Yes, self-harm can trigger release of both dopamine and endogenous opioids (endorphins). The sharp pain of an injury activates endogenous opioid circuits in the brainstem and midbrain, while the relief and anticipated reward that follow engage dopamine pathways in the ventral tegmental area and nucleus accumbens. The opioid signal is generally the more direct and better-supported of the two, but dopamine release is downstream and secondary rather than a primary cause, and neither fully explains why nonsuicidal self-injury (NSSI) persists. This explainer traces the mechanism from nociceptors to neurotransmitter, quantifies what research actually shows versus popular oversimplification, and separates evidence from speculation.

Nociception and the endogenous opioid surge

Tissue injury does more than signal damage to the cortex. It triggers a stress-response cascade that begins in the periphery and ascends through the spinal cord, brainstem, and midbrain. Specialized nociceptors release substance P and glutamate at the dorsal horn. In response, the periaqueductal gray (PAG) and rostral ventromedial medulla mobilize descending inhibitory controls that release endogenous opioids at multiple levels.

The key molecules are beta-endorphin, enkephalins, and dynorphins. Beta-endorphin is cleaved from pro-opiomelanocortin and acts preferentially at mu-opioid receptors. Enkephalins bind delta receptors. Dynorphins act at kappa receptors and, counterintuitively, often produce dysphoria rather than relief. The mu-opioid receptor is the dominant player in analgesia and reward, and it is densely expressed in the ventral tegmental area (VTA), where it disinhibits dopaminergic neurons.

This is the mechanistic hinge: mu-opioid activation in the VTA removes GABAergic brake signals on dopamine neurons. That disinhibition increases dopamine release in the nucleus accumbens and prefrontal cortex. Endorphins therefore do not merely dull pain. They can indirectly generate the dopamine signal associated with relief and reward. This is why the two systems are coupled rather than separate.

Visualize a fire alarm that also unlocks a reward cabinet. The alarm is nociception. The reward cabinet is the opioid-dopamine circuit that opens alongside it. That coupling is what makes self-injury feel, for some people, like relief rather than pure harm.

Quantified context: what studies actually measure

Human evidence is thinner than popular claims suggest, so treat any specific percentage with caution. The most commonly cited evidence comes from cerebrospinal fluid (CSF) measurements of beta-endorphin and met-enkephalin in people with a history of self-injury, alongside controlled pain-analgesia paradigms.

Measure What it reflects Typical finding in self-injury Limit of interpretation
CSF beta-endorphin Central opioid tone Altered versus controls in small samples Invasive; small n; not diagnostic
CSF met-enkephalin Delta-opioid activity Reported differences in some cohorts Direction varies by study
Pain threshold (pressure/thermal) Endogenous analgesia Often elevated during or after injury Confounded by arousal and expectation
Pain tolerance Descending inhibition Frequently increased Not specific to opioid release
Mu-opioid receptor PET Receptor availability Altered binding in some psychiatric samples Scarce NSSI-specific data
Dopamine metabolite (homovanillic acid) Dopamine turnover Indirect and nonspecific Cannot localize release

The honest summary: there is credible evidence of opioid-system involvement, particularly elevated pain thresholds and altered CSF opioid levels, but no study has cleanly measured real-time dopamine release during an act of self-injury in a naturalistic setting. PET and fMRI work cannot ethically observe active self-harm. Researchers rely on proxies. That gap between mechanism and measurement is where thin explainers overreach.

A second quantification matters: temporal. Pain-evoked opioid release peaks within seconds to a few minutes. Dopaminergic reward signaling follows within tens of seconds to minutes and is shaped by prediction error, not injury per se. This means the relief is fast and the reward learning is gradual. Repeated episodes may sensitize the association between pain and relief, which is a learning mechanism, not a simple chemical flood.

The endogenous opioid system map

The opioid system comprises three receptor families (mu, delta, kappa) distributed across pain, reward, stress, and autonomic circuits. Practically, three nodes matter for self-injury:

Periaqueductal gray: central hub for descending pain control and defensive responses. Ventral tegmental area: mu-opioid disinhibition of dopamine neurons; the bridge to reward. Nucleus accumbens: dopamine and opioid integration for salience and relief.

The kappa-opioid system is the counterintuitive piece. Kappa activation by dynorphins is aversive and can drive stress-related relapse. Some researchers hypothesize that self-injury may also function to regulate kappa-mediated dysphoria rather than only to seek mu-mediated pleasure. That reframes the behavior as stress regulation, not only reward-seeking.

The opioid system also interacts with the hypothalamic-pituitary-adrenal (HPA) axis. Beta-endorphin is co-released with ACTH under stress. Self-injury during acute distress may ride an existing stress-hormone surge, amplifying opioid availability. Again, the effect is contextual, not automatic.

Plain-language version: the body has a built-in painkiller chemistry, and hurting yourself can switch it on, which then nudges the brain's reward chemistry.

Dopamine does more than signal pleasure. It encodes prediction error, salience, and motivation to repeat. If self-injury reliably reduces unbearable tension, the brain learns to anticipate relief before the act begins. That anticipation is dopamine's territory.

The ventral tegmental area receives mu-opioid input and projects to the nucleus accumbens. Mu-opioid activation there disinhibits dopamine neurons, producing a dopamine signal that coincides with analgesia. The result can feel like "the noise stopped" rather than euphoria. Many people describe relief and calm, not pleasure. That distinction matters clinically because it reframes the behavior as negative reinforcement (removing distress) more than positive reinforcement (adding pleasure).

Dopamine's role also explains the compulsion to repeat under stress. Prediction-error learning is powerful: a cue (emotional pain, dissociative numbness) predicts relief (self-injury), and the brain updates its model. Over time, the cue alone can trigger craving for the relief. This is not a moral failure. It is ordinary learning applied to an extraordinary stress response.

