Face Perception and Visual Agnosia PDF

Ch.3 Perception Sept. 3rd Face perception Face inversion More difficult to recognize face if inverted Relative to other inverted objects Low-level properties still the same as upright face E.g., Contrast Why is it more difficult? Disrupts holistic process...

Ch.3 Perception Sept. 3rd Face perception Face inversion More difficult to recognize face if inverted Relative to other inverted objects Low-level properties still the same as upright face E.g., Contrast Why is it more difficult? Disrupts holistic processing Face perception Face perception Holistic perception Perceived as a “whole” not parts Integration of features across face Perception of entire configuration of face i.e., relationships between features Features not perceived individually, then synthesized i.e., perceive nose, mouth, etc., then integrate into a face Face perception Focusing on single feature might be unreliable Others may share same feature Changes in conditions (e.g., lighting) Features “change” shape (e.g., mouth, eyes) Perspective varies Face perception Face perception Able to recognize an individual face... Despite faces sharing features and configuration Approx. 200-400 msec. to recognize face as familiar Why is it considered as holistic processing? What is the evidence for it? Is it only holistic? Do individual features play a role? Is the top half the same for both faces? Is the top half the same for both faces?  Face perception Composite task Report identity of selected half Ignore the other half Greater difficulty in identifying cued half Disruption caused by other half Difficult to ignore because of holistic processing Disruption holistic processing with offset Perception of facial features not done independently Combined/integrated into whole  Face perception Composite task Report identity of selected half Ignore the other half Greater difficulty in identifying cued half Disruption caused by other half Difficult to ignore because of holistic processing Disruption holistic processing with offset Perception of facial features not done independently Combined/integrated into whole Face perception What are the impacts of holistic processing on memory? Face perception (Tanaka & Sengco, 1997) What are the impacts of holistic processing on memory? Part/whole task 1. Participants study a series of faces  Face perception (Tanaka & Sengco, 1997) A Part/whole task 1. Study faces 2. Tested over a feature (e.g., nose, mouth, B eyes) A. In isolation B. As part of original face C. As part of “new” face Altered configuration (e.g., distance b/w eyes) C  Face perception (Tanaka & Sengco, 1997) A Tested over a feature (e.g., nose, mouth, eyes) A. In isolation B B. As part of original face C. As part of “new” face Altered configuration (e.g., distance b/w eyes) If features as processed holistically, C then memory for feature will be better when seen as part of the whole face (relative to alone)  Part/whole task (Tanaka & Sengco, 1997) A Results Original configuration (B) > New configuration (C) > In isolation (A) New Configuration disrupted recognition B of features not altered by change Different eye spacing  lower recognition of nose, mouth Altered spatial info can interfere with memory of features But, individual features are encoded C Memory for features in isolation (70%) above chance  Face perception Face perception not entirely holistic... But more so relative to perception of other objects For example, recognition of a house feature (e.g., door) the same despite of testing condition Face Inversion Potential reduction of “perceptual field” Greater field when looking at upright faces Able to take in holistic configuration of features When field is reduced... Less likely to take into account other features outside of perceptual field Makes it more difficult to recognize face Can then attempt a feature-by- feature analysis of face Thatcher Illusion Face inversion disrupts holistic processing... but retains local (e.g., feature) processing Makes it difficult to detect “unaltered” features Prosopagnosia Prosopagnosia Inability to recognize faces Even those of close family, friends, or themselves! Able to... Detect & describe individual facial features Recognize objects Although sometimes associated with object agnosia Read Prosopagnosia Prosopagnosia What is the deficit? Inability to recognize the whole or gestalt Try to piece individual features together May use other cues Voices, clothing, hair Prosopagnosia Accounts of prosopagnosia (Farah, 2004) “I can see the eyes, nose, and mouth quite clearly, but they just don’t add up. They all seem chalked in, like on a blackboard.... I