Quick summary: Scientists propose that dopamine and opioids primarily manage the body’s energy budget rather than directly generating pleasure or desire, with motivation arising from increased physiological activity to meet demands and satisfaction from the subsequent reduction in energetic costs. This reframing links reward to measurable metabolic gains across systems such as digestion, respiration and immune function, suggesting that reinforced behaviours optimise energy regulation and that social bonding enhances efficiency through proximity. The findings imply that conditions including depression, addiction and obesity may stem from disruptions in these energy management mechanisms, with consequences for mental health understanding and healthcare practice focused on physiological balance.
Scientists have long described dopamine and opioids as the brain’s reward chemicals, the substances behind pleasure, desire, and the drive to seek out food, sex, or social connection. A new study proposes something more fundamental: that these two systems are primarily tools for managing the body’s energy budget, with feelings of motivation and pleasure emerging as byproducts rather than the main event.
The research, published in Neuroscience and Biobehavioral Reviews, was conducted by Matan Cohen and Shir Atzil, PhD at the Hebrew University of Jerusalem. Atzil, an associate professor at the university, says the work began with a basic question about biology. “From an evolutionary perspective, the brain regulates the body’s energy, and every action carries a metabolic cost. We propose that reward reflects improvements in this regulation, reframing dopamine and opioids as core components of a system that manages the body’s energetic budget.”
The authors argue that dopamine and opioids function as opposing physiological regulators, with dopamine activating bodily systems and opioids calming them back down once a disturbance has passed. According to the framework, dopamine does not encode pleasure directly. Instead, it raises the body’s physiological activity in response to a demand, such as hunger, stress, or physical exertion, and the rising energetic cost of that response generates the experience of motivation to act. When the demand is resolved, opioids are released to dampen the physiological response, and the resulting reduction in energetic burden is what a person subjectively experiences as relief or satisfaction.
The researchers draw on evidence from digestion, respiration, immune function, fluid regulation, sleep and temperature control, all areas where dopamine consistently activates bodily processes and opioids consistently suppress them. “This pattern appears across physiological systems, suggesting these are not specialised ‘reward’ agents but fundamental regulators of physiology,” Atzil notes.
The study also reframes how the brain learns. Behaviours and physiological responses are reinforced not because they feel good in an abstract sense, but because they reduce the body’s energetic costs. The authors describe this as metabolic gain, a measurable reduction in the physiological burden that follows a successful adaptation.
The implications for understanding addiction, depression and schizophrenia are significant. Atzil suggests that familiar psychiatric conditions may need to be reconsidered through a physiological lens. “Disorders such as depression, addiction, and obesity could reflect disruptions in the neural mechanisms that regulate energy regulation.”
The framework also extends to social bonding. “In humans, social bonding optimises metabolic efficiency. We show that physiological regulation is more efficient in social proximity, a phenomenon we term Social Physiology. This provides an evolutionary mechanism for bonding: individuals come to prefer specific others to the extent that they reliably provide metabolic gains.” The team is now testing how these gains reinforce romantic and parental attachments.
The researchers acknowledge that other neurotransmitter systems, including serotonin, norepinephrine and GABA, are not yet incorporated into the model, and they note several findings in the existing literature that do not fit neatly into the proposed sequence. Further empirical work will be needed to test whether metabolic effort and metabolic gain can reliably predict behaviour and learning across species.
