Biological and Cognitive Mechanisms of Depression: What Research Shows

Depression is sometimes explained as a chemical imbalance, a faulty stress response, an inflamed brain, a pattern of negative thinking, or a problem in particular brain circuits. Each explanation points toward a real area of research. None is a complete account of depression.

The strongest current model is multifactorial. Genetic differences, life circumstances, stress systems, sleep, physical health, reward learning, cognition, and brain function can interact in different ways. These interactions can influence vulnerability, trigger an episode, maintain symptoms, or change as a consequence of being depressed.

This distinction matters. A biological difference found across a research group does not necessarily identify what caused one person’s depression. It may not even be present in that person.

🧬 Depression has a genetic component, but there is no single “depression gene.”
🧠 Brain and neurotransmitter systems are involved, but depression is not established as a simple chemical deficiency.
🌍 Stress and adversity matter, but they do not produce the same response in everyone.
🎯 Reward and cognitive processes can change, but these changes are neither chosen nor universal.
🛏️ Sleep and circadian disruption can contribute to and result from depression.
🩺 Inflammation may be relevant for some people, but it is not a universal explanation or diagnostic test.

🔬 The Short Answer: There Is No Single Depression Mechanism

Depression is a syndrome: a recognizable combination of emotional, cognitive, physical, and motivational symptoms. It is currently identified through symptoms, their duration and course, functional change, and clinical context—not through one laboratory result or scan.

People who meet the same diagnostic criteria may have arrived there through different routes. One person’s episode may follow prolonged stress and sleep loss. Another person may have a strong familial vulnerability and recurrent episodes without one obvious trigger. Someone else may develop depression in the context of pain, illness, isolation, medication effects, hormonal changes, or sustained environmental demands.

Several processes can then begin reinforcing one another:

Poor sleep can reduce emotional and cognitive flexibility.
Reduced energy can make activity and social contact harder.
Fewer rewarding experiences can deepen anhedonia.
Rumination can prolong distress and interfere with recovery.
Practical problems may accumulate as functioning falls.
The resulting stress can further disrupt sleep, motivation, and mood.

This is why researchers increasingly study interacting systems and feedback loops instead of looking for one broken component.

It also helps to separate three questions:

1. Does a factor increase the probability of developing depression?
2. Does it help maintain symptoms once depression has started?
3. Is it a consequence of living through a depressive episode?

A single study—especially a cross-sectional comparison of depressed and non-depressed groups—often cannot distinguish between these possibilities.

🧬 Genetics: Vulnerability, Not Destiny

Family and twin research indicates that depression is moderately heritable. In population research, genetic differences help explain some variation in who develops depression. Substantial variation remains connected to individual environments, experiences, health, and measurement.

Heritability is a population statistic. It does not mean that a particular percentage of one person’s depression was genetically caused. It also does not show that an outcome is fixed or untreatable.

Depression is highly polygenic. This means that many genetic variants may each make extremely small contributions to probability. Researchers have identified numerous associated regions, but no individual variant explains depression on its own.

Genetic findings are also affected by how depression is defined. Studies based on brief self-reports, medical records, or carefully assessed recurrent major depression may not identify exactly the same genetic patterns.

At present:

🧬 There is no single genetic test that can confirm depression.
🧬 A polygenic risk score cannot determine whether a particular person will become depressed.
🧬 Family history can be clinically relevant without making depression inevitable.
🧬 Genetic vulnerability can interact with stress, development, physical health, sleep, and social conditions.

Genes influence probabilities within environments. They do not write an unavoidable personal forecast.

🌍 Stress, Adversity, and Person–Environment Effects

Stressful events and chronic adversity are among the most consistent predictors of depression, but stress is not one uniform exposure.

Bereavement, poverty, discrimination, unsafe housing, relationship conflict, caregiving pressure, chronic pain, job insecurity, social exclusion, and relentless daily demands can affect people through different combinations of psychological, physiological, and practical pathways.

