Best-of-n decision making by human groups

Area of Science:

• Behavioral Ecology
• Social Psychology
• Virtual Reality Studies

Background:

• Collective decision-making is crucial in animal and human groups, often involving consensus on the best option using evidence.
• Human collective behavior studies typically lack the embodied, nonverbal scenarios common in animal research.

Purpose of the Study:

• To investigate human collective decision-making in an embodied, nonverbal virtual environment mimicking animal nest-site selection.
• To analyze how perceptual difficulty, information availability, and social dynamics influence consensus formation.

Main Methods:

• Groups of up to 10 participants used a 3D virtual environment to select the largest of four sites without verbal communication.
• Movement-based interactions and observations of empirical features guided consensus in the online multi-participant experiment.

Main Results:

• Consensus speed and accuracy were significantly affected by perceptual difficulty and information availability, with no speed-accuracy trade-off observed.
• Participants adapted their use of social information based on task difficulty, location, and time invested in an option.
• A minority of informed individuals independently influenced the group towards faster and more accurate consensus.

Conclusions:

• Human collective decision-making can be effectively studied in nonverbal, embodied virtual environments, offering parallels to social insects.
• Group decision strategies flexibly adapt to environmental and social factors, highlighting the importance of information integration.
• Informed minorities play a key role in enhancing group efficiency and accuracy in collective decision-making scenarios.