Psyco 258 (October 7 Lecture) Chapter 3 – Perception
Subjective experience of reality… Mind produces what is reality
These subjective experiences of reality are very important to carrying out
Perception is all about taking in information from the world
But our view of reality is based on more than this
React with the incoming stimulation
Phone ringing has no meaning … but if waiting for someone special to call
you or if someone calls at 4am... Then the phone ringing would have more
Some animals can perceive sounds that humans cannot. It’s just a different
perceptual world that the dog brain creates.
Dogs see little color but our world is full of color -> for us… very functional
for our personal experiences -> to see which fruit is ripe so we can eat it
Damage to other regions of the brain as well
Blindsight is sight out of consciousness
o “Can you see what is in front of you?” “No.” “I want you to reach out
and pick up the water bottle.” The person picks up the water bottle.
Ask: “Did you see it in front of you?” “No.” “How did you pick it up?” “I
don’t know I just did.”
This is quite expensive information processing that happens behind the
scenes. It is still available to people
Primary visual cortex is necessary for conscious visual perception
Sensation and Perception
We can see from very very far away if nothing is our way.
We can hear a watch ticking in a very very very quiet room.
Sensation: From the environment and it is the initial encoding
Perception: Process of interpreting and understanding the sensation
Arbitrary distinction between sensation and perception-> It’s all part of one
task. Cant really draw a line in the middle
Picture of rose:
o Sensation: pink, green against background
o Perception: rose , smell of a rose (perfume)
Vision’s 2 tasks:
Object recognition: Process of identifying that object.
o Key to learning
o Have to combine new info with old info, but we must categorize
everything Slide #6 picture of cube
Gestalt psychologists: The person who is receiving an object must therefore
be organizing it.
The whole is greater than the sum of its parts -> the whole is different than
the sum of its parts
Ambiguous figure: More than one way to perceive this cube. Top or bottom
Why is perception not neutral? You’ve seen one more than the other in your
previous experience. -> most of us look at the cube from above.
Theories of Visual Object Recognition
Information comes in (sensation) -> we organize it, we give meaning to it,
can be based on our past history of dealing with the same thing or same
Which elements go with which
Then we interpret, if visual then we interpret the orientation of the object.
Then we identify the object, and then we recognize a tremendous number of
different patterns, actions, running vs. falling, etc.
Template Matching Theory
Retinal image of object that gets transmitted faithfully to the visual centers of
the brain and then we compare it other objects
Eg. If two lines that meet at right angles this will be a T
Multiple-trace memory model
o A trace of every single experience that we have will remain
somewhere in the brain. Stored in memory.
o Primary memory or working memory will activate the Long term
mem store/ 2ndry memory and the memory traces of the prototype
are matched up on the basis of the characteristics of the image itself.
o Efficient...Only need to find a small number of images.
o Completely breaks down if there is any sort of variant -> eg. If looking
at a book a diff way, it wont be recognized as a book.
o If you have not even seen your name written out in calligraphy before,
you should not be able to recognize it but you are able to recognize it.
If your name is flipped 90 degrees you can still recognize it
o This model is tooooo simple.
Recognition by components theory
Structural descriptions describe parts of objects
Similar to the 3 model.
Difference in types of parts that are assembled to compose the object though
Our ability to recognize many objects only comes down to a number of basic
Basic shapes are put together in certain ways and forms to represent
everything. Pay attention to where the components match together.
The basic idea is that we recognize objects by breaking them down, look up
combination of components and see which best matches the combination.
Some objects are easily recognized if pieces of a picture are missing. If
missing info of where the two components join then we will struggle with the
recognition of that object.
Penguin would be 9 geons. Much more info is contained… The more complex
a figure is, the easier to recognize. The more info provided, yes its more
complex, but easier to process because of all the extra info that is present.
Problem: Too much focus on the object itself. Context matters and so does
experience. If teacher pops out a flashlight… It will take us 0.1 ms later to
recognize it’s a flashlight compared to if we were on a camping sight. A dog
expert can tell diff dogs apart… This theory doesn’t take this into account, it
says that anyone can tell any dog apart.
Feature Detection Theories
Key idea is that all objects are made up of separable distinct parts called
Refer to their shared properties to see how things are similar.
T = horizontal line on top of a vertical line vs. L = horizontal line at the
bottom right of a vertical line
R= 3 d