One talk I was particularly interested in seeing at this year’s Develop, having attended one of his at last year’s Develop, was Graham McAllister’s on biometrics. Last year, Graham, director at Vertical Slice, discussed the principles of biometrics, and the methods used to detect players’ reactions to games.
This year he widened his talk to discuss other methods of measuring enjoyment and how they compare with biometrics. All games developers want to maximise the enjoyment players get out of a game, so measuring this provides a starting point for them to improve it. Four types of measurement were discussed:
- Data metrics: A traditional method studios often use to measure this is data metrics. Many metrics are collected to measure what players do in the game and how long they spend doing it. This leads to two problems. One is that this data is often collected retrospectively, when it is too late to change the game, and the other is that it doesn’t actually show whether someone liked or disliked the game.
- Interviews: After playing a game, players can also be interviewed about what they liked or disliked about it. However, research has shown that this does not always match up with their physiological reactions to a game and is therefore quite unreliable. Memory is also a drawback here. Players only reliably report back on the beginning, end and anything dramatic that happened in the middle.
- Facial expressions: Researchers have also tried recording and measuring facial expressions during the game. However, although players may experience excitement and surprise, they are unlikely to experience sadness or anger, and the emotions are often not mirrored in their facial expressions. So facial coding is not really applicable to gamers.
- Biometrics: Biometrics provides an automatic method for gathering the data needed. It is tested by galvanic skin response (GSR). If a player is enjoying a particular part of a game, GSR increases (a change in temperature leading to the production of tiny amounts of sweat). In this way researchers can accurately measure which parts of a game players are enjoying. This can then be compared with what developers wanted them to experience, and the game can consequently be improved.
Although a combination of methods is often used, biometrics therefore provides the most accurate data concerning enjoyment. Using these techniques, developers can create games with what Peter Molyneux terms the ‘golden line’ (the path of most enjoyment).
If you develop games, what techniques have you found most effective for assessing how enjoyable the game is? Have you ever taken part in any playtesting too? Share your experiences in the comments below.
Filed under: Event, Game development Tagged: | Biometrics, Develop 2011, game development
