Huberman LabDr. Jennifer Groh on Huberman Lab: How Gaze Reshapes Hearing
Groh shows thoughts are multi-sensory simulations running in sensory cortex; the superior colliculus links gaze direction to sound maps, altering what you hear.
CHAPTERS
- 0:00 – 2:30
Defining Thought as Multi-Sensory Simulation
Groh introduces a theory of thought as the brain running internal simulations in sensory and motor cortices. Using the example of thinking about a cat, she illustrates how concepts may be instantiated as visual, auditory, and other sensory replays. This framing explains everyday phenomena like why verbal distraction can impair visually demanding tasks such as driving in heavy traffic.
- •Thought may be implemented as simulations in visual, auditory, and sensory-motor cortex.
- •Concepts like “cat” likely trigger multi-sensory activity (appearance, sound, even smell).
- •Cognitive resources are shared across modalities; a demanding visual-motor task (merging in traffic) competes with processing speech.
- •This theory links subjective experience, psychology, and neural implementation into a unified account of thinking.
- 2:30 – 8:30
Introduction to Groh’s Work and Sensory Integration
Huberman formally opens the episode, introduces Groh’s background, and frames the importance of multi-sensory integration for perception, attention, and learning. He highlights Groh’s clarity in explaining what thoughts are at the neural level and previews applications for focus and smarter thinking.
- •Groh is a Duke neuroscientist studying how the brain represents space and merges different senses.
- •Eye movements are central to what we pay attention to and what the brain can do at any moment.
- •The episode will cover how thoughts are built, how to control them, and how to improve learning.
- •Huberman emphasizes the podcast’s educational mission and independence from his Stanford roles.
- 8:30 – 17:00
Where Vision and Hearing First Meet: Superior Colliculus and Moving Maps
Groh recounts how a college seminar introduced her to the superior colliculus, a midbrain structure where visual and auditory information interact. She explains dynamic receptive fields that shift with eye position, and how this challenges simple ideas of static brain maps. The conversation connects eye movement-induced map shifts to the stability of our experience despite rapid eye motion.
- •Superior colliculus neurons respond to both visual and auditory stimuli.
- •Auditory spatial receptive fields change with eye position, forming dynamic spatial maps.
- •This computation is simple on paper (adding angles) but nontrivial for brain circuits.
- •Despite large retinal image shifts with each eye movement, perception feels stable, implying substantial under-the-hood processing.
- 17:00 – 31:00
Everyday Sensory Integration: Phones, Screens, and Ventriloquism
Using examples like reading on a train, watching movies, and ventriloquism, they show how the brain flexibly binds or separates sights and sounds. The brain expects audio-visual events to co-occur in plausible ways and will often trust vision over raw auditory cues, even when sound physically comes from elsewhere.
- •When engrossed in a phone or laptop, the brain collapses the multisensory world into a small spatial box.
- •We automatically re-expand our spatial mapping when unexpected sounds (e.g., a conductor asking for a ticket) occur.
- •In films and with earbuds, sound rarely originates from where we see mouths moving, yet we still localize it to faces.
- •Ventriloquists exploit visual dominance and timing to reassign perceived sound source from their own mouth to a puppet.
- •These effects highlight learned probabilistic rules for when to merge or separate sensory cues.
- 31:00 – 46:00
How the Brain Locates Sounds in Space
Groh explains the physics and neurobiology of sound localization, focusing on interaural timing and level differences and the role of ear anatomy. She notes that babies must relearn localization as their heads grow, that one-eared hearing complicates but does not abolish localization, and that precise neural machinery is required to resolve sub-millisecond delays.
- •Key cues: interaural time differences (up to ~0.5 ms), interaural level differences, and spectral filtering by ear folds.
- •Head size changes in development alter timing cues, requiring continuous recalibration through learning.
- •Even with one deaf ear, people can localize somewhat using spectral cues from pinnae and reflections.
- •Detecting sub-millisecond timing differences requires specialized neural circuits with high temporal precision.
- •Damage or flattening of ear folds (e.g., from wrestling) likely impacts localization but may be partially compensated over time.
- 46:00 – 1:02:00
Self-Voice, Hearing Safety, and Bone Conduction
They explore why our recorded voice sounds strange to us, the brain’s active dampening of self-generated sound, and the role of bone conduction. The conversation expands into practical hearing health, including volume thresholds, hearing loss, and the differences between ‘inside-the-head’ headphone listening and external speakers.
- •Recorded voices sound odd partly because recordings miss full frequency content, and we normally hear ourselves via air and bone conduction.
