Why the Basal Ganglia Are Important in Neurofeedback Practice

Most of us learned the basal ganglia, a cluster of structures buried beneath the cortex, as the brain's movement machinery. Dopamine falls, movement slows, and tremor appears. That account is not wrong. It is too small. A large 2026 study in Nature argues that Parkinson's disease damages a network that ties movement to arousal, internal organ function, and motivation (Ren et al., 2026).

Neurofeedback professionals have a particular stake in this literature. Every training session you run is an operant learning trial, and the striatum is where operant learning is consolidated (Kerson et al., 2025; Sitaram et al., 2017). These structures are not a distant neurological topic that belongs to movement disorder clinics. They are the machinery your protocol recruits each time a reward tone fires.

A newly described action network sits at the core of the circuit that Parkinson's disease disrupts (Gordon et al., 2023; Ren et al., 2026). The caudate nucleus earns its reward reputation only when the trainee's own activity predicts the outcome (Tricomi et al., 2004).

The Circuit Does Three Jobs

You can run most clinical thinking on three facts. The basal ganglia receive, they gate, and they learn. Everything else is detail you can look up.

First, they receive. The striatum is the input desk. Its upper portion, the dorsal striatum, consists of the caudate nucleus and the putamen. Its lower portion, the ventral striatum, includes the nucleus accumbens. Together they collect signals from most of the cortex and from the thalamus.

Second, they gate. Two output nuclei hold a standing brake on the thalamus: the inner segment of the globus pallidus and one division of the substantia nigra. Releasing that brake lets an action proceed, and tightening it holds the action back. A third structure, the subthalamic nucleus, drives that brake and has become a standard surgical target.

Third, they learn. A separate division of the substantia nigra supplies dopamine to the striatum, and that dopamine teaches the striatum which actions were worth taking. Degeneration of those dopamine neurons is the pathological hallmark of Parkinson's disease. This teaching signal is why the circuit matters to anyone who delivers contingent feedback.

We adapted this graphic from Medico Plasma on Instagram.

Those three jobs explain a pattern you already see in the clinic. The same circuit that gates limb movement also gates working memory updates, eye movement, and the pursuit of reward (Alexander et al., 1986; Middleton & Strick, 2000). So a client referred for attention training may also show low initiative, blunted response to reward, and slowed thinking. Those features are not separate problems that happen to co-occur; they share a substrate with the presenting complaint.

Go and Stop, and Why the Textbook Diagram Is Changing

Two routes leave the striatum. The direct pathway releases the brake and lets a chosen action through. The indirect pathway tightens the brake and holds competing actions back (Albin et al., 1989). Generations of students learned this as a go switch and a stop switch.

Recording from identified cells has complicated that story. Both routes fire together at the moment of movement rather than opposing each other cleanly (Calabresi et al., 2014). In behaving mice, the direct route scales choice in a fairly straight line, while the indirect route exerts a more complicated influence that depends on how strong the input is and what the rest of the network is doing (Li & Jin, 2023).

We adapted this graphic from Medico Plasma on Instagram.

This revision matters more than it first appears. The basal ganglia do not work like a switch that is either on or off, so treatment does not flip a broken circuit back into place. It shifts the balance of a system in which both routes stay active (Calabresi et al., 2014; Li & Jin, 2023). That is also a fair description of what neurofeedback does across 20 or 40 sessions. Describe gains to clients as graded and dose-dependent rather than as a repair.

The Basal Ganglia Work Through Whole Networks

Alexander et al. (1986) proposed that the basal ganglia run parallel loops with the cortex, one for movement, one for eye movement, one for thinking, and one for emotion. Modern imaging has tested that idea in single people rather than in group averages.

Greene et al. (2020) used precision functional mapping, which means scanning one person many times so that their individual network boundaries can be drawn with confidence. They measured resting-state functional connectivity, a method that asks whether two regions rise and fall together while a person lies quietly in the scanner doing nothing. Regions that fluctuate in step are probably working as one circuit. The result was a set of integration zones, subcortical spots where two or more cortical networks converge. In the thalamus, control and movement networks met at the site where deep brain stimulation reliably relieves essential tremor. In the caudate nucleus, the default mode network met several control networks. Put plainly, the caudate is where inward-facing and outward-facing systems talk to each other.

That convergence is the reason a single protocol can move more than one domain. Neurofeedback trains cortical signals, yet the learning that stabilizes those signals passes through subcortical territory shared by several networks (Greene et al., 2020).

Dosenbach et al. (2025) make the same point from the cortical side, renaming the control network the action-mode network because it describes a state rather than a place. A client who sits down, attends to a display, and works to move a signal is holding the brain in exactly that state.

The Homunculus Was Never the Whole Story

For nearly a century we have taught the primary motor cortex as a body map. The famous homunculus drapes a distorted little human across the motor strip, with oversized hands and lips. It is a powerful teaching image and it is not wrong. It is incomplete.

Homunculus.

Gordon et al. (2023) scanned individuals repeatedly and found the map interrupted. Woven between the hand, foot, and mouth patches lies a second system. These in-between regions do not control any single muscle group. They activate during action planning and during whole-body movements, and they connect to the control network and to internal organs. The authors named the system the somato-cognitive action network, or SCAN.

Redrawn SCAN graphic from Nico Dosenbach on X.

