Abstract
Haptic angle discrimination is an important paradigm for studying human perceptual decision-making. Traditional research has mostly focused on the measurement of perceptual sensitivity (such as thresholds), while neglecting the crucial role of cognitive strategies during task execution. This study aims to explore how different cognitive decision-making strategies affect behavioral performance (discrimination accuracy and decision-making efficiency) in the task of sorting from the tactile perspective. We recruited 18 healthy subjects and asked them to complete the same Angle ranking task using two strategies respectively: (1) the "successive minimum Angle" strategy stipulated in the experiment; (2) the optimal strategy chosen by the subjects themselves. We fitted the perceived uncertainty parameter (σ) from the ranking results through the maximum likelihood estimation method and used it as an indicator of discrimination accuracy. At the same time, we recorded the number of adjustments during the decision-making process as an efficiency indicator. The results show that, compared with the fixed "successive minimum" strategy, when the self-selected strategy is adopted, the decision-making efficiency (number of adjustments) of the subjects is significantly improved (p < 0.001), while there is no significant loss in the discrimination accuracy (Sigma value) (p > 0.05). Further qualitative analysis revealed that the vast majority of the subjects (16/18) spontaneously adopted the advanced cognitive strategy of "double-boundary-limit transition," which maximizes efficiency by dynamically managing the working memory load and the number of comparisons. This study quantified for the first time the optimization effect of strategy flexibility on the decision-making process in the tactile sequencing task, revealing that the human cognitive system can adaptively select strategies according to task requirements to achieve the best trade-off between efficiency and accuracy. This discovery holds significant theoretical importance for constructing more comprehensive perceptual decision-making computing models and understanding human adaptive behaviors in complex environments.