Analyzing Risk-Taking Behaviors on a Computer
According to a study conducted at the Ohio State University, the risk-taking attitudes assumed by humans (open to risk-taking/avoiding risk-taking), may differ from the risk-related behaviors that they show.
The team that conducted the research analyzed a total of 215 selections, which were made by 652 individuals on games that required them to make risky decisions like in a gamble, by using the movements of a mouse cursor. In this study, where the magnitudes of the risks varied, a great number of data, such as the indecisiveness experienced by participants prior to the decisions that they turned into actions, or being sure of the decisions that they made, were analyzed based on the movements of the mouse cursor.
During the experiment, the participants were shown two boxes, and they were asked to pick one of those boxes. One of the boxes contained a risky choice, which, for example, stated that there would be an earning of 10 Dollars by a possibility of 50%, and a loss of 5 Dollars by a possibility of 50%. The other box, on the other hand, offered a choice in which generally no specific amount could be earned or lost.
Throughout this process, the movements of the mouse cursor were monitored to see how the participants made that specific choice before they had made their final decisions to make a choice.
The conclusions drawn from the research:
- The more the mouse cursor moves towards the choice beyond the decision made, the harder the choice gets for that specific person.
- Individuals, who directly gravitated towards to the decisions that they made, had experienced less inner conflicts, and they were more sure of the decision that they made during the experiment.
- Those who headed towards the safe choice before they made their decisions, could have been abstaining from more risk than their choices showed, even if they had selected a risky choice.
- On the other hand, those who gravitated towards a risky choice before making their decisions, may have tended to take even more risks than their choices demonstrated, even if they marked the reliable option.
As reported by Prof. Ian Krajbrich, one of the administrators of the study, the experiment administrators who monitored the decision-making process for once, could accurately estimate the other decisions they would make during the research to a large extent. Reporting that it is extremely rare to foresee the other decisions by observing the initial decision in such experiments, Prof. Krajbich says “The data obtained only from choices fall short for interpretation for a lot of reasons. For instance, you cannot see how strong the decision of a person is, or how prone they are to make the other decision, based on these data. Analyzing the movements of the mouse cursor also provides us with these data.”
Hence, during the experiment, it is possible to guess how many of the participants, who made the same decisions in the game that included the same offer, would gravitate towards the other choice in other games with the same magnitude of risk. The participants, who made the opposite decision of the ones that they made in the decision-making phase of the first game, could make a selection that they gravitated towards but did not prefer to choose in games with similar risks.
Paul Stillman, the head administrator of the experiment, interprets the significance of the study by stating that “it enables us to analyze the risk-taking behaviors of individuals from a richer perspective by showing the dichotomies in the decision-making process”.
– Measuring Risk-Taking by Watching People Move Computer Mice. Jeff Grabmeier – Ohio State University.
– Stillman, P. E., Krajbich, I., & Ferguson, M. J. (2020). Using dynamic monitoring of choices to predict and understand risk preferences. Proceedings of the National Academy of Sciences.