Implicit learning of geometric eigenfaces
| Autor: | Xiaoqing Gao, Hugh R. Wilson |
|---|---|
| Rok vydání: | 2014 |
| Předmět: |
Adult
Male Implicit learning Computer science 050105 experimental psychology Discrimination Learning 03 medical and health sciences Young Adult 0302 clinical medicine Encoding (memory) Visual Objects Feature (machine learning) Face learning Humans Learning 0501 psychology and cognitive sciences computer.programming_language Communication Analysis of Variance Principal components Models Statistical business.industry 05 social sciences Pattern recognition Recognition Psychology Prototype Sensory Systems Ophthalmology Eigenface Face Principal component analysis Human visual system model Visual Perception Female Artificial intelligence Summary statistics business computer 030217 neurology & neurosurgery Photic Stimulation |
| Zdroj: | Vision Research. 99:12-18 |
| ISSN: | 0042-6989 |
| DOI: | 10.1016/j.visres.2013.07.015 |
| Popis: | The human visual system can implicitly extract a prototype of encountered visual objects (Posner & Keele, 1968). While learning a prototype provides an efficient way of encoding objects at the category level, discrimination among individual objects requires encoding of variations among them as well. Here we show that in addition to the prototype, human adults also implicitly learn the feature correlations that capture the most significant geometric variations among faces. After studying a group of synthetic faces, observers mistook as seen previously unseen faces representing the first two principal components (eigenfaces, Turk & Pentland, 1991) of the studied faces at significantly higher rates than the correct recognition of the faces actually studied. Implicit learning of the most significant eigenfaces provides an optimal way for encoding variations among faces. The data thus extend the types of summary statistics that can be implicitly extracted by the visual system to include several principal components. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect