The mood swings, impulsivity, grandiosity, and anhedonia (loss of the capacity to feel pleasure or interest in activities) that define bipolar disorder do not come from a single broken brain region or a simple chemical imbalance. The contemporary understanding of bipolar disorder operates at multiple levels simultaneously: circuit-level dysfunction, cellular pathology, and genetic architecture. The core model is one of failed regulation: hyperactive "bottom-up" limbic and reward-processing structures are inadequately controlled by hypoactive "top-down" cortical regions (Oglodek et al., 2025, Int J Mol Sci).

Understanding this circuit model matters for three practical reasons. First, it explains why bipolar disorder produces such different clinical pictures at different times in the same person, from the expansive grandiosity of mania to the paralytic anhedonia of depression. Second, it helps explain why certain treatments work (lithium restores prefrontal volume; antipsychotics dampen limbic hyperactivation) and why others are dangerous (antidepressant monotherapy can further destabilize an already unregulated circuit). Third, it gives you a framework for explaining the illness to patients and families that is both accurate and destigmatizing: this is a wiring problem, not a willpower problem.

2.1 The Prefrontal Cortex: When the Brakes Fail

The prefrontal cortex (PFC) serves as the brain's executive control center. Its ventrolateral subdivision (vlPFC) handles behavioral response inhibition and cognitive reappraisal of emotions. Its dorsolateral subdivision (dlPFC) manages working memory and sustained goal-directed attention.

In the bipolar brain, both regions show structural thinning and pervasive functional hypoactivation. Functional neuroimaging meta-analyses consistently demonstrate that patients with bipolar disorder under-activate the right vlPFC during executive cognitive tasks and emotional processing (Chen et al., 2011, Bipolar Disord). During acute manic states, there is a marked failure to engage the vlPFC during response inhibition tasks (Go/No-Go, Stroop). This specific regional hypoactivation maps directly onto the clinical hallmarks of mania: impulsivity, disinhibition, and risk-taking.

Two findings make this clinically important beyond the acute episode. First, even during euthymia (clinical remission), patients show significantly reduced vlPFC activation when viewing emotional faces. This was measured using functional MRI (fMRI), a neuroimaging technique that tracks blood flow changes to identify which brain regions are active during specific tasks or at rest. The finding indicates a persistent, trait-like deficit in automatic emotion regulation that does not resolve between episodes (Chen et al., 2011, Bipolar Disord).

Second, meta-analytic data show a negative correlation between vlPFC activation and the number of previous manic episodes, suggesting progressive degradation of cortical control over time *(Brooks et al., 2009, Am J Psychiatry, 166(4), 461-469).* This is one piece of evidence supporting the neuroprogression hypothesis (the concept that the illness worsens biologically over time, with cumulative brain changes), discussed in Section 4.

The dlPFC shows a complementary pattern. In euthymic individuals, resting-state fMRI reveals abnormally weak functional connectivity within the frontal-limbic system. During cognitive tasks and mania, the dlPFC shows general hypoactivation. During sadness induction paradigms, however, patients with bipolar depression occasionally show dlPFC hyperactivation, which researchers interpret not as intact functioning but as highly inefficient compensatory effort: the cortex straining and failing to regulate overwhelming negative emotion from the limbic system *(Townsend et al., 2010, Biol Psychiatry, 68(5), 428-435).*

In plain language for families:

The prefrontal cortex is the brain's control center for impulses, decision-making, and emotional regulation. In bipolar disorder, this area is structurally thinner and less active than usual, especially during mood episodes. During mania, the prefrontal cortex essentially cannot keep up with the emotional and motivational signals flooding in from deeper brain regions. This is why a person in a manic episode may make decisions that seem completely out of character: the part of the brain that would normally say "wait, think about this" is functionally offline. This deficit does not fully correct between episodes, which is part of why ongoing treatment matters even when someone feels well.

2.2 The Amygdala and Anterior Cingulate Cortex: The Accelerator Without a Brake

If the prefrontal cortex represents the brakes, the limbic system acts as an overly sensitive accelerator. You have likely seen this in practice: the patient whose emotional response to a minor interpersonal slight escalates to rage within seconds, or whose tearfulness during a therapy session feels disproportionate to the content being discussed. The amygdala, a deep temporal lobe structure central to threat detection, emotional salience processing, and generating emotional responses, shows consistent structural and functional hyperactivation in bipolar disorder (Townsend & Altshuler, 2012, Psychiatr Clin North Am).

Across both manic and depressive episodes, the amygdala shows exaggerated responses to emotional stimuli (fearful or angry faces, for example). This amygdalar hyperreactivity provides the biological substrate for the intense emotional volatility, irritability, and affective lability that clinicians observe across mood states (Oglodek et al., 2025, Int J Mol Sci).