A useful way to hold this: the body-part patches are individual instrumentalists, each playing their own part, and SCAN is the conductor. The conductor shapes timing, intensity, and fit to context rather than producing any single note. An independent team using depth electrodes found a comparable non-body-mapped region in the same territory, which strengthens the case (Jensen et al., 2023).

SCAN also solves an old puzzle. Researchers had long noticed that one movement activates several unrelated patches of motor cortex at once, which made no sense if the strip were purely a body map. Those extra activations are network nodes that plan and contextualize the action rather than execute it (Gordon et al., 2023; Ren et al., 2026).

One anatomical detail turns this from theory into a practical problem. SCAN nodes do not sit in one corner of the motor strip. They alternate with body-part zones along the central sulcus, the deep groove running down the side of the brain, at a scale of millimeters (Ren et al., 2026).

So when anyone stimulates or records over motor cortex, the network being engaged depends on millimeters of positioning. Older studies that treated the motor strip as one uniform band were averaging across two different systems.

Why Parkinson's Symptoms Never Fit a Simple Motor Pathway

Parkinson's disease is taught as a movement disorder, and patients do develop tremor, rigidity, and slowness. But the disease also fragments sleep, destabilizes blood pressure control, and impairs mood and thinking (Armstrong & Okun, 2020; Bloem et al., 2021). Those features have never sat comfortably inside a limb-control story.

The clinical oddities are stranger still. Symptoms worsen under stress and time pressure. Familiar music or a steady rhythm can temporarily restore fluent walking (Nutt et al., 2011). In one striking report, patients with advanced disease and severe gait failure escaped an earthquake unaided, and several kept the improvement for months (Bonanni et al., 2010).

If Parkinson's disease damaged only a circuit for moving an arm or a leg, why would anxiety make walking worse, and why would a familiar melody give it back? The SCAN framework supplies a coherent answer. The circuit that degenerates is not a limb-specific pathway but a system that couples movement with arousal, internal regulation, and motivation, which is precisely what SCAN is built to do (McGregor & Nelson, 2019; Ren et al., 2026).

What the Connectivity Data Show

Ren et al. (2026) assembled imaging from 863 people, including patients with Parkinson's disease, healthy controls, and patients with other movement disorders for comparison. They mapped how deep structures connect to the cortical surface.

The first finding reverses an assumption. The deep structures implicated in Parkinson's disease, including the substantia nigra and the subthalamic nucleus, connect preferentially to SCAN nodes rather than to the hand, foot, or mouth zones. When these structures talk to motor cortex, they are addressing the conductor, not the instrumentalists.

The second finding is about specificity. In patients, coupling between those deep structures and SCAN is not merely present; it is abnormally elevated, a pattern called hyperconnectivity. It does not appear in healthy controls, and it is absent in essential tremor, dystonia, and amyotrophic lateral sclerosis. The pattern is therefore not a generic consequence of having a movement disorder, although the authors caution that it may not be unique to Parkinson's disease either.

How well does that coupling track how sick patients are? Weakly, and not in one direction. Stronger coupling went with slightly worse motor scores, but its small associations with cognition, anxiety, and depression all ran the other way, leaving their clinical meaning unresolved. Read the hyperconnectivity finding as a robust group difference rather than as a severity gauge you could apply to a person in front of you.

What Happens When the Circuit Is Treated

A connectivity signature could be a bystander. The investigators therefore asked whether four different treatments move it. Convergence across treatments strengthens causal inference without settling it, and most of these cohorts were small and observational.

Surgery came first. In deep brain stimulation, surgeons implant electrodes in a deep target and deliver continuous pulses. The electrode placements that produce the best outcomes, the so-called sweet spots, show stronger connectivity to SCAN than to body-part zones. Over months, effective stimulation reduces the abnormal coupling, as though it is partly normalizing an over-coupled circuit.

Direct recording tells the same story. Using electrocorticography, which places electrodes on the brain surface to record electrical activity with high precision, the team found that stimulating the subthalamic nucleus produces its strongest cortical responses at SCAN locations. The signal travels along the hyperdirect pathway, a fast one-step connection between cortex and the subthalamic nucleus. Medication fits the pattern as well, since levodopa reduced the abnormal coupling, and the patients whose coupling fell the most improved the most.

Surgical lesioning adds a third line. In MRI-guided focused ultrasound, converging sound beams create a small precise lesion without opening the skull. The authors calculated where the ideal thalamic target would sit based on SCAN connectivity, then measured how far each patient's actual lesion landed from it. Lesions closer to the SCAN-defined spot produced greater motor improvement, and the relationship vanished when body-part connectivity was used as the criterion instead.

The most translatable result used transcranial magnetic stimulation, which delivers magnetic pulses through the scalp without surgery. Thirty-six patients were randomly assigned to receive intermittent theta-burst stimulation at either their own SCAN nodes or their own body-part zones. Both groups received patient-specific functional mapping, so this compared two personalized targets rather than a personalized target against a generic one. The SCAN group improved roughly twice as much, and only SCAN targeting reduced the abnormal coupling. The authors describe this as a small, single-center study needing multicenter replication.

The Caudate Nucleus and the Currency of Reward

Ask most clinicians which structure signals reward, and they will name the nucleus accumbens.

The caudate does something different and, for feedback training, more useful.

Caudate nucleus graphic © Kateryna Kon/Shutterstock.com.

Tricomi et al. (2004) delivered monetary gains and losses three ways: at random, after a warning cue, and after the participant pressed a button. Caudate activity distinguished reward from punishment only when people believed their own response had produced the outcome. Money that simply arrived produced no such response. The caudate's trigger is action contingency, the perceived link between what you did and what happened next.

