Sensory Systems · Skim version

Sensory Systems

About 37 min to read · 30 lessons

Each lesson in order: its guiding idea, then the take-homes from its films. All 30 have film take-homes.

Back to the course page →

Only this unit will print. Press Ctrl+P or ⌘P to print it. Show the whole course again

Transduction and sensory cell types

Print this unit

1. Five Ways to Turn a Stimulus into a Current

Guiding idea. With no enzyme between stimulus and pore, a current can start as fast as the channel changes shape; with enzymes between, one event changes thousands of molecules, usually at a cost in time.

First film Nothing in Between, or Enzymes in Between

  • Every sensory cell turns a stimulus into a current, and what sits between the stimulus and the pore sets what the cell can do.
  • Where no enzyme sits in between, as in a hair cell, the current can start very fast; in turtle hair cells channels can activate with a time constant as short as about 50 microseconds.
  • Where enzymes sit in between, as in a rod, one stimulus event can change thousands of molecules, usually at a cost in time.
  • Ask two questions of every sensory cell: what sits between stimulus and pore, and how does the signal leave the cell?

Second film Forty Microseconds: What a Fast Number Measures

  • The 40-microsecond delay is a tissue response latency, the corrected summed response of the bullfrog sacculus at 22 degrees C, not the opening time of one channel.
  • Channel activation was measured directly later, with time constants as short as about 50 microseconds in turtle hair cells.
  • A tissue latency, a channel time constant, a molecular gain and a behavioural latency are different kinds of number, so check what kind before comparing two speeds.

Third film Where the Gain Sits, and What It Costs

  • In mouse rods, about 2000 cGMP molecules are hydrolysed per absorbed photon, so a rod cascade can multiply one event many times.
  • The receptor step need not amplify: in frog olfactory neurons one bound receptor rarely activates even one G protein, and the gain is thought to come at later steps.
  • The price of a cascade is time, because every rung is a molecular encounter or an enzyme turning over, and a cascade can also amplify spontaneous receptor activity.

2. Piezo Channels and the Physics of Force

Guiding idea. Tension pays for area: a channel that takes up more membrane when open is pushed open by membrane tension, and in a two-state model the area gained sets the steepness of the switch.

First film Which Protein Carries the Poke?

  • Silencing a gene in a cell line showed it was required for the poke-evoked current there, not that its protein is the channel.
  • Expressing the Piezo genes in other cultured cells gave mechanically activated currents, which showed sufficiency in those cells.
  • Necessity, sufficiency and pore identity are three separate claims, and later work on purified protein and structures earned the third.

Second film A Channel Shaped Like a Dome

  • Mouse PIEZO2 has 38 transmembrane helices per subunit and 114 per channel, and the two numbers are kept apart.
  • In the leading model, still debated, the curved blades bow the membrane into a dome.
  • For an idealized smooth cap, flattening gains about 310 square nanometres of projected area, which depends on how deep the dome is, not how wide.

Third film Tension Pays for Area

  • Tension pays for area: if a channel takes up more membrane when open, tension lowers the energy of the open state by tension times the area gained.
  • In a two-state model, the half-activation tension is the resting preference for closed divided by the area gained, and the slope factor is thermal energy divided by the area gained.
  • More area gained means a steeper switch.

Fourth film Reading a Patch

  • Read as a two-state switch, the slope factor of a current-tension fit implies an effective area.
  • For mouse Piezo1 in a human cell line the fit implied about 5 square nanometres, an effective area that can underestimate the true one.
  • A channel's sensitivity depends on the tension it was already under, since a prepulse lowered the half-activation tension.

Fifth film The Sixty-Fold Gap

  • Cryo-EM showed about 300 square nanometres of in-plane expansion, but a two-state reading of the current gives about 5, so the slopes differ about sixty-fold.
  • The dome may not be a two-state switch, patch curves may pool uneven tensions, or the flattened class may not be the open channel.
  • Which part of the structural expansion couples to opening is still open.

Sixth film Lipid, Tether and People

  • Purified mouse Piezo1 in droplet bilayers was activated by force, so protein plus lipid is sufficient there, which does not exclude a tether route in intact cells.
  • Two PIEZO2 variants from distal arthrogryposis families both sped recovery from inactivation, but only one slowed inactivation during the pulse, so each variant must be read on its own.
  • Mouse and human loss-of-function findings should be kept apart: two people without working PIEZO2 had selective deficits in touch and limb position.

3. Sensory Neurons by Their Genes: the Atlas Era

Guiding idea. Single-cell sequencing refined the broad fibre classes of mouse dorsal root ganglion neurons into many molecular subtypes: eleven classes in 2015, seventeen in 2018 and eighteen harmonized subtypes across six species in 2024.

First film From Cluster to Cell Type

  • Atlas cluster counts depend on the data set, resolution, preparation and cell state.
  • A transcriptomic cluster is a candidate type, not a functional conclusion.
  • Genetic access lets researchers connect a candidate cluster to anatomy or physiology.
  • An independent anchor strengthens a cluster's claim to be a functional type.

Second film When Identity Changes

  • Subtype identity emerges as broad transcription factor expression narrows during mouse development.
  • After peripheral nerve injury, mouse sensory neurons can suppress subtype markers and enter a shared injury state.
  • Subtype expression returns over a time course associated with axonal regrowth; timing alone does not prove cause.
  • Mouse markers and cell classes do not automatically transfer to human sensory neurons.

Only this unit will print. Press Ctrl+P or ⌘P to print it. Show the whole course again

Touch

Print this unit

4. The Low-Threshold Mechanoreceptors: One Touch, Several Signals

Guiding idea. Low-threshold mechanoreceptors (LTMRs) split gentle touch into separate signals.