Researchers often study the hypothalamic–pituitary–adrenal axis, or HPA axis, which helps coordinate responses to challenge. Some depression studies find differences in cortisol patterns or stress reactivity. Results vary with time of measurement, age, symptom profile, trauma history, medication, physical health, and other factors.

Depression therefore cannot be described accurately as a state in which everyone’s cortisol is simply “too high.” Some groups show elevated activity, some show blunted responses, and many findings overlap with those of people who are not depressed.

Stress responses are also bidirectional. Prolonged strain may increase vulnerability to depression, while depression itself can create additional stress through impaired functioning, financial problems, conflict, withdrawal, and interrupted routines.

A person–environment model asks more useful questions than “Is this biological or situational?”

It asks:

What has the person been carrying?
Which demands are ongoing?
What recovery opportunities are missing?
Which biological and psychological systems may be affected?
What can be changed, supported, treated, or accommodated?

Social conditions are not somehow less real because they affect biology. Biology is one of the ways lived experience is embodied.

🧠 Brain Chemistry and Brain Networks

Neurotransmitters are involved—but “chemical imbalance” is too simple

Serotonin, dopamine, noradrenaline, glutamate, GABA, and other signalling systems participate in mood, learning, attention, sleep, motivation, and stress regulation. That does not mean depression is caused by a measurable shortage of one chemical.

The traditional claim that depression results from abnormally low serotonin has not been supported as a general causal explanation. There is no routine test showing that a depressed person has a serotonin deficiency that treatment must correct.

This finding does not mean antidepressants cannot help. A treatment’s effects do not have to reveal the original cause of a condition. Antidepressants can influence neurotransmission and downstream adaptation across multiple systems without correcting a proven pre-existing chemical deficit.

Medication response also varies. A drug helping one person does not prove that their depression was caused by the neurotransmitter system the drug affects. A drug not helping does not prove that system was irrelevant.

For evidence-based discussion of medication alongside therapy and practical adaptations, see Therapy, Medication, and Adaptations for Depression in Autistic and ADHD Adults.

Brain scans show research patterns, not a depression signature

Imaging studies have reported average differences involving networks associated with reward, self-referential processing, attention, emotion, memory, and cognitive control. Structural studies have also identified small average differences in some brain regions.

These are group-level findings with substantial overlap between depressed and non-depressed people. Results vary across samples, methods, symptom profiles, episode histories, medication exposure, and co-occurring conditions.

A scan usually cannot tell whether a difference:

🧠 existed before the episode
🧠 contributed to vulnerability
🧠 developed during depression
🧠 reflects treatment or another condition
🧠 has any clinical meaning for that individual

There is currently no standard brain scan that diagnoses depression, identifies its cause, or reliably selects the best treatment for an individual. Imaging remains valuable for research and for investigating other medical concerns when clinically indicated.

It is also misleading to describe every research difference as “brain damage.” The brain is dynamic. Activity and connectivity can vary with learning, sleep, stress, medication, illness, and current mental state.

🎯 Reward Learning, Motivation, and Anhedonia

Depression often changes how accessible, worthwhile, or motivating rewards feel. Researchers refer to several partially distinct reward processes:

Anticipating something positive
Judging whether it is worth the effort
Initiating action to obtain it
Experiencing pleasure when it happens
Learning from rewarding outcomes

Meta-analytic research finds small-to-moderate average differences in some reward-processing tasks among depressed groups. The size and consistency of those differences vary by the process being measured.

This does not establish a universal “broken reward circuit” or dopamine shortage. It also does not mean a person has stopped caring.

Someone may still value an activity intellectually while being unable to anticipate its emotional payoff. They may enjoy an experience once it begins but be unable to generate the motivation required to start. High effort costs, fatigue, executive dysfunction, anxiety, and reduced reward expectation can all look like “not trying” from the outside.

That distinction has practical value. Support may need to reduce start-up effort, make rewards more immediate, or help activity happen before motivation returns. For a fuller explanation, see Anhedonia in ADHD and Autism: Why Nothing Feels Good and How to Rebuild Pleasure.