Two further findings sharpen this. Caudate responses are largest early in learning, while a person is still working out which choices pay (Delgado et al., 2005). And in monkeys, silencing the pathway from orbitofrontal cortex to the caudate caused a reproducible loss of sensitivity to how much a cue was worth (Oyama et al., 2022). Value information reaches behavior through this structure.

The translation to the training room is direct. In that task, caudate reward responses depended heavily on perceived action contingency, supporting a role in assigning outcomes to actions rather than registering reward receipt alone. One monetary paradigm cannot prove the caudate never encodes pleasure, but it does show that contingency is what recruited the structure here. Any protocol that blurs the link between what the trainee's brain did and what the screen did weakens a signal that drives learning. Clear, immediate, and honest contingency is one of the few learning conditions you fully control.

Kerson et al. (2025) reach a compatible conclusion from the learning-theory side. Their update of the 2011 framework treats reinforcement schedules, timing, and signal integrity as determinants of what the nervous system actually learns rather than as technical housekeeping. Conditions that make reinforcement effectively random are the ones most likely to defeat training. Both literatures give the same instruction, which is to protect the contingency.

Effort, Vigor, and Why Apathy Is Not Laziness

Reward is not one thing. Berridge and Robinson (2003) separated liking from wanting and showed that dopamine matters far more for wanting. Schultz (2007) described the brief dopamine burst that signals a reward prediction error, the gap between what was expected and what arrived. That signal reaches the striatum and updates the next choice (Wise, 2004).

Salamone and Correa (2012) added the cost side. Dopamine in the nucleus accumbens, part of the ventral striatum, supports effort-related decision-making. Animals with less accumbens dopamine still prefer the bigger reward. They simply stop paying a high effort price to get it and take the cheaper, smaller option. The failure is in exertion, not in preference.

That distinction changes the conversation you have with clients. Apathy in Parkinson's disease and anhedonia in depression involve a shifted cost-and-benefit calculation rather than a character flaw, although neither syndrome reduces to that single computation. In the training room, this is the client who values the goal, attends every session, and still cannot start home practice. Telling that person their brain is currently overcharging them for effort is a useful metaphor rather than an established individual diagnosis, and it removes the blame that usually rides along with the symptom.

Practical steps follow. Break home practice into short intervals so that starting costs less. Schedule the most demanding activity inside the client's best medication window. Set early goals around starting and later goals around finishing, because the evidence implicates mobilization rather than valuation.

Gating, Thinking, and Session Structure

The same architecture shapes cognition. Frank et al. (2001) modeled the striatum as a gate on working memory. The gate decides when new information gets in and when current contents stay protected. Open it too readily and the person is distractible. Open it too rarely and the person perseverates. Middleton and Strick (2000) confirmed that basal ganglia output reaches thinking regions of the frontal lobe, not only motor cortex.

Robbins (2007) reviewed the same circuits behind shifting and stopping, including the ability to halt an action already underway. Da Cunha et al. (2012) added that these structures help select the actions that express both facts and skills. Grahn et al. (2008) placed deliberate, goal-directed selection in the caudate and automatic execution in the putamen, which supports habit learning.

This is why cognitive complaints in basal ganglia disease differ from those in cortical dementia. Clients often hold the information but cannot retrieve or switch it on demand, and they do better with cues and structure than with open-ended prompts (Robbins, 2007). Build that structure into the session itself. Explicit instructions, a fixed session sequence, and cued transitions substitute for a gate that no longer opens on schedule, and they free capacity that would otherwise compete with learning.

Anhedonia and an Expanded Salience Network

Striatal circuits also carry mood. Reduced ventral striatal activity has long been linked to anhedonia, the loss of pleasure and interest (Haber & Knutson, 2010; Price & Drevets, 2010). Those findings were modest and hard to replicate until scanning methods improved.

A 2024 study in Nature scanned individuals with depression repeatedly over many months. The salience network, the frontostriatal system that flags what deserves attention, occupied nearly twice its usual cortical territory in depression. Its size stayed stable across mood swings, which makes it a trait feature rather than a state marker. The expansion was already present in children scanned at ages 10 and 12 who went on to develop clinically significant depressive symptoms at 13 or 14.

Graphic by Schmimmelpfennig and colleagues (2023).

Within that stable map, one connection did fluctuate. In two deeply sampled individuals, coupling between the nucleus accumbens and the anterior cingulate cortex weakened as anhedonia worsened. It predicted anhedonia about a week ahead in one of those two but not the other, and it did not explain differences in anhedonia across the larger sample of 135. Treat anhedonia as a circuit-level target rather than a residual symptom you wait out, and treat the idea that neurofeedback and behavioral activation engage the same machinery as a working hypothesis rather than a finding.

Three Other Striatal Disorders in Brief

Huntington's disease preferentially and severely degenerates striatal projection neurons, producing chorea, cognitive decline, and psychiatric symptoms that often precede the motor diagnosis. Recent single-cell work explains the long silent period. The inherited repeat is unstable and keeps lengthening inside individual cells, a process called somatic CAG repeat expansion. Below roughly 150 repeats, sampled striatal projection neurons showed no abrupt disruption of gene expression, which is not the same as being unaffected in every respect. Above that length, they lose their identity and die.