First film One Press, Several Reports

  • Each class of low-threshold mechanoreceptor is most sensitive to a different part of a touch.
  • The ending and the neuron's own properties shape what a class responds to best, and axon size and myelin set how fast the report travels.
  • No class reports the whole touch, so the nervous system reads the pattern across classes.

Second film One Hair, Three Classes

  • One movement of an awl/auchene hair can be reported in parallel by three classes, at three speeds.
  • The combinations of endings repeat across the skin and differ between hair types.
  • A single neuron branches to many hairs, so endings at a hair are not a count of neurons.

Third film Your Own Skin: The Slow Stroke

  • In people, C-tactile afferents fire most in the slow stroke range, in an inverted U across stroking speeds.
  • These afferents support gentle, affective touch, but not every affective touch relies on them.
  • The human C-tactile ending is still unknown, so the mouse hair map cannot be transplanted onto human hair.

5. Merkel Cells: Touch Receptors That Are Not Neurons

Guiding idea. In mouse touch domes the ending mainly feels the change and the Merkel cell mainly holds the press.

First film Two Receptors, One Nerve

  • In macaque fingertips, SA1 firing carried the spatial layout of millimetre-scale features, with a dynamic onset burst and a static hold phase.
  • In mouse touch domes, the ending mainly feels the change and the Merkel cell mainly holds the press.
  • The word mainly matters, because rat whiskers give a different reading in which Merkel cells drive both phases.

Second film Take One Route Away

  • Taking one route away, the Merkel cell's, and watching which phase is left shows what each receptor contributes.
  • In mouse skin, light on Merkel cells was enough to make the afferent fire, and silencing them suppressed sustained firing during a held press.
  • Deleting epithelial Piezo2 left the onset similar and the hold weakened, but developmental loss of Merkel cells also lowered onset, so each route acts mainly, not only, on one phase.

Third film A Brief Current That Holds a Press

  • Compare charge, not peaks: in a thought experiment, a small steady current through a one-second hold carried most of the charge.
  • Irregular spike timing in the hold is a clue, not proof, that packets of transmitter help sustain firing.
  • How the Merkel cell signals to the nerve is still argued, with noradrenaline on beta-2 receptors in mouse touch domes and protons acting on ASIC channels in rodent whiskers.

6. Meissner and Pacinian Corpuscles: Flutter and Vibration

Guiding idea. A corpuscle reports change, not level, and the brevity of its response can arise in three places: the capsule, the terminal's channels and the spike generator.

First film Reporters of Change

  • Meissner and Pacinian corpuscles report change, not level: their afferents fire when the skin starts to move, stops or keeps moving, and go quiet while a press is held.
  • Meissner afferents serve low-frequency flutter and Pacinian afferents high-frequency vibration, and the best frequency differs between humans and mice.
  • The brevity of the response can arise in three places: the capsule, the terminal's channels and the spike generator.

Second film Peel the Onion

  • Brevity can arise in the capsule, the terminal's channels or the spike generator, and the classical account put the filtering in the capsule.
  • In cat corpuscles with most layers peeled, the generator potential decayed more slowly but the impulses did not persist, so brevity had two stages.
  • In the duck preparation, the terminal stayed fast without the outer core, so fast adaptation does not need the onion there.

Third film The Cells That Help It Feel

  • In mice, force opens Piezo2 in the sensory axon, but the wrapping cells also feel force and contribute to detection; how they normally do so is not yet known.
  • Switching off Pacinian wrapping cells in mice meant a vibration had to be about five times larger to be detected.
  • Mice without Meissner corpuscles needed about twice the push to detect a touch yet still detected stronger pushes, so these corpuscles help set sensitivity.

7. Hairy Skin, Stroking and Affective Touch

Guiding idea. In human hairy skin, the C-tactile (CT) afferent fires most at caress speeds and pleasantness ratings follow its curve, making it the prime candidate for the pleasant side of touch.

First film The Fibre That Prefers a Caress

  • In human hairy skin, the C-tactile afferent fires most at caress speeds of about 1 to 10 cm/s, while the myelinated touch fibres recorded fire harder the faster the brush moves.
  • Average pleasantness ratings follow the same inverted U, so C-tactile input is the prime candidate for the pleasant side of touch.
  • A candidate is not a proven cause: only taking the input away can show whether it is needed.

Second film Take the Input Away

  • In carriers of a nerve growth factor mutation with fewer thin fibres, caress-speed stroking was rated less pleasant, though the mutation affects more than C-tactile afferents.
  • After cordotomy, pain, temperature and itch were markedly disrupted but pleasantness ratings were not significantly changed.
  • So thin-fibre input seems to contribute and the anterolateral route is not needed, while C-tactile input remains a candidate with its route unresolved.

Third film A Net, a Lineage and a Mouse That Cannot Say

  • In mouse hairy skin, at least three afferent classes are candidates for detecting a stroke.
  • A field afferent may respond to gentle stroking by summing many weakly sensitive endings, as the authors proposed.
  • Mouse outcomes such as place preference are not feelings, and which mouse cell, if any, matches the human C-tactile afferent is unsettled.

8. Where Touch Is First Processed: Dorsal Horn and Dorsal Column Nuclei

Guiding idea. In mice, a touch reaches the gracile nucleus by two routes, a direct afferent axon that keeps precise timing and an indirect dorsal horn route, rebuilt by local circuits, that reports sustained pressure.

First film Two Routes, One Patch of Skin

  • A touch on the mouse hindpaw reaches the gracile nucleus by two routes: the afferent's own axon, and a route through the dorsal horn.
  • For thalamus-projecting gracile neurons, the two inputs had highly aligned receptive fields, so both report the same patch of skin.
  • The direct route keeps precise timing, while the indirect route reports sustained pressure, up to high forces.