The professional caveat sits here: "dopamine release" is a mechanism-level explanation, not a diagnosis. It does not make self-injury an addiction in the DSM-5 sense, though some researchers describe behavioral features that overlap with substance-use models. Overlap is not equivalence.

Edge cases: when the simple model fails

Several findings complicate the clean dopamine-endorphin story.

First, not everyone reports relief. Some people report increased distress, shame, or even no change. If the behavior produced reliable reward for everyone, prevalence patterns would look different. Individual differences in opioid receptor genetics (such as OPRM1 variants), trauma history, and psychiatric comorbidity likely shift the response.

Second, pain offset matters. Animal work and human studies suggest that the relief after pain cessation involves endogenous opioid activity that dampens subsequent pain perception. This offset relief may be more important than the act itself. The mechanism is about the end of pain, not the injury per se.

Third, method and severity vary. Different methods involve different tissue damage, nociceptor recruitment, and stress-axis engagement. Grouping all NSSI into one category hides real variance in opioid and dopamine response.

Fourth, medications modulate the system. Opioid antagonists such as naltrexone have been studied in NSSI and related behaviors, with mixed results. Dopamine-blocking antipsychotics can blunt reward learning. The fact that blocking these systems does not uniformly stop the behavior is itself evidence that the mechanism is necessary but not sufficient.

Counterintuitive detail: some evidence suggests endogenous opioid activity is not uniformly elevated in people who self-injure. Baseline differences may be small or in the opposite direction, and the effect may be a dynamic response to stress rather than a static trait.

Common misconceptions

Misconception one: "Self-harm is about attention." The literature consistently contradicts this. Most NSSI occurs in private, is concealed, and functions as emotional regulation, self-punishment, or dissociation management.

Misconception two: "Endorphins make self-harm feel good." For many, it feels like relief, not pleasure. The opioid-dopamine coupling is about restoring equilibrium.

Misconception three: "Dopamine makes it an addiction." Reward learning and addiction overlap, but they are not identical. Self-injury lacks the defining features of substance dependence in most clinical descriptions.

Misconception four: "Pain causes a single chemical rush." The response is layered across nociception, descending inhibition, HPA-axis activity, and reward learning, each with its own time course.

Misconception five: "If you understand the chemistry, you can override the behavior." Knowing the mechanism can reduce shame, but it does not replace clinical care.

Time course of the chemical response

Phase Approximate timing Dominant system What it may feel like
Anticipation Seconds to minutes before Dopamine (prediction) Tension, urgency, craving for relief
Injury and nociception Immediate (milliseconds to seconds) Nociceptors, glutamate, substance P Sharp pain, alarm
Descending inhibition Seconds to several minutes Beta-endorphin, enkephalins, PAG Pain dulling, narrowing attention
Relief and offset Seconds to tens of minutes Mu-opioid and dopamine interaction Calm, quiet, emptiness filling
Reward learning Minutes to hours later Dopamine prediction error Memory linking cue to relief
Consolidation Hours to days Synaptic and circuit changes Strengthened habit over repetition

Reconcile the thresholds directly. Pain offset relief appears in studies within seconds, whereas the learned association requires repeated pairings over multiple episodes. The neurochemical cascade is fast. The behavioral pattern is slow. Both are true, and the difference explains why someone can know the mechanism and still struggle.

Practical actions without overclaiming

Understanding the chemistry is not a treatment plan. It is a way to reduce shame and guide the right conversations. These actions are concrete and appropriate for most people seeking help.

Track triggers and timing with a simple log: what preceded the urge, what the urge felt like (tension, numbness, anger), and what happened after. Patterns matter more than single events. Raise it with a clinician directly: "I want to understand the relief I feel after injury and whether medication or therapy targeting opioid or dopamine systems is appropriate." Clinicians can assess, but they should not prescribe based on a single explanatory article. Use harm-reduction questions if behavior is ongoing: ask about safe wound care, infection signs (redness, warmth, pus, fever), and when to seek urgent care. Identify a replacement that provides relief without injury: intense cold, vigorous exercise, paced breathing, or grounding techniques. These engage overlapping arousal systems without tissue damage. Escalate to urgent help for red flags: thoughts of suicide, injuries needing stitches, injuries near arteries or the neck, fever with a wound, or escalating frequency.

Red flags to treat as urgent: any suicidal ideation, any injury with uncontrolled bleeding, any wound showing infection signs, and any rapid escalation in frequency or severity. Call or text 988 in the US for crisis support. The Suicide and Crisis Lifeline is available 24/7.

One unit system applies here: all timings above are approximate and measured in seconds, minutes, or hours as labeled. No costs are asserted because the evidence base does not establish a price for any intervention in this context.

Limitations of the current evidence

The mechanistic case for opioid involvement is stronger than the case for dopamine, but both rest on indirect evidence. The field lacks naturalistic human measurement of neurotransmitter release during self-injury. Ethical constraints and technical limits explain why. That means claims of a precise "dopamine hit" are oversimplifications of a more complex stress and reward response.

Small sample sizes, heterogeneous definitions of self-injury, and confounding psychiatric comorbidity limit generalizability. Medication studies targeting these systems show mixed and modest results, which argues against a single-neurotransmitter explanation. The next decade of research may clarify genetic moderators and receptor-level differences, but current guidance should not be treated as definitive.

The most defensible conclusion: self-injury can engage endogenous opioids and, downstream, dopamine, producing relief and reinforcing repetition in vulnerable people. The chemistry explains part of the behavior, not all of it, and understanding it is a starting point for compassionate, evidence-based care rather than a substitute for it.