Cognitive Psychology 2101 Week 4 Notes (9/20 and 9/22)
Cognitive Psychology 2101 Week 4 Notes (9/20 and 9/22) PSY 2101 - 001
Popular in FOUNDATIONS OF COGNITIVE PSYCHOLOGY
Popular in Psychology (PSYC)
This 6 page Class Notes was uploaded by Asmaa Abdullah on Thursday September 22, 2016. The Class Notes belongs to PSY 2101 - 001 at Temple University taught by Pamela J. Shapiro (P) in Fall 2016. Since its upload, it has received 26 views. For similar materials see FOUNDATIONS OF COGNITIVE PSYCHOLOGY in Psychology (PSYC) at Temple University.
Reviews for Cognitive Psychology 2101 Week 4 Notes (9/20 and 9/22)
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 09/22/16
9/20: Chapter 4: Recognizing Objects Object Recognition ● the process through which the object is identified ● It is crucial for learning and applying knowledge ● We can recognize objects ○ Even if they are incomplete ○ Through context ○ From different views ○ Of the same type presented in different ways ● Our understanding of language is important with context to recognize unusual figures even if the stimuli are the same ○ Example of figure that shows “TAE CHT” but is actually “THE CAT” written awkwardly ● There are 2 types of processing in object recognition ○ Bottom Up ■ Datadriven processing ■ Stimulusdriven effects ■ Analysis of form ■ Based on feature ○ TopDown ■ Conceptdriven ■ Knowledge or expectation ■ E.g. your knowledge of the English language guides your interpretation of ambiguous letters ■ Based on context Features ● The small elements out of which more complicated patterns are composed ● How quickly you connect features and context suggests parallel processing ● Features are inferred in the visual system’s organized perception of form ● Advantages of featurebased system: ○ Features are the building blocks of object perception and recognition ○ Helps us distinguish common vs. variable/uncommon ○ Experimental tasks suggest that features have priority in perception ● Conjunctive search (where an object has combined features of distinct objects around it) takes more time and is harder than feature search (where an object has one distinct feature that makes it easier to find in visual search) ○ Also, FINDING A TARGET is easier than NOT FINDING IT ■ To find it, you distinguish its features (whether single or conjunctive) ■ When you don’t find it right away, you make sure that it’s not there so you take more time ● Other data suggest that detection of features is a distinct step in object recognition ○ Integrative agnosia ■ Is caused by parietal cortex damage ■ Is a disorder of attention ■ Can detect features ■ Cannot detect how features are bound ○ Similar finding of distinction found using TMS Word Recognition ● Some methodology for studying it ○ Tachistoscope ■ Device for presenting stimuli for precisely controlled amounts of time ■ Computers used today ○ Mask ■ A stimulus designed to disrupt further sensory processing of the word ● E.g. random letters over the word itself ● In theory, finding a letter with no shared features with another letter in the midst (Os in the midst of Ns) is easier than finding a letter that has similar features (Zs in the midst of Ns) ● Visual words can be recognized with extremely brief presentations under the right conditions: ○ More frequent in the language ○ Recently seen (phenomenon of repetition priming) ○ Words (generally) rather than random strings of letters (phenomenon of wordsuperiority effect) ■ Briefly present word → ask participant if it is a certain letter → if it is an actual word, participants are more accurate in the guess OR if it is a random string of letters, they are less accurate ■ Word superiority represents the probability of the word appearing in English (how Englishlike the words are) Feature Nets ● Possibility for how the visual system recognizes words ● Initial layer (at the bottom/base) comprises of detectors for features ○ Detectors have receptive fields. They fire a signal when a threshold of stimulation is reached ((NEURAL NETWORK)) ○ More likely complex assemblies of neurons rather than single neurons ● Next layers detect more complex patterns (letters → words → e.g. sentences) ● Explains experimental results of word recognition ○ Frequency of word and recency of word ■ In both situations, detectors that fired more recently have a higher starting activation level (which