Second film Silence One Route

  • Silencing one route at a time showed that the direct input carries vibration with high temporal precision.
  • The indirect route carries sustained pressure, and gentle displacement uses both routes.
  • The high-force result rests on a drug that is not specific to PSDC neurons, so it carries that limit.

Third film Why Later Is Fine, and What the Cord Does

  • In a simple model the indirect route arrives a few milliseconds later, but that lag is illustrative, not a measurement.
  • A fixed delay does not blur a route's own timing; random jitter at each synapse could, which would cost fast vibration more than a sustained press.
  • Losing feedforward or presynaptic inhibition enlarged the receptive fields of PSDC neurons in anaesthetized mice.

9. Touch Over-reactivity in Autism Models

Guiding idea. Sensory features are part of the diagnostic criteria for autism, and in mouse models carrying mutations in autism-associated genes a tactile phenotype can begin in the peripheral sensory neuron itself.

First film Delete Here, Learn Where

  • In these mouse experiments, a fixed air puff measured movement, not perception.
  • Gene loss in the Advillin-targeted population was sufficient for selected mouse outcomes.
  • Restoring Mecp2 mainly in that population rescued selected outcomes but left others unchanged.
  • A restricted manipulation can locate one contribution without explaining the whole phenotype.

Second film One Touch Behaviour, Several Routes

  • The same tactile endpoint can arise from different defects in these mouse models.
  • Mecp2 and Gabrb3 models support a reduced presynaptic GABA-A inhibition account.
  • The Shank3 model instead showed reduced Ih and greater sensory-neuron excitability.
  • Locus, mechanism and developmental timing must be tested separately.

Only this unit will print. Press Ctrl+P or ⌘P to print it. Show the whole course again

Temperature, pain and itch

Print this unit

10. Hot, Warm, Cool and Cold

Guiding idea. Skin temperature is read from a few afferent lines at once, and a cool line that fires at rest can report warming by slowing down.

First film A Fibre That Fires at Rest

  • Skin temperature is read from a few afferent lines at once, not one channel per feeling.
  • A cool line that fires at rest can report warming by slowing down.
  • In the mouse forepaw, losing a channel that supplies a line's firing can cost warmth, though the circuit that reads the lines has not been identified.

Second film Take Away the Cool Line

  • Without TRPM8, mice failed to report forepaw warming in this task, even after 10 days of training.
  • TRPM8 loss removes the cool fibres' resting firing in isolated skin, so warming has no ongoing firing left to silence.
  • Loss of heat channels blunted warmth detection, whereas TRPM8 loss abolished it in this task.
  • Whether gentle warmth rests mainly on warm fibres or on quieter cool fibres is not settled.

Third film Three Channels, One Burn

  • In mice, acute withdrawal from noxious heat survives the loss of any one of TRPV1, TRPM3 or TRPA1.
  • The three channels act in parallel: only mice lacking all three lacked acute withdrawal in the reported assays.
  • The same mice still preferred moderate temperatures, so withdrawal and preference are different behaviours.

11. Nociceptors and the Sodium Channels of Pain

Guiding idea. A sodium-channel change can alter a nociceptor's spike initiation, so what the neuron does depends on how many channels are available at rest, not only on how easily they open.

First film Three Families, One Gene

  • Three consanguineous families in northern Pakistan had members who never experienced pain, and the mutant SCN9A channels passed no current above background in cultured cells.
  • Linkage says a variant travels with a phenotype, and only a recording can say whether the channel conducts.
  • People with the same loss of function also cannot smell, and the blocker that reached patients targets a different channel.

Second film Three Channels and the Availability Budget

  • Nav1.7 and Nav1.8 contribute to the spike itself, while Nav1.9 mainly sets the resting potential and subthreshold excitability.
  • Availability is the fraction of sodium channels not already inactivated, and it falls as the resting potential becomes more depolarized.
  • Depolarization closes the gap to threshold but costs availability, so ask what happened to the channel, then to rest and availability, then to firing.

Third film The Same Gene, Too Much Pain

  • Variants in the same gene that alter gating can cause too much pain, as in primary erythermalgia.
  • Pain-associated variants can open too easily, close too slowly, or fail to inactivate.
  • The gating measurements came from channels in cultured cells, so a gating change still has to pass through the neuron before it reaches behaviour.

Fourth film Where More Current Breaks

  • More sodium current does not always mean more pain, because Nav1.9 carries a standing inward current near rest that depolarizes the neuron.
  • Gain-of-function Nav1.9 variants raised firing in one study and impaired it in another, so the current alone cannot predict what the neuron does.
  • The outcome depends on the size and time course of the depolarization, subtype-specific availability, channel densities, other conductances and the stimulation.

Fifth film Strong Nav1.8 Block, Still Firing

  • Strong Nav1.8 block generally reduced but did not eliminate repetitive firing in adult human sensory neurons recorded at 37 °C.
  • A channel-current calculation is not neuronal silence, and it is not a dose.
  • Suzetrigine beat placebo in two surgical trials, but neither showed superiority over the opioid comparator, and a failed superiority test is not equivalence.

Sixth film Labels, Classes and States

  • A molecular label, a conduction class and a functional state are different kinds of measurement, so nociceptor categories overlap.
  • In mice, removing one labelled population in adults raised mechanical thresholds while sparing heat and cold in the ranges tested, and a TRPV1-positive lesion did the opposite.
  • Deleting the same neurons from birth showed no detectable mechanical deficit, so a germline deletion is a developmental experiment.

