Did you know that games (card, board or video) were sometimes used in scientific research?
Researchers use games to study brain functioning and social behavior.
In June 2024 the newspaper Nature Human Behavior published this article by Allen K et al. aimed at encouraging the use of games as experimental paradigms to increase both the ecological (i.e., contextual) validity, scale, and robustness of research on brain functioning.
Advances in psychological and cognitive science have been made possible primarily through the development of simple experiments with controllable factors used again and again in numerous studies. This is perfect for obtaining reproducible results as well as precise statistical modeling, but it greatly limits the type of questions we can ask, and therefore the answers we obtain!
The authors therefore propose here games as an alternative path. The goal is to expand the repertoire of classic psychological tasks to both verify that psychological theories developed through simple experiments can explain people’s behavior in more contextual situations, but also to ask (and therefore answer) new questions. on the functioning of the brain.
But actually, what is a game?
Yes, yes, it is a serious question, and to which there is this astonishing answer from the Austrian philosopher Wittgenstein: according to him, gaming is the perfect example of a concept that does not require a rigorous definition to understand it, since there is no element that is common to all games. It is therefore probably impossible to find a definition that can cover all existing and future cases!
However, the authors have chosen here the following definition of “games”, which certainly has its limits, but allows it to respond to the context of the article: “facilitators which structure the player’s behavior and whose main goal is pleasure”. In fact, games are intuitive environments that reflect aspects of the real world that can be easily interacted with. This is what makes them particularly interesting for studying the inductive biases that lead to a complex action in a given context.
The other advantage of the game: on s’amuse, and that is intrinsically motivating! Believe me, this is a significant element for what is probably the most laborious part of setting up a population research protocol: the recruitment of subjects (in number but also in diversity). It is often difficult to obtain a sufficiently high number of subjects to have interesting statistical power.
We are therefore the opposite of classic experiences which are (generally) both not at all intuitive and rather boring.
And concretely, how does it work? To give an example, van Opheusden et al. developed a two-player game similar to tic-tac-toe or gomuku, in which players take turns placing tokens until one player connects four of their colored tokens into a line.
By working in collaboration with a mobile application company, the authors managed to collect data from more than 1.2 million players online and in the laboratory!
Their results demonstrate that humans’ planning depth increases with their level of gaming expertise. In addition, they showed that online gamers started with significantly worse search strategies than lab participants. It may therefore be interesting to study how subjects improve their search strategies by observing the more general online population, since lab volunteers seem to start with a higher level already.
Observe social behaviors
Another example : Overcooked et Codenames are perfect games for studying social behaviors that rely on shared knowledge (such as social transmission, collective research, or any other large-scale social phenomenon), since victory relies on sharing inductive biases with other members of the group. the team.
One of the main benefits of games is that they can help us answer questions in a rather unique way. which concern our deepest motivations. For example, why do we pursue goals despite the absence of extrinsic reward? This is what we observe with games like Minecraftwhich do not offer a specific goal to achieve but rather offer a sandbox in which the player can act freely.
What do people decide to do in this kind of scenario? Do they set themselves goals to achieve and, if so, what time scale or difficulty do they give themselves? These are questions that are not generally found in classic psychological experiments and which games can give the chance to answer.
Other games like Zelda require the player to use multiple cognitive aspects simultaneously like planning, exploration and memorywhich are traditionally studied separately with classical methods, thus ignoring potential interaction effects.
Obviously, the use of games in the context of Research encounters limits, starting with the difficulty of controlling the different parameters or even inter-individual variability concerning, for example, the experience that subjects have with the same game before take part in the test. However, these limitations are not insurmountable with a good study design, according to the authors.
So what about you, would you agree to play for Science?
