieval, in a patient with severe amnesia caused by rare bilateral hippocampal damage. We performed three separate behavioral experiments using immersive virtual reality. The patient showed consistently more difficulties recollecting information encoded in embodied versus disembodied states, particularly when asked to recall her perspective experienced at encoding. These results contrasted with the usual beneficial effect of BSC on EAM, and significantly differed from controls. These data provide consistent evidence that BSC impacts encoding and later reliving, and shows that the hippocampus is not just a critical structure for EAM, but also for effects of embodiment on memory. Additional fMRI data extend these findings by revealing that hippocampal-parietal connectivity mediates BSC-EAM coupling. Our findings plead for an important role of BSC in EAM, mediated by the hippocampus and its connectivity, leading to embodied memories that are experienced as belonging to the self.

BACKGROUND:Opioid use disorder remains a critical health care challenge because current therapeutic strategies have limitations that result in high recurrence and deaths. We evaluated the safety and feasibility of focused ultrasound (FUS) neuromodulation to reduce substance cravings and use in severe opioid and co-occurring substance use disorders. METHODS:This prospective, open-label, single-arm study enrolled 8 participants with severe, primary opioid use disorder with co-occurring substance use. Participants received a 20-minute session of low-intensity FUS (220 kHz) neuromodulation targeting the bilateral nucleus accumbens (NAc) with follow-up for 90 days. Outcome measures included safety, tolerability, feasibility, and effects of FUS neuromodulation by assessment of adverse events, substance craving, substance use (self-report, urine toxicology), mood, neurological examinations, and anatomical and functional magnetic resonance imaging (fMRI) at 1, 7, 30, 60, and 90 days post-FUS. RESULTS:No serious device-related adverse events or imaging abnormalities were observed. Following FUS, participants demonstrated immediate (p < .002) and sustained (p < .0001; mean 91%) reductions in cue-induced opioid craving, with median ratings on a scale from 0 to 10 as follows: 6.9 (pre-FUS) versus 0.6 (90-day post-FUS). Craving reductions were similar for other illicit substances (e.g., methamphetamine [p < .002], cocaine [p < .02]). Decreases in opioid and co-occurring substance use were confirmed by urine toxicology. Seven participants remained abstinent at 30 days; 5 participants remained abstinent throughout 90 days post-FUS. Resting-state fMRI demonstrated decreased connectivity from the NAc to reward and cognitive regions post-FUS. CONCLUSIONS:NAc FUS neuromodulation is safe and a potential adjunctive treatment for reducing drug cravings and use in individuals with severe opioid and co-occurring substance use disorders. Larger, sham-controlled, randomized studies are warranted.