GRACE – ANR Project

The GAME-ON dataset

Introduction

The GAME-ON (Group Analysis of Multimodal Expression of cohesiON) dataset is a multimodal dataset specifically designed for studying group cohesion and for controlling its changes over time. The GAME-ON design was theoretically based on the Severt and Estrada’s integrative framework of cohesion. More specifically, GAME-ON focuses on the social and task dimensions of the instrumental function of cohesion. GAME-ON is the result of a scientific collaboration with the Department of Computer Science, Bioengineering, Robotics and Systems Engineering (DIBRIS) of the University of Genoa (Prof. Gualtiero Volpe is a member of the Advisory and Liaison Board of GRACE) and it is co-owned by LTCI, IMT-TΓ©lΓ©com Paris and DIBRIS University of Genoa.

Overview

GAME-ON consists of over 11 hours of synchronized multimodal recordings (video, audio and motion capture data) of 17 small groups of 3 persons playing an escape game. This game consisted of 5 tasks conceived to elicit changes on the social and/or task dimension of cohesion. The members of each group were asked to fill in questionnaires at the beginning and at the end of the game, and also after each task to provide their assessment on cohesion and emotions, as well as opinions. Here below, the timeline of the flow of the game. The expected changes on cohesion are indicated at the bottom of each task, the questionnaires are indicated in the coloured circles.

Download

Motion capture data, audio features, and self-assessment through questionnaires are available only for research purposes for academics under a Licence Agreement.

To obtain a copy of the Licence Agreement please contact: giovanna.varni (at) telecom-paris.fr

The Licence Agreement must be returned signed before having the access to the data.

Reference

If you use this data, please cite the following paper:

GAME-ON: A Multimodal Dataset for Cohesion and Group Analysis. Maman, L., Ceccaldi, E., Lehmann-Willenbrock, N., Likforman-Sulem, L.,  Chetouani, M.,  Volpe, G.,  and Varni, G. IEEE Access, 8, 2020.