Look and learn: A model of gaze-contingent learning

doi:10.1109/DEVLRN.2016.7846833

Paper

Look and learn: A model of gaze-contingent learning

Published Sep 1, 2016 · M. Murakami, J. Bolhuis, T. Kolling

2016 Joint IEEE International Conference on Development and Learning and Epigenetic Robotics (ICDL-EpiRob)

UNKNOWN SJR score

0

Citations

0

Influential Citations

Full text Semantic Scholar

Abstract

We present a computational model of action discovery, which reproduces the functioning bias that infants exhibit during gaze-contingent learning. This first result encourages us to apply the model to ongoing gaze-contingent learning tasks in which aspects like memory, control, and learning progress are studied. The model allows us to explore effects of learning speed, habituation, and prior knowledge on simulated behavior. By comparing simulations with empirical data, we can then devise mechanistic explanations about the observed behavior on the level of groups of subject as well as the individual. Additionally, the model allows us to develop testable predictions about gaze behavior in experimental scenarios that have not been realized yet. We believe that by capturing the essence of behavioral experiments, the model may help to bridge the gap between neuronal processes and human behavior.

Study Snapshot

Key takeawayThis computational model reproduces the functioning bias that infants exhibit during gaze-contingent learning, allowing us to explore effects of learning speed, habituation, and prior knowledge on simulated behavior.

PopulationOlder adults (50-71 years)

Sample size24

MethodsObservational

OutcomesBody Mass Index projections

ResultsSocial networks mitigate obesity in older groups.

Look and learn: A model of gaze-contingent learning

References

Citations