The Endocannabinoids — The Brain's Backward Signal

Most of this series has described neurotransmitters that work in one direction: a presynaptic neuron releases a chemical, a postsynaptic neuron receives it. The endocannabinoids break that rule. They are made by the receiving cell and travel backward across the synapse to the cell that just spoke, where they tell it to quiet down. It is the closest thing the brain has to a listener raising a hand to say that's enough. This retrograde signaling is not a curiosity at the margins — it is one of the most widespread regulatory mechanisms in the nervous system.

The system was found by working backward from a plant. Researchers isolated THC as the active compound in cannabis in the 1960s, which raised the obvious question of why a mammalian brain would carry receptors for a molecule made by a flowering plant. It doesn't, of course. The receptors were there first, and by the early 1990s the brain's own ligands had been identified: anandamide, named from the Sanskrit ananda, meaning bliss, and 2-arachidonoylglycerol, usually shortened to 2-AG. The receptors they act on, CB1 in the nervous system and CB2 largely on immune cells, turned out to be extraordinarily abundant. CB1 is among the most numerous G-protein-coupled receptors in the mammalian brain. Cannabis did not invent a state of mind; it borrowed one.

Endocannabinoids are also unusual in how they are stored, which is to say they aren't. Classical neurotransmitters are packaged into vesicles and held ready for release. Endocannabinoids are lipids, built on demand out of the neuron's own membrane when calcium floods in, released immediately, and broken down within minutes by enzymes — fatty acid amide hydrolase for anandamide, monoacylglycerol lipase for 2-AG. This makes the system fundamentally reactive rather than anticipatory. It does not initiate activity; it responds to activity that has already occurred and modulates what comes next. Functionally, it is a dimmer switch layered on top of both glutamate and GABA transmission, which is why its effects show up nearly everywhere: appetite, pain, sleep, memory, immune regulation, and above all the recovery from stress.

That last function is the one most relevant here. Endocannabinoid signaling in the amygdala and prefrontal cortex is central to fear extinction — not the erasure of a threat memory but the learning that a previously dangerous cue is now safe. Animals with impaired CB1 signaling learn threat perfectly well and struggle to unlearn it. The system also helps terminate the stress response, applying a brake to the HPA axis once a stressor has passed. Chronic stress, in turn, downregulates the system, which suggests an uncomfortable feedback loop: the machinery for recovering from stress is itself degraded by sustained stress. This has become a leading framework for thinking about why prolonged adversity produces disproportionate anxiety and why some people struggle to return to baseline long after a threat resolves.

The endocannabinoids also revised a story most people still tell wrong. Runner's high has been attributed to endorphins for decades, but endorphins are large peptides that cross the blood–brain barrier poorly, and blood measurements of them may say little about what is happening centrally. Anandamide, a small lipid, crosses easily and rises reliably with sustained moderate exercise. A frequently cited mouse study found that blocking endocannabinoid receptors abolished the post-exercise reduction in anxiety, while blocking opioid receptors did not. Human work is harder and less conclusive, and the honest position is that both systems likely contribute. But the endocannabinoid account has become the better-supported explanation for the calm, low-anxiety state that follows a long run.

This connects directly to work covered elsewhere on this site. If exercise recruits endocannabinoid signaling, and if that signaling underlies stress recovery and fear extinction, then the repeated finding that physical activity in green space outperforms the same activity indoors starts to look mechanistically tractable rather than merely pleasant. The candidate pathways are additive: movement raising anandamide, natural settings lowering sympathetic arousal, sunlight and sleep regulation stabilizing the whole system. Nothing here is settled, and the field has a well-earned reputation for outrunning its data. But it does suggest that the biology of feeling restored by an afternoon outdoors is neither vague nor sentimental — and that the steady loss of accessible green space is, among other things, the removal of a working piece of neurochemical infrastructure.