Research & Ideas / 2026-07-31 / 11 min read
Why Things Smell Good or Bad: From Jasmine to Feces
Fragrant and foul are not properties of molecules but stamps the brain applies. Recognizing an odor and liking it run on separate tracks: genes decide what you detect, experience decides how you judge it.
- Science
- Biology
- Olfaction
- Cognitive science
Why do humans find feces foul while dogs are indifferent — or even eager to sniff it? Why is indole a jasmine fragrance at low concentration but a fecal stench once concentrated? Why do some people crave stinky tofu while others bolt at the first whiff? The answers all converge on one point: fragrant and foul are never properties of the molecule; they are stamps the brain applies.
At which stage does the smell appear?
Your sense of smell runs a six-stop pipeline: volatile molecules → matching with roughly 400 olfactory receptors in the human nose (dogs have 800-plus) via lock-and-key pairing → one molecule activates several receptors, producing a “receptor fingerprint” → the olfactory bulb maps that fingerprint into a spatial pattern → cortical areas read out “what smell is this” → pleasantness evaluation and behavioral response.
The first four stages only answer “what is this” — they are entirely neutral. “Fragrant/foul” emerges only after stage five, generated by the amygdala and orbitofrontal cortex with input from concentration, experience, context, and expectation. The molecule supplies the raw material; the brain applies the label.
Two tracks: recognizing and liking are separate paths
The brain handles “what smell is this” (identity track) and “do I like it” (valence track) separately, and the two tracks can light up at the same time. That explains a seemingly contradictory phenomenon — “smells bad, tastes good”: stinky-tofu lovers still recognize the “stink,” but their valence track has been flipped to “delicious” by learning and culture. Cheese, durian, natto, and surströmming are all the same story: the same molecules, with evaluation rewritten by experience.
Genes set the amplifier; experience passes the verdict
Cilantro is the textbook case. People who dislike it are actually smelling “soap” — cilantro’s aroma is dominated by aldehydes, which are precisely the components of soap. A variant near the gene rs72921001 (close to the olfactory receptor OR6A2) makes that soap note easier to detect (Eriksson et al., 2012). But genes explain less than 10% of preference differences: among people raised on cilantro in the Middle East, only about 3% dislike it; among East Asians, about 21%. Genes determine what you smell; experience determines how you judge it.
Pleasantness is not all experience either — an experiment across nine cultural groups found that molecular structure alone predicts about 40% of pleasantness differences, while culture accounts for only 6% (Arshamian et al., 2022); and mice’s avoidance of predator urine is hardwired by a single receptor gene (Dewan et al., 2013). Nature and nurture are both present; only their weight shifts from odor to odor.
A harder example is androstenone — a steroid in sweat and also a pig sex pheromone. Within the same population, some people cannot smell it at all, some find it foul, and some find it sweet like a flower. Variation in the receptor gene OR7D4 explains roughly 20–40% of the perceptual differences (Keller et al., 2007): genes directly rewrite subjective experience. Yet repeated exposure can turn “can’t smell” into “can smell.” Genes set the initial state; experience can move it.
Where do dogs and humans differ?
The sensory end genuinely differs: dogs have about twice as many receptor types (Quignon et al., 2003) and an olfactory epithelium roughly 20 times larger, letting them resolve far more odor combinations. But “dogs think feces smell good” is a misreading — dogs’ interpretive end treats feces and urine as information (territory, health, estrus), which is “important,” not “delicious.” And much of humans’ aversion to feces is itself learned: young children do not naturally find feces disgusting.
A third system: the sense hardest to “learn”
Beyond smell and taste (sweet, sour, bitter, salty, umami), there is a third perceptual system — “chemesthesis”: the heat of chili, the pungency of mustard, and the coolness of mint are all mediated by TRP channels on the trigeminal nerve. Menthol activates TRPM8 — a receptor originally tuned to detect “cold” — so your mouth feels cool without any drop in temperature. Nociceptive signals are the hardest to learn: you can tolerate them, but rarely like them. That may explain why someone can learn to eat cilantro yet spend a lifetime finding mustard vaguely “plastic” — the raw material of that plastic-ness (sulfur + bitter + irritant) is partly inborn, while the “plastic” label is a gift of culture.
An underrated myth
“Humans have a poor sense of smell” is actually a 19th-century inference: the anatomist Broca classified humans as “microsmatic” based on brain proportions alone, without ever running a single smell test. That label suppressed olfactory research for nearly a century. Modern evidence says the opposite: the human olfactory bulb still holds tens of millions of neurons, people can track scent trails along the ground like dogs (Porter et al., 2007), and humans are more sensitive than dogs to some molecules (McGann, 2017). Dogs having more receptors means they smell more kinds of things — a difference in range, not in sophistication.
References
- Quignon, P. et al. (2003). Comparison of the canine and human olfactory receptor gene repertoires. Genome Biology, 4:R80. doi:10.1186/gb-2003-4-12-r80
- Eriksson, N. et al. (2012). A genetic variant near olfactory receptor genes influences cilantro preference. Flavour, 1:22.
- Keller, A. et al. (2007). Genetic variation in a human odorant receptor alters odour perception. Nature, 449:468–472. doi:10.1038/nature06162
- Dewan, A. et al. (2013). Non-redundant coding of aversive odours in the main olfactory pathway. Nature, 497:486–489.
- Arshamian, A. et al. (2022). The perception of odor pleasantness is shared across cultures. Current Biology, 32(9):2061–2066. doi:10.1016/j.cub.2022.02.062
- Porter, J. et al. (2007). Mechanisms of scent-tracking in humans. Nature Neuroscience, 10:27–29. doi:10.1038/nn1819
- McGann, J. P. (2017). Poor human olfaction is a 19th-century myth. Science, 356(6338):eaam7263. doi:10.1126/science.aam7263