Class Notes (1,100,000)
CA (650,000)
UTSC (30,000)
Psychology (9,000)
PSYB51H3 (300)
Lecture 3

PSYB51H3 Lecture Notes - Lecture 3: Stretch Marks, Receptive Field, Simple Cell

Course Code
Matthias Niemeier

This preview shows pages 1-3. to view the full 10 pages of the document.
Week #3 Lecture #3: Spatial Vision
Spatial Vision
Spatial vision: refers to our ability to resolve or discriminate spatially defined features.
- Pinhole camera or better
- “pinhole” eye allows finer directional sensitivity and limited imaging
- important because it allows us express info with the light that tells us about spatial info
Spatial Vision - Overview
• Why does the visual system use filters for lines, edges and gratings? (Fourier transform)
• What is spatial frequency – or what is the secret of Mona Lisa’s smile? … and related visual
• What is visual acuity?
• The primary visual pathway ***MIDTERMMM****
• Functional properties of the striate cortex (tuning function, retinotopy, cortical column)
• Deficits of spatial vision
Why lines, edges and gratings?
• Why does the visual system use filters for lines, edges and gratings?
Gutenberg’s printing press – a metaphor
An independent component analysis of the visual world
Gratings and their characteristics
What is a Fourier transform?
Printing press metaphor
1439 Johannes Gutenberg invented a printing press with movable types.
Huge impact on 15th century Europe (1424 Cambridge 122 books!)
- incredible wealth, all books were handwritten
“Greatest invention in the last 1000 years”.
Similar technology in ancient China & Korea had no comparable impact. Why?
- difference bwt Western and Eastern way to write, 26 letters in English and
1000s in Chinese, only need Western alphabet to create, not efficient in Chinese,
• What does the large number of characters indicate?
• Mandarin: e.g., pictophonetic compounds
hé "river
hú "lake
with the 三點水 semantic indicator of something related to a river, i.e., water.
• More redundancy
- not economical, too many
- Language can be represented as text and…
• …Text can be pieced together from a small set of simple and independent (non-redundant)
• Similarly, visual images can be described/ represented as compositions of simple and
independent elements.
- different words/ letters
• Each of these elements can be filtered out separately.
find more resources at
find more resources at

Only pages 1-3 are available for preview. Some parts have been intentionally blurred.

Bell & Sejnovski (1996) used “independent component analysis” (ICA) to identify the
components of photos of natural images
found the components to be edges and lines (i.e., pieces of gratings).
• Efficiency?
- very efficient
• Biological relevance?
- yes, sth to do with brain, there should be neurons that send out all these info
• Natural images have certain statistical properties
• The visual system is plastic (evolution & learning)
We should expect:
A certain specialization of the visual system for gratings and spatial
Filters specialized for edges and lines of various orientations.
• Rectangular grating
• Sine wave (sinusoidal) grating
• Gabors
Sine wave gratings & gabors as building blocks of vision (& audition)
• Three characteristics of gratings
• Cycles per second = Hz
• Cycles per visual degree = cpd
- visual angle
- how light/ big the contrast is (for spatial frequency)
- where does the curve start (does it start by going up/down)
Spatial Frequency: The number of cycles of a grating per unit of visual angle (usually
specified in degrees)
Cycles per degree: The number of dark and bright bars per 1 degree of visual angle
Fourier transform: an operation that breaks down a function/an image into sine waves of
different frequencies.
• Music can be decomposed into sine waves with different temporal frequencies.
- can break down any kind of signal into sine waves (speakers, etc.)
From spatial frequencies to Mona Lisa’s smile
• Visual images can be decomposed into sine waves with different spatial frequencies
a) complete image
b) low-frequency component
- blurred picture
c) high-frequency component
- can see transition from black and white
• Different pieces of information (e.g. look from a distance vs. stick your nose into the slide)
• What do you see?
- a pixelated image
- step away easier to see, walk up close, hard to see
find more resources at
find more resources at

Only pages 1-3 are available for preview. Some parts have been intentionally blurred.

• Lincoln’s face is difficult to recognize because irrelevant high SF information tells you
that you see squares.
• Interferes with the relevant low SF information.
- high frequency square, low frequency - lincoln
Mona Lisa’s secret
• One reason for Mona Lisa’s fame is the ‘secret’ about her smile…
- many layers of painting, to put LF frequency of smiling and HF of not smiling
How the illusion works
• Look at her face: does she smile?
• Look to the side: does she smile now?
• What do you see & why?
From spatial frequencies to Mona Lisa’s smile
• Still not convinced that different spatial frequencies carry different kinds of spatial
• How about this:
- bottle from condensation
- honeycomb
- coral
- woodpecker (holes in tree)
- holes in stone that look like a sponge
- holes in pancake (blueberries)
- strawberry with all the seeds taken out
- melon with zoomed in seed portion
- xray of child’s teeth (1st & 2nd teeth)
- garlic
- stretch mark (on woman)
- child knees kneeling on peas
- sea pond photoshopped on tongue, on entire body
- 16% of the pop has fear of holes
From spatial frequencies to Mona Lisa’s smile
• Trypophobia is a common form of phobia elicited by images of holes.
Cole & Wilkins (2013): fearsome images of holes show similar spatial frequencies as images
of dangerous animals.
• Simple warning system.
- based on spatial frequency, if had holes on tongue, seen as not healthy
Visual Acuity
• What is the path of image processing from the eyeball to the brain?
How good are we at seeing stripes?
Which are the parts of the visual pathway?
How does the striate cortex organize processing of orientation, width, colour,
find more resources at
find more resources at
You're Reading a Preview

Unlock to view full version