- •The brain actively turns down auditory gain via middle ear muscles and cochlear mechanisms before we speak.
- •High-volume headphone use, especially if others can hear your audio, likely causes permanent hearing damage.
- •Noise-cancelling or bone-conduction devices can help reduce the need for dangerous volume levels.
- •Spatial audio through headphones requires recreating timing, level, and ear-specific filtering, which is hard because ears are idiosyncratic.
- 1:02:00 – 1:17:00
Echoes, Distance Cues, and Constructed Auditory Reality
Groh discusses how the brain infers distance and environment from loudness, thunder/earthquake timing, and room reflections. Rather than hearing multiple copies of sounds from walls, ceilings, and tables, the brain combines them into a single percept, using delays and frequency changes as subtle distance and room-size cues.
- •Distance cues include loudness (relative to expected source volume) and delay between direct sound and echoes.
- •Thunder and lightning timing is a natural example: seconds between flash and boom map to distance.
- •In rooms, multiple reflected copies of a voice arrive at different times, yet are integrated as one sound unless delays get very long (true echoes).
- •Surfaces (carpet vs. hard floors vs. high ceilings) fundamentally alter perceived clarity and spaciousness.
- •Humans likely also pick up low-frequency vibrations through structures (e.g., floor or bones), akin to elephants using ground vibration.
- 1:17:00 – 1:42:00
Frequency, Emotion, and the Evolutionary Role of Music
The discussion shifts to music’s universality and unclear evolutionary function. Groh highlights rhythm as the core invariant across cultures and introduces the idea that coordinated loud rhythmic behavior could have evolved to amplify group presence. They link this to modern military bands, war dances, and political or sporting rituals like the Maori haka.
- •Music is universal, yet its adaptive purpose is less obvious than language.
- •Rhythm underpins melody and harmony; without stable timing, tunes become unrecognizable.
- •A prominent theory: rhythm and group sound enable synchronized group action to intimidate predators/competitors (e.g., scaring off hyenas).
- •Music may also be a sexual selection signal, akin to peacock tails—attractive traits that don’t directly enhance survival.
- •Modern examples: haka in rugby, political chants, and concerts show music’s power to convey intention, unity, and vigor.
- 1:42:00 – 1:50:00
Music, Memory, and the Brain’s Use of Song Structure
They examine how melodies and rhythmic structure dramatically aid memory, from the ABC song to professional songwriters recalling lyrics. The brain appears to use the first words of a verse as a retrieval cue that triggers the rest, showing how musical structure scaffolds language and sequential recall.
- •Songs make long sequences of words far more memorable than prose of the same length.
- •Musicians often need only the first two or three words of a verse to recall the entire lyric sequence.
- •The brain likely stores and retrieves language within a musical-temporal framework.
- •This structure can be leveraged for learning lists, concepts, or educational material via melody and rhythm.
- 1:50:00 – 2:07:00
Beyond the Superior Colliculus: Eye Position in the Auditory Pathway and the Ear
Groh describes her lab’s search for where eye position is integrated into auditory processing. Finding eye-movement effects in multiple auditory brain regions, they hypothesized that modulation might reach all the way to the ear itself. Using microphones in the ear canal, they discovered eardrum movements synchronized to eye saccades, suggesting an ultra-early stage of multimodal alignment.
- •Eye position modulates auditory responses not just in superior colliculus but throughout the auditory pathway.
- •The middle ear muscles and outer hair cells are controlled by descending brain signals and can move the eardrum.
- •By measuring otoacoustic emissions via ear-canal microphones, Groh’s lab showed that every saccade drives tiny, phase-opposed eardrum motions in the two ears.
- •These eardrum movements carry information about eye movement direction and magnitude and may be the first step in visual–auditory spatial integration.
- •This challenges the strict bottom-up model of hearing and shows strong top-down influence even at the sensory periphery.
- 2:07:00 – 2:23:00
Architectures of Sound: Grand Central, Cathedrals, and Room Design
Huberman brings up acoustic phenomena in Grand Central Terminal and churches to illustrate how architecture channels sound. Whispering galleries and high-ceiling cathedrals create dramatic effects through geometry and materials, changing reflections, delays, and reverberation patterns. Groh explains how these physical features alter perception, intelligibility, and the kind of music that works well in such spaces.
- •Whispering galleries exploit curved surfaces so quiet speech can travel long distances along ceilings or arches.
- •Such experiences violate naive expectations that quiet sounds can’t travel far, revealing limits of our intuitive sound model.
