Different dynamics of performance and brain activation in the time course of perceptual learning

Abstract

Perceptual learning is regarded as a manifestation of experience-dependent plasticity in the sensory systems, yet the underlying neural mechanisms remain unclear. We measured the dynamics of performance on a visual task and brain activation in the human primary visual cortex (V1) across the time course of perceptual learning. Within the first few weeks of training, brain activation in a V1 subregion corresponding to the trained visual field quadrant and task performance both increased. However, while performance levels then saturated and were maintained at a constant level, brain activation in the corresponding areas decreased to the level observed before training. These findings indicate that there are distinct temporal phases in the time course of perceptual learning, related to differential dynamics of BOLD activity in visual cortex.

Figure 1

Experimental design. We conducted four fMRI sessions. Experiment 1 (n=6) involved six training sessions conducted until post 2 (each session is represented as a black bar). The average time intervals (± standard errors) between the initial training and post 1, between posts 1 and 2, and between posts 2 and 3 were 19.3 (± 2.7) hours, 11.3 (± 0.6) days and 14.8 (± 1.1) days, respectively. In Experiment 2 (n=5), eight additional training sessions (each of which is represented as a white bar) were conducted between posts 2 and 3. The average time interval between the initial training and post 1, between posts 1 and 2, between posts 2 and 3 were 26.8 (± 1.3) hours, 10.0 (± 0.4) days, and between posts 2 and 3, 14.0 (± 0.4) days, respectively, in Experiment 2. Experiment 3 (n=4) consisted of the same numbers of training and “fMRI sessions”. Unlike Experiment 1, the “fMRI sessions” in Experiment 3 was conducted in a mock scanner and BOLD signals were not measured.

Figure 2

Results. (A) The averaged threshold SOA (± standard errors) across all subjects in Experiment 1. (B) Mean performance (± standard errors) in the trained (circles) and untrained (squares) locations in Experiment 1. (C, D) Mean location-specific learning indices (+ a standard error) for fMRI response in V1 as “V1 index” (filled squares) and for performance as “performance index” (open circles with dashed lines) in Experiment 1 (C) and Experiment 2 (D).

Figure 3

Reaction time to the orientation task, activated region size, and correct response ratio for the fixational letter task in Experiment 1. (A) The reaction time to the orientation task was defined as the time interval from the onset of the target stimulus to the button press for reporting the orientation. Black and white bars represent the averaged reaction time (± standard errors) to the trained and untrained locations, respectively. (B) Activated region size is defined as the mean number of voxels that were activated (p<0.01) within the V1 cortical quadrants that corresponded to the trained (or untrained) visual field quadrants while the subjects were performing the TDT at the trained (or untrained) location in the pre-training phase, subtracted from that in each of posts 1, 2 and 3. Black circles and white squares represent the activated region size in the trained and untrained V1 cortical quadrants, respectively. Vertical bars represent standard errors. (C) Mean correct response ratio (± standard errors) for the fixational letter task.

Figure 4

The mean activation indices (± standard errors) for fMRI responses in V1 (A), V2 and V3 (B), and IPS (the intraparietal sulcus), SPG (the superior parietal gyrus) and MFG (the middle frontal gyrus) (C), from the results from Experiment 1 combined with those in Experiment 2. IPS and SPG are parts of the parietal lobe, and MFG is a part of the prefrontal area. An asterisk indicates that the index is significantly larger than the baseline (p<0.05).