Taste and Vision Neuroscience: Gustation, Retina & Cortical Pathways

Posted on Feb 12, 2026 in Psychology

Chemical Senses: Taste (Gustation)

Function of Taste

• Identify nutritious substances (e.g., sugars, proteins, salts).
• Avoid toxins and poisons, often detected as bitterness.
• Taste is distinct from olfaction, but both are chemical senses.
• Helps evaluate food after ingestion (versus olfaction, which acts at a distance).

Anatomy and Pathway

• Taste starts on the tongue in taste buds containing taste receptor cells.
• Taste receptor cells are not neurons; they do not fire action potentials but release neurotransmitters.
• Taste receptor cells synapse onto gustatory neurons in sensory ganglia.
• Gustatory neurons do fire action potentials to the brainstem → thalamus (VPM) → gustatory cortex (insula).
• Distinct from olfaction, which can bypass the thalamus.

Types of Taste

Transduction

• Ion channels (salt & sour) open directly when tastants bind → depolarize the cell.
• GPCRs (sweet, umami, bitter) bind ligands → activate G-proteins → intracellular signaling → depolarization.
• Depolarization → Ca²⁺ influx → vesicle release (e.g., glutamate) → activates gustatory neuron.

Special Topics

Taste Receptor Knockouts

• Knockout T1R2 → lose sweet.
• Knockout T1R1 → lose umami.
• Knockout T1R3 → lose both sweet & umami.
• These results show both use T1R3 as a shared partner.

Dimerization

• GPCRs form heterodimers to function:
• T1R2 + T1R3 = Sweet receptor.
• T1R1 + T1R3 = Umami receptor.

Artificial Sweeteners

• Aspartame = non-sugar dipeptide that mimics sweetness.
• Activates the sweet receptor (even though it is not a sugar).
• Some sugars (e.g., sucralose) cannot be digested → sweet but zero-calorie.

Miracle Fruit

• Contains miraculin (a protein) → binds the sweet receptor.
• Activated only in acidic environments (e.g., lemons) → makes sour foods taste sweet.

Taste-Like Sensations (Adjuncts)

These are not taste per se, but thermosensory / pain sensations mediated by TRP ion channels, not GPCRs.

Flavor ≠ Taste

• Flavor = Taste + Olfaction + Somatosensory (texture) + temperature.
• Full flavor experience is lost with anosmia (e.g., blocked nose, COVID).

Vision Lecture One: Retina to Higher Visual Processing

Why Vision Is Special

• Most information-dense sense (100 million+ photoreceptors per eye).
• One third of cortex devoted to vision.
• Perception ≠ objective reality → brain reconstructs meaning from visual input.
• Vision evolved for survival: food, tool use, social recognition.

Retina: Structure & Function

• Inverted structure: light passes through ganglion and bipolar layers before hitting photoreceptors.
• Three neuron layers:
  • Photoreceptors (rods + cones): no action potentials.
  • Bipolar cells: analog, can flip the signal.
  • Ganglion cells: fire action potentials → brain via the optic nerve.
• Fovea: high cone density → highest acuity, no rods.
• Cells are arranged so light reaches photoreceptors effectively.
• Peripheral retina: low acuity, more rods → motion and night vision.
• Blind spot: where optic nerve exits → no photoreceptors.

Rods vs Cones

Color cones:

• S (blue) ~420 nm
• M (green) ~530 nm
• L (red) ~560 nm

Molecular Transduction (Phototransduction)

Photopigment = GPCR + retinal.

• Opsin (GPCR) + retinal (from Vitamin A → β-carotene).
• Retinal switches cis → trans when light hits → changes GPCR structure.

Signal Cascade (in Rods)

1. Photon → retinal isomerization.
2. Activates rhodopsin → G-protein (transducin).
3. Activates PDE → breaks down cGMP.
4. ↓ cGMP → closes Na⁺ channels → hyperpolarization.
5. ↓ Glutamate release.

Strange Features of the Retina

• Photoreceptors are active in the dark (depolarized at rest, ~–40 mV).
• Light → hyperpolarization → reduces neurotransmitter (glutamate).
• Glutamate is inhibitory in ON bipolar cells (via metabotropic receptors).
• Bipolar cells invert the signal → ganglion cells depolarize and spike in response to light.

Why Analog?

• Photoreceptors & bipolar cells do not use action potentials.
• They use high-fidelity, graded (analog) signals over short distances.
• Only ganglion cells spike.

Adaptation & Evolution

• Amplification: 1 photon → activates 100,000+ molecules downstream.
• Humans can detect single photons.
• Convergent evolution: Cephalopod eyes are similar to vertebrate eyes but evolved independently.
• Compound eyes (e.g., insects) = many ommatidia (mini-eyes).

Visual Illusions & Subjective Reality

• Perception is constructed by the brain.
• Visual illusions (checkerboard, color context) show how context shapes experience.
• Color is subjective, even though wavelength is objective.

Key Terms

• Fovea: central retina, high acuity.
• Blind spot: no photoreceptors at optic nerve exit.
• Receptive field: region of space that drives a neuron’s response.
• Retinal: light-sensitive molecule derived from Vitamin A.
• Phototransduction: process of converting light to electrical signal.

Big Picture Ideas

1. Why Vision?

   Vision is one of our most developed senses and relatively self-contained, which makes it a great system to “go deep” on in this course.

   It helps demonstrate how neuroscience works from retina → perception.