- •High, reflective ceilings create long-delay reflections—great for slow, sustained music (e.g., Gregorian chant) but problematic for rapid note changes.
- •Carpet, wood, and stone all shape sound by absorbing or reflecting different frequency ranges.
- •Room acoustics are therefore integral, not incidental, to how we perceive speech and music.
- 2:23:00 – 2:48:00
Thought, Attractor States, and the Mechanics of Focus
They zoom out from sensory systems to discuss cognitive dynamics: how thoughts cluster, why free association is constrained, and how focus emerges and deepens into ‘flow’. Huberman introduces an attractor-state metaphor (ball bearing in progressively deeper trenches) to describe how context and time stabilize mental states, and both explore why most people misjudge their own attention capacities.
- •Free thought isn’t random; people struggle to generate words unrelated to recent context, suggesting strong attractors.
- •Huberman’s attractor model: initial cognition is shallow and easily perturbed; sustained engagement digs a deeper ‘trench’ of focus.
- •Phones and abundant stimuli continually jostle the ball bearing, preventing deep attractor formation.
- •Flow states are marked by loss of self-consciousness and immersion, often requiring uninterrupted time and restricted inputs.
- •Most people who think they ‘have no attention’ often haven’t engineered conditions (sensory constraints, time, environment) to reach deep focus.
- 2:48:00 – 3:07:00
Practical Attention Management: Interval vs. Endurance Cognition
Groh describes her own working style as cognitively interval-based: write a sentence, then briefly check something, then return, using those micro-breaks as background incubation time. They compare mental work to sports—some are sprinters, others endurance types—and note how acetylcholine and norepinephrine shape the attentional spotlight. They stress designing individualized systems instead of chasing an unrealistic ideal of nonstop concentration.
- •Groh accepts that deep writing feels effortful and uses short, structured breaks instead of fighting her process.
- •Background processing continues during breaks; clarity for the next sentence often emerges after stepping away.
- •High-stakes pressure (e.g., large rewards) can cause ‘choking’ by over-recruiting motor units and disrupting fluid execution.
- •Cognition mirrors sport: sprinter-like short bursts vs. marathon-like sustained effort, both valid if matched to tasks.
- •Neurochemicals like acetylcholine (spotlight) and norepinephrine (arousal) are finite resources, replenished largely in sleep.
- 3:07:00 – 3:24:00
Chickens, Vergence, and Visual Control of Brain State
A seemingly whimsical detour into Groh’s backyard chickens turns into a concrete demonstration of vision’s power over attention. They discuss farm tricks for ‘hypnotizing’ chickens by drawing a line from their beak, which likely exploits vergence eye movements to lock a narrow focus cone. Huberman connects this to human practices in classrooms and to the broader idea that eye position and focus geometry (near vs. panoramic) strongly modulate global brain state.
- •Farmers can ‘hypnotize’ chickens by placing their beaks on a line in the dirt and drawing the line, after which the birds remain fixated.
- •Birds with lateral eyes converge their gaze during pecking to create a precise visual cone for targeting seeds or insects.
- •Human protocols in some countries use fixed-point staring before lessons to improve subsequent attention, mirroring this principle.
- •Near, convergent focus is associated with narrow, high-arousal attention; panoramic viewing is linked to relaxation.
- •Vision is not just input; it is a direct control knob on arousal and attentional system configuration.
- 3:24:00
Phones, Partial Disconnection, and Designing a Modern Cognitive Environment
The episode closes with a frank look at smartphones and constant connectivity. Groh and Huberman reject simplistic ‘just get rid of phones’ advice and instead discuss realistic strategies: controlled accessibility, using phones for bounded tasks, creating off-grid workspaces, and outsourcing some monitoring of the world to others or to simple tools. They emphasize defining what you want from the phone at any given moment and avoiding endless, frictionless engagement loops.
- •Groh relies on a satellite communicator on wilderness trips so she can relax knowing she’s reachable for true emergencies.
- •Huberman physically segregates social media to a separate device to reintroduce access friction and prevent seamless scrolling.
- •The riskiest digital experiences have seamless on-ramps to full attention and no natural endpoints (endless feeds).
- •Groh builds early-morning routines around bounded apps (Duolingo, single-round games) that have clear completion points.
- •Both advocate phone-free schools or stricter school policies while recognizing adults need more nuanced, context-specific containment, not total abstinence.
- •The overarching principle: engineer your sensory and digital environment so it serves your cognitive goals instead of exploiting your attention.