Obsessive-compulsive disorder implicates a loop running from orbitofrontal cortex and anterior cingulate through the striatum and thalamus, known as the cortico-striato-thalamo-cortical circuit (Pauls et al., 2014). That loop governs habit and the inhibition of unwanted action, and both fail in the disorder. Recent reviews argue the loop alone cannot explain the clinical variety and point to amygdala, hippocampal, and cerebellar contributions (Jijimon et al., 2026).

Schizophrenia implicates different striatal territory. The associative striatum, spanning medial caudate and dorsal putamen, shows the largest elevation of dopamine production, which revised the older view that psychosis was mainly a limbic dopamine problem (Howes & Kapur, 2009; McCutcheon et al., 2019). Blocking dopamine receptors relieves hallucinations and delusions while leaving negative and cognitive symptoms largely untouched.

Pharmacology has finally moved past that ceiling. Xanomeline combined with trospium chloride became the first schizophrenia treatment approved in the United States that does not act directly on brain dopamine receptors, working instead on muscarinic receptors (Dean, 2024). The general point matters more to you than the specific drug. Striatal circuits can be reached through several transmitter systems, so a client's medication regimen shapes the reinforcement learning capacity your protocol depends on, and medication changes belong in the training record.

Designing Neurofeedback Around the Striatum

Neurofeedback can recruit corticostriatal reinforcement learning within a distributed self-regulation network. It is not only striatal learning. Contemporary models describe parallel systems operating on different timescales, including explicit strategy learning, interoceptive learning, cortical plasticity, attention, and motivation. A major review of neurofeedback mechanisms nonetheless named the basal ganglia as core components of that network, an early study of slow cortical potential training found that successful learning coincided with basal ganglia activation, and animal work shows that blocking a key receptor type in these structures disrupts neurofeedback learning.

If corticostriatal reinforcement learning is part of what your protocol recruits, contingency quality becomes a central learning variable and one of the few you fully control. The caudate responds when the trainee's own activity predicts the outcome and stays quiet when reward arrives independently of behavior. A drifting threshold, artifact that leaks into the reward signal, or a laggy display degrades the intended contingency and may impair learning of the target signal. That is a protocol defect worth eliminating before anything else gets adjusted.

Four protocol decisions follow. Keep feedback latency short and the contingency visible, so the trainee can perceive the link between their state and the reward. Guard signal integrity, because artifact that triggers reward quietly converts a contingent schedule into a partly random one (Kerson et al., 2025). Set thresholds that produce frequent early success, since caudate responses are largest while the contingency is still being learned (Delgado et al., 2005). Move toward automaticity later in a series, because the putamen supports habitual execution once the outcome no longer needs evaluating (Grahn et al., 2008).

Session and intake practices follow the same logic. Use shaping, adjusting thresholds as performance improves, so reinforcement tracks current capability rather than a fixed standard. Track starting and effort separately from performance, since these rest on different striatal mechanisms. Ask about apathy, sleep, and autonomic symptoms at intake, because these are core network features rather than incidental findings. Recommend aerobic exercise to clients with early Parkinson's disease, since six months of home-based aerobic training attenuated off-state motor signs in a double-blind randomized trial.

The targeting lesson needs care. That trial compared personalized SCAN targeting with personalized body-part targeting, so it validates individualized functional targeting for TMS. It does not validate symptom-to-protocol matching in EEG, and it says nothing about qEEG-guided protocol selection.

A hard physical limit follows. The conventional scalp EEG has spatial resolution measured in centimeters rather than millimeters, so C3, Cz, or C4 training cannot selectively engage a SCAN node rather than the body-part region beside it. The present relevance of these findings to scalp-EEG practice is conceptual, not evidence of spatially selective targeting. Say that plainly when a client or referring physician asks. The CRED-nf consensus checklist remains a useful filter when weighing outcome claims, since it asks whether trainees actually learned to regulate the target signal.

What We Still Do Not Know

The basal ganglia are not a motor circuit with cognitive side effects. They receive from the whole cortex, gate what gets through, and learn from what happens next (Alexander et al., 1986; Greene et al., 2020). The caudate sits where inward-facing and outward-facing networks meet, and its reward signal appears only when action and outcome are linked (Tricomi et al., 2004). The ventral striatum sets the price a person will pay in effort (Salamone & Correa, 2012).

Disease follows that architecture. The 2026 authors propose that Parkinson's disease may be better conceptualized and treated as a disorder of the somato-cognitive action network, which would explain why motor, autonomic, sleep, and motivational symptoms arrive together. Huntington's disease is decades of silent DNA expansion inside striatal neurons. Obsessive-compulsive disorder and schizophrenia implicate different striatal territory and different transmitters. Depression shows an enlarged frontostriatal salience network detectable years before symptoms.

The practical message is that your work touches this circuit. Contingency design plausibly engages corticostriatal reinforcement learning whether or not you frame it that way, which makes signal integrity, feedback latency, and threshold policy clinical decisions rather than software preferences. Precision targeting is gaining evidentiary support in stimulation research, though a single small trial in TMS does not transfer to EEG protocol selection. Treat effort, initiation, and anhedonia as measurable outputs, and you will find targets that a purely attentional or purely mood framework hides.

Bringing the Circuit Together

The basal ganglia are not a motor circuit with cognitive side effects. They receive from the whole cortex, gate what gets through, and learn from what happens next. The caudate sits where inward-facing and outward-facing networks meet, and its reward signal appears only when action and outcome are linked. The ventral striatum sets the price a person will pay in effort.

