![]() only once viewing also contaminates memory and decrease accuracy). Results: ID accuracy was lower for those who created the composite and they were more confident (so neg. View target 60 s -> half of participants made composition. ability to recall decreases rapidly over time (1/3 after an 1h) Effect of constructing and viewing facial composites on memory accuracy / misinfo effect DV: memory accuracy, IV: creating or not a composite (2), x 3: repeated viewing: viewing composite once vs. free recall: describing a face in detail can interfere with face recognition (verbal overshadowing effect) > faces are better matched by their external features -> not diagnostic here what is the verbal overshadowing effect? verbal memories/description in the CI influence visual memories Limitations of Modern feature based systems -differences in importance of internal and external f. ![]() History of Facial composites 70s systems: sketch artist, modern feature based composite systems (selection of individual facial features in the context of a complete face by CI), holistic composite system. Retrieving individual features from memory is unnatural. influence on witnesses and police, guilt bias. Not effective for getting someone's identity, neg. What is a facial composite? effective? A facial composite is constructed by an EW of a person seen to commit a crime. face processing biases: inversion effect, female same sex bias, race effect, within-person variability. This is like matching a photo to a person. What is the face recognition paradigm? Asked participants to match a target face to a 10 face line up, which might or might not involve a different image of the target himself. Participants got a composite -> lineup matching + say rate of similarity -> the more u think a person is guilty, the more similarities u find. ability to match the similarities of 2 faces affected by guilt bias. Charman (2009): composites are used for ID and evidence of guilt and innocence. ![]() diagnostic value of composite experiment. (Frowd experiment, 2017) Face composite paradigm II. therefore it involves both unfamiliar and familiar face perception. then they give this composite to people who are familiar with the person by name. What's the facial composite paradigm? ask witnesses to describe in full detail using CI an unfamiliar target and then to construct a composite. Burton: the more pictures u average, the higher the hit rate, the lower the RT, the fewer the errors. The process of familiarisation: clarifying which stored representations are diagnostic or not of an identity. Different views together to create view-independent representations of familiar faces. seems to rely disproportionally on external features How do unfamiliar faces become familiar? the more familiar a face gets, the more independent of image-based properties.Įxperiment: averaging faces with PCA. eyes) but not unfamiliarĮxternal: hair, shape of face. What do you know about internal and external features? internal: eyes, nose, mouth are more likely to remain constant over time. distinctiveness helps both (the more typical a face, the harder) movement facilitates recognition for both inverted photos are harder to recognize for both Effects on familiar and unfamiliar faces -changes in viewpoint, expression, context neg effect on recogn. pictorial codes underlying unfamiliar face proc. Dissociation between familiar and unfamiliar face processing: fam face (more widespread activity during fMRI), different underlying brain mechanisms. people based on their faces (some are able to perform face-matching). Prosopagnosia: inability to recognize fam. unfamiliar faces They are represented differently. : we don't know how many dimensions there are and what are their labels. ![]() distinctive faces are recognized more easily! Strengths and Limitations of the valentine face space model + :can explain things like the cross race effect and caricature (cause distinctiveness helps) Cross-race effect: better at recognizing faces of the same race as you (bc you can compare it). Faces are encoded relative to the prototypical. The space has dimensions that correspond to attributes. Unfamiliar: on the basis of pictorial -> context-limited info Explain what is the valentine face space model each face is coded as a point in a multidimensional perceptual space and the prototypical face at the centre. On what basis are familiar and unfamiliar faces recognized? familiar: on the basis of abstract view-invariant -> more stable representations
0 Comments
Leave a Reply. |