Abstract
Humans initially learn about objects through the sense of touch, in a process called “haptic exploration.” In this paper, we present a neural network model of this learning process. The model implements two key assumptions. The first is that haptic exploration can be thought of as a type of navigation, where the exploring hand plays the role of an autonomous agent, and the explored object is this agent's “local environment.” In this scheme, the agent's movements are registered in the coordinate system of the hand, through slip sensors on the palm and fingers. Our second assumption is that the learning process rests heavily on a simple model of sequence learning, where frequently-encountered sequences of hand movements are encoded declaratively, as “chunks.” The geometry of the object being explored places constraints on possible movement sequences: our proposal is that representations of possible, or frequently-attested sequences implicitly encode the shape of the explored object, along with its haptic affordances. We evaluate our model in two ways. We assess how much information about the hand's actual location is conveyed by its internal representations of movement sequences. We also assess how effective the model's representations are in a reinforcement learning task, where the agent must learn how to reach a given location on an explored object. Both metrics validate the basic claims of the model. We also show that the model learns better if objects are asymmetrical, or contain tactile landmarks, or if the navigating hand is articulated, which further constrains the movement sequences supported by the explored object.
Highlights
How do we acquire knowledge about the objects we encounter in the world? While vision is probably the dominant source of information for most mature adults, our primary source of information about objects comes from the sense of touch
In earlier work with a simple unarticulated agent (Yan et al, 2018a,b), we showed that an MSOM model can learn some knowledge about 3D objects through the agent’s tactile exploration
To investigate the performance of the proposed articulated goal-oriented learning model, we compare it with a baseline, in which the same articulated agent performs a random walk on the 2 × 2 × 2 cube
Summary
How do we acquire knowledge about the objects we encounter in the world? While vision is probably the dominant source of information for most mature adults, our primary source of information about objects comes from the sense of touch. Articulated Tactile Learning primary, while those derived from vision are only “secondary”: visual representations of objects only acquire their meaning through associations with touch-based representations. This argument is still largely unchallenged, but there are relatively few models of how infants acquire touch-based representations of objects. We know this process involves active exploration of objects, because touch only delivers information about an object serially, at the points of contact (Dahiya et al, 2009, 2013). The second is that the mechanism controlling haptic exploration can be reduced in large part to a domain-general circuit for learning regularities in sequences We will introduce these ideas in turn
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