12. Spinal Circuits That Gate Pain

Guiding idea. Inhibitory circuits in the spinal dorsal horn normally keep light touch from reaching pain pathways, and losing that control can make touch painful.

First film Who Opens the Touch Gate?

  • In mice, inhibitory dorsal-horn neurons can keep touch input from recruiting mechanical-pain relays.
  • Removing somatostatin-lineage relays reduces mechanical nocifensive behaviour in the assays tested.
  • Ablation removes cells, while silencing can remove transmitter output from surviving cells.
  • Behaviour identifies a failed gate; circuit measurements help locate the failure.

Second film When Inhibition Changes Sign

  • In adult male rats, nerve injury lowered KCC2 and shifted lamina I anion reversal.
  • A synapse can remain present while its postsynaptic effect changes.
  • A depolarizing anionic response is not automatically excitatory because it can still shunt current.
  • Cell counts, evoked output and gramicidin recordings test different failure points.

13. Itch: A Sense of Its Own

Guiding idea. Itch is a sense of its own: identifiable sensory neurons detect itch-producing stimuli, histamine is only one of several triggers, a peptide relay in the dorsal horn carries the signal on, and the pathway overlaps with pain without being weak pain.

First film Which Line Carries the Itch?

  • Deleting a receptor route and ablating its whole neuron population answer different questions.
  • MrgprA3-marked neurons receive several itch-producing inputs in the mouse assays shown.
  • Restricted rescue shows what those neurons can drive when they alone respond.
  • The behavioural assays distinguish scratching from wiping, not a reported sensation.
  • Each perturbation supports only the link and assay it tests.

Second film Order the Relay, Find the Brake

  • Downstream rescue and bypass experiments can order functions in a mouse spinal relay.
  • The supported order is sensory Nppb, NPRA cells, GRP and GRPR.
  • Functional order does not prove direct synapses or one source of GRP.
  • Removing spinal inhibitory populations can raise scratching without strengthening sensory input.
  • Chemical and mechanical scratching can use partly different gates.

14. When Pain Persists

Guiding idea. After a nerve injury, pain can persist because the rules of the system change, and the alarm wiring is part of what changes.

First film A Touch That Hurts

  • After a nerve injury, pain can persist because the rules of the system change, and the alarm wiring is part of what changes.
  • In rodent models, the rodent sign of allodynia is a withdrawal threshold that falls and stays low.
  • Most of this comes from rodent nerve-injury models; in people the evidence is association.

Second film How Microglia Let a Touch Through

  • In rats, nerve injury lowers KCC2 in lamina I neurons and shifts the anion reversal potential up, which weakens inhibition without turning it into excitation.
  • In rats, a chain of tested links runs from microglial P2X4 to BDNF, TrkB and that shifted potential.
  • In mice, the immune route to hypersensitivity can differ with sex, so ask which cell, in which sex, in which compartment.

Third film A Reflex Is Not the Hurt

  • A withdrawal reflex is easy to mistake for pain, because the reflex and the affective behaviour are two measurements that can come apart.
  • In male mice, silencing a basolateral amygdala ensemble reduced escape and avoidance while withdrawal reflexes were preserved.
  • In people, the evidence is prediction and association: baseline prefrontal-accumbens connectivity predicted persistence, which is not a cause.

Only this unit will print. Press Ctrl+P or ⌘P to print it. Show the whole course again

Proprioception and interoception

Print this unit

15. Proprioception: Sensing the Body's Own Position

Guiding idea. Limb position is an estimate built mainly from muscle spindle length signals, which carry no label of cause, so vibrating a tendon of an unseen, still limb usually biases it toward stretch of that muscle.

First film Read as Stretch

  • Limb position is an estimate built mainly from muscle spindle signals of muscle length.
  • A spindle's firing carries no label of its cause, so vibrating one tendon of a still, unseen limb usually biases the estimate toward that muscle being stretched.
  • When proprioceptive input is severely impaired, accurate movement depends heavily on vision.

Second film The Channel on the Line

  • Deleting Piezo2 with two Cre lines, neither specific to proprioceptors, sharply reduced stretch-evoked firing in mouse muscle, so Piezo2 is the principal transducer on the stretch line.
  • A small stretch response remained, which may be Piezo2 left by incomplete deletion; it neither proves nor rules out a second channel.
  • How a channel whose current fades within milliseconds in cultured cell bodies supports a held stretch is still open.

Third film When the Line Goes Quiet

  • When proprioceptive input is lost, vision does much of the work of knowing where the arm is, and without sight reaching becomes far less accurate.
  • Two people with inactivating PIEZO2 variants and a man without large-fibre sensation show this, though neither case silences spindles alone.
  • In mice, deleting Piezo2 with a parvalbumin driver curved the spine in some animals by postnatal day sixty, and how the signal reaches the skeleton is not known.

16. Vagal Sensory Neurons: The Body Reports to the Brain

Guiding idea. To learn what a vagal sensory population does, ask three separate questions: what stimulus the cells answer, what channel turns that stimulus into current, and what output changes when you drive or remove them.

First film Three Questions, Not One

  • To learn what a vagal sensory population does, ask three separate questions: what stimulus it answers, what channel turns that stimulus into current, and what output changes when you drive or remove it.
  • Evidence for one question does not by itself answer the others.
  • A genetic label is a handle on a population, not an answer about its stimulus, channel or output.

Second film Two Ways to Stop Breathing

  • Light on P2ry1 afferents suppressed breathing and light on Npy2r afferents produced rapid shallow breathing in anaesthetized mice; this shows what each population can do when driven, not what drives it in life.
  • Airway protection and lung inflation are separate routes to a similar output, and most stretch-responsive neurons were Piezo2 positive and P2ry1 negative.
  • An output alone does not identify which pathway produced it, so measure stimulus and output separately.

