The default mode network has occupied a central place in cognitive neuroscience since its discovery in the early 2000s, when positron emission tomography and, later, functional magnetic resonance imaging revealed a set of regions that consistently decrease their activity during demanding external tasks and increase it during quiet rest. This network, anchored in the medial prefrontal cortex, the posterior cingulate cortex, the precuneus, and the inferior parietal lobule, is thought to support autobiographical memory retrieval, envisioning the future, self-referential processing, and mind-wandering. In schizophrenia, decades of imaging work have documented disruptions in this network’s connectivity, which have been linked to disturbances in self-monitoring, hallucination severity, and disorganized thought. What the new study adds is a bridge between this circuitry and one of the most consequential health behaviors in the disorder: smoking.

The researchers approached the question using resting-state functional connectivity analysis, a technique that measures the degree to which spatially distributed brain regions fluctuate together in their blood-oxygen-level-dependent signals while participants lie quietly in the scanner. Synchronized low-frequency fluctuations are interpreted as a signature of functional coupling, even though they do not directly measure anatomical wiring. By parcellating the cortex and extracting connectivity profiles associated with the default mode network, the team was able to quantify how strongly parietal nodes of this circuit communicated with the rest of the network and with other major systems, including the frontoparietal control network and the salience network, which are implicated in cognitive control and in switching between internal and external attention.

Across the study sample, the strength of parietal default mode connectivity emerged as a statistically reliable correlate of tobacco use measures, which for many participants included biologically verified indicators such as cotinine levels, the primary metabolite of nicotine, rather than relying solely on self-report. This methodological point matters enormously. Self-reported smoking in psychiatric populations is notoriously unreliable, shaped by stigma, recall difficulty, and cognitive impairment, and studies that depend on it risk both overestimation and underestimation of true exposure. By anchoring the smoking phenotype in objective biochemical measures where available, the analysis strengthens the claim that the brain-behavior association is genuine rather than an artifact of reporting bias.

Why should a network associated with daydreaming and self-referential thought care about nicotine? One plausible explanation lies in the interplay between the default mode network and dopaminergic signaling. Nicotine acts on nicotinic acetylcholine receptors that modulate dopamine release in the mesolimbic pathway, the reward circuitry that reinforces drug-taking. In psychosis, this dopaminergic system is already dysregulated, with the prevailing neurobiological models of schizophrenia positing aberrant striatal dopamine synthesis and release as a proximate cause of positive symptoms. Nicotine’s ability to transiently normalize aspects of this signaling, or to dampen sensory gating deficits, has long been cited in self-medication accounts. The new findings suggest that individual differences in the intrinsic organization of the default mode network may reflect, or even partly determine, the degree to which nicotine exerts reinforcing and normalizing effects in a given brain.

A complementary interpretation comes from the cognitive domain. The default mode network and the frontoparietal control network are engaged in a dynamic antagonist relationship: when the former is active, the latter is typically suppressed, and effective cognitive performance requires the orchestration of switching between internally and externally directed states. Smoking initiation and maintenance depend on executive functions, including the capacity to inhibit impulses, delay gratification, and weigh long-term health consequences against immediate relief. If parietal default mode connectivity indexes the rigidity of internal focus or the difficulty of disengaging from internally generated thought, then individuals with stronger or atypical coupling may find external, health-protective control processes harder to deploy, making tobacco use more likely to persist. In this framing, connectivity is not a cause of smoking in a simple causal chain but a marker of the neurocognitive soil in which the behavior takes root.

The psychosis context amplifies both the scientific and clinical significance of these results. Roughly three-quarters of people with schizophrenia who smoke do so heavily, and smoking accounts for the majority of the excess mortality observed in the disorder. Yet smokers with psychosis are less likely to receive smoking cessation counseling, less likely to be prescribed pharmacotherapy such as varenicline or bupropion, and more likely to relapse after quitting attempts. If neural measures such as default mode connectivity could stratify patients by the likely neurobiological drivers of their smoking, clinicians might eventually tailor interventions accordingly, deploying more intensive combined behavioral and pharmacological strategies for those whose circuit profiles indicate a strongly entrenched pattern. The present study does not yet support such clinical deployment, but it supplies the kind of mechanistic correlate that personalized approaches require.

