ProfBrain

Research Study Title:

The robustness of using economic games to cluster players: a case study in computational psychiatry

Investigators: (From the School of Computing / Insight Centre for Data Analytics)

· Dr. Darragh Walsh (Post-doctoral researcher)
· Lili Zhang (PhD candidate)
· Prof. Tomas Ward (AIB Chair of Data Analytics)

Research summary:
There is a growing recognition that computational methods, such as machine learning (ML), could be usefully applied in psychiatric research. Numerous mental illnesses could be classified as disorders that affect decision making and how rewards influence our behaviour through learning. To assist psychiatrists in making diagnoses more objective, researchers have begun to explore how classic economic games could be used to cluster players into distinct groups using ML. These games examine a players’ ability to think rationally to maximise their reward, whether that requires cooperating or acting selfishly. Typical games include the prisoner’s dilemma and the stag hunt.

Though using ML techniques to cluster players into groups is promising, it is important that the clustering is meaningful and stable across different contexts/situations. This project will examine the reliability of these games to cluster healthy adults only. Healthy adults will be clustered according to their performance playing classic economic games. The clustering will be performed by ML techniques following methodology developed in Poncela-Casasnovas et al. (2016)[1].

If you agree to participate in this project:

You will need to download an app and play a game that requires you to choose an option that gives you the best chance of earning the most points, taking into account that your opponent (the computer player 2) will be doing the same thing. You need to make 16 decisions in total (there are 16 rounds of the game). You will then be asked to answer a short questionnaire (the mini-IPIP questionnaire developed by Donnellan et al. (2006)[2]) and finally play another game which has 16 rounds. In total, you should be able to complete all tasks in approximately 20-30 minutes.

No personal information is collected:

We do not ask for your:

· Name
· Email address or
· Phone number

We do ask for your:

· Gender
· Age-range (for example 21 – 30, or 31 – 40)
· Employment status (e.g. Student, employed, unemployed, other)

Anticipated benefits from participating in this project:

Participating in this project will help determine how much information can be reliably concluded from studies using economic games to examine decision making.

We do not foresee any risks to participants from completing these tasks. Participants will be free to change their mind and withdraw from the study at any time until they submit their final response and tick to consent to their
responses being collected.

Since no personal data is collected, participants will not be informed of the results of this project. However, the completed work will be submitted for publication where the anonymised results will be displayed (similar to the results in Poncel-Casasnovas et al.[1]).

This research is funded by AIB.

Further details on the project may be obtained from Lili Zhang(lili.zhang 27@mail.dcu.ie).

If participants have concerns about this study and wish to contact an independent person, please contact:

The Secretary, Dublin City University Research Ethics Committee, c/o Research and Innovation Support, Dublin City University, Dublin 9. Tel 01-7008000, e-mail rec@dcu.ie

[1]Poncela-Casasnovas, J., et al.Humans display a reduced set of consistent behavioral phenotypes in dyadic games. Sci Adv 2, e1600451 (2016).
[2]Donnellan, M.B., et al. The mini-IPIP scales: tiny-yet-effective measures of the big five factors of personality. Psychological assessment 18, 192-203, (2006).