Third film Which Channel, and What the Test Can Prove

  • Deleting Piezo2 in adult mouse sensory neurons impaired the lung inflation reflex.
  • For the baroreflex, deleting Piezo1 or Piezo2 alone showed no statistically detectable difference, while deleting both essentially abolished the drug-evoked reflex.
  • That shows the channels are required collectively in the targeted population, not that each neuron carries both.

17. Sensing the Gut: Enteroendocrine Cells and Nutrient Signals

Guiding idea. Cells in the lining of the intestine, which are not neurons, sense chemicals in the gut and pass the signal to sensory nerves.

First film How to Prove a Gut Synapse

  • Anatomy can show a gut cell contacts a nerve, but it cannot identify the transmitter.
  • A fast current in a cocultured vagal neuron tests transmission directly.
  • A glutamate-receptor blocker that silences the current, followed by washout recovery, supports glutamate transmission.
  • In mice, silencing and activating CCK-labelled gut cells test necessity and sufficiency for a vagal sugar response.

Second film How the Gut Teaches Preference

  • In mice, learned sugar preference can develop without sweet-taste signalling.
  • Initial attraction and learned preference are different measurements.
  • Intestinal SGLT1 function is needed for the gut sugar signal in these mouse studies.
  • Sugar and fat have separable receptor requirements and partly distinct vagal populations.

Only this unit will print. Press Ctrl+P or ⌘P to print it. Show the whole course again

Vision

Print this unit

18. Parallel Channels Out of the Eye

Guiding idea. In the mouse, the optic nerve carries more than 40 overlapping, differently filtered maps of the scene: one ganglion cell type, one mosaic, one map.

First film Many Maps, Not One Picture

  • The optic nerve carries many overlapping, differently filtered maps of the scene, not one picture.
  • In the mouse, each of more than 40 ganglion cell types reports the scene through its own filter, built by retinal circuits before any spike leaves the eye.
  • One type, one mosaic, one map: the cells of a type usually keep their distance from one another.

Second film The Cell That Ignores the World Moving

  • A W3 cell fires for a small object moving over a still background and falls nearly silent when the whole scene moves.
  • Part of that silencing is presynaptic and part is postsynaptic, and the VG3 amacrine cell supplies much of the excitation.
  • The JAM-B cell is a second channel: in isolated retina, its lopsided cells preferred movement from the cell body toward the dendrites.
  • Calling W3 cells alarm neurons for overhead predators is a hypothesis, because the experiments used isolated retina.

Third film One Type or Two? Four Ways to Count

  • A real type usually spaces its cells out in its own mosaic, so merging two types usually makes spacing patchy and coverage higher.
  • A mosaic is strong support for a type, not a strict rule, because some real types are only weakly regular.
  • Imaging, RNA sequencing, spike recording and electron microscopy gave different counts because each reflects a method, a preparation and a resolution, not rival answers.
  • The macaque fovea needs its own census, since midget clusters made up about 85% of sampled foveal ganglion cells.

19. Computing Motion in the Retina

Guiding idea. In mouse and rabbit retina, each starburst dendrite prefers outward motion, and a direction-selective ganglion cell is wired mainly to the dendrites pointing in its null direction, so null-direction motion mostly drives inhibition.

First film A Veto From the Null Side

  • A direction-selective ganglion cell fires to its preferred direction and little to its null direction, and Barlow and Levick found a deficit for null-order flashes, so the main operation they inferred is a veto.
  • Each starburst dendrite is its own small direction detector, most active when motion runs outward from cell body to tip.
  • The dendrites that give a ganglion cell its strongest inhibition prefer roughly the direction that cell ignores.

Second film Reading the Wiring

  • In rabbit starburst cells, dendritic calcium was larger for outward motion, while the somatic voltage did not show that preference, so the branch is the unit.
  • In mouse retina, most starburst input to a direction-selective ganglion cell came from branches pointing roughly in its null direction.
  • An electron microscopy contact shows wiring, not synaptic strength.

Third film Built by Strengthening, and Where the Idea Stops

  • In paired recordings in mouse retina, null-side inhibition grew while preferred-side inhibition stayed about constant, so development strengthened one side and did not delete the other.
  • The authors concluded that null-side pairs had more inhibitory synapses, not stronger single synapses.
  • When only starburst GABA release was cut in mice, inhibition lost its direction tuning but selectivity survived in a subset of ON-OFF cells, so tuned inhibition is important but not indispensable.

20. Melanopsin Cells: Light for the Clock and the Pupil

Guiding idea. A mouse with no working rods or cones still sets its body clock and narrows its pupils.

First film A Blind Mouse That Knows It Is Day

  • A small set of retinal ganglion cells, the ipRGCs, make their own light response with melanopsin and also relay rod and cone signals.
  • In mice, these cells are the main route by which light reaches the clock and the pupil.
  • Each ipRGC pools two light signals into one output.

Second film The Light Response That Survived the Block

  • A ganglion cell that still responds to light once rod and cone synapses are blocked is most likely making its own response.
  • The cell's own response is usually slow to start, needs brighter light and is long-lasting.
  • Slow and sustained suits a cell that reports how bright the world is, not what is in it.

Third film Two Inputs, One Exit

  • An ipRGC pools the rod and cone relay and its own melanopsin response into one output.
  • In M1 cells, melanopsin signals through a G protein and phospholipase C beta 4 to open TRPC6 and TRPC7 channels, which depolarizes the cell.
  • Light hyperpolarizes a rod but depolarizes an M1 cell.