As with all resting-state connectivity research, important caveats frame the interpretation. Functional connectivity is correlational; the cross-sectional design of the analysis cannot determine whether atypical parietal connectivity predisposes individuals to smoking, whether chronic nicotine exposure reshapes the network over time, or whether both are downstream of a third factor such as illness severity, medication exposure, or shared genetic risk. Longitudinal designs, within-person repeated imaging, and causal modeling techniques, including studies in animal models where nicotine exposure can be experimentally controlled, will be needed to disentangle these possibilities. Sample heterogeneity, medication effects, and the modest effect sizes typical of brain-wide association studies further caution against overreading any single result. Still, the consistency of the default mode network’s involvement across cognitive, symptomatic, and now behavioral domains in psychosis builds a cumulative case that this circuit is a genuine hub of individual difference in the disorder.

For the broader field, the study exemplifies a trend in psychiatric neuroscience toward connecting large-scale intrinsic brain organization with real-world health behaviors, rather than with abstract laboratory measures alone. Smoking is among the most modifiable risk factors in severe mental illness, and understanding its neural correlates is a step toward interventions that could meaningfully extend lives. The finding that the brain’s daydreaming circuitry carries information about tobacco use in psychosis is a reminder that even the most habitual and seemingly volitional behaviors are embedded in the biology of the disorders themselves, and that dismantling smoking’s grip on this vulnerable population may ultimately require working with, rather than around, the architecture of the psychotic brain.

Subject of Research: Resting-state parietal default mode network functional connectivity and its association with tobacco use in psychotic disorders

Article Title: Parietal default mode network connectivity is associated with tobacco use in psychosis

Article References: Parietal default mode network connectivity is associated with tobacco use in psychosis. (n.d.). https://doi.org/10.1038/s41537-026-00797-0

Image Credits: AI Generated

DOI: 10.1038/s41537-026-00797-0

Keywords: schizophrenia, psychosis, default mode network, tobacco use, nicotine, functional connectivity, resting-state fMRI, neuroimaging, dopamine, smoking cessation, parietal cortex, severe mental illness

The study, published in Nature Communications, analyzed 1,957 fMRI sessions from 1,199 people. The work was led by Cory Shain, an assistant professor of linguistics at Stanford University, in collaboration with Evelina Fedorenko of the Massachusetts Institute of Technology. Researchers used functional connectomics, an approach that maps relationships between brain areas by measuring how their activity fluctuates over time. Unlike a conventional MRI, which produces a structural image of the brain, fMRI records changes in blood oxygenation that indirectly reflect local neural activity. When separate regions repeatedly become more or less active together, scientists can infer that they are functionally connected, even if they are not physically adjacent.

Rather than beginning with a narrowly defined language task, the researchers first examined the timing of activity across many brain regions without considering what participants were doing in the scanner. This allowed them to identify several candidate networks based on their internal patterns of coordination. One of these networks occupied areas previously linked to language, particularly regions in the left frontal and temporal lobes. The researchers then tested the candidate networks against participants’ responses during language tasks. The network identified from its spontaneous activity responded strongly when people read, listened to, or produced language, while showing relatively little response to nonlinguistic activities. This pattern provided evidence that the network was functionally specialized rather than simply reflecting general attention or sensory processing.

The most striking test came when the scientists removed all scans collected during language-related activities. They then attempted to reconstruct each participant’s language network using only data from nonlinguistic tasks or periods of rest. The same network could still be detected in individual brains with high reliability. In other words, the researchers did not need to observe a participant speaking or understanding sentences in order to locate the system that supports those abilities. The result indicates that the network’s ongoing fluctuations carry a recognizable signature. Its activity rises and falls continuously, and those fluctuations appear to preserve enough information about the network’s organization to distinguish it from other functional systems in the brain.

The findings also highlight a tension between shared human biology and individual variation. Across participants, the language network generally appeared in similar parts of the left hemisphere, including frontal and temporal regions involved in speech production, comprehension, vocabulary, and the integration of meaning. Yet its precise boundaries, shape, and internal connectivity differed considerably from one person to another. Averaging these brains together could blur or even conceal those differences. By analyzing individuals rather than relying only on group-level maps, the researchers showed that each person possesses a language network with a distinctive architecture that remains remarkably consistent within that individual across different tasks.

This individual stability may help explain why earlier debates about language and the brain have persisted for so long. Since the nineteenth century, cases of aphasia—selective impairments in speaking, understanding, reading, or writing after brain damage—have demonstrated that language depends on organized biological structures. However, the exact boundaries and roles of those structures have remained controversial. Language is distributed across multiple regions and depends on interactions among sound processing, memory, motor control, attention, and conceptual knowledge. The new study does not reduce language to one isolated “language center.” Instead, it supports the existence of a coordinated, language-selective network while showing that the network is embedded within the broader architecture of each individual brain.

