Purchase this article with an account.
Anna Byers, John T Serences; Learning to attend transfers across spatial locations. Journal of Vision 2012;12(9):1135. doi: 10.1167/12.9.1135.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Early studies of perceptual learning argued that the behavioral improvements associated with learning are largely orientation-specific, retinotopic, and monocular, and likely the result of low level sensory changes (Karni & Sagi, 1991; 1993). However, real-world instances of perceptual expertise, such as a radiologist searching for a fracture of unknown size, location, orientation, or contrast, are inherently non-specific and must transfer across low-level feature values. We have previously shown that subjects can learn a general attentional strategy when trained on nine different orientations, and that this improvement is supported by an increase in the feature-selectivity of V1 responses (Byers & Serences, VSS 2011). Because feature-based attentional modulations spread across the whole visual field (Treue and Martinez-Trujillo, 1999), we hypothesized that the benefits associated with learning to attend may generalize beyond a single spatial location. To test this, subjects learned to discriminate nine different orientations over five training sessions. Maintaining central fixation, each subject trained with a grating in one of the four quadrants on the screen and switched to a new quadrant on the sixth session. Performance improved across the training sessions and this improvement was maintained after transferring to the new location. Thus, subjects learned a generalizable attentional strategy that enhanced their ability to discriminate various orientations, irrespective of spatial location. This type of generalized learning is likely mediated by feature-selective attentional modulations that improve the precision of sensory representations of stimuli at all locations. The spatially global spread of learning in this context also provides one possible functional explanation for why feature-based attention spreads across the visual field.
Meeting abstract presented at VSS 2012
This PDF is available to Subscribers Only