🧩 Cognition: Biases, Rumination, and Reduced Flexibility

Cognitive research shows that depression can be associated with changes in attention, memory, interpretation, self-evaluation, decision-making, and cognitive control.

Some people become more likely to notice threatening or negative information, interpret ambiguity pessimistically, retrieve negative memories, or struggle to disengage from self-critical material. Positive information may make less of an impact.

Rumination can keep attention circling around causes, mistakes, losses, or feared consequences without reaching a useful decision. Executive slowing may make it harder to shift perspective, organize a response, or interrupt the loop.

These findings should not be translated as “depression is caused by thinking badly.” Cognitive processes are not simply chosen, and they interact with sleep, stress, energy, learning history, physical state, and environment.

They can also occupy different positions in the causal chain. A negative processing style may increase vulnerability for some people. During an episode, depressed mood may itself make negative information more accessible. Repeated experiences of failure, exclusion, or overload may also produce conclusions that are understandable within the person’s circumstances.

Therapy can help people examine thoughts, reduce rumination, improve cognitive flexibility, and test predictions. Its usefulness does not mean depression was imaginary or purely psychological.

See Rumination Loops in Neurodivergent Depression for practical ways to distinguish reflection from repetitive, non-resolving thought.

🛏️ Sleep and Circadian Processes

Sleep and depression have a strongly bidirectional relationship.

Insomnia, persistent sleep disruption, some circadian disturbances, and several sleep disorders are associated with an increased probability of later depression. Depression can also produce difficulty falling asleep, repeated waking, early-morning waking, oversleeping, irregular sleep timing, or sleep that does not feel restorative.

Possible connecting processes include emotional regulation, cognitive control, stress reactivity, reward learning, inflammation, light exposure, activity patterns, and the timing of internal biological rhythms. No single sleep pattern appears in everyone with depression.

This means sleep should not be treated as a minor symptom. It can be a contributor, a maintaining factor, a consequence, or all three.

It also means “fix your sleep” is not an adequate response. Sleep may be affected by pain, medication, sensory conditions, delayed circadian timing, sleep apnea, restless legs, caregiving, shift work, housing, anxiety, or a depressive episode itself.

A proper assessment considers both sleep quantity and timing, as well as whether another sleep condition needs treatment. See Sleep, Circadian Drift, and Depression in ADHD and Autism for the detailed sleep-focused discussion.

🩺 Inflammation, Metabolism, and Physical Health

Some people with depression have higher average levels of particular inflammatory markers. Research also suggests that clinically meaningful low-grade inflammation may be present in a subgroup rather than across all depression.

These findings are important, but easy to overinterpret.

Inflammatory markers such as C-reactive protein are nonspecific. They can be influenced by infection, autoimmune conditions, body composition, smoking, sleep, medication, chronic stress, physical activity, and many other factors. An elevated result does not show that inflammation caused someone’s depression. A normal result does not make their depression less biological or less serious.

Likewise, “inflammatory depression” is not currently a simple diagnosis that can be confirmed with one standard blood test. Research is investigating whether inflammatory profiles may eventually help identify subgroups or predict treatment response, but routine personalized use remains limited.

Physical health still deserves attention. Pain, endocrine conditions, nutritional deficiencies, medication effects, sleep disorders, reproductive transitions, and other medical factors can resemble, precipitate, or intensify depressive symptoms.

Testing should be guided by the person’s symptoms, history, risk factors, and clinical examination—not by ordering every possible biomarker.

Self-treating depression with anti-inflammatory drugs or unregulated supplements can create interactions and medical risks. Suspected inflammation or physical illness should be discussed with a qualified clinician.

♾️ What About Depression in ADHD and Autism?

There is not yet evidence for one separate biological mechanism that explains depression in all autistic people, all people with ADHD, or all people with both.

Many mechanism studies exclude complex co-occurring conditions or do not report neurodevelopmental status. Their average findings therefore cannot automatically be treated as specific explanations for neurodivergent adults.

