Class 3, Lecture 5, Chapter 3: Recognizing Objects
McClelland and Rumelhart‟s (1981) model of word recognition included two
o Excitatory and inhibitory connections between detectors.
Inhibitory connects occur if one letter is detected it might excite letters
that are common in the bigram for it and inhibit those that are not.
Certain letter combinations are more commonly seen.
o Top-down connections from words to letters and letters to features.
Once the features of a letter are presented, the features not included are
You can have activation going to from the word level to the letter level
to the feature level.
This is the reason why we can read text where the words are jumbled,
es are changed to 3s, upside down text.
Similar feature nets may underlie our perception of objects.
The recognition by components (RBC) model includes an intermediate layer that is
sensitive to geons, basic shapes proposed as the building blocks for all three-
o When we see a certain combination of geons we identify a certain object as
being said object.
One piece of evidence supporting the representation of geons is that perceptually
degraded pictures are better recognized if geons are preserved
o When you take away the geons, individuals are less likely to know what the
object is when distortion occurs
The object-superiority effect is analogous to the word-superiority effect.
Lines are perceived more accurately if presented in the context of an object.
o the position of them on the screen
o Getting a deeper recognition of objects when we see it as being part of a geon
Models of object recognition differ on whether object recognition depends on
o In the recognition by components model, geons result in viewpoint-
Does not matter what side you see the object from you will be able to
recognize the object. Do not mentally rotate anything to recognize the
Response should have quicker response time o Other proposals are viewpoint-dependent, requiring the remembered
representation to be “rotated” into alignment with the current view.
Require more of a matching, see something and search our memory for
something that matches it.
The more the object has to be rotated, the longer it takes to recognize.
Different Objects, Different Recognition Systems?
Some evidence suggests that certain categories of objects are perceived using
o Not all objects are necessarily recognized using the same process
In particular, the recognition of faces may involve principles different from those
discussed thus far.
o Fusaform area where the face recognition occurs
One source of evidence for specialized face-processing mechanisms comes from
prosopagnosia, a type of agnosia in which the visual object-recognition deficit is
specific to faces.
o Agnosia = the inability to recognize an object
o Prosopagnosia = the inability to recognize faces
o The fact that you can lose the inability to recognize faces but not objects
indicates that the idea that there i