Disease follows that architecture. The 2026 authors propose that Parkinson's disease may be better conceptualized and treated as a disorder of the somato-cognitive action network, which would explain why motor, autonomic, sleep, and motivational symptoms arrive together. Huntington's disease is decades of silent DNA expansion inside striatal neurons. Obsessive-compulsive disorder and schizophrenia implicate different striatal territory and different transmitters. Depression shows an enlarged frontostriatal salience network detectable years before symptoms.

The practical message is that your work touches this circuit. Contingency design plausibly engages corticostriatal reinforcement learning whether or not you frame it that way, which makes signal integrity, feedback latency, and threshold policy clinical decisions rather than software preferences. Precision targeting is gaining evidentiary support in stimulation research, though a single small trial in TMS does not transfer to EEG protocol selection. Treat effort, initiation, and anhedonia as measurable outputs, and you will find targets that a purely attentional or purely mood framework hides.

Five Key Takeaways

1. Neurofeedback can recruit corticostriatal reinforcement learning within a distributed self-regulation network, so these deep structures are working equipment in every session rather than a topic reserved for movement disorder clinics.

2. Caudate reward responses depend heavily on perceived action contingency, which makes signal integrity, short feedback latency, and honest thresholds central learning variables you control.

3. Motor cortex is not one uniform strip. SCAN nodes alternate with body-part zones at a scale of millimetres, and in one small randomized trial personalized SCAN targeting produced roughly twice the motor improvement of personalized body-part targeting.

4. Parkinson's disease is better understood as a network disorder coupling movement with arousal, internal regulation, and motivation, which offers a coherent explanation for why stress, music, and emergencies change how a patient moves.

5. Apathy and anhedonia involve a shifted effort-versus-reward calculation, so screen for them, offer the mechanism as a metaphor that removes blame, and track starting separately from performance.

Glossary

action contingency: the perceived link between a person's own response and the outcome that follows, which is what activates the caudate nucleus.

action-mode network: the cingulo-opercular network renamed for what it does, namely holding the brain in a state of heightened arousal and outward attention during goal-directed behavior.

anhedonia: reduced capacity to experience pleasure or interest in previously rewarding activities.

associative striatum: the medial caudate and dorsal putamen, the striatal territory showing the greatest elevation of dopamine synthesis capacity in schizophrenia.

basal ganglia: a group of subcortical nuclei that convert goals, value, and physiological state into action, comprising the striatum, globus pallidus, subthalamic nucleus, and substantia nigra.

caudate nucleus: the medial component of the striatum, involved in goal-directed action selection and in learning which actions produce which outcomes. central sulcus: the deep groove separating frontal motor regions from parietal sensory regions, along which SCAN nodes and body-part zones alternate.

chorea: involuntary, rapid, irregular movements characteristic of Huntington's disease.

cortico-striato-thalamo-cortical circuit: the orbitofrontal, cingulate, striatal, and thalamic loop long implicated in obsessive-compulsive disorder.

deep brain stimulation: implantation of electrodes that deliver electrical stimulation to a subcortical target such as the subthalamic nucleus, internal globus pallidus, or ventral intermediate thalamus.

direct pathway: the striatal output route projecting to the internal globus pallidus and substantia nigra pars reticulata, which facilitates a selected action.

dorsal striatum: the upper portion of the striatum, made up of the caudate nucleus and the putamen.

effort-related decision-making: the calculation weighing what an action will cost in effort against what it is expected to return, supported by dopamine in the nucleus accumbens.

electrocorticography: recording of electrical activity from electrodes placed directly on the brain surface, offering high spatial and temporal precision.

globus pallidus: a basal ganglia nucleus divided into external and internal segments, the latter serving as a major output to the thalamus.

habit learning: acquisition of behavior that runs automatically without consulting the current value of the outcome, supported by the putamen.

Huntington's disease (HD): an inherited neurodegenerative disorder in which striatal projection neurons preferentially and severely degenerate, producing involuntary movements, cognitive decline, and psychiatric symptoms. hyperconnectivity: abnormally elevated coupling between brain regions, reported between deep structures and SCAN in Parkinson's disease.

hyperdirect pathway: the monosynaptic projection from cortex to the subthalamic nucleus that bypasses the striatum.

indirect pathway: the striatal output route passing through the external globus pallidus and subthalamic nucleus, which suppresses competing actions.

integration zone: a subcortical region where two or more cortical functional networks converge, identified through precision functional mapping.

intermittent theta-burst stimulation: a patterned form of transcranial magnetic stimulation designed to produce longer-lasting changes in cortical excitability.

levodopa: a dopamine precursor medication that remains the mainstay of Parkinson's disease treatment.

