Learning-induced uncertainty reduction in perceptual decisions is task-dependent

Abstract

Perceptual decision-making in which decisions are reached primarily from extracting and evaluating sensory information requires close interactions between the sensory system and decision-related networks in the brain. Uncertainty pervades every aspect of this process and can be considered related to either the stimulus signal or decision criterion. Here, we investigated the learning-induced reduction of both the signal and criterion uncertainty in two perceptual decision tasks based on two Glass pattern stimulus sets. This was achieved by manipulating spiral angle and signal level of radial and concentric Glass patterns. The behavioral results showed that the participants trained with a task based on criterion comparison improved their categorization accuracy for both tasks, whereas the participants who were trained on a task based on signal detection improved their categorization accuracy only on their trained task. We fitted the behavioral data with a computational model that can dissociate the contribution of the signal and criterion uncertainties. The modeling results indicated that the participants who were trained on the criterion comparison task reduced both the criterion and signal uncertainty. By contrast, the participants who were trained on the signal detection task only reduced their signal uncertainty after training. Our results suggest that the signal uncertainty can be resolved by training participants to extract signals from noisy environments and to discriminate between clear signals, which are evidenced by reduced perception variance after both training procedures. Conversely, the criterion uncertainty can only be resolved by the training of fine discrimination. These findings demonstrate that uncertainty in perceptual decision-making can be reduced with training but that the reduction of different types of uncertainty is task-dependent.

Keywords: Glass pattern; categorization; learning; perceptual decision; uncertainty.

Figure 1

Stimulus sets for the criterion comparison and signal detection tasks. For the criterion comparison task, the signal level of the stimuli was set to 100%, and the spiral angles varied from 0° to 90°. For the signal detection task, Glass patterns of 0° (radial) and 90° (concentric) spiral angles were presented, and the signal level varied between 0 and 100%. Modified from Li and Yang (2012).

Figure 2

Schematic illustration of the computational model. An example of the Double Model is shown here. The decision process can be considered as comparing the perceived spiral angle of a stimulus (p_i) with the implicit decision criterion (c_i). If p_i > c_i, the stimulus was categorized as a concentric pattern. Otherwise, if p_i < c_i, the stimulus was categorized as a radial pattern. The red curve represents the decision criterion. The two blue curves represent the perceived spiral angle of two stimuli with different signal strengths: the left blue curve represents the perception of a low-signal-strength Glass Pattern whose spiral angle was 30°, and the right blue curve represents the perception of a high-signal-strength Glass Pattern whose spiral angle was 70°.

Figure 3

The categorization accuracy in the test sessions. The performance is shown for (A) the criterion comparison group in the criterion comparison task and the signal detection task, (B) the signal detection group in the criterion comparison task and the signal detection task. Error bars represent the standard errors of the means.

Figure 4

The slopes of psychometric functions. The slopes of the psychometric functions are shown for (A) the criterion comparison group in the criterion comparison task, (B) the criterion comparison group in the signal detection task, (C) the signal detection group in the criterion comparison task, and (D) the signal detection group in the signal detection task. Each dot in the scatter plot represents one participant's slope of psychometric function in the post-test session vs. the pre-test session. The dashed line is the equal slope line on which the post-test slope is equal to the pre-test slope. The bar figures on the bottom-right are the averaged group results in both the pre-test and post-test. Error bars represent standard errors of the means. ^*p < 0.05.

Figure 5

Model fitting results for the signal uncertainty. The signal uncertainty is indexed by the variance of perception. The fitting results are shown for (A) the individual data and group average of the criterion comparison group, (B) the individual data and group average of the signal detection group. Scatter plots show individual results of the model fitting. Each dot denotes one participant's perception variance at one signal strength condition. Error bars represent the standard errors of the means.

Figure 6

Model fitting results for the criterion uncertainty. The criterion uncertainty is indexed by the variance of decision criterion. The fitting results are shown for (A) the variance of the decision criterion distribution of criterion comparison group, (B) the mean of the decision criterion of the criterion comparison group, (C) the variance of the decision criterion distribution of signal detection group, and (D) the mean of the decision criterion of the signal detection group. Error bars represent the standard errors of the means. ^**p < 0.01; ^*p < 0.05; n.s.: not significant.