2. Neuroscience’s Goal Here

   Show how sensory input (light) gets transformed into meaning (perception).

   The key theme is computation: the brain doesn’t add information—it interprets it based on the signals coming from the retina.

Retina (revisited)

• Three layers:
  • Photoreceptors (rods + cones): transduce light to voltage.
  • Bipolar cells: some flip the sign of the signal.
  • Ganglion cells: the first cells to fire action potentials.
• Their axons form the optic nerve.

Receptive Fields and Edge Detection

• Receptive fields: ganglion cells have a center-surround structure.
• On-center, off-surround (or vice versa) helps filter spatial information and enhance edges.
• This setup helps detect contrast, not uniform light, and is made possible by lateral inhibition via horizontal cells.

Thalamus (LGN)

• Still has center-surround receptive fields.
• No major transformation — it is primarily a relay.
• Important for gating (e.g., sleep, attention).
• Inputs from each eye are segregated into different layers.

Visual Cortex (V1)

• Major transformation happens here:
• Neurons go from circular center-surround fields → orientation-selective receptive fields.
• This is the canonical cortical computation, built by summing inputs from aligned LGN neurons.
• Orientation tuning: neurons fire most to bars of light at specific angles; tuning curves show response strength vs orientation.
• Edge detection: V1 neurons are essentially edge detectors; the brain compresses input and retains edges, which carry significant information.
• Other features encoded: direction of motion, disparity (depth), color, binocularity.
• Maps on maps: retinotopic map, orientation columns, ocular dominance columns — overlaid feature maps in cortex.

Binocular Vision & Depth

• Disparity: the offset between the image in each eye.
• The brain compares disparities to compute depth (stereopsis).
• Only possible because cortical neurons can integrate inputs from both eyes.
• Strabismus & amblyopia: misalignment of eyes leads to poor integration; during a critical period, visual experience shapes cortex; covering the “good” eye can help rewire to use the “bad” eye.

Plasticity & Competition

• Hubel & Wiesel’s critical period discovery: depriving one eye in kittens leads to permanent loss of cortical response to that eye.
• Shows how experience shapes wiring during development.
• Neurons compete — the more active eye “wins” cortical territory.
• Even in frogs, introducing competition leads to column formation, suggesting a deep principle in brain organization.

Higher-Level Visual Processing Cheat Sheet

Face Inversion Effect

• Humans and monkeys recognize upright faces better than inverted ones.
• Inverting eyes or mouth is not noticeable when the face is upside down, but is disturbing when upright.
• Demonstrates we process faces holistically, not just feature-by-feature.

Two Higher Visual Functions Covered

1. Object processing (e.g., faces).
2. Motion processing (e.g., optic flow).

Motion Perception and MT (V5)

• Motion is critical to survival: tracking prey, avoiding predators, and navigating space.
• MT (middle temporal area) is specialized for processing motion.
• MT neurons encode both direction and speed of motion.

Hubel & Wiesel Review

• V1 neurons tuned to orientation of bars (e.g., vertical, tilted).
• Some are direction-selective (e.g., prefer a bar moving up-right vs down-left).

Self-Generated Motion / Optic Flow

• Caused by your own movement, e.g., walking or driving.
• On the retina, everything appears to move outward from center.
• Term coined by J. J. Gibson.
• MT neurons are sensitive to types of motion, including optic flow.

MT Microstimulation Experiment (Causal Evidence)

Random Dot Motion Task: dots move with varying coherence (some aligned, some random).

• MT neurons are tuned to direction of motion, not orientation.
• Key question: can activating MT neurons change what an animal perceives?

Experiment Design

• Show monkey dots with varying coherence.
• Microstimulate MT neurons preferring “down” motion.
• Monkey presses a lever indicating perceived direction.

Result

Stimulating “down” neurons biases the monkey toward perceiving “down” even at 0% coherence. The psychometric curve shifts — perception is causally altered.

Takeaway: perception can be controlled by neural activity; the brain operates like a democracy of neurons voting on what we perceive.

Dorsal vs Ventral Visual Streams

Ventral Stream (“What” Pathway)

• V1 → V2 → V4 → IT (inferotemporal cortex).
• Object and face recognition, color, form.
• Inspiration for convolutional neural networks (CNNs).

Dorsal Stream (“Where / How” Pathway)

• V1 → MT (V5) → parietal cortex.
• Motion, spatial location, guiding movement.
• Performs coordinate transformations (retinal → egocentric → allocentric).

IT Cortex and Face Recognition

Face Cells in IT (Fusiform Gyrus)

• Some neurons fire only for faces.
• Not sensitive to contrast, missing features, or color.
• Columnar organization — neurons nearby respond to similar stimuli (e.g., face-preferring).

Prosopagnosia

Damage to the fusiform face area → inability to recognize faces (patients can still see faces but cannot identify them).

Jennifer Aniston Neuron (Hippocampus)

• Neurons fire to specific individuals, even just to the name.
• Context-dependent (e.g., may not fire for Jennifer Aniston pictured with Brad Pitt).
• Shows abstract, semantic representation at the top of the visual hierarchy.

Object Agnosias (Due to Lesions)

1. Apperceptive agnosia: cannot perceive whole objects.
2. Associative agnosia: can draw the object but cannot name or identify it.

Blindsight

Patients with V1 damage claim they are blind. But when forced to guess, they perform far above chance. This shows a dissociation between awareness and perception and suggests other visual pathways can guide behavior without conscious awareness.