Modeling crowdsourcing as collective problem solving

TL;DR

This paper presents a modeling framework to analyze how the effectiveness of crowdsourcing depends on group size and problem difficulty, revealing optimal group sizes for collective problem solving.

Contribution

It introduces a novel modeling approach to study the impact of collectivism and group size on crowdsourcing efficiency, providing insights into optimal group configurations.

Findings

Optimal group size varies with problem difficulty

Effectiveness peaks at certain group sizes

Relationship between group size and success is complex

Abstract

Crowdsourcing is a process of accumulating the ideas, thoughts or information from many independent participants, with aim to find the best solution for a given challenge. Modern information technologies allow for massive number of subjects to be involved in a more or less spontaneous way. Still, the full potentials of crowdsourcing are yet to be reached. We introduce a modeling framework through which we study the effectiveness of crowdsourcing in relation to the level of collectivism in facing the problem. Our findings reveal an intricate relationship between the number of participants and the difficulty of the problem, indicating the optimal size of the crowdsourced group. We discuss our results in the context of modern utilization of crowdsourcing.