Dr Robert Malenka How Your Brains Reward Circuits Drive Your Choices

High-Level Overview

Andrew Huberman sits down with Dr. Robert Malenka, a psychiatrist and neuroscientist at Stanford whose lab helped uncover core mechanisms of neuroplasticity and then brought that lens to the brain's reward circuitry. Many of today's leading labs in neuroplasticity and reward trace back to his training, and their conversation spans nearly three hours. They cover what dopamine really signals, why context can turn a craved stimulus into an aversive one, what makes some drugs so addictive, the difference between wanting and liking, and how the same reward circuitry drives social connection, generosity and even behaviors resembling empathy in mice. From there they turn to autism, oxytocin's disappointing clinical record, MDMA's split effects on dopamine and serotonin, and the promise and risks of psychedelic research.

The tone is warm, collegial and frequently funny, full of donuts, Thanksgiving dinners, Paris cigarettes and a roulette-table confession, and punctuated by moments of real candor, including Malenka's account of overcoming deep self-doubt early in his career. The conversation moves between hard molecular neuroscience, personal stories and big societal questions like addiction, social media and the need for more compassion in the world.

The Discussion Explores:

• Dopamine's reward circuitry, from the ventral tegmental area to the nucleus accumbens, and why it evolved
• How context, cues and memory reshape what feels rewarding from moment to moment
• Addictive liability: why the amount and speed of dopamine release matter, including route of administration
• How drugs of abuse change synapses, and why some people become addicted while others do not
• Reinforcing versus rewarding, and wanting versus liking
• Malenka's gradual path from shy, insecure trainee to confident scientist
• Social reward, oxytocin and serotonin, and mouse models of empathy and generosity
• Autism, MDMA's pro-social effects through serotonin, and a cautious view of psychedelics

Deep Dive into Main Ideas

What the reward circuitry is for

Malenka frames dopamine as one of the brain's major neuromodulators, best known for its role in the reward circuitry, and starts with evolution. Every species needs a mechanism that flags stimuli as important for survival, so it is no accident that sugary, fatty foods, sex, warmth when you are cold, and water when you are thirsty feel so good. The system says: this matters, repeat the behavior that led here. It is equally important for the brain to flag danger so you avoid it next time. Malenka admits a weakness for donuts, which he loves so much he rarely lets himself eat them.

Anatomically, dopamine neurons in the ventral tegmental area (VTA) of the midbrain project to the nucleus accumbens, part of the ventral striatum. What happens next, he says, is still somewhat mysterious: we do not have a deep mechanistic understanding of why dopamine release there feels the way it does. It is more than simple reward. It signals that something important is happening, engages arousal and attention, and ties closely to memory systems, because you want to remember both powerfully rewarding and powerfully painful events. Subdivisions of the dopamine system are activated by painful and aversive stimuli too, which is why reducing dopamine to pleasure is an oversimplification.

Context changes everything

Huberman asks whether a small piece of a glazed old-fashioned donut releases as much dopamine as a whole one, and how fullness or a goal of abstaining changes things. Malenka agrees the system is "staggeringly simple, simultaneously staggeringly complex." His favorite example is Thanksgiving: in the morning the smells of turkey and apple pie and the prospect of Uncle Joe's visit are highly reinforcing, building anticipation that is almost craving. By evening, the same smells, the same pie and Uncle Joe himself may be aversive. Similarly, "the love of your life can turn to the bane of your existence in 20 seconds." None of this happens by magic; the brain is being modified by context.

The reason, Malenka explains, is that VTA dopamine neurons receive inputs from all over the brain, which he has mapped with Stanford colleague Liqun Luo, and the nucleus accumbens receives inputs from the hippocampus (memory), the amygdala (emotion) and the prefrontal cortex (planning and decision making). Huberman adds that the prefrontal cortex is only one connection away from these structures and can set rules for context, deciding whether we are in a mode of loving, arguing or reconciling. As for the donut, Malenka says it depends on your history: for people who love donuts, even a small piece activates all the memory systems.

Addictive liability, dose and speed

Malenka treats addiction as a continuum and prefers the concept of addictive liability. Caffeine produces tolerance, but no one robs a bank for coffee, so its liability is low. Psychostimulants like cocaine and methamphetamine and opioids have very high liability. Answering Huberman's question, he confirms that a substance's addictive liability roughly tracks two things: how much dopamine is released in the nucleus accumbens and how fast. He cites work by Nora Volkow, director of the National Institute on Drug Abuse, alongside animal studies.

