Understanding the Dynamics of Analogy-Making: New Analytical Techniques from Eye-Tracking
Robert M. French, Research Director, CNRS, Université de Bourgogne, France
Venue | Senior Common Room, Level 2 (2D17), Priory Road Complex
Date | 31st January 2019
In recent years eye-tracking has begun to be used to study the dynamics of analogy making. There are numerous scanpath-comparison algorithms and machine-learning techniques that can be applied to the raw eye-tracking data. We show how scanpath-comparison algorithms, combined with multidimensional scaling and a classification algorithm, can be used to resolve an outstanding question in analogy making — namely, whether or not children’s and adults’ strategies in solving analogy problems are different. (They are.) We show which of these scanpath-comparison algorithms is best suited to the kinds of analogy problems that have formed the basis of much analogy-making research over the years. Further, we use machine-learning classification algorithms to examine the item-to-item saccade vectors making up these scanpaths. We show which of these algorithms best predicts from very early on in a trial, based on the frequency of various item-to-item saccades, whether a child or an adult is doing the problem. Analyses of this type can also be used to predict, based on the item-to-item saccade dynamics in the first third of a trial, not only whether the person solving the analogy problem was an adult or a child, but also whether the problem was complex (i.e. difficult to solve) or simple (i.e., straightforward solution) and whether or not the a problem will be solved correctly.
All Welcome | Tea, coffee and cakes will be available after the seminar.