Class Notes (807,938)
Canada (492,936)
Psychology (2,615)
PSYC 2700 (115)

Part 1: Lecture 3

9 Pages
Unlock Document

Carleton University
PSYC 2700
Chris Herdman

LECTURE #3 OUTLINE 1. Pattern Recognition (Textbook Chapter #3, pages 90-113) a. Template Models b. Feature Models c. Object Recognition i. RBC Theory ii. Agnosia 2. Selective Attention (Textbook Chapter #4, pages 138-149) a. Filter Model b. Attenuation Model c. Multimode Model of Attention I. PATTERN RECOGNITION Recap: Sensory Memory SM is large. Masking: process by which info from sensory memory is masked.  SM codes are primitive (features)  SM duration is short (~250 ms)  SM has large capacity How can we recognize patterns? We must select for further processing. A. Template Models For everything we experience, we form a mental template (ex. Everything we do; everyone we meet, etc.). An example of this is a Scantron. The scanner has a template that knows the correct answers.  Match stimulus to template in memory  Support for idea: computers  Limits: inefficient o Irregular world o Strict match B. Feature Models Fundamental features are usually regular.  Features more regular than patterns  Complex objects composed of single features  Gibson (1969) o Features of letters o We don’t actually read letters. Our brain knows different features and puts that knowledge together to recognize letters  Neisser (1964) (Fig. 3.8, page 96) o High-speed scanning  Features extracted and noted  “count” (compare) “Z” to item o Harder (slower) if target (Z) shares features with distracters  Pritchard (1961) o Psychological nystagmus (jitter 30-70/sec)  Refreshes retinal receptors  Eyes are constantly jittering  What if we stabilize the image?  Image fades  Chunks (lines, curves)  Suggests higher-order feature detectors  Lettvin et al. (1959) o Microelectrodes into cells of frog retina o Recorded activity  Edge detectors  Moving edge  Dimming  Convex edge (small, circular dot moves) o Shows existence of feature detectors  Hubel & Weisel (1963) o Cat vision o Simple cells:  Simple patterns of light:  Location specific  Edge, slit, line  Complex:  Same as simple but NOT location specific  Hypercomplex:  Moving lines  W,X,Y:  Speed of transmission  Y (movement speed) Beyond Features: Top-down (conceptually driven) pattern recognition  Pattern recognition is influenced by knowledge  Breaking our world into features and using these patterns to recognize  Avant & Lydall o Masking of BOY vs. YOB  Results: a shorter interval is required to erase (mask) BOY than YOB  Reading o Average of 7,500 features per page (100 per sec) o Role of top-down?  Not feature-by-feature  Word Superiority Effect (WSE) o Reicher (1969), Wheeler (1970) D WIND D G WIND WING o Performance better with WIND than D o Top-down influences pattern recognition  McClelland & Rumelhart (1981) o Word level o Letter level o Feature level o Parallel, feedback  Repetition Blindness (Textbook page 98) o Where we fail to accurately see something if it is repeated o Use of RSVP paradigm (Morris & Harris, 2002)  Rapid Serial Visual Presentation  “When she spilled ink there was ink all over.” o People fail to see the stimulus “ink” the second time o Top-down influence: Cognitive system has identified the stimulus, so it “expects” NOT to see it again  “I broke a wine class in my class yesterday.” o People read “wine glass” (top-down context biases) nd o Also, no repetition blindness to the 2 class (top-down overrides effect) C. OBJECT RECOGNITION i) RBC theory (Beiderman, 1987, 1990) (Textbook page 105-109)  Recognition by component theory  Objects made up of combinations of “geons”  Recognition involes: o Parse objects into component geons o Note where geons join (find edges) o Match geon combinations to representations in memory o Very bottom-up model  Geon idea is good, but RBC has pro
More Less

Related notes for PSYC 2700

Log In


Don't have an account?

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.