Because fMRI measures blood flow rather than neurons directly, the researchers are not literally watching individual language neurons fire. The technique captures a slower physiological consequence of neural activity: changes in oxygenated and deoxygenated blood associated with local energy use. Functional connectivity therefore describes statistical coordination between brain regions, not a direct wiring diagram or proof that one region causes another to activate. Even so, the scale of the dataset and the consistency of the findings strengthen the conclusion that the language system has a detectable functional organization. The fact that the network can be recovered from resting or nonlinguistic data suggests that its signature is robust enough to survive changes in attention, sensory input, and immediate behavioral demands.

The method could eventually become valuable in clinical neuroscience. After a stroke, tumor, traumatic injury, or neurosurgical procedure, parts of a person’s language network may be damaged or reorganized. Traditionally, clinicians often use language tasks during brain imaging to determine which areas remain active, but such tasks can be difficult for patients who are unconscious, severely impaired, very young, or unable to cooperate. If a patient’s language network can be estimated from resting-state activity or from simple nonlinguistic tasks, doctors may gain a new way to assess the organization of intact tissue. Researchers could compare the surviving network with its expected individual pattern, investigate how damage alters communication between regions, and potentially identify pathways that support recovery from aphasia.

The study may also influence how scientists think about brain imaging more broadly. Many neuroimaging results are based on averages across groups, producing maps that describe a hypothetical typical brain. The new analysis demonstrates the value of preserving individual patterns instead of treating variation as noise. A person’s functional connectome may act like a biological signature, revealing stable features of cognition even when behavior changes from moment to moment. The researchers emphasize that language is complex and difficult to define, but their strategy deliberately set aside assumptions about what language should look like. By allowing spontaneous brain activity to reveal the network first, then testing its response to language, they found evidence for a system that is both specialized and continuously present—quietly active even when no words are being spoken.

Subject of Research: The individual organization and spontaneous functional connectivity of the human brain’s language network.

News Publication Date: 13-Aug-2026

Web References: Nature Communications article; DOI: 10.1038/s41467-026-75745-8

References: Nature Communications, DOI: 10.1038/s41467-026-75745-8

Keywords: language network, brain imaging, fMRI, functional connectivity, functional connectome, neurolinguistics, language processing, language comprehension, neuroscience, aphasia, brain structure, human brain, neurophysiology, psychological science

Alpha oscillations, typically ranging from 8 to 12 Hz, have long been recognized as critical indicators of the brain’s resting state and attentional processes. Traditionally, neurofeedback protocols aim to train individuals to consciously enhance these rhythms, resulting in improved cognitive performance, reduced anxiety, and beneficial effects on various neurological conditions. What sets this study apart is the demonstration that alpha power can amplify spontaneously, even without explicit instructions or strategic efforts from participants, hinting at endogenous regulatory mechanisms at play.

The experimental design involved participants undergoing standard neurofeedback sessions where real-time electroencephalographic (EEG) signals were monitored and fed back visually. Interestingly, the researchers observed a consistent trend: alpha wave enhancement occurred naturally over the course of the sessions, irrespective of whether participants were directed to modulate their brain activity or simply allowed to passively engage with the feedback loop. This observation suggests intrinsic neural adaptation processes that operate alongside, or possibly independently of, conscious control efforts.

Delving deeper, the team employed advanced time-frequency analysis and spectral decomposition methods to discern the specific temporal dynamics underlying the alpha power increase. The results indicate that the spontaneous elevation is not a fleeting artifact but a slowly building process that consolidates over repeated neurofeedback trials. The sustained nature of this increase hints at neuroplastic changes, potentially mediated through synaptic modulations within thalamo-cortical circuits known to generate alpha rhythms.

These insights hold profound implications for the field of neurofeedback therapy, which has historically grappled with inconsistent efficacy across individuals. Understanding that alpha power can augment autonomously suggests that neurofeedback protocols could be optimized to harness these innate oscillatory plasticity mechanisms, possibly reducing the cognitive load on participants and improving overall treatment outcomes. This opens a new vista wherein neurofeedback becomes a blend of active training and passive neurological facilitation.

