Observation Study: Biofeedback in Gaming

Does adding biofeedback to a video game improve the gameplay experience?

That was the main perspective for a research conducted during a course from Tilburg University's Marc Swerts and Emiel Krahmer. In order to conduct this research, a setup was prepared that generated artificial data and streamed the output to the players. The players assumed that they heard each others' heart beat and saw a visualization of the opponents galvanic skin response. After each round the players answered a questionnaire, from which we gathered the necessary data.

Concept development

The main focus of this research was games. During the brainstorm and concept development, multiplayer gaming was the starting perspective. Some ideas were generated, and in the end researching biofeedback in multiplayer games was favored by all team members. When we agreed on the main focus of our research, we conducted a literature study to find out what had already been done regarding biofeedback in games. We stumbled on some interesting research projects that provided good inspiration for our own research. In the end, we decided that we would find out whether adding biometric data of the opponent would enhance the gaming experience.

Experiment setup

We had set up an experiment environment to test this hypothesis. We designed the experiment in a way that 2 players would play a one versus one shooter game, while sitting in the same room. Each participant would play 2 rounds of 10 minutes, 1 round with biofeedback, the other without. We set up the experiment in a way that results would not be compromised by bias. The recruitment focused on players who do have some gaming experience, but are not too skilled. We used a 2 by 2 experiment, design, depicted in figure 3. We would have 2 groups of participants, of which the first group would start with the biofeedback, and the other group would first play without.

Biofeedback

The hardware was self-constructed, and built as a preparation for a study with real biofeedback. The setup we used generated artificial data. As input factors we decided that heart beat rate and galvanic skin response (GSR) are the most appropriate biometric data. The biofeedback setup consisted of 2 components: hardware and software.

Hardware

2 sets of hardware were designed to process the biometric data that is generated by the players. We strapped the fingers with conductive tin foil, and attached a clip to the player's ear lobe, to capture the heart beat rate. This hardware is connected with the player's laptop, where a piece of software processed the incoming data.

Software

The software consisted out of a client and server. The client processed the data gathered by the hardware, and sent this to the server. The server exchanged the data, and submitted the opponent's data to the player. This data was represented through a graphical gauge (GSR) and a heart beat sound (heart beat rate). Due to the experimental nature, the data that was presented to the players was artificial.

Results

After each 10 minute round, the players were asked to fill out a questionnaire. We used both the Game Experience Questionnaire and the Game Experience Questionnaire - Social Presence to gather our quantitative data. After the session we conducted an informal interview, of which the data was not used in the results. When the data of the biofeedback sessions were pooled and compared to the pooled data of the no-biofeedback session, one main effect on the flow scale of the GEQ questionnaire was found.