That is why route of administration matters. Snorting cocaine differs from smoking or injecting it; smoked crack cocaine and smoked methamphetamine reach the brain almost instantly and produce a rapid, powerful dopamine surge lasting tens of seconds to a few minutes, followed by an overwhelming urge to do it again, even if the feeling itself is not especially happy. Malenka notes that for most of human history, the only addictive substance around was alcohol; our brains were not designed for cocaine, methamphetamine or synthetic opioids like fentanyl. He tells students that using a drug does not guarantee addiction, but you are taking a risk, and "it is impossible to become addicted to a substance if you've never used it by definition," a line Huberman flags as profound.

Nicotine, delivered in cigarettes, also has very high liability, and tobacco companies worked out the dose that produces a feeling lasting just a few minutes so you want to repeat it. Malenka shares that he smoked for a few weeks in Paris around age 20, then stopped on returning home, yet 40 years later he still gets cravings whenever he goes back to Paris, a vivid example of how cues tied to powerful experiences keep their pull.

How drugs rewire the reward circuit, and who gets hooked

Malenka's lab, among others, showed that drugs of abuse such as cocaine, methamphetamine, morphine and heroin cause powerful plasticity at synapses onto dopamine neurons and onto nucleus accumbens neurons, changes similar to those that evolved for adaptive learning and memory. In rodent models, a single dose of cocaine or morphine can alter these connections for several days to a few weeks; they are not permanent, and repeated doses make the changes stronger and longer-lasting.

Why some people become addicted and others do not is one of the field's biggest questions. The answer is always a mix of genetics, the environment people grew up in and their current environment, including whether they have other, healthy sources of reward. Huberman notes that friends who are recovered alcoholics describe their first drink as a magic elixir, while both men say alcohol mostly makes them sleepy. Malenka points to evidence that alcohol use disorder runs in families and has a genetic component. Exercise, they note, is almost the inverse of cocaine: unpleasant in the moment, followed by feeling great.

Reinforcing versus rewarding, wanting versus liking

Huberman recounts a recovered cocaine addict who said that after his first use he thought he hated it and could not wait to do it again. Malenka calls this the perfect illustration: "I hate it, but I want to do it again." In behavioral terms, something is reinforcing if it makes you repeat the behavior that led to it, while rewarding means it actually felt good. Cocaine can be highly reinforcing without being enjoyable. Kent Berridge and Terry Robinson captured the same split as wanting versus liking, which also describes destructive relationships where you want someone you do not necessarily like.

Huberman suggests that 12-step programs, which Anna Lembke has told him have high success rates when followed, work partly by splitting wanting from liking, building a liking for sobriety and rewarding abstinence through social connection and a new self-concept. Malenka agrees, noting that helping people feel they deserve credit for not drinking at a party creates new sources of reward.

On opioids, Malenka explains that psychostimulants and opioids share a final common action, massive dopamine release in the nucleus accumbens, but by different mechanisms. Cocaine and methamphetamine block the dopamine "vacuum cleaner" that clears dopamine and even reverse it, while opioids act mainly where dopamine neurons live, indirectly boosting their activity. Their subjective effects differ greatly because of actions elsewhere in the brain, stimulation and teeth grinding versus dreamy sedation. Fentanyl has a much larger addictive liability because of its molecular properties. Both men dislike opioids like Vicodin for post-surgery pain, and Huberman describes an accidental encounter with blue lotus flower tea at a friend's house, which he says contains morphine-like compounds and left him with bizarre dreams and 14 hours of sleep.

From shy trainee to confident scientist

Malenka, who worked with Allan Hobson as a Harvard undergraduate, reveals that as a young man he was very shy and insecure, assuming his ideas must be obvious and never voicing them, even in medical school. Huberman asks how that changed. For Malenka, it was a gradual process. As a postdoc with Roger Nicoll at UCSF, an intellectually intense and forceful mentor, he entered a field of vigorous debate and had to develop a thicker skin and argue for his hypotheses. As an assistant professor he realized he could do science and write papers colleagues cared about, and confidence built over the next decades.

