Bottom-up and top-down influences at untrained conditions determine perceptual learning specificity and transfer
- Peking University, China
Figures
Figure 1
Orientation discrimination learning and the effect of bottom-up stimulation of the transfer orientation on learning transfer.
(a) The baseline. Training was conducted at one orientation and the transfer of learning was tested at an orthogonal orientation to reveal orientation specificity. (b) CFS configurations for subconscious presentation of a Gabor at the orthogonal transfer orientation. The flashing noise was presented from 0 to 530 ms in the dominant eye. From 212 to 318 ms, a Gabor at the transfer orientation was also presented in the non-dominant eye. (c) Stimuli for bottom-up stimulation. Bottom-up stimulation was achieved by different CFS stimuli for instruction presentations before the experiment and for actual experimental presentations. Instruction stimuli: Before the experiment, the observers were instructed to report the color of a dot centered on the noise pattern in the dominant eye. A blank screen was shown to the non-dominant eye. Actual stimuli: In the actual experiment, a Gabor at the orthogonal transfer orientation was also flashed in the non-dominant eye. The observers were not told, neither were they aware of, the presence of the orthogonal Gabor. Here orientation discrimination training and bottom-up stimulation of the orthogonal transfer orientation were performed in different blocks of trials. (d) The mean and individual learning and transfer data with training and bottom-up stimulation of the transfer orientation. (e) Control experiment. Same as 1d except that there was no orientation discrimination training. (f) Control experiment. Same as 1d except that there was no actual presence of the orthogonal Gabor. (g) A summary of learning and transfer in the baseline, training plus bottom-up stimulation, bottom-up stimulation alone, and training plus noise-only conditions. (h) A summary of the transfer indices in the baseline, training plus bottom-up stimulation, and training plus noise-only conditions. Error bars indicate ± 1 standard error of the mean. DE - dominant eye. *p<0.05; **p<0.01; ***p<0.001. See Figure 1—source data 1 for raw data.
-
Figure 1—source data 1
The first data sheet summarizes the mean and individual data presented in figure panels 1a, 1d, 1e, and 1f.
The other data sheets contain raw staircase data for 1a, 1d, 1e, and 1f.
- https://doi.org/10.7554/eLife.14614.004
Figure 2
Orientation discrimination learning and the effect of top-down attention to the transfer orientation on learning transfer.
(a) CFS configurations for the top-down attention condition. Instruction stimuli: Before the experiment the observers were instructed through a demo that an orthogonal Gabor or an uppercase letter C would be presented to the non-dominant eye, and that they needed to report whether it was a Gabor or a C even if they could not perceive it. Actual stimuli: In actual testing the Gabor/C was replaced by a blank screen. Orientation discrimination training and the top-down attention condition were conducted in different blocks of trials. (b) The mean and individual learning and transfer data with training and top-down attention to the transfer orientation. (c) Control experiment. Same as 2b except that there was no orientation discrimination training. (d) A summary of learning and transfer in the top-down attention and control conditions. (e) A summary of the transfer indices in the current training plus top-down attention condition and the previous baseline (replotted from Figure 1a) and training plus noise-only (replotted from Figure 1f) conditions. Error bars indicate ± 1 standard error of the mean. DE - dominant eye. *p<0.05; **p<0.01; ***p<0.001. See Figure 2—source data 1 for raw data.
-
Figure 2—source data 1
The first data sheet summarizes the mean and individual data presented in figure panels 2b and 2c.
The other data sheets contain raw staircase data for 2b and 2c.
- https://doi.org/10.7554/eLife.14614.006
Figure 3
Orientation discrimination learning and the effect of combined bottom-up stimulation and top-down attention on learning transfer.
(a) CFS configurations for the combined bottom-up stimulation and top-down attention condition. It only differed from Figure 2a in that the orthogonal Gabor was present in 80% of the trials in the actual stimuli, and an uppercase letter C was present in the remaining 20%. (b) The mean and individual learning and transfer data with training and combined bottom-up stimulation and top-down attention to the transfer orientation. (c) Control experiment. Same as 3b except that there was no orientation discrimination training. (d) A summary of learning and transfer in the combined bottom-up and top-down condition and the control condition. (e) A summary of the transfer indices in the current combined bottom-up and top-down condition and the previous baseline (replotted from Figure 1a) and training plus noise-only (replotted from Figure 1f) conditions. Error bars indicate ± 1 standard error of the mean. DE - dominant eye. *p<0.05; **p<0.01; ***p<0.001. See Figure 3—source data 1 for raw data.
-
Figure 3—source data 1
The first data sheet summarizes the mean and individual data presented in figure panels 3b and 3c.
The other data sheets contain raw staircase data for 3b and 3c.
- https://doi.org/10.7554/eLife.14614.008
Figure 4
The effects of bottom-up stimulation of the transfer location on Vernier learning transfer.
(a) CFS configurations for the simultaneous Vernier training and bottom-up stimulation of the diagonal transfer location. Instruction stimuli: The observer was instructed before the experiment to perform the Vernier task in the dominant eye, while flashing noise patterns covered the opposite visual hemifield across the horizontal meridian. A blank screen was shown to the non-dominant eye. Actual stimuli: A horizontal Gabor was simultaneously flashed in the non-dominant eye at the diagonal transfer location. The observers were neither told, nor were they aware of, the presence of the Gabor stimulus. (b) The mean and individual learning and transfer data with training and bottom-up stimulation of the transfer location. (c) Control experiment. Same as 4b except that there was no Vernier training. (d). Control experiment. Same as 4b except that there was no presence of the Gabor stimulus at the transfer location. (e). A summary of learning and transfer in the training plus bottom-up stimulation condition, the bottom-up stimulation alone condition, and the training plus noise-only condition. (f). A summary of the transfer indices in the training plus bottom-up stimulation condition and the training plus noise-only condition. Error bars indicate ± 1 standard error of the mean. DE - dominant eye. *p<0.05; **p<0.01; ***p<0.001. See Figure 4—source data 1 for raw data.
-
Figure 4—source data 1
The first data sheet summarizes the mean and individual data presented in figure panels 4b, 4c, and 4d.
The other data sheets contain raw staircase data for 4b, 4c, and 4d.
- https://doi.org/10.7554/eLife.14614.010
Figure 5
The effect of top-down spatial attention to the transfer location on Vernier learning transfer.
(a) CFS configurations for a top-down spatial attention trial. The observers were instructed through a demo to report a Gabor or an uppercase C in the non-dominant eye at the diagonal transfer location, which was also dichoptically covered by the flashing noise in the dominant eye. In actual experiment the Gabor/C stimulus was absent. The Vernier trials and the Gabor/C trials alternated within the same block of trials. (b) The mean and individual learning and transfer data with training and top-down attention to the transfer location. (c) Control experiment. Same as 5b except that there was no Vernier training. (d) A summary of learning and transfer in the training plus top-down attention condition and the top-down attention alone condition. (e) A summary of the transfer indices in the training plus top-down attention condition and the previous training plus noise-only condition (replotted from Figure 4d). Error bars indicate ± 1 standard error of the mean. DE - dominant eye. *p<0.05; **p<0.01; ***p<0.001. See Figure 5—source data 1 for raw data.
-
Figure 5—source data 1
The first data sheet summarizes the mean and individual data presented in figure panels 5b and 5c.
The other data sheets contain raw staircase data for 5b and 5c.
- https://doi.org/10.7554/eLife.14614.012
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Bottom-up and top-down influences at untrained conditions determine perceptual learning specificity and transfer
eLife 5:e14614.
https://doi.org/10.7554/eLife.14614