MRI-guided focused ultrasound: a minimally invasive technique using focused sound energy under magnetic resonance guidance to create a therapeutic lesion in a deep brain target.

nucleus accumbens: a ventral striatal structure central to incentive motivation, effort allocation, and reinforcement learning.

obsessive-compulsive disorder (OCD): a psychiatric condition marked by intrusive unwanted thoughts and repetitive behaviors the person feels driven to perform, linked to dysfunction in a loop connecting frontal cortex, striatum, and thalamus. Parkinson's disease (PD): a neurodegenerative disorder characterized by progressive degeneration of nigrostriatal dopamine neurons alongside broader distributed pathology, producing tremor, rigidity, and slowness together with sleep, autonomic, mood, and cognitive symptoms.

precision functional mapping: scanning one person repeatedly so that their individual network boundaries can be defined with confidence rather than averaged across a group.

primary motor cortex: the frontal strip historically drawn as a body map, now known to contain SCAN nodes interleaved among the body-part zones.

putamen: the lateral component of the striatum, supporting motor control and habit learning.

resting-state functional connectivity: the temporal correlation of spontaneous blood-oxygen-level-dependent signals between brain regions during rest.

reward prediction error: the difference between an expected and an obtained outcome, signaled by phasic dopamine bursts and used to update future action selection.

salience network: a frontostriatal and insular-cingulate network that identifies which internal and external events warrant attention and action.

schizophrenia: a psychiatric disorder involving hallucinations, delusions, blunted motivation, and cognitive impairment, associated with elevated dopamine production in the associative striatum.

shaping: the practice of progressively adjusting a reinforcement threshold so that reward tracks the trainee's current capability rather than a fixed standard.

slow cortical potential: a low-frequency shift in cortical electrical activity that can be brought under voluntary control through feedback training.

somatic CAG repeat expansion: the progressive lengthening of an inherited trinucleotide repeat within individual cells across the lifespan.

somato-cognitive action network: a system of inter-effector regions inside primary motor cortex that integrates goals, arousal, organ physiology, and whole-body action rather than coding specific movements.

striatal projection neuron: the inhibitory output neuron of the striatum, the cell type that degenerates in Huntington's disease.

striatum: the input structure of the basal ganglia, comprising the dorsal striatum (caudate nucleus and putamen) and the ventral striatum (including the nucleus accumbens).

substantia nigra: a midbrain structure with two divisions: one supplies dopamine to the striatum and degenerates in Parkinson's disease, and the other is a principal basal ganglia output nucleus.

subthalamic nucleus: a small nucleus below the thalamus that excites basal ganglia output nuclei and serves as a standard deep brain stimulation target.

transcranial magnetic stimulation: a noninvasive technique that delivers magnetic pulses through the scalp to change cortical excitability.

ventral striatum: the ventral portion of the striatum, including the nucleus accumbens, central to reward and motivation.

References

Albin, R. L., Young, A. B., & Penney, J. B. (1989). The functional anatomy of basal ganglia disorders. Trends in Neurosciences, 12(10), 366–375. https://doi.org/10.1016/0166-2236(89)90074-X

Alexander, G. E., DeLong, M. R., & Strick, P. L. (1986). Parallel organization of functionally segregated circuits linking basal ganglia and cortex. Annual Review of Neuroscience, 9, 357–381. https://doi.org/10.1146/annurev.ne.09.030186.002041

Armstrong, M. J., & Okun, M. S. (2020). Diagnosis and treatment of Parkinson disease: A review. JAMA, 323(6), 548–560. https://doi.org/10.1001/jama.2019.22360

Berridge, K. C., & Robinson, T. E. (2003). Parsing reward. Trends in Neurosciences, 26(9), 507–513. https://doi.org/10.1016/S0166-2236(03)00233-9

Bloem, B. R., Okun, M. S., & Klein, C. (2021). Parkinson's disease. The Lancet, 397(10291), 2284–2303. https://doi.org/10.1016/S0140-6736(21)00218-X

Bonanni, L., Thomas, A., & Onofrj, M. (2010). Paradoxical kinesia in parkinsonian patients surviving earthquake. Movement Disorders, 25(9), 1302–1304. https://doi.org/10.1002/mds.23075 Calabresi, P., Picconi, B., Tozzi, A., Ghiglieri, V., & Di Filippo, M. (2014). Direct and indirect pathways of basal ganglia: A critical reappraisal. Nature Neuroscience, 17(8), 1022–1030. https://doi.org/10.1038/nn.3743

Da Cunha, C., Gomez-A, A., & Blaha, C. D. (2012). The role of the basal ganglia in motivated behavior. Reviews in the Neurosciences, 23(5–6), 747–767. https://doi.org/10.1515/revneuro-2012-0063

Dean, B. (2024). IUPHAR review on muscarinic M1 and M4 receptors as drug treatment targets relevant to the molecular pathology of schizophrenia. Pharmacological Research, 210, 107510. https://doi.org/10.1016/j.phrs.2024.107510

Delgado, M. R., Miller, M. M., Inati, S., & Phelps, E. A. (2005). An fMRI study of reward-related probability learning. NeuroImage, 24(3), 862–873. https://doi.org/10.1016/j.neuroimage.2004.10.002

Dosenbach, N. U. F., Raichle, M. E., & Gordon, E. M. (2025). The brain's action-mode network. Nature Reviews Neuroscience, 26(3), 158–168. https://doi.org/10.1038/s41583-024-00895-x

Frank, M. J., Loughry, B., & O'Reilly, R. C. (2001). Interactions between frontal cortex and basal ganglia in working memory: A computational model. Cognitive, Affective & Behavioral Neuroscience, 1(2), 137–160. https://doi.org/10.3758/cabn.1.2.137

Gordon, E. M., Chauvin, R. J., Van, A. N., Rajesh, A., Nielsen, A., Newbold, D. J., Lynch, C. J., Seider, N. A., Krimmel, S. R., Scheidter, K. M., Monk, J., Miller, R. L., Metoki, A., Montez, D. F., Zheng, A., Elbau, I., Madison, T., Nishino, T., Myers, M. J., . . . Dosenbach, N. U. F. (2023). A somato-cognitive action network alternates with effector regions in motor cortex. Nature, 617(7960), 351–359. https://doi.org/10.1038/s41586-023-05964-2

