Purchase this article with an account.
Jianwei Lu, Zhong-Lin Lu, Barbara A. Dosher; Perceptual learning in peripheral vision with attention reflects (mostly) template retuning. Journal of Vision 2002;2(7):68. doi: https://doi.org/10.1167/2.7.68.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Under divided attention, perceptual learning of Gabor orientation identification in peripheral vision reflects a mixture of stimulus enhancement and template retuning with nearly equal amount of threshold reduction in a full range of external noise (1). In this study, we ask: what is the mechanism of perceptual learning in peripheral vision under focal attention? On each trial, two Gabors with equal contrast but independent orientations were displayed simultaneously at two spatial locations: upper-left (3.1 deg, 2.3 deg), and lower-right (−3.1 deg, −2.3 deg). The eccentricity and the parameters of the Gabors were identical to those in Dosher & Lu (1998). A central pre-cue occurred 217 ms before the onset of the Gabor. The observer was instructed to report the orientation of the Gabor only at the cued location, which varied randomly across trials. Six levels of external noise were added to the Gabors. The contrast of the noise in the two spatial regions was always the same in each trial. Two staircases, 3/1 and 2/1, were used to measure thresholds at two performance levels (79.3% & 70.7% correct) for each external noise level and each spatial location. A total of 1,296 trials were run in each of 20 sessions of training by each observer. All three observers showed a larger magnitude of learning in high noise (46% reduction in contrast threshold) than in low noise (25% reduction in contrast threshold). PTM modeling identifies a primary template retuning, and a secondary stimulus enhancement mechanism of perceptual learning. This contrasts with essentially equal learning in high and low noise with divided attention in (1). We conclude that perceptual learning in periphery with spatial attention, as in fovea (2), reflects mostly template retuning.
DosherLu PNAS, 1998.
LuDosher ARVO, 2001
This PDF is available to Subscribers Only