His takeaway is not that all his ideas are great, but that it is okay to voice them and explain why he believes one thing and not another. Huberman recalls assuming, as a young scientist, that Nicoll worked for Malenka, because they quickly became peers. Malenka says the one confidence he did have was wanting to be treated as an equal: even with mentors, he always felt he was working for himself.

The social reward circuit

About 13 to 15 years ago, Malenka began asking what else modifies the reward circuitry beyond drugs and depression (where anhedonia, the inability to feel reward, is a core symptom). Positive, non-aggressive, non-sexual social interaction is highly reinforcing, and for him, time with close friends is the most meaningful part of life besides his children. Evolutionarily, sociability aids mating, protection from predators and, Huberman adds, child rearing.

With postdoc Gül Dölen, now at Johns Hopkins, his lab studied oxytocin in the nucleus accumbens of mice, building on Larry Young's and Tom Insel's classic prairie vole work on pair bonding (while noting a recent paper that questioned oxytocin's role, set against 30 years of prior evidence). They found oxytocin in the accumbens promotes sociability and, surprisingly, enhances serotonin release there, which launched Malenka's work on serotonin and MDMA. His lab and a Northwestern group also showed oxytocin modulates dopamine neurons in the VTA. The lesson, both men stress, is that neuromodulators never work in isolation.

The nucleus accumbens has long been modeled as two cell types, one acting as an accelerator promoting behaviors and one as a brake, differentially modulated by dopamine and serotonin. Huberman frames social behavior as choosing some actions and suppressing others depending on context, hugging close friends but watching your words with a near stranger, and Malenka agrees the accumbens shapes the probability of which behaviors you pursue.

Social media, gambling and intermittent reward

Asked whether social media has high addictive liability, Malenka says he is not sure an experiment is needed; it clearly does. Though not on social media himself, he finds checking email and texts compulsive: "It's like a lever press for a mouse." Messages from friends mean they are thinking of him, activating reward circuitry, though nowhere near as strongly as cocaine or opioids. Both agree these platforms, like pornography and gambling, capitalize on ancient mechanisms that evolved for in-person interaction and reproduction.

Huberman shares that gambling addicts told him the unique sting is believing the next bet really could change everything, which no alcoholic believes about the next drink. Malenka confesses to winning a meaningful sum at roulette, forcing himself back to his room with only one brief stop at another table (which he lost), and leaving Las Vegas quickly. Huberman notes the reward system remembers what led to the win, and nobody ever won by getting up and leaving. Casinos, Malenka suspects, employ experts in behavioral economics to tune payouts, because intermittent rewards are so powerful.

Studying empathy in mice

Malenka's motivation for studying empathy is deeply personal. Recently back from lecturing in Israel and reflecting on the Israeli-Palestinian conflict, he asks "what is more important for the survival of the human species than empathy and compassion?" Without a political platform, he believes his contribution is to study its neurobiology. With Monique Smith, now an assistant professor at UC San Diego, his lab uses what he calls behavioral antecedents of empathy: behaviors showing one animal is influenced by another's emotional state.

In one assay, a bystander mouse that simply spends an hour with a mouse in modest pain begins showing pain behaviors itself for roughly 4 to 20 hours, likely partly through an odor cue, since bedding from mice in pain can do it. In a new assay published in Science, the social transfer of pain relief, a mouse in pain given morphine can reduce pain behaviors in a cage mate that is still in pain. Malenka links this to human research on social buffering of pain, and Huberman connects it to Noam Sobel's work on chemical signals between people and the power of a nod or touch. The lab is also developing generosity assays, testing whether a mouse will work to give another a reward or spare it a shock, with promising early results. These behaviors appear to involve the anterior cingulate cortex, active in human empathy studies, and its projections to the nucleus accumbens.

Huberman suggests testing how inconvenience affects helping, recalling his mother, who would stop to help homeless people and even put them up in hotels, making the family late everywhere. Malenka candidly admits he might not stop if late for a plane, and says they are already asking whether mice help familiar versus stranger mice, or ones they previously fought and beat or lost to.

Autism, oxytocin and serotonin

The two discuss how to talk about autism respectfully. Malenka describes autism spectrum disorder as highly heterogeneous, from people with severe intellectual and social impairments who clearly need help to high-functioning people who see it as a different style of living and interacting, and says both deserve respect. Studies in humans and mice suggest the reinforcing component of social interaction is reduced in some individuals, though whether that is genetic or shaped by early experience is unknown. In his mouse models, deficits in social and empathy-like behaviors can be rescued by manipulating the serotonin system, including with MDMA.