Grahn, J. A., Parkinson, J. A., & Owen, A. M. (2008). The cognitive functions of the caudate nucleus. Progress in Neurobiology, 86(3), 141–155. https://doi.org/10.1016/j.pneurobio.2008.09.004

Greene, D. J., Marek, S., Gordon, E. M., Siegel, J. S., Gratton, C., Laumann, T. O., Gilmore, A. W., Berg, J. J., Nguyen, A. L., Dierker, D., Van, A. N., Ortega, M., Newbold, D. J., Hampton, J. M., Nielsen, A. N., McDermott, K. B., Roland, J. L., Norris, S. A., Nelson, S. M., . . . Dosenbach, N. U. F. (2020). Integrative and network-specific connectivity of the basal ganglia and thalamus defined in individuals. Neuron, 105(4), 742–758.e6. https://doi.org/10.1016/j.neuron.2019.11.012

Haber, S. N., & Knutson, B. (2010). The reward circuit: Linking primate anatomy and human imaging. Neuropsychopharmacology, 35(1), 4–26. https://doi.org/10.1038/npp.2009.129

Handsaker, R. E., Kashin, S., Reed, N. M., Tan, S., Lee, W.-S., McDonald, T. M., Morris, K., Kamitaki, N., Mullally, C. D., Morakabati, N. R., Goldman, M., Lind, G., Kohli, R., Lawton, E., Hogan, M., Ichihara, K., Berretta, S., & McCarroll, S. A. (2025). Long somatic DNA-repeat expansion drives neurodegeneration in Huntington's disease. Cell, 188(3), 623–639.e19. https://doi.org/10.1016/j.cell.2024.11.038

Howes, O. D., & Kapur, S. (2009). The dopamine hypothesis of schizophrenia: Version III. The final common pathway. Schizophrenia Bulletin, 35(3), 549–562. https://doi.org/10.1093/schbul/sbp006

Jensen, M. A., Huang, H., Valencia, G. O., Klassen, B. T., van den Boom, M. A., Kaufmann, T. J., Schalk, G., Brunner, P., Worrell, G. A., Hermes, D., & Miller, K. J. (2023). A motor association area in the depths of the central sulcus. Nature Neuroscience, 26(7), 1165–1169. https://doi.org/10.1038/s41593-023-01346-z

Jijimon, F., Mehak, S. F., Mohammed, Z., Praharaj, S. K., & Gangadharan, G. (2026). Rewiring the OCD brain: Insights beyond cortico-striatal networks. Neurobiology of Disease, 218, 107216. https://doi.org/10.1016/j.nbd.2025.107216

Kerson, C., Sherlin, L. H., & Davelaar, E. J. (2025). Neurofeedback, biofeedback, and basic learning theory: Revisiting the 2011 conceptual framework. Applied Psychophysiology and Biofeedback. Advance online publication. https://doi.org/10.1007/s10484-025-09756-4

Lanciego, J. L., Luquin, N., & Obeso, J. A. (2012). Functional neuroanatomy of the basal ganglia. Cold Spring Harbor Perspectives in Medicine, 2(12), a009621. https://doi.org/10.1101/cshperspect.a009621

Li, H., & Jin, X. (2023). Multiple dynamic interactions from basal ganglia direct and indirect pathways mediate action selection. eLife, 12, RP87644. https://doi.org/10.7554/eLife.87644

Lynch, C. J., Elbau, I. G., Ng, T., Ayaz, A., Zhu, S., Wolk, D., Manfredi, N., Johnson, M., Chang, M., Chou, J., Summerville, I., Ho, C., Lueckel, M., Bukhari, H., Buchanan, D., Victoria, L. W., Solomonov, N., Goldwaser, E., Moia, S., . . . Liston, C. (2024). Frontostriatal salience network expansion in individuals in depression. Nature, 633(8030), 624–633. https://doi.org/10.1038/s41586-024-07805-2

McCutcheon, R. A., Abi-Dargham, A., & Howes, O. D. (2019). Schizophrenia, dopamine and the striatum: From biology to symptoms. Trends in Neurosciences, 42(3), 205–220. https://doi.org/10.1016/j.tins.2018.12.004

Middleton, F. A., & Strick, P. L. (2000). Basal ganglia and cerebellar loops: Motor and cognitive circuits. Brain Research Reviews, 31(2–3), 236–250. https://doi.org/10.1016/S0165-0173(99)00040-5

Nutt, J. G., Bloem, B. R., Giladi, N., Hallett, M., Horak, F. B., & Nieuwboer, A. (2011). Freezing of gait: Moving forward on a mysterious clinical phenomenon. The Lancet Neurology, 10(8), 734–744. https://doi.org/10.1016/S1474-4422(11)70143-0 Oyama, K., Hori, Y., Mimura, K., Nagai, Y., Eldridge, M. A. G., Saunders, R. C., Miyakawa, N., Hirabayashi, T., Hori, Y., Inoue, K., Suhara, T., Takada, M., Higuchi, M., Richmond, B. J., & Minamimoto, T. (2022). Chemogenetic disconnection between the orbitofrontal cortex and the rostromedial caudate nucleus disrupts motivational control of goal-directed action. The Journal of Neuroscience, 42(32), 6267–6275. https://doi.org/10.1523/JNEUROSCI.0229-22.2022