The critical finding is what happens between the prefrontal cortex and the amygdala at the circuit level. In a healthy brain, as vlPFC activity increases, amygdala activity decreases: this is successful emotion regulation. In the bipolar brain, this inverse relationship is decoupled. Advanced neuroimaging shows a marked decrease in the normally negative functional connectivity between the vlPFC and the amygdala *(Foland et al., 2008, Am J Psychiatry, 165(4), 440-449).* The regulatory circuit is physically and functionally disconnected.

The anterior cingulate cortex (ACC) detects emotional conflict, integrates cognitive and affective information, and mediates autonomic responses. In bipolar disorder, it shows its own pathological alterations. During manic phases, localized neuroinflammatory processes, specifically microglial activation, are pronounced in the ACC. These molecular changes disrupt the ACC's ability to filter affective information, contributing to emotional flooding and cognitive distractibility *(Haarman et al., 2014, JAMA Psychiatry, 71(4), 425-433).*

In plain language for families:

The amygdala is the brain's alarm system. In bipolar disorder, it is set too sensitive: it reacts too strongly and too quickly to emotional situations, whether positive or negative. At the same time, the connection between the alarm system and the control center (the prefrontal cortex) is weakened. In a healthy brain, the control center can calm the alarm down. In bipolar disorder, the alarm rings and the control center cannot reach the switch. This is why your loved one may react to a small frustration with intense anger, or to a minor disappointment with overwhelming sadness. It is not an overreaction by choice; it is an overreaction by wiring.

2.3 The Ventral Striatum: A Reward Circuit That Swings Between Extremes

The ventral striatum, including the nucleus accumbens, is the central hub of the brain's mesolimbic dopamine reward and motivation circuitry (the network connecting deep midbrain dopamine-producing areas to the frontal cortex, responsible for driving motivation, pleasure, and the feeling that something is "worth doing"). Bipolar disorder is uniquely characterized by cyclical fluctuations in goal-directed behavior, motivation, and energy that trace directly to states of ventral striatum dysregulation (Alloy et al., 2016, J Abnorm Psychol).

Research using fMRI reward-anticipation paradigms (the Monetary Incentive Delay task) reveals a consistent neurobiological pattern:

During mania and hypomania, patients with bipolar disorder show abnormally elevated, often left-lateralized, activation of the ventral striatum and orbitofrontal cortex (the region of the prefrontal cortex involved in evaluating rewards and making value-based decisions) during reward anticipation *(Caseras et al., 2013, Am J Psychiatry, 170(5), 533-541).* One study found that patients with Bipolar II demonstrate significantly greater ventral striatal activity during reward anticipation than both Bipolar I patients and healthy controls (Caseras et al., 2013) This excessive reward-circuit activation provides the neurobiological fuel for the extreme approach-motivated behavior seen in hypomania and mania: the ambitious projects, spending sprees, sexual indiscretions, and persistent goal-pursuit. Structural MRI adds a complementary finding: BD-II patients have significantly larger left putamen volumes (the putamen is a structure within the striatum that helps translate motivation into action), which correlate positively with heightened ventral striatal functional activity, establishing a structure-function relationship in bipolar reward pathology (Caseras et al., 2013)

During bipolar depression, the reward circuitry enters a refractory, depleted state, showing marked hypo-responsiveness to normally rewarding stimuli. This neurobiological shutdown shows up clinically as anhedonia (loss of the ability to experience pleasure), avolition (loss of motivation), and psychomotor retardation (a visible slowing of physical movement and speech, as though the body is moving through molasses) *(Redlich et al., 2015, JAMA Psychiatry, 72(11), 1188-1196).*

This polarity in reward-circuit function, hypersensitive during mania and hypo-responsive during depression, is one of the neurobiological signatures that distinguishes bipolar depression from unipolar depression. In unipolar MDD, the ventral striatum shows persistent hypo-responsiveness without the episodic hypersensitivity. This difference has treatment implications that Section 5 develops further.

In plain language for families:

The brain has a built-in reward and motivation system. In bipolar disorder, this system swings between two extremes. During mania or hypomania, it is turned up too high: everything feels exciting, achievable, and worth pursuing right now, which is why your loved one may take on enormous projects, spend recklessly, or make impulsive decisions that seem driven by an almost unstoppable internal engine. During depression, the same system shuts down: nothing feels rewarding, nothing feels worth doing, and the motivation to engage with life drops dramatically. Understanding this helps explain why bipolar depression feels qualitatively different from ordinary sadness. It is not just feeling sad; it is the reward system going offline.

2.4 Two Emerging Models: Desynchrony Cascade and Nexus

Recent conceptual frameworks have attempted to integrate the circuit-level findings with cellular and molecular pathology.

