'/Command' and Conquer: Analysing Discussion in a Citizen Science Game

Abstract

Citizen science is changing the process of scientific knowledge discovery. Successful projects rely on an active and able collection of volunteers. In order to attract, and sustain citizen scientists, designers are faced with the task of transforming complex scientific tasks into something accessible, interesting, and hopefully, engaging. In this paper, we examine the citizen science game EyeWire. Our analysis draws up a dataset of over 4,000,000 completed game and 885,000 chat entries, made by over 90,000 players. The analysis provides a detailed understanding of how features of the system facilitate player interaction and communication alongside completing the gamified scientific task. Based on the analysis we describe a set of behavioural characteristics which identify different types of players within the EyeWire platform.