L3: From LGN to Cortex

GO OVER THE SLIDE ABOUT NASAL AND LATERAL BC WTF??

What is the Lateral Geniculate Nucleus (LGN)

• Subcortical structure, first relay station between eyes and cortex

• Takes information from the eye though long axons

• They also have centre-surround receptive fields

How many layers are in an LGN

6

What makes the cells in the LGN monocular

Each layer gets inputs from only one eye

Retinotopic map

• How neurons are arranged in LGN

• Means the neighbouring relationships that exist in the retina (cells that sit next to each other) are retained in the LGN (stay in same organisation)

Layers 1 and 2 (of LGN)

• Large cell bodies

• Magnocellular layers 

• Receive input from M ganglion cells

• High contrast sensitivity and low spatial resolution

• Process coarse features and motion

Layers 3 to 6 (of LGN)

• Small cell bodies

• Parvocellular layers

• Receive input from P ganglion cells

• Low contrast sensitivity and high spatial resolution

• Process fine features and colours

What does each layer of an LGN consist of

Each layer contains monocular cells that form a retinotopic map of half a visual field

What is the function of LGN

Still not fully understood but:

• Richly connected to many other parts of the brain (vs. retina mainly connected to LGN and has no feedback i.e. input, only output)

• A locus at which retinal information can be modulated by brain areas

• More than just a relay station

What does the LGN project to

V1 (primary visual cortex)

What are most neurons in V1

Binocular, they receive input from both eyes (around 70%)

• Where brain starts to form image

How is the retinotopic map in V1 distorted

Cortical magnification

• More neurons dedicated to interpreting fovea than peripheral (centre of visual space is magnified)

Why do we not use the same amount of neurons for the whole visual field?

If we would have the same acuity as in fovea in the whole visual field, we would need eyes and brains multiple times the size of our skull

What is key to get a large visual field and high acuity (overcome limitations of peripheral vision)

Rapid eye movements

What do retinal ganglion cells/LGN neuron receptive fields respond best to

Spots of light

Receptive Fields in V1

• Elongated receptive fields (the rectangle images for simple cells)

• Respond best to lines, bars, and edges

• Receptive fields → Ganglion cells/LGN → V1

Simple Cell Receptive Fields

Can appear different, but all respond to orientation

Orientation Tuning (and Size Tuning)

Cells respond (tuned) best to certain orientations, also tuned to size (thin or large bars)

Graded response

Some orientations that don’t stimulate the cell at all

What is activation of one signal cell always

Ambiguous

Why are populations of cells tuned to different orientations

It allows to disambiguate intensity (brightness) and orientation - features are coded in a population code (same basic code in facial perception but a bit simpler)

Why is the activation of one single cell always ambiguous

Optimal but faint stimulus activates cell EQUALLY as sub-optimal intense stimulus

• Why population code is used

Complex cells

Don’t have positive/negative cell distinction, will respond the same wherever a light is shone on it

Hypercomplex Cells

• If bar (of light) extends beyond cell, cell response starts to decline to baseline 

• Plays important role in object recognition