Individual curiosity modulates exploration in sequential book selection.

Area of Science:

• Cognitive Science
• Information Science
• Computational Neuroscience

Background:

• Information seeking is often viewed as curiosity-driven exploration, akin to foraging for resources.
• It's unclear if decision mechanisms like reward generalization and directed exploration apply to knowledge and information domains.

Purpose of the Study:

• To investigate how people select sequential reading materials in large, unfamiliar content environments.
• To determine if foraging mechanisms generalize to semantic spaces of knowledge.

Main Methods:

• Analysis of a large-scale real-world dataset of sequential book selections.
• A controlled behavioral experiment to observe reading choices.

Main Results:

• Book selection is influenced by structured generalization within a semantic embedding space.
• Readers engage in directed exploration towards options with high uncertainty.
• Individual curiosity levels impact the exploration-exploitation tradeoff, affecting reading enjoyment.

Conclusions:

• Computational mechanisms of foraging extend to epistemic domains, guiding media selection.
• Understanding these mechanisms provides insight into how people navigate and learn from information.