A new approach to understanding human behaviour

My name is Susan Fennell and I am a first year PhD student.

My area of research is social psychology, the study of how a person’s feelings and behaviour are affected by society. Up until recently, it was thought that a person’s behaviour is affected by the norms present in society but that they do not have any effect on these norms, in other words that people conform. This view is changing. Recent developments in social psychology suggest that social identities and norms develop through interaction. People do not always conform, they can rebel against and change social norms.

VIAPPL, the Virtual Interaction APPLication, is a software platform which allows researchers to conduct experiments that incorporate this dynamic aspect of social interaction. There are typically 8 to 14 participants in each experiment who are represented as avatars in a virtual environment. Participants interact by exchanging tokens over a number of rounds. The figure below shows what a participant in this experiment would see during a round of the game. The node corresponding to the participant is in bold and in this case he has chosen to give a token to the purple player on the left of the circle.

I must admit that when I first heard about this setup I was a little confused. What do these tokens mean? Do people not just choose other people at random to give tokens to? What is the aim of this “game”? I am sure that at first the participants must ask themselves these questions. At the start, people presumably do give at random. However, certain behaviours develop as the game progresses. For example, there are two groups in the experiments and from looking at the data we can see that participants give significantly more to people in their own group than the other group. People feel some sense of loyalty or connection to members of their own group.

In the experiments, participants must choose between competing actions at each round. Will I give to that player because he gave to me last round even though he is not in my group? Or will I give to the player who has significantly less tokens than everyone else? Expectations of how to act develop as the rounds progress to produce norms such as in-group favouritism and fairness.

VIAPPL is a flexible environment which allows for many different research questions to be answered. For example the nodes corresponding to each player can be made more individual, either by randomly assigning a different image to each node or by attaching a participant’s photo, which may change how attached participants feel to their group. The wealth (in tokens) of each group can be different at the start of the game (for example participants in group A might each start with 20 tokens while participants in group B start with 40), which allows us to investigate how inequality affects people’s behaviour.

While VIAPPL has the potential to revolutionise our understanding of how humans interact, it is necessary to develop methods of analysing the experimental data in order to answer the questions we have, which is my job as mathematician and statistician. This has proved challenging (which is not a bad thing!) but having such an interesting application to the work I do is great motivation and it is exciting to be involved in a project which is at the forefront of research in social psychology.

Susan Fennell is a PhD student at MACSI working with Professor James Gleeson and Dr Kevin Burke (Dept. Mathematics and Statistics) and Dr. Michael Quayle (Dept of Psychology)