means a lower threshold and frequent firing) ○ Idea that words in general are easier to detect leads to the use of a level of bigrams (letter pairs) to the feature nets ■ Bigram system also helps the system recover from confusion about individuals letters ● Location of letters make a difference ● Bi or multilingualism are less likely to affect word recognition ● Downside: ○ Overregularization ■ Word detected is not the stimulus presented ■ We detect features we are used to ■ Our brains favor efficiency over accuracy ● The network sacrifices a small amount of accuracy for a great deal of efficiency ■ These errors are not easily detected because of context ● These errors are produced by the same mechanism responsible for its advantages (the ability to deal with ambiguous inputs and to recover from errors) ○ The network’s expectations are not locally represented in any single detector ■ It is rather a property of the network as a whole ● Distributed knowledge OR distributed parallel processing ○ Word detectors (final) Bigram detectors Letter detectors Feature detectors Stimulus Input (the visual stimulus of seeing the word) (initial) 9/22: Continuation of Feature Nets: ● McClelland and Rumelhart’s (1981) model of feature nets ○ 2 additions: ■ Excitatory and inhibitory connections between detectors ■ TopDown connections from words to letters and letters to features ○ Similar feature nets for object recognition: ■ RBC (recognition by components) model ● Intermediate layer that is sensitive to geons (geometric ions) ○ Basic shapes proposed as the building blocks for all threedimensional forms ○ There are 36 geons → 30,000 discriminable objects ○ Objects are defined as relationships between geons ○ Feature net detects edges → axes → positions → geon → geon assemblies → objects ○ Evidence supporting the representation of geons is that perceptually degraded pictures are better recognized if geons are preserved Face Recognition ● Facial recognition is different from feature recognition ○ You may not be able to recognize a whole face from only one feature ● Holistic Processing ○ The composite face effect ■ Two different parts of two different faces are combined ■ From this stimulus presented, people cannot distinguish the faces when they are aligned together ■ People distinguish faces when they are separated/fragmented ○ Face recognition does not depend on an inventory of a face’s parts ■ It is not featuredependent ■ Depends on holistic perception of the face ■ Complex relationships created by the face’s overall configuration Viewpoint Dependence ((Orientation Matters)) ○ The Inversion Effect ■ Much harder to recognize inverted faces than other inverted objects ■ Inverting faces causes a greater disruption in memory performance compared to inverting houses ■ Inverting face + inverting certain features = might not detect difference from upright perspective ■ Upright face but inverted features = MONSTER!!! Data from Neuroimaging ● FFA is active when viewing faces ○ Active in bird expert viewing ■ Suggests evidence of fine distinctions ● PPA is active when viewing houses Disorders of Face Recognition ● Prosopagnosia ○ Face blindness ○ Type of visual agnosia ○ Prosopon = face, agnosia = without knowledge ○ Patients can’t ■ Recognize familiar faces ■ Distinguish one face from another ■ Recognize their own faces ○ Acquired Prosopagnosia ■ Previously normal face recognition disrupted by stroke, head injury, or neurodegenerative disease ○ Developmental Prosopagnosia ■ never develop normal face recognition ability ■ Genetic ■ Early brain insultprenatal, during infancy or childhood ○ How do we test for it? ■ Test ability to learn and distinguish new faces ■ Test ability to recognize famous faces ■ FMRI during face, scrambled face, and nonface recognition tasks ■ Learning New Faces ● Cambridge Face Memory Test: ○ Tests your ability to learn and recognize unfamiliar faces. 1. Study 3 views of same face ● S elect that face from an array ○ s tudied images ○ n ovel images ○ i mages with noise 2. Study 6 different faces ● S elect the studied face from array ○ s tudied images ○ n ovel images ○ d egraded images ○ Hair is masked when testing for face recognition ■ Because prosopagnosia patients might identify the face by hair or other nonfacial features ((testmybrain.org/setup?b=100))
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'