Huberman LabEssentials: How Smells Influence Our Hormones, Health & Behavior | Dr. Noam Sobel
CHAPTERS
- 0:00 – 4:35
How smell works: receptors, epithelium, and the olfactory bulb map
Dr. Noam Sobel explains the core pathway of orthonasal olfaction, from sniffing airborne molecules through the nose to transduction in the olfactory epithelium. He describes how signals converge in the olfactory bulb into glomeruli-based patterns that can be thought of as an odor “map.”
- •Primary smelling happens through the nose via sniffing (orthonasal olfaction)
- •Airborne molecules travel to the olfactory epithelium, which contains millions of receptors across many receptor types
- •Odorants are transduced into neural signals that travel through the olfactory nerve into the brain
- •Receptor subtypes converge onto specific olfactory bulb locations called glomeruli, creating distinct activation patterns for odors
- •Signals then project broadly to multiple brain targets; coding beyond the bulb is described as less clear
- 4:35 – 5:36
Humans as ‘olfactory animals’ and why we might underestimate our sense of smell
Sobel argues that humans have a remarkable sense of smell, even if we don’t behave like animals that obviously sniff one another. He frames his lab’s approach as looking at human behavior through an olfactory lens to uncover subtle smell-driven actions.
- •Humans may seem less smell-driven than dogs, but that assumption can be misleading
- •Sniffing loved ones (like babies) is culturally acceptable, while sniffing strangers is not
- •The lab investigates hidden or socially masked ways humans may sample odors
- •Reframing everyday behavior can reveal olfaction’s role in social interaction
- 5:36 – 10:03
Handshakes as covert smell sampling: evidence people sniff after greeting
Sobel describes research suggesting that handshaking prompts increased hand-to-face movements consistent with smelling. The work includes baseline measurements, handshake vs. no-handshake conditions, and nasal airflow monitoring to test whether post-handshake behavior is olfactory in nature.
- •Participants were covertly recorded to measure baseline face-touching behavior before any greeting
- •A new experimenter entered and either shook hands or did not, allowing comparison between conditions
- •After a handshake, there was a marked increase in bringing the hand to the nose/face area
- •Nasal airflow measurements indicated participants were sniffing during these hand-to-face moments
- •People also smelled the non-handshaking hand, interpreted as possible self-to-other comparison
- 10:03 – 12:35
‘Click friendships’ and body-odor similarity: testing with an electronic nose
Sobel outlines a study led in his lab investigating whether friends who “click” quickly have more similar body odors than expected by chance. Using an electronic nose and human ratings, the work links perceived and sensor-measured odor similarity to these rapid friendships.
- •Pairs of same-sex, non-romantic “click friends” were recruited and their body odor was sampled via a controlled T-shirt protocol
- •An electronic nose (sensor array) measured odor profiles and found click friends were more similar than random pairs
- •Human smellers also judged click friends’ odors as more similar compared to non-click pairs
- •The electronic nose is presented as a rough technological analog of biological olfaction
- 12:35 – 14:04
From correlation to prediction: odor similarity forecasting social preference among strangers
To probe causality, the lab measured strangers’ odor similarity first and then observed a structured nonverbal interaction. While one motor-performance prediction did not hold, odor similarity still predicted social ratings and desire for friendship after the interaction.
- •Total strangers were profiled with the electronic nose before interacting
- •Participants engaged in a nonverbal “mirror game,” moving hands in close proximity
- •The prediction that odor similarity would improve mirror-game performance did not pan out
- •After round-robin interactions, participants rated how much they wanted to be friends and other traits
- •These social ratings were predicted by electronic-nose similarity, supporting a role in forming social bonds
- 14:04 – 16:22
Chemosignals vs. ‘pheromones’: framing and the ‘smell of fear’
Huberman asks about pheromone-like effects, and Sobel emphasizes his lab’s preference for the term “chemosignals.” He discusses evidence that humans emit fear-related body odors that can influence others’ autonomic arousal, contributing to a form of fear contagion.