Neurodivergence may still shape the routes into and through depression. Relevant interactions can include:

Sensory overload and inadequate recovery
Sleep and circadian difficulties
Executive demands that repeatedly exceed available capacity
Social exclusion, rejection, bullying, or discrimination
Masking and sustained self-monitoring
Disrupted routines or loss of stabilizing interests
Barriers to healthcare and misunderstood communication
Co-occurring anxiety, pain, or physical conditions

These are not merely “external” influences. They can interact with learning, stress regulation, sleep, reward, cognition, and physical health over time.

At the same time, clinicians should not assume every difficulty is caused by autism or ADHD. The key question is often whether there has been a meaningful change from the person’s usual baseline.

For a broader account of presentation, see Neurodivergent Depression: Signs, Overlap, and Support. The evidence concerning elevated risk and its possible contributors is covered in Why Depression Is More Common in ADHD and Autism.

🧭 How Mechanism Research Can Help in Real Life

Mechanism research is most useful when it generates questions rather than assigning a totalizing explanation.

A clinician cannot usually determine that someone has “serotonin depression,” “cortisol depression,” or “inflammatory depression.” They can, however, examine the processes and conditions that may be relevant to this particular episode.

A useful assessment may consider:

🧭 Symptom pattern, severity, duration, and baseline change
🧭 Episode history, family history, and possible bipolar features
🧭 Sleep timing, sleep quality, and possible sleep disorders
🧭 Pain, illness, hormonal factors, medication, and substance use
🧭 Recent events, chronic adversity, relationships, and practical pressures
🧭 Reward loss, withdrawal, rumination, and reduced activity
🧭 Sensory, communication, and executive-function needs
🧭 Which treatments have helped, harmed, or been inaccessible before

Treatment can then target several parts of the system at once. Medication may reduce symptoms enough to make other changes possible. Therapy may address rumination, avoidance, trauma, interpersonal patterns, or rigid self-judgements. Sleep treatment may improve both rest and mood. Environmental support may reduce an ongoing source of strain. Practical help may interrupt the consequences of lost functioning.

Progress is often better judged through observable changes than through a presumed mechanism:

Is starting becoming slightly easier?
Is sleep becoming more stable?
Are thoughts less sticky or punishing?
Is there more access to interest or pleasure?
Is functioning returning without unsustainable effort?
Are side effects or demands making treatment harder to use?

If symptoms are persistent, severe, unusual, or worsening, reassessment is appropriate. A clinician may need to revisit the diagnosis, co-occurring conditions, physical contributors, medication effects, treatment fit, and barriers to implementation.

For an overview of symptoms and when evaluation is warranted, see What Is Depression? Symptoms, Types, and When to Seek Help.

If depression includes immediate danger, an inability to stay safe, or an intention to act on suicidal thoughts, contact local emergency services or a crisis service now. Do not wait for a perfect explanation of the mechanism before seeking help.

🌱 Conclusion

Depression is real, embodied, and potentially disabling. Its biological reality does not depend on proving one chemical deficiency, abnormal scan, inflammatory marker, or stress-hormone profile.

Research instead supports a more complex picture: many small vulnerabilities and processes can interact with circumstances over time. The relevant combination may differ between people and between episodes in the same person.

That uncertainty is not evidence that nothing is known. It is a reason to avoid simplistic stories, assess the whole person, and use treatment as an informed, adjustable process rather than a test of one presumed cause.

❓ Frequently Asked Questions

Is depression caused by a chemical imbalance?

A simple chemical-imbalance explanation is not supported as a general account of depression. Neurotransmitter systems participate in mood and other relevant functions, but researchers have not established that depression is caused by a universal deficiency of serotonin or another single chemical.

This does not mean antidepressants cannot be effective. Their clinical effects do not prove that they correct a pre-existing chemical shortage.

Can a brain scan show whether someone has depression?

