Lassonde Professor part of Research team that Published a Paper in the Journal of Vision
An interdisciplinary research team comprised of Department of Psychology Ph.D. student Audrey Wong-Kee-You and her primary supervisor, Professor Scott A. Adler, collaborating with Lassonde’s own Professor John Tsotsos, from the Department of Electrical Engineering & Computer Science, recently published a paper titled Development of spatial suppression surrounding the focus of visual attention in the Journal of Vision on a breakthrough discovery relevant both to child development and to machine learning algorithms.
A main component of the research involved experiments held at the Ontario Science Centre with 400 kids participating, designed and conducted by Wong-Kee-You. The purpose of the experiment was to determine if young children can suppress their visual surroundings, when attempting to inspect a specific detail in front of them (i.e. in order to recognize if someone is wearing a hat do human brains have the innate ability to suppress other details of the person like their hair colour), or if this ability develops over time.
Wong-Kee-You’s research concluded that this quality only develops by 16 – 18 years of age and is in fact, not an innate characteristic of the human brain. Thus, children have difficulty suppressing irrelevant visual details when having to resolve other specific details.
The research can be applied in countless ways such as informing educational practices by enabling educators to better align teaching methods and educational requirements to age-appropriate attentional skills or influencing how environments for children are designed.
In the future we may be better equipped to recognize difficulties in children and determine how to best deal with them to improve their visual function.
This is also a revelation for those in the field of Machine Learning as the principles of machine learning theory are built on the idea that one need only feed an artificial brain with sufficient data. These results show that the human brain takes years to develop and does so while it learns about its world.
This insight from human visual development suggests that a different approach to machine learning and to Artificial Intelligence is needed if it is to truly behave as humans do.
The interdisciplinary nature of this research is a key part of its success.
“I had spent a good portion of my PhD reading and learning about descriptive models of visual attention but meeting John [Professor Tsotsos] and working on this project gave me the opportunity to familiarize myself with his computational theory, the Selective Tuning model of attention. This gave me a different perspective and allowed me to consider the strengths of theories with an algorithmic mechanism,” says Audrey Wong-Kee-You.