Fourth film Remove the Pigment, or Remove the Cell

  • In mice, either input alone is enough to entrain the clock and to constrict the pupil, and losing both leaves nothing.
  • Removing the ipRGCs gave larger deficits than removing melanopsin alone, close to the triple mutants.
  • A single knockout shows what the other input can do alone, and it does not show each input's share.

Fifth film Separate Cells for the Clock and the Pupil

  • The clock and the pupil depend on partly different ipRGCs.
  • Removing only the Brn3b-positive cells severely impaired the pupil reflex but spared photoentrainment.
  • The names M1 to M6 are for mouse cells, and they are not six matching human types.

Sixth film Beyond the Clock: Mood, Learning, Seeing and People

  • In mice, ipRGCs contribute to mood-like and learning effects of abnormal light cycles, through separate outputs.
  • Whether a matching circuit exists in humans was not tested.
  • Two blind adults show that human light responses can run without rods and cones, but that is an existence proof and not a population result.

21. Calcium Feedback Sets Photoreceptor Gain, and When Photoreceptors Die

Guiding idea. Calcium feedback turns a photoreceptor's gain down so that a steady background switches off less of its limited dark current, keeping current in reserve and trimming each new response.

First film A Current That Light Can Use Up

  • Light switches off part of a photoreceptor's dark current, so the current still flowing limits how big the next response can be.
  • Calcium feedback turns the gain down as the background brightens, keeping current in reserve and trimming each response, mainly in mouse rods and cones.
  • Over a range of bright backgrounds, sensitivity falls roughly as one over the background, so the peak response to a small, brief flash of fixed contrast and duration stays about the same.

Second film Cut the Loop

  • Without the GCAPs, a mouse rod's single-photon response in darkness is about five times larger, and on a background its adaptation is severely compromised.
  • The calcium feedback trims each response and keeps current in reserve, so cutting it leaves the cell short of current sooner.
  • Some adaptation survives deleting the GCAPs and recoverin, and what supplies it is still unresolved.

Third film Rods First, Then Cones

  • In typical retinitis pigmentosa, cones die too, usually after most nearby rods have gone, although the faulty gene is often not expressed in cones.
  • Why cones die is not settled; lost rod support through RdCVF is one candidate, shown in chicken cone cultures and a mouse model, not in people.
  • A randomized trial supports a benefit on its mobility test, while a small single-arm study supports observed safety and a preliminary signal only.

Only this unit will print. Press Ctrl+P or ⌘P to print it. Show the whole course again

Hearing and balance

Print this unit

22. The Hair Cell's Transduction Channel

Guiding idea. The mean current from a hair bundle is I = N × i × P_open, so losing a protein that any factor needs silences the cell without saying which factor was lost.

First film One Current, Three Factors

  • The mean current from a hair bundle is the product of the number of working channels, the current through one open channel and the open probability.
  • A lost current cannot tell you which of those three factors was removed.
  • Evidence about permeation, a residue reached in a working channel and activity of the protein on its own speaks more directly to the pore, and the case rests on several rungs together.

Second film Change It While It Works

  • Loss and rescue of TMC1 and TMC2 put them in the pathway but cannot separate a pore from its helpers.
  • A point mutation that changes which ions pass, as in the Beethoven mouse, narrows the candidates, though a partner beside the pore can also change what passes.
  • A reagent that alters permeation in a channel already carrying current rules out altered synthesis or delivery, but the native partners were still present.

Third film A Pore Inside a Complex

  • The native worm TMC-1 complex solved by cryo-EM held two copies each of TMC-1, CALM-1 and TMIE, and the human complex in that paper was a model.
  • In mice, without CIB2 or TMIE, TMC1 was not detected in the stereocilia, and TMIE is needed for more than getting TMC1 there.
  • The current you measure belongs to the complex, not to TMC1 alone.

23. The Cochlear Amplifier and Otoacoustic Emissions

Guiding idea. In a living mammalian cochlea, outer hair cells add a gain that is large for soft sounds, small for loud ones and gone after death.

First film A Gain That Death Erases

  • The cochlear amplifier is the gain of the living cochlea over the dead one, and in mammals it depends on outer hair cells.
  • That gain is large for soft sounds, small for loud ones and gone after death.
  • Sensitivity, sharp tuning, compressive growth of motion and the returning sound are four signatures of that one gain.

Second film From a Bent Bundle to a Pushing Cell

  • The outer hair cell turns sound into force through a chain: bundle bending, transduction current, voltage, prestin shape change, and a push on the organ of Corti.
  • Nonlinear capacitance is the electrical fingerprint of prestin's charge movement, and it does not measure the force or power a hair cell produces.
  • In a prestin knock-in mouse amplification was lost while transduction and cell mechanics were normal, so cell-body motility is necessary; how the work is split with the bundle in mammals is not established.

Third film The Sound That Comes Back

  • A fast nonlinearity in a living ear makes new tones, such as 2f1 minus f2, that a microphone in the ear canal can record.
  • At screening levels the distortion product fails when outer hair cells fail, which is what ties it to the amplifier.
  • A passed emission screen shows outer hair cells can make an emission, not that hearing is intact, because auditory neuropathy can present with a normal emission.

24. Hidden Hearing Loss, Gene Therapy and Regeneration

Guiding idea. A hearing test can come back normal while a cochlea has lost many of its synapses.

First film A Threshold Is Not a Count

  • In these mice, a cochlea lost many synapses and still passed a hearing test.
  • A threshold depends mostly on the most sensitive fibres and does not count them.
  • A normal threshold is good news about sensitivity, but it is not a census of synapses.

Second film The Link That Must Fire, and How to Put It Back

  • A knockout shows that a protein is needed, and it does not by itself show that the protein is the sensor.
  • In severe otoferlin deafness the inner hair cells are present, but their synapses release almost nothing.
  • Replacing a missing link can work because the cells that need it are still alive.

