NICOP - Affordance Landscape Viewer (ALV): Learning new skills by visualization object affordances that only experts can see
Abstract
Affordances are interactive possibilities offered by the environment, such as the ~sit-~~ability~ of a chair or the ~grasp-~~ability~ of a climbing hold. Perceiving the affordances of an ordinary object like a chair or a cup is relatively simple, but only experts can fully perceive the affordances of complex objects like a climbing wall or the control panel of an airplane ~ which implies that not only experts and novices have different skills, but they also perceive the same objects differently. For example, only expert climbers can understand the interactive possibilities offered by a climbing wall, and how these interactive possibilities ~ or the wall~s ~affordance landscape~ ~ change dynamically as climbers move, e.g., which climbing holds become available after a movement, and how to grasp them to reach the next hold and ultimately the top of the wall. This advanced affordance perception ability is part and parcel of expertise, over and above better motor skill. Is it possible to show to novices the affordance landscapes that only experts can see? And would help novices learn new skills (e.g., climbing)? This 3-~~year project explores the possibility of transferring experts~ advanced affordance perception abilities to novices, to help the latter learning new skills (e.g., climbing). We will use augmented / virtual reality to show novices the affordances that only an expert usually sees (e.g., the dynamical affordance landscape or a climbing wall), and systematically test whether ~seeing the world through an expert~s eyes~ would facilitate skill learning and generalization to novel situations. Besides addressing fundamental questions about the ways experts and novices perceive affordances, this project will pave the way for a novel, possibly disruptive skill training technology.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Jan 23, 2019
- Source ID
- N629091912017
Entities
People
- Giovanni Pezzulo
Organizations
- Consiglio Nazionale delle Ricerche
- Office of Naval Research
- United States Navy