The Desynchrony Cascade Model proposes a sequential, feed-forward trajectory beginning at the genomic level (McIntyre et al., 2020, Lancet). The model works like a chain of falling dominoes. Polygenic risk, concentrated in genes governing synaptic function, calcium channels (CACNA1C), and circadian rhythms, predisposes an individual to early circadian disruptions. Those circadian disruptions trigger mitochondrial dysfunction and cellular stress responses. The resulting metabolic damage provokes HPA axis hyperactivity and systemic immune activation. This neuroinflammatory and oxidative state ultimately produces the large-scale network dysfunction that manifests as mood and cognitive symptoms.

The Nexus Model offers a non-linear, bidirectional alternative. Rather than a sequential cascade, it proposes that the circadian, mitochondrial-calcium, and HPA-immune systems are coupled through interdependent feedback loops. A perturbation at any single node, whether from a psychosocial stressor, an immune challenge, or a single night of severe sleep disruption, propagates dynamically across the entire system (Vieta et al., 2018, Nat Rev Dis Primers, 4, 18008) This framework is more effective at explaining the heterogeneous, variable, and often unpredictable clinical presentations of bipolar disorder.

Both models converge on a critical clinical implication: the systems involved in bipolar disorder are multiply interconnected, meaning that interventions targeting one node (circadian rhythm stabilization, immune modulation, calcium channel function) may have cascade effects across the entire system. This has direct relevance for the chronobiological interventions discussed in Section 4 and the treatment approaches covered in Modules 3 and 4.

2.5 Cellular Pathology: Calcium, Mitochondria, and Neuroinflammation

At the cellular level, three interconnected pathological processes underlie the circuit-level dysfunction.

Calcium signaling dysregulation. Induced pluripotent stem cell (iPSC) studies using patient-derived neurons demonstrate that bipolar neurons are inherently hyperexcitable compared to healthy controls, driven by disturbances in intracellular calcium homeostasis linked to risk-allele variants in the CACNA1C gene (Harrison et al., 2018, Trends Neurosci). This calcium dysregulation is coupled with impaired oxidative phosphorylation and excessive production of reactive oxygen species (ROS), compromising mitochondrial buffering capacity and creating chronic oxidative stress (Oglodek et al., 2025, Int J Mol Sci).

Neuroinflammation and glial dysfunction. Two types of immune-related cells play central roles in bipolar pathology. Pro-inflammatory cytokines (TNF-alpha, IL-6, IL-1beta) are chemical messengers that activate the immune system and promote inflammation. When persistently elevated in the brain, they damage neurons. Microglia are the brain's resident immune cells. When chronically activated, they shift from protective surveillance to a state that damages the tissue they are supposed to protect. Both cytokine elevation and microglial activation are consistently observed in the blood and central nervous system of patients with bipolar disorder *(Rosenblat & McIntyre, 2017, Psychiatr Clin North Am, 40(1), 125-137).*

In the prefrontal cortex, activated microglia release neurotoxic levels of nitric oxide and glutamate. Glutamate is the brain's primary excitatory neurotransmitter, and in excessive amounts, it overstimulates neurons to the point of damage, a process called excitotoxicity. Normally, astrocytes (support cells between neurons) clean up excess glutamate from the spaces between neurons. Postmortem studies in bipolar disorder reveal significantly reduced expression of critical astrocytic markers *(Rajkowska & Stockmeier, 2013, Curr Drug Targets, 14(11), 1225-1236).* With astrocytes failing to do their job, glutamate accumulates, overstimulating and eventually killing local neurons and destabilizing the fronto-limbic networks described above.

Why this matters clinically: These cellular processes explain, in part, why certain treatments work. To understand the mechanism, two enzymes matter.

GSK-3beta is an enzyme involved in cell survival and gene regulation. Lithium inhibits it. Histone deacetylases (HDACs) are enzymes that control how tightly DNA is wound, which determines whether specific genes can be "read" by the cell. When HDACs are overactive, genes responsible for brain repair and resilience get silenced. Lithium inhibits these as well.

The clinical relevance: preclinical evidence suggests that by inhibiting HDACs, lithium may reopen access to protective genes, potentially counteracting some of the epigenetic modifications associated with the kindling process described in Section 4. Epigenetic modifications are changes to how genes are expressed without altering the DNA sequence itself, often caused by environmental stress or repeated illness episodes. Direct clinical evidence for this specific pathway remains limited. Lithium also increases prefrontal gray matter volume, normalizes calcium excitability, and provides measurable neuroprotection in neuroimaging studies (Harrison et al., 2018, Trends Neurosci). Valproate similarly acts as an HDAC inhibitor in preclinical models. These mechanisms suggest that mood stabilizers may function not only as symptom managers but as potential neuroprotective agents, an area of active research. Module 3 develops the pharmacological implications in full detail.

PsychHQ

The Neurobiology of Mood Dysregulation