- •Sobel avoids the term “pheromone” in publications and uses “chemosignals” instead
- •Humans emit chemosignals that can influence other humans’ behavior
- •A major focus in the field has been the “smell of fear,” where fear-state body odor affects others
- •Exposure to fear-related odor can increase autonomic/sympathetic arousal in others
- •He notes research indicating both dogs and humans can detect human fear-related odor differences
- 16:22 – 21:39
Hexadecanal and aggression: opposite effects in men and women, and a baby-head link
Sobel recounts research on hexadecanal, a molecule studied as a social chemosignal, using a validated aggression task and later fMRI replication. The striking finding is sex-divergent effects on aggression, which is then interpreted through an offspring-centered evolutionary logic and linked to baby-head emissions.
- •Hexadecanal was introduced to the lab following animal work suggesting social effects and conservation across mammals
- •A standard aggression measure (the Tyler Aggression Paradigm) was used to quantify behavioral effects
- •In the task, hexadecanal reduced aggression in men but increased aggression in women, and this replicated in an fMRI version
- •Sobel relays the interpretation that this pattern could benefit mammalian offspring by reducing male aggression while increasing maternal protectiveness
- •He states hexadecanal is highly abundant among semi-volatiles emitted from baby heads and frames this as “chemical warfare” by babies
- 21:39 – 25:37
Emotional tears as chemosignals: odorless tears, testosterone reduction, and brain effects
Sobel describes experiments testing whether emotional tears function as chemical signals. He reports that although the tears are subjectively odorless, exposure reduced men’s testosterone and altered brain activity in specific regions during arousal-related contexts.
- •The work was motivated by the idea that tears might have a functional role as nonverbal chemical communication
- •Emotional tears were collected and presented to participants; Sobel emphasizes they were not consciously detectable by smell
- •Men exposed to women’s emotional tears showed a reduction in free testosterone within roughly 20–30 minutes
- •Sobel says an independent group later replicated the testosterone effect with similar numbers
- •Brain imaging showed dampened activity including in the hypothalamus (and also in the fusiform gyrus, with uncertainty about why)
- 25:37 – 29:49
From ‘Google Smell’ to olfactory metamers: predicting perceptual similarity of mixtures
Huberman raises the concept of generating smells digitally, and Sobel discusses efforts to digitize smell. He describes his lab’s algorithmic framework that predicts perceptual similarity between molecular mixtures and enables creation of olfactory metamers—distinct mixtures that smell the same.
- •The conversation frames digitizing smell as an emerging technological effort, popularized by the idea of “Google Smell”
- •Sobel describes a Nature paper from his lab presenting an algorithm that predicts perceptual similarity between mixtures
- •The framework can be used not just to predict similarity but to design mixtures with targeted similarity relationships
- •They created olfactory metamers: mixtures with no overlapping molecules that nevertheless smell the same
- •An example described is generating a metamer for Chanel No.5 without using components of Chanel No.5
- 29:49 – 33:43
Transmitting odors over IP and the path toward medical diagnostics
Sobel explains how similarity prediction enables practical digitization: mapping mixtures into a perceptual space and recreating odors using a set of “odor primaries.” He shares early proof-of-concept attempts at recreating specific odors remotely and outlines a longer-term vision where high-resolution olfactory digitization supports continuous medical diagnostics.
- •Similarity prediction is positioned as sufficient to estimate what a mixture will smell like by projecting it into a perceptual map
- •Sobel describes ongoing work to identify “odor primaries” that could be mixed to reproduce many perceivable odors
- •A collaboration measured odorants remotely (via analytical tools) and transmitted data over IP for reconstruction
- •In proof-of-concept reconstructions, attempts at rose and anise were perceived differently (e.g., bubblegum or cherry), gasoline was modestly above chance, and violets were correctly identified by most participants
- •He imagines future high-resolution digitized olfaction enabling routine disease-related odor diagnostics (e.g., an electronic nose doing ongoing checks), while emphasizing it is not near-term
- 33:43 – 34:55
Closing reflections and thanks
Huberman thanks Sobel for the discussion and highlights the range of topics covered, from olfactory system architecture to hormones and digitization of smell. Sobel responds with appreciation for the warm closing remarks.
- •Huberman recaps themes including hormones, smell, and olfactory system architecture
- •Huberman expresses gratitude for Sobel’s contributions and approach to science
- •Sobel thanks Huberman for the closing words
