Perceptual learning improves efficiency by re-tuning the decision 'template' for position discrimination

Visual position discrimination improves with practice; however, the mechanism(s) underlying this improvement are not yet known. We used positional noise to explore the underlying neural mechanisms and found that position discrimination improved with practice over a range of noise levels. This improvement can be largely explained by an increasing efficiency with which observers used positional information in the stimulus. In a second experiment, we tested the hypothesis that the improved efficiency reflects a re-tuning of the observers' perceptual 'template'--the weightings of inputs from basic visual mechanisms--to more closely match the ideal template required to perform the perceptual task. Using a new technique to measure which parts of the stimulus influenced the observer's performance, we were able to record the re-tuning of the decision template across training sessions; we found a robust and steady increase in template efficiency during learning.