Not in routine clinical practice. Research can detect average differences between groups, but there is substantial overlap between depressed and non-depressed individuals. No standard scan currently confirms depression, determines its cause, or reliably chooses the best treatment for one person.

Imaging may still be used when a clinician is investigating another neurological or medical concern.

Is depression genetic?

Depression has a moderate genetic component at the population level, but it is not caused by one gene. Many genetic variants may make very small contributions to vulnerability, interacting with experiences, environment, health, and development.

Family history can increase concern about risk or recurrence, but it does not make depression inevitable.

Does inflammation cause depression?

Inflammation may contribute to depression in some people, and some depressed groups have higher average inflammatory markers. The relationship is heterogeneous and potentially bidirectional.

Inflammatory markers are nonspecific, and there is no single routine blood test that establishes inflammation as the cause of an individual’s depression.

If there is no single cause, how can treatment be selected?

Treatment is chosen through the person’s symptoms, severity, history, preferences, health, previous responses, current circumstances, and access needs. Often, more than one maintaining process is addressed.

The plan can then be adjusted according to benefits, adverse effects, functional change, and feasibility. Uncertainty about the original mechanism does not prevent evidence-based treatment.

➡️ Where to Go Next

🧭 What Is Depression? Symptoms, Types, and When to Seek Help
♾️ Neurodivergent Depression: Signs, Overlap, and Support
📊 Why Depression Is More Common in ADHD and Autism
🎯 Anhedonia in ADHD and Autism
🧰 Therapy, Medication, and Adaptations for Depression

📚 References

Marx W, Penninx BWJH, Solmi M, et al. (2023). Major depressive disorder. Nature Reviews Disease Primers, 9, 44.

Flint J. (2023). The genetic basis of major depressive disorder. Molecular Psychiatry, 28, 2254–2265.

Kendler KS, Ohlsson H, Lichtenstein P, Sundquist J, Sundquist K. (2018). Population-based estimates of heritability shed new light on the genetic architecture of major depression. American Journal of Psychiatry, 175, 1102–1109.

Herrman H, Patel V, Kieling C, et al. (2022). Time for united action on depression: A Lancet–World Psychiatric Association Commission. The Lancet, 399, 957–1022.

Moncrieff J, Cooper RE, Stockmann T, et al. (2023). The serotonin theory of depression: A systematic umbrella review of the evidence. Molecular Psychiatry, 28, 3243–3256.

Cipriani A, Furukawa TA, Salanti G, et al. (2018). Comparative efficacy and acceptability of 21 antidepressant drugs for the acute treatment of adults with major depressive disorder. The Lancet, 391, 1357–1366.

Pilmeyer J, Huijbers W, Lamerichs R, et al. (2022). Functional MRI in major depressive disorder: A review of findings, limitations, and future prospects. Journal of Neuroimaging, 32, 582–595.

Ho TC, Gutman B, Pozzi E, et al. (2022). Subcortical shape alterations in major depressive disorder: Findings from the ENIGMA Major Depressive Disorder Working Group. Human Brain Mapping, 43, 341–351.

Halahakoon DC, Kieslich K, O’Driscoll C, et al. (2020). Reward-processing behavior in depressed participants relative to healthy volunteers: A systematic review and meta-analysis. JAMA Psychiatry, 77, 1286–1295.

LeMoult J, Gotlib IH. (2019). Depression: A cognitive perspective. Clinical Psychology Review, 69, 51–66.

LeMoult J. (2020). From stress to depression: Bringing together cognitive and biological science. Current Directions in Psychological Science, 29, 592–598.

Zhang MM, Ma Y, Du LT, et al. (2022). Sleep disorders and non-sleep circadian disorders predict depression: A systematic review and meta-analysis of longitudinal studies. Neuroscience & Biobehavioral Reviews, 134, 104532.

Osimo EF, Baxter LJ, Lewis G, Jones PB, Khandaker GM. (2019). Prevalence of low-grade inflammation in depression: A systematic review and meta-analysis of CRP levels. Psychological Medicine, 49, 1958–1970.