Moreover, the spontaneous dynamics of alpha waves elucidated by Maaz and colleagues dovetail with emerging theories about the brain’s intrinsic activity patterns during rest and task engagement. Alpha oscillations have been implicated in functional inhibition, gating of sensory input, and modulation of attentional focus. The natural propensity for these rhythms to enhance under neurofeedback may reflect a fundamental homeostatic process aimed at balancing cortical excitability and maintaining optimal information processing states.

From a neurophysiological perspective, the findings raise intriguing questions about the loci of alpha wave generation during neurofeedback. The authors speculate on the involvement of distributed neural networks, including parietal and occipital regions, interacting with deep thalamic nuclei and cortical layer-specific circuits. These complex interactions could facilitate a feedback loop where initial minor fluctuations in oscillatory power get amplified through recurrent connectivity and synaptic potentiation, culminating in the observed spontaneous increase.

Clinically, the study’s revelations may translate into significant advancements for disorders linked to dysregulated alpha activity, such as anxiety, depression, and attention-deficit/hyperactivity disorder (ADHD). Harnessing spontaneous alpha increases might offer therapeutic windows where minimal intervention suffices to recalibrate dysfunctional brain rhythms, marking a paradigm shift from effort-intensive biofeedback training to subtler, perhaps even subconscious, modulation techniques.

In cognitive neuroscience, the implications extend to enhancing human performance and creativity. Given alpha waves’ association with states of relaxed alertness and creative ideation, the discovery that alpha power can grow naturally during neurofeedback suggests that mental states conducive to innovation might be more accessible than previously thought. This could pave the way for non-invasive methods to foster cognitive flexibility, problem-solving skills, and stress resilience without taxing mental resources.

Importantly, the study also underscores methodological considerations for future neurofeedback research. The spontaneous alpha increase phenomenon necessitates refined experimental controls to disambiguate volitional from automatic neural enhancements. Consequently, the authors advocate for integrative approaches combining behavioral measures, subjective reports, and electrophysiological data to capture the nuanced interplay between conscious effort and automatic brain dynamics.

The team employed rigorous statistical modeling, including mixed-effects designs, to validate the robustness of their observations across diverse participant samples and session parameters. This methodological rigor provides strong confidence that the spontaneous alpha power increase is a replicable and generalizable phenomenon, not merely an idiosyncratic finding or statistical anomaly.

From a mechanistic perspective, the researchers propose that spontaneous alpha power enhancement might stem from intrinsic brain rhythms’ resonance properties. Neural circuits could be entering an optimized state by leveraging feedback-induced plasticity, whereby neurofeedback acts not just as an external signal, but as a catalyst for endogenous rhythmic stabilization. This reframing positions neurofeedback less as a ‘training’ paradigm and more as a facilitatory context for intrinsic brain self-organization.

The broader scientific community is already taking notice, with experts heralding the study as a pivotal contribution that bridges cognitive neuroscience, clinical neuropsychology, and neuroengineering. As neurofeedback devices become more accessible through advances in wearable technology and artificial intelligence, insights into spontaneous alpha dynamics are expected to guide the development of next-generation brain-computer interfaces.

Ethical considerations also emerge in this context, especially regarding passive neurofeedback-induced brain state changes. The unintentional modulation of brain waves could raise questions about autonomy and informed consent, particularly if deployed in commercial wellness products or augmentative devices. The authors emphasize the importance of transparency and rigorous ethical frameworks accompanying neurofeedback technology deployment.

Future research directions outlined by Maaz and colleagues will explore the boundary conditions fostering spontaneous alpha upregulation, including individual differences in baseline alpha power, neurochemical modulators like GABA and acetylcholine, and interaction effects with environmental factors such as sensory input and cognitive load. Longitudinal studies will be critical to assess whether spontaneous increases translate into durable cognitive and emotional benefits.

Finally, this seminal work enriches our fundamental understanding of the brain’s capacity for self-regulation and dynamic adaptation. By revealing that alpha power enhancement is not solely a product of volitional control but also an innate neurophysiological process, the study invites us to reconsider how mental states can be cultivated and optimized, potentially heralding a new era of brain health interventions grounded in the brain’s own spontaneous rhythms.

Subject of Research: Neural dynamics of alpha oscillations during neurofeedback sessions

Article Title: Alpha power increases spontaneously during a neurofeedback session

Article References: Maaz, J., Waroquier, L., Dia, A. et al. Alpha power increases spontaneously during a neurofeedback session. Commun Psychol (2026). https://doi.org/10.1038/s44271-026-00431-w

Image Credits: AI Generated