Third film Reading the Trial, and When the Cell Is Gone

  • Nine of twelve is a share of children meeting an endpoint, not a share of hearing.
  • When the hair cell itself is gone, a new hair cell has to be made, and the adult mammalian cochlea does not do this.
  • In newborn mice, reprogramming works less well a week later, so age is a barrier.

25. Balance: Two Codes for Every Head Movement

Guiding idea. The vestibular nerve reports each head movement on two overlapping channels, whoever caused it: a steady, precise firing rate and a higher-gain, well-timed signal.

First film Report It Twice

  • The vestibular nerve reports each head movement on two overlapping channels, whoever caused it.
  • In the canals, regular fibres follow how fast the head is turning in a steady rate, while irregular fibres add higher gain, more response to how quickly the motion changes, and well-timed spikes.
  • In the head-posture neurons tested, your own turn is largely cancelled when neck feedback matches the prediction, which may help the head reflex spare the turns you intend.

Second film A Synapse That Does Not Wait

  • Type I hair cells release glutamate from vesicles, as type II cells do, though sustained release in mature mouse tissue was about tenfold smaller.
  • In mouse tissue in a dish, type I cells lacking otoferlin or a needed calcium channel could still make the calyx fire.
  • Cleft potassium and cleft potential can move the calyx without vesicles; the animal recordings are measurements, while the model's order of events is a prediction.

Third film Pass On the Surprise

  • In the neurons tested in monkeys, the response to a head turn is cancelled only when the turn is self-made and neck feedback matches the motor command's prediction.
  • This is not the ear switching off: canal afferents kept firing, so any cancelling happens in the brain.
  • A vestibular implant supplies a motion-modulated rotation signal without the nerve's natural two channels or an otolith signal, and in a small cohort hearing in the implanted ear fell in all but one.

Only this unit will print. Press Ctrl+P or ⌘P to print it. Show the whole course again

Smell and taste

Print this unit

26. One Receptor per Neuron: How Olfactory Neurons Choose

Guiding idea. In mouse olfactory neurons, one rule carries the lesson: silence every receptor gene, let a few compete, and lock in the winner.

First film One in Two Thousand

  • A mature mouse olfactory neuron expresses one receptor gene from one allele, out of about two thousand candidates, and keeps using it.
  • A simple model of rare, independent, unregulated escapes from silence gives at most about 37 percent of neurons with exactly one receptor, so it is insufficient.
  • The rule is to silence every gene, let a few compete, and lock in the winner.

Second film Silence First

  • Receptor genes carry repressive marks, and the silence appears to come before detectable receptor expression, so the winner did not start the broad silence by itself.
  • Comparing the two alleles of one receptor gene is an association, so it does not say which change starts the choice.
  • Early loss of LSD1 left about a third as many receptor genes detected, which shows dependence, not mechanism.

Third film Hubs That Compete

  • Enhancers from different chromosomes form hubs, and LHX2 and EBF help build them.
  • Single-cell data inferred several competing hubs in developing neurons before one gene wins.
  • Receptor RNA suppressed competing receptors at a distance, and translation of the receptor then stabilized transcription at the winning hub.

Fourth film The Lock

  • In the model, the receptor made by the winning gene closes the lock, so a neuron whose chosen gene cannot make a working receptor often expresses a second one.
  • The lock runs through the unfolded protein response, Perk, ATF5 and adenylyl cyclase 3, which helps turn down LSD1 and makes later escapes unlikely.
  • A lock need not be perfect, because other steps such as the RNA competition may still limit expression.

Fifth film The Choice Becomes an Address

  • In mice, receptor identity is instructive in axon targeting but not the only factor.
  • Receptor-derived cyclic AMP signals and receptor-dependent stress signals link to guidance and adhesion gene expression.
  • The choice and the address can share a signal, but these mouse reports show a link, not the full rules for building the map.

Sixth film What the Choice Buys: a Pocket

  • A receptor's binding pocket sets which molecules a neuron responds to, but one pocket does not mean one odour.
  • In cultured cells, a pocket mutant lost potency for propionate and gained responses to longer-chain fatty acids, so its selectivity shifted.
  • A flat response would not prove a ligand cannot bind, and these are functional values, not binding constants.

27. From Glomeruli to Piriform Cortex

Guiding idea. In a simple model, each piriform neuron listens to several scattered glomeruli and fires when enough are active together, so odours that share more glomeruli share more neurons, wherever those neurons sit.

First film What the Bulb Sends

  • The olfactory bulb sends piriform cortex a map of which receptor types are active, and in the cortex one odour switches on scattered neurons.
  • An odour is characterized by the set of glomeruli it activates at a given concentration, and higher concentrations often recruit more of them.
  • A simple model explains why a scatter can keep which odours are alike, and the cortex's own network and experience are candidates for anything tighter.

Second film Following One Glomerulus

  • In mice, one glomerulus sent axons diffusely across piriform cortex but to stereotyped patches in the cortical amygdala.
  • In piriform cortex an odour activates a dispersed ensemble, and a single neuron can respond to chemically dissimilar odours.
  • The scatter belongs to piriform cortex, not to the whole bulb, and diffuse is not the same as random.

Third film How a Scatter Keeps Similarity

  • In a simple model, a piriform neuron fires only when enough of its scattered glomerular inputs are active together, and where it sits does not matter.
  • Random wiring of this kind keeps the expected order of odour similarity.
  • The same wiring makes related odours less alike than in the bulb, so anything tighter needs more than this model.

