Information sampling behavior with explicit sampling costs.

Area of Science:

• Cognitive Psychology
• Decision Science
• Information Gathering

Background:

• Decisions to gather information involve balancing its value against accrual costs (time, energy, money).
• Understanding how individuals weigh monetary costs and benefits in information gathering is crucial for decision-making models.

Purpose of the Study:

• To investigate how people balance monetary costs and benefits when gathering information in a perceptual-motor estimation task.
• To compare human information gathering behavior against optimal observer models.

Main Methods:

• Participants performed a touch-screen task, sampling cues to estimate a hidden target's location.
• They decided when to stop sampling and attempt a touch, balancing reward against sampling costs.
• Two conditions varied initial reward and per-cue cost to calculate optimal sampling.

Main Results:

• Participants over-sampled information in one condition and showed no significant deviation in the other, challenging claims of consistent under-sampling.
• Unlike ideal observer models, participants utilized current sample dispersion to guide their sampling decisions.
• This suggests a reliance on sample dispersion, potentially due to imperfect learning of underlying distributions.

Conclusions:

• Human information gathering in decision-making tasks can deviate from normative models, sometimes involving over-sampling.
• The use of sample dispersion indicates a heuristic or learning process influencing stopping decisions.
• Further research is needed to understand the mechanisms behind imperfect learning and its impact on information gathering strategies.