Abstract
Practicing a perceptual task, such as distinguishing small changes in the orientation of a line, often leads to improvements in performance that are highly specific to the trained stimulus. Practice with images presented in the left visual field, for instance, may not yield performance benefits when the same stimuli are shown to the right of fixation. Does this form of location-specific perceptual learning also result in structural changes within the human brain? After one month of practice discriminating differences in orientation between successively presented peripheral gratings, subjects' orientation thresholds improved from an average of 2.4 degrees prior to training to 1.2 degrees after training. Thresholds did not change significantly for stimuli presented at an untrained location in the opposite visual field, or for stimuli of an orientation orthogonal to that which was practiced. Structural MRI and cortical thickness measurements revealed a 3–5% increase in cortical thickness after training across a range of higher visual regions, including MT+ and FFA. A group random effects analysis performed across the cortical surface showed a single large region of significantly increased thickness (p < 0.001 clusterwise) in lateral and ventral occipital cortex. This increase in thickness was restricted to the hemisphere contralateral to the trained visual field location, with no significant effects observed in the untrained ipsilateral hemisphere. These lateralized structural changes thus show specificity resembling that of the similarly lateralized improvements in behavioral performance. These results indicate that extensive practice, even of relatively simple perceptual tasks, can trigger widespread structural changes in the adult brain over comparatively short time scales. The observed increase in cortical thickness may reflect either neuronal plasticity, such as synaptic proliferation and increased dendritic arborization, or a response to greater metabolic demands through mechanisms such as enlargement of glial cells or increased vascularization.
NIH F32 EY019448, F32 EY019802, R01 EY017082,P30 EY008126, and a Rubicon grant from the Netherlands Organisation for Scientific Research.