Tracking mouse movement in feature inference: category labels are different from feature labels.

Area of Science:

• Cognitive Psychology
• Computational Neuroscience
• Human-Computer Interaction

Background:

• Existing research offers conflicting views on whether category labels function as regular features or possess unique influence in inductive inference.
• Understanding the precise role of category labels is crucial for models of human cognition and decision-making.

Purpose of the Study:

• To investigate the distinct influence of category membership information on inductive inference compared to other feature information.
• To differentiate between two prominent theories regarding the role of category labels in cognition.

Main Methods:

• Developed an online judgment measure incorporating mouse-tracking, analogous to eye-tracking, to capture fine-grained cognitive processes.
• Presented participants with arbitrary labels conveying either category membership or non-membership information.
• Analyzed both explicit judgments and implicit mouse-movement data.

Main Results:

• Joint analysis of judgment and mouse-tracking data supported a unique role for category labels.
• Participants viewed labels conveying category membership more frequently and earlier than those conveying non-membership.
• This suggests category labels actively guide the inductive inference process.

Conclusions:

• Category membership information exerts an influence on inductive inference that is qualitatively different from that of standard features.
• Category labels serve as cognitive guides, directing attention and inference processes.
• An ecological rationale for this distinct processing of category membership information is proposed.