Pauls, D. L., Abramovitch, A., Rauch, S. L., & Geller, D. A. (2014). Obsessive-compulsive disorder: An integrative genetic and neurobiological perspective. Nature Reviews Neuroscience, 15(6), 410–424. https://doi.org/10.1038/nrn3746

Price, J. L., & Drevets, W. C. (2010). Neurocircuitry of mood disorders. Neuropsychopharmacology, 35(1), 192–216. https://doi.org/10.1038/npp.2009.104

Ren, J., Zhang, W., Dahmani, L., Gordon, E. M., Li, S., Zhou, Y., Long, Y., Huang, J., Zhu, Y., Guo, N., Jiang, C., Zhang, F., Bai, Y., Wei, W., Wu, Y., Bush, A., Vissani, M.,

Wei, L., Oehrn, C. R., . . . Liu, H. (2026). Parkinson's disease as a somato-cognitive action network disorder. Nature, 651(8107), 1030–1038. https://doi.org/10.1038/s41586-025-10059-1

Robbins, T. W. (2007). Shifting and stopping: Fronto-striatal substrates, neurochemical modulation and clinical implications. Philosophical Transactions of the Royal Society B: Biological Sciences, 362(1481), 917–932. https://doi.org/10.1098/rstb.2007.2097

Ros, T., Enriquez-Geppert, S., Zotev, V., Young, K. D., Wood, G., Whitfield-Gabrieli, S., Wan, F., Vuilleumier, P., Vialatte, F., Van De Ville, D., Todder, D., Surmeli, T., Sulzer, J. S., Strehl, U., Sterman, M. B., Steiner, N. J., Sorger, B., Soekadar, S. R., Sitaram, R., . . . Thibault, R. T. (2020). Consensus on the reporting and experimental design of clinical and cognitive-behavioural neurofeedback studies (CRED-nf checklist). Brain, 143(6), 1674–1685. https://doi.org/10.1093/brain/awaa009

Ross, C. A., & Tabrizi, S. J. (2011). Huntington's disease: From molecular pathogenesis to clinical treatment. The Lancet Neurology, 10(1), 83–98. https://doi.org/10.1016/S1474-4422(10)70245-3

Salamone, J. D., & Correa, M. (2012). The mysterious motivational functions of mesolimbic dopamine. Neuron, 76(3), 470–485. https://doi.org/10.1016/j.neuron.2012.10.021

Schimmelpfennig, J., Topczewski, J., Zajkowski, W., & Jankowiak-Siuda, K. (2023). The role of the salience network in cognitive and affective deficits. Frontiers in Human Neuroscience, 17:1133367. https://doi.org/10.3389/fnhum.2023.1133367

Schultz, W. (2007). Multiple dopamine functions at different time courses. Annual Review of Neuroscience, 30, 259–288. https://doi.org/10.1146/annurev.neuro.28.061604.135722

Sherlin, L. H., Arns, M., Lubar, J., Heinrich, H., Kerson, C., Strehl, U., & Sterman, M. B. (2011). Neurofeedback and basic learning theory: Implications for research and practice. Journal of Neurotherapy, 15(4), 292–304. https://doi.org/10.1080/10874208.2011.623089

Sitaram, R., Ros, T., Stoeckel, L., Haller, S., Scharnowski, F., Lewis-Peacock, J., Weiskopf, N., Blefari, M. L., Rana, M., Oblak, E., Birbaumer, N., & Sulzer, J. (2017). Closed-loop brain training: The science of neurofeedback. Nature Reviews Neuroscience, 18(2), 86–100. https://doi.org/10.1038/nrn.2016.164

Tricomi, E. M., Delgado, M. R., & Fiez, J. A. (2004). Modulation of caudate activity by action contingency. Neuron, 41(2), 281–292. https://doi.org/10.1016/S0896-6273(03)00848-1

U.S. Food and Drug Administration. (2024, September 26). FDA approves drug with new mechanism of action for treatment of schizophrenia [Press release]. https://www.fda.gov/news-events/press-announcements/fda-approves-drug-new-mechanism-action-treatment-schizophrenia

van der Kolk, N. M., de Vries, N. M., Kessels, R. P. C., Joosten, H., Zwinderman, A. H., Post, B., & Bloem, B. R. (2019). Effectiveness of home-based and remotely supervised aerobic exercise in Parkinson's disease: A double-blind, randomised controlled trial. The Lancet Neurology, 18(11), 998–1008. https://doi.org/10.1016/S1474-4422(19)30285-6

Wise, R. A. (2004). Dopamine, learning and motivation. Nature Reviews Neuroscience, 5(6), 483–494. https://doi.org/10.1038/nrn1406

About the Author

Fred Shaffer earned his PhD in Psychology from Oklahoma State University. He earned BCIA certifications in Biofeedback and HRV Biofeedback. Fred is an Allen Fellow and Professor of Psychology at Truman State University, where he has taught for 50 years. He is a Biological Psychologist who consults and lectures in heart rate variability biofeedback, Physiological Psychology, and Psychopharmacology. Fred helped to edit Evidence-Based Practice in Biofeedback and Neurofeedback (3rd and 4th eds.) and helps to maintain BCIA's certification programs. He is a recipient of AAPB's Distinguished Scientist Award and BFE's Lifetime Impact Award.

Support Our Friends

The Basal Ganglia Are More Than a Motor Circuit_ What Deep Nuclei Mean for Neurofeedback P.

Download • 16.42MB