Fourth film What the Cortex Adds

  • In mice, related odours gave more similar patterns than unrelated ones in both bulb boutons and piriform cortex.
  • Piriform neurons clustered related odours more tightly than their input did, so the cortex adds structure that the random model does not give.
  • Blocking transmitter release from excitatory piriform neurons moved the cortex's odour relationships toward the bulb's, but this does not show that local connections alone produce the change.

Fifth film Teaching an Arbitrary Ensemble, and a Code That Drifts

  • Activating small ensembles of piriform neurons with light, paired with reward or shock, was enough for mice to learn an association, which does not show that they smelled anything.
  • In recordings over days to weeks, a decoder trained on the first day approached chance later, so odour responses drifted.
  • Drift does not mean the animal's smell changed, because a downstream circuit may read a moving population consistently.

Sixth film More Than a Map: Timing in the Bulb

  • The bulb reshapes its output with recurrent and lateral inhibition through reciprocal dendrodendritic synapses, and granule cells are not the only source.
  • The relative timing of activity in the bulb conveys odour information, though how downstream circuits use it has been debated.
  • In anaesthetized mice, sister cells shared an input and correlated rates, but an odour shifted their timing against the breath differently, so a shared glomerulus is a shared address, not a shared message.

28. Taste from Tongue to Cortex

Guiding idea. In the mouse, a taste's identity begins with which taste-cell population is driven, and its value, how much the animal wants it, can be pulled apart by silencing the amygdala.

First film Which Taste, and How Good?

  • In the mouse, identity (which taste) and value (how much the animal wants it) are two readouts that can be pulled apart.
  • In mice, type II taste cells release ATP through channels instead of a conventional vesicular synapse.
  • In mice, silencing the amygdala leaves recognition of sweet intact but removes the attraction produced by driving sweet cortex.
  • Value also changes with what the body needs.

Second film Narrow at Threshold, Broader with Strength

  • In anaesthetized mice, most responding geniculate neurons were strongly activated by only one of the tested tastes.
  • For the bitter-sour class of fibre, acid drives the neuron through bitter cells, so the fibre reports the cell population and not the chemical.
  • Tuning broadened as concentration rose in most neurons tested at both strengths, though not all.
  • Whether a neuron looks narrow or broad depends on the panel, the concentration and the criterion for a response.

Third film Opening the Package

  • In mice, driving the sweet cortical field drew animals in and driving the bitter field made them turn away.
  • Blocking either field impaired recognition of its taste.
  • Silencing the amygdala in mice spared trained recognition but abolished the attraction to sweet-field light.
  • Value shifts with need: a sodium-depleted animal drinks salt concentrations it would normally reject, in these assays.

Only this unit will print. Press Ctrl+P or ⌘P to print it. Show the whole course again

Integration and repair

Print this unit

29. Combining Senses and Predicting Them

Guiding idea. A signal is worth what it tells you about the world: the more reliable of two senses counts for more, and some pathways subtract the predictable result of your own movement.

First film Which Sense Do You Believe?

  • When two senses report one event, each is weighted by its reliability, one over its variance, so the more reliable sense counts for more.
  • If vision is blurred until it is less reliable than touch, the combined estimate moves toward touch.
  • Part of what you sense is the predictable result of your own movement, and some pathways subtract a prediction of it.

Second film A Fish That Learns to Cancel Its Own Pulse

  • In the electric fish, a copy of the motor command is used to learn a timed negative image of the expected input, which cancels it.
  • The negative image was learned, because it followed whatever stimulus had been paired with the command.
  • The evidence comes from paralysed fish plus a model built on recordings.

Third film Only When the Feedback Matches

  • In alert monkeys, canal afferents report active and passive turns alike, so the receptors pass self-caused input on.
  • In the monkey central vestibular-only neurons tested, the subtraction of active turns ran only while neck feedback matched the prediction.
  • Self-produced touch is rated less ticklish than the same touch from outside, and a forward model is one candidate explanation, not a settled one.

30. Restoring Senses: Implants, Prostheses and Optogenetic Vision

Guiding idea. A prosthesis delivers detail only as far as it drives distinct groups of neurons, each in the right place on the map, so a count of contacts or pixels is not a count of usable channels.

First film Contacts Are Not Channels

  • A prosthesis delivers detail only as far as it drives distinct groups of neurons, so usable channels have to be measured by testing what the listener can tell apart.
  • Current spread makes neighbouring contacts excite overlapping neurons, one main reason speech recognition plateaued at 8 to 10 channels in 18 implant users.
  • There is no single channel number, since it depends on the listener, the array placement and the test, and the implant still gives large gains.

Second film Streaks and Pixels

  • An epiretinal electrode often gives a streak rather than a dot, because current excites axons of passage that serve other places in the visual field, as a model of axon pathways accounts for.
  • PRIMA places pixels beneath the retina with local return electrodes, and its mean acuity of about 1.32 logMAR sits near the pixel limit, which is agreement rather than a precise prediction.
  • A zoomed score adds a magnifier, not pixels, and serious device- or procedure-related events came with the benefit.
  • In one optogenetic patient the response depended on the goggle light, and one patient cannot show how often this works.

Third film Writing Touch: Addresses, Then Patterns

  • In one man with a spinal cord injury, stimulating electrodes in sensory cortex evoked touch in regions of the hand, so each electrode is an approximate address, not a guaranteed separate channel.
  • With touch feedback his median task time with a robotic arm fell from 20.9 s to 10.2 s, about 51 percent, which shows feedback helped him and not a population effect.
  • Patterns across electrodes evoked tactile edges and apparent motion, and biomimetic pulse trains felt more like a real press on some electrodes, which supports biomimicry as a strategy but is not a law.

Back to the course page →