Nervous System

1
Nervous System
Perception Memory Planning
Sensory inputs light, sound, skin pressure, odor

Motor responses limb movement, facial
expression, speech
2
Sensory Systems
3
Light Projection on Retina
4
Transduction
5
Transduction
6
(No Transcript)
7
Transduction
8
Cone Response
9
Fechners Law
Membrane Response V
Subjective Intensity I
Log S
Log S
Log So
Log So
Fechners law I A log (S/So)
for S gt So I 0
for S lt So S Physical stimulus
intensity So Threshold stimulus intensity A
Constant
V A log (S/So) for the linear range S
Physical stimulus intensity So Threshold
stimulus intensity A Constant
10
Webers Law
I
I
dI
dI
Log S
S
Log So
dS
dS
Fechners law I A log (S/So) Differentiat
e dI A dS/S dS needed to get a fixed
dI dS (dI/A) S Webers Law dS K S
where K dI/A
11
Visual pathway
12
Decussation
13
Hierarchical and Parallel Processing
14
(No Transcript)
15
Receptive Field
16
(No Transcript)
17
(No Transcript)
18
(No Transcript)
19
Geometry of Projection
20
Geometry of Projection
Retinal image size is inversely proportional to
distance
21
Left eye
Right eye
22
Ponzos Illusion
23
(No Transcript)
24
Muller-Lyer Illusion
25
(No Transcript)
26
(No Transcript)
27
Circular World of the Zulus (South Africa)
28
David Marrs Concept of a Computational Theory
for Understanding an Information Processing Task
in the Brain
We cannot understand how a bird flies by only
studying its wings, but need, in addition, an
aerodynamic theory of lift generation by the flow
patterns around the wings. We cannot understand
how a computer works by only studying the
transistors on the circuit boards and their
connections, but need, in addition, concepts of
operating system, data structure, and application
programs.
29
David Marrs Concept of a Computational Theory
for Understanding an Information Processing Task
in the Brain
Therefore, even if some day we had complete
knowledge of every molecule in the brain, and
could record the electrical activities of every
cell at any time, we would still not understand
how the brain processes information. We need, in
addition, a computational theory which specifies
how the electrical signals carried by a large
number of neurons could act in concert to solve a
certain perceptual problem.