Search

Group Discussions: When We Do and Don’t Use Them

Image courtesy of Pixabay.

Picture this. You have no clue how to solve an expert-level Sudoku puzzle and are given two options: you can either ask a group of five people to work together on it or ask five individuals to solve the puzzle independently. Which do you choose? 

At Yale, Professor of Psychology and Linguistics Frank Keil and PhD student Emory Richardson presented people with scenarios just like this, contrasting group discussion with crowdsourcing–getting information from a large group of people. “You can think of this as a choice between two network structures: in the group, everyone can influence everyone else. In the crowd, you get independent answers, which means that you avoid groupthink, but you also miss out on the benefits of sharing information,” Richardson said. 

Richardson and Keil showed kids and adults three types of questions: questions they could answer conclusively through explicit reasoning (e.g., solving a sudoku puzzle), and two kinds of questions for which reasoning would not be enough—popularity (e.g., what most people in the world say their favorite fruit is) and hard perceptual discrimination (e.g., determining whether an opaque box contains thirty or forty marbles just by listening to a recording of it being shaken). They found that even six-year-olds prefer reasoning in groups just as strongly as adults. However, while they are more likely to crowdsource to answer the “non-reasoning” questions, younger kids are less sensitive to the risks of thinking as a group than adults. “Intuitions about when to collaborate in groups and how to structure our groups could be part of what makes our species so successful,” said Richardson.