have to tell by the clothes or by the voice whether it is a man or a woman.... The hair may help a lot, or if there is a mustache.... ” “At the club I saw someone strange staring at me, and asked the steward who it was. You’ll laugh at me. I’d been looking at myself in a mirror.” Prosopagnosia Acquired Prosopagnosia Acquired from brain damage (to fusiform face area) E.g., stroke Developmental/Congenital Prosopagnosia (~2% of pop.) No brain damage Fusiform face area might not be fully developed Genetic influence Face recognition is heritable Face Perception Fusiform Face Area (FFA) Selective neural activation to faces In contrast to activation to objects https://www.youtube.com/watch? v=O7AQ8NjSnTo Visual Agnosias Apperceptive Agnosia Impaired perceptual processing “Visual acuity could not be measured with a Snellen eye chart, as he could Impaired recognition/description of... neither identify letters of the alphabet Objects nor describe their configuration. ” Figures (e.g., letters, numbers) (Farah, 2004) p. 11 Some unable to distinguish X’s from O’s Difficulty drawing objects Visual Agnosias Apperceptive Agnosia Impaired perceptual processing Impaired recognition/description of... Objects Figures (e.g., letters, numbers) Some unable to distinguish X’s from O’s Difficulty drawing objects Farah, 2004  Visual Agnosias Apperceptive Agnosia Sometimes also impaired recognition of people Some able to recognize based on other cues Tactile, sound (voice) Can be from occipital cortex damage Stroke Farah, 2004 Anoxia From carbon monoxide poisoning Visual Agnosias Apperceptive Agnosia Some can name object if in prototypical view E.g., Side view of bucket But have impaired recognition of objects in atypical views E.g., Top-down view of bucket Lack of salience of prototypical features hurts recognition Visual Agnosias Associative Agnosia Normal (or close to normal) perception of objects Difficulty naming/identifying them May guess visually similar objects E.g., see a baseball bat, guess paddle Or, Key  a file or tool Sometimes may recognize object through another modality Farah, 2004 E.g., Touch https://www.youtube.com/watch?v=T1qnP xwalhw Visual Agnosias One patient shown (Farah, 2004)... Stethoscope “a long cord with a round thing at the end” Can opener “could be a key” Cigarette lighter Didn’t know, but identified when lit Visual Agnosias Associative Agnosia May be unable to group objects based on semantic categories E.g., Object not belonging to category E.g., pine, palm, pyramid E.g., Unable to select two objects (out of three) with similar functions E.g., closed umbrella, cane, open umbrella Unable to access semantic knowledge Visual Agnosias Associative Agnosia May be able to draw objects... Piecemeal & very slowly But, can’t identify them Farah, 2004 Visual Agnosias Associative Agnosia Sensitive to quality of visuals recognition best for (ordered best to worst)... Objects Photographs Line drawings Damage temporal and occipital cortex E.g., from a stroke Farah, 2004 Visual Agnosias Patients can display... Low-level visual functions preserved E.g., brightness, color, motion Other cognitive functions intact E.g., intelligence, language Feature detection not the same as perception Perception (Davenport & Potter, 2004) Perception is helped by semantic knowledge Semantically consistent images more likely to be recognized When perceiving scene knowledge can be accessed B A C (Even if images flashed by in 80 ms) i.e., (foreground) objects are processed interactively with scene (influence each other) Objects Backgrou Knowledge & Perception Face & computer images Averages of hundreds of pictures Contexts help interpret scenes/images i.e., knowledge & experience helps interpret stimuli  Top-down processing The (blurred) “car” is the same image as the (blurred) “person” Spatial location within the context of a street can lead to interpreting blur as a person Top-down processing Knowledge, memory, expectations influence how info in processed Higher-order processes impact how sensory info is interpreted E.g., influence on object recognition, reading Bottom-up processing In contrast to... Bottom-up processing Processing determined by physical characteristics or pattern of stimulus E.g., detect simple features and integrated into more complex features Top-down processing (Bar et al., 2006) Top-down influence on perception E.g., knowledge can make it easier to recognize objects Influence from the frontal cortex (orbitofrontal cortex, OFC) before object recognition is completed i.e., before recognizing an object through the ventral stream Early visual areas signal to frontal areas Frontal cortex, then helps