Flexible human collective wisdom.

Area of Science:

• Cognitive Science
• Social Psychology
• Decision Science

Background:

• Group decisions often surpass individual judgment, but the mechanisms for combining individual inputs remain debated.
• Existing models typically use static rules like majority voting or weighted averaging.
• It is unclear if groups adapt these rules to evolving information landscapes.

Purpose of the Study:

• To investigate whether human groups dynamically adjust their collective decision-making strategies in response to changing information environments.
• To test if groups can move beyond simple majority rules when optimal strategies shift.
• To explore the relationship between adaptive strategy use and collective decision accuracy.

Main Methods:

• A novel experimental paradigm was developed where information distributions changed mid-experiment without participant awareness.
• Participants engaged in perceptual and cognitive signal-detection tasks.
• Group decision rules were analyzed as information environments shifted from favoring majority to minority opinions.

Main Results:

• Groups gradually shifted from relying on majority opinions to favoring minority opinions with high confidence.
• This adaptation occurred even when the majority rule did not show a performance decline.
• A greater ability to abandon the majority rule correlated with higher collective decision accuracy.

Conclusions:

• Human collective decision-making is not governed by static rules but involves dynamic adaptation of decision algorithms.
• Groups infer information distributions and adjust their strategies to optimize collective wisdom.
• This research proposes a new theoretical framework for understanding adaptive collective intelligence.