The attraction of yellow corn: reduced attentional constraints on coding learned conjunctive relations. Rappaport, S. J., Humphreys, G. W., & Riddoch, M. J. Journal of Experimental Psychology. Human Perception and Performance, 39(4):1016--1031, August, 2013.
doi  abstract   bibtex   
Physiological evidence indicates that different visual features are computed quasi-independently. The subsequent step of binding features, to generate coherent perception, is typically considered a major rate-limiting process, confined to one location at a time and taking 25 ms per item or longer (A. Treisman & S. Gormican, 1988, Feature analysis in early vision: Evidence from search asymmetries, Psychological Review, Vol. 95, pp. 15-48). We examined whether these processing limitations remain once bindings are learned for familiar objects. Participants searched for objects that could appear either in familiar or unfamiliar colors. Objects in familiar colors were detected efficiently at rates consistent with simultaneous binding across multiple stimuli. Processing limitations were evident for objects in unfamiliar colors. The advantage for the learned color for known targets was eliminated when participants searched for geometric shapes carrying the object colors and when the colors fell in local background areas around the shapes. The effect occurred irrespective of whether the nontargets had familiar colors, but was largest when nontargets had incorrect colors. The efficient search for targets in familiar colors held, even when the search was biased to favor objects in unfamiliar colors. The data indicate that learned bindings can be computed with minimal attentional limitations, consistent with the direct activation of learned conjunctive representations in vision.
@article{rappaport_attraction_2013,
	title = {The attraction of yellow corn: reduced attentional constraints on coding learned conjunctive relations},
	volume = {39},
	issn = {1939-1277},
	shorttitle = {The attraction of yellow corn},
	doi = {10.1037/a0032506},
	abstract = {Physiological evidence indicates that different visual features are computed quasi-independently. The subsequent step of binding features, to generate coherent perception, is typically considered a major rate-limiting process, confined to one location at a time and taking 25 ms per item or longer (A. Treisman \& S. Gormican, 1988, Feature analysis in early vision: Evidence from search asymmetries, Psychological Review, Vol. 95, pp. 15-48). We examined whether these processing limitations remain once bindings are learned for familiar objects. Participants searched for objects that could appear either in familiar or unfamiliar colors. Objects in familiar colors were detected efficiently at rates consistent with simultaneous binding across multiple stimuli. Processing limitations were evident for objects in unfamiliar colors. The advantage for the learned color for known targets was eliminated when participants searched for geometric shapes carrying the object colors and when the colors fell in local background areas around the shapes. The effect occurred irrespective of whether the nontargets had familiar colors, but was largest when nontargets had incorrect colors. The efficient search for targets in familiar colors held, even when the search was biased to favor objects in unfamiliar colors. The data indicate that learned bindings can be computed with minimal attentional limitations, consistent with the direct activation of learned conjunctive representations in vision.},
	language = {eng},
	number = {4},
	journal = {Journal of Experimental Psychology. Human Perception and Performance},
	author = {Rappaport, Sarah J. and Humphreys, Glyn W. and Riddoch, M. Jane},
	month = aug,
	year = {2013},
	pmid = {23668254},
	keywords = {Adult, Attention, Color Perception, Eye Movement Measurements, Eye Movements, Female, Form Perception, Humans, Learning, Male, Recognition (Psychology), Visual Perception, Young Adult},
	pages = {1016--1031}
}

Downloads: 0