Transcript PY202 Overview

PY202 Overview
Meta issue
How do we internalise the world to enable
recognition judgements to be made, visual
thinking, and actions to be executed.
Basic issues
Methodologies for studying perception and
cognition.
Neurophysiology of low- (early) and high-level
(late) vision (including retina, LGN, cortical
processing).
Construction and representation (including
surface perception, figure-ground, object and
face processing).
Attention.
Imagery.
Methods and Issues
Neurophysiology and low-level vision
(simple properties, small receptive fields).
Neurophysiology and high-level vision
(complex properties, large receptive
fields).
Psychophysics and neurophysiology.
Cognitive Psychology and the needle-inthe-haystack.
Possible Exam Questions
Define the concept of the receptive field (RF) of a neuron
in the visual system. Describe the evidence for the RF
properties of neurons in the primate visual system
between the level of retinal ganglion cells up to primary
visual cortex (V1).
The receptive field of a visual neuron is the region of the
retina within which light affects the firing rate of the
neuron. The RFs of cells with well-defined “on” and “off”
regions, which includes retinal ganglion cells, LGN cells,
and simple cells in primary visual cortex, are mapped
using small spots of light and observing which locations
affect the firing rate of the cell. The responses of these
cells to more complex patterns of light can then be
predicted by thinking of any pattern as being composed
of small spots of light and “adding up” the effects of all
these spots. The RFs of complex and hypercomplex
cells are mapped using moving bars or edges of
particular orientations.
Describe the evidence for the trichromatic theory of color
vision.
Psychophysics. By varying the amounts of a fixed set of
three different wavelengths in the visible spectrum people
can find a match for the color of any given wavelength of
light. For example, by varying the amounts of red, green,
and blue people can match any given color.
Physiology. The cone receptors in the retina come in three
different types: Some fire most strongly when illuminated
by long wavelengths of light (red), others fire most strongly
when illuminated by middle wavelengths of light (green),
and still others fire most strongly when illuminated by short
wavelengths of light (blue). Also, the pigment molecules in
the cells that fire most strongly when illuminated by long
wavelengths of light selectively absorb long wavelengths of
light (rather than the middle or short wavelengths).
Similarly, the pigment molecules in the cells that fire most
strongly when illuminated by middle (short) wavelengths of
light selectively absorb middle (short) wavelengths of light.
‘Are faces just another class of object.’ Discuss.
Consider the ways in which face processing might differ
from object processing – in functional (from cognitive
experiments) and anatomical (from neurophysiological
and neuropsychological experiments) terms.
In functional terms how do representations of objects
differ from those of faces? (e.g. describe the effects of
inversion on recognition)
In anatomical terms, are there areas of cortex that
encode only faces/objects? (e.g. the fusiform face area)
Are there some objects that might act like faces –
functionally and anatomically? (e.g. what happens when
objects become highly familiar.
Describe the different roles that have been suggested
for the ventral and dorsal pathways in visual cortex,
and the evidence that demonstrates these different
roles.
Ventral - "WHAT" pathway - identification of objects
independently of location; Neurons in inferior temporal
areas that respond to faces, hands, and other complex
stimuli; These neurons have very large receptive
fields; they are not very selective for location; Lesion
studies with monkeys; When presented with a pair of
objects, they can no longer select the one that they
have seen before; Still perform well on a localisation
task; Lesions in humans produce identification deficits;
Visual Agnosia - inability to recognise objects;
Prosopagnosia - inability to recognise faces; difficulty
in making age and gender judgements about pictures
of faces.
Dorsal - "WHERE" or "HOW" pathway - representing
objects' locations and/or allowing their manipulation;
Lesion studies with monkeys; Deficits in selecting
location near a landmark object; Still perform well on
recognition task; Lesions in humans; spatial neglect Unable to attend to any object on one side of space,
regardless of its shape or identity; Difficulty in learning
path through a maze.
Although there are important differences between these
two pathways, there are a multitude of connections
between them. The processing in each pathway is
obviously strongly affected by the other pathway.