On oxytocin nasal spray, Malenka says most clinical trials have been pretty disappointing, though he would not close the door, and a Stanford colleague is seeing possible benefit with the related peptide vasopressin. There is no FDA-approved drug for autism's social deficits, and well-done trials of SSRIs and SNRIs have not shown efficacy. In full disclosure, he co-founded the biotech Maplight Therapeutics with Karl Deisseroth, which has a drug in phase two trials targeting a serotonin receptor subtype, and he sits on the scientific advisory board of MindMed, which is preparing a trial of one mirror-image form (enantiomer) of MDMA. Research by other labs suggests the S form interacts more with dopamine and the R form more with serotonin.

MDMA: dopamine for liking it, serotonin for connection

MDMA, an amphetamine derivative, blocks and reverses the vacuum cleaners for both serotonin and dopamine, making them spew out their contents. It affects serotonin more than dopamine, perhaps 70 to 30. Huberman highlights the Malenka lab paper showing that serotonin release in the nucleus accumbens, acting through serotonin 1B receptors, drives MDMA's pro-social effects, while oxytocin plays a less prominent role. Malenka's hypothesis is that MDMA's addictive liability and fun factor come mostly from dopamine, while its pro-social and possibly empathogenic effects come from serotonin. He says some colleagues dislike him saying so, but MDMA does carry some addictive liability, which matters if the FDA approves it as an adjunct to therapy for PTSD. He adds that serotonin neurons in the dorsal raphe influence dopamine neurons, and that serotonin in the accumbens modulates dopamine release.

Huberman, who participated in an MDMA trial, stresses that MDMA remains a Schedule I drug, that people are going to prison for large seizures, and that street supplies can be contaminated with fentanyl, so they are not encouraging its use. Malenka credits MAPS and founder Rick Doblin for a 30-year effort to make rigorous, ethical MDMA research possible. What fascinates him is that a drug with specific molecular targets has such a relatively specific effect on social behavior; it does not make you want more donuts. Huberman notes that psilocybin and LSD act mainly through serotonin 2A receptors with very different, often mystical effects, and Malenka points out there are 16 serotonin receptor types, a daunting complexity that also creates therapeutic opportunities.

A cautious view of psychedelics

Malenka, a self-described child of the 60s and 70s who experimented with psychoactive substances and credits them partly with sparking his interest in neuroscience, is excited but not evangelical. He separates classical hallucinogens (LSD, psilocybin) from entactogens and empathogens like MDMA, and mentions ibogaine, ayahuasca and peyote as complex cases. These drugs deserve rigorous, sophisticated and above all ethical research, and may have therapeutic potential, but "I do not think they're going to be miracle cures."

He recalls the Timothy Leary era and a bad LSD trip or two in the 70s, which he says was the closest he has come to understanding the hopeless darkness of suicidal depression. Well-controlled trials exclude many people through strict criteria, and he worries that bad outcomes from careless use could set back legitimate research. Huberman mentions Kentucky's plan to put $42 million from an opioid settlement toward ibogaine research, which Malenka supports as long as it is done thoughtfully and ethically. Malenka closes by saying he was nervous to come on and now wants to come back.

Conclusion

Malenka's central message is that the reward circuitry is both elegantly simple and endlessly complex: a system built to flag what matters for survival, constantly reshaped by context, memory and experience. That plasticity is what lets drugs with fast, large dopamine surges hijack it, what separates wanting from liking, and what lets cues like a Paris cafe keep their pull for decades. The same circuits, tuned by dopamine, serotonin and oxytocin, make friendship and social connection so rewarding, and may underlie early forms of empathy and generosity that can be studied in mice. That research could one day help people with autism who want more social connection, and it explains why MDMA and psychedelics are such intriguing tools. His recurring caution, whether about addiction, social media or psychedelic therapy, is to respect the brain's power and complexity and to pursue these questions with rigor and care.

This is a summary of "Dr Robert Malenka How Your Brains Reward Circuits Drive Your Choices" from Huberman Lab, giving you the GIST so you can decide if you want the REST. Listen to the full episode.