Intrinsically Motivated Learning of Visual Motion Perception and Smooth Pursuit
This work addresses the challenge of unified perception/action loops in computational neuroscience, offering a potentially foundational but incremental extension of existing frameworks.
The authors tackled the problem of modeling the co-development of sensory processing and behavior by extending efficient coding to the perception/action cycle, combining sparse coding and reinforcement learning to optimize neural encoding fidelity. They applied this to an active eye model, resulting in the development of smooth pursuit behavior and neurons resembling primary visual cortical motion-selective neurons.
We extend the framework of efficient coding, which has been used to model the development of sensory processing in isolation, to model the development of the perception/action cycle. Our extension combines sparse coding and reinforcement learning so that sensory processing and behavior co-develop to optimize a shared intrinsic motivational signal: the fidelity of the neural encoding of the sensory input under resource constraints. Applying this framework to a model system consisting of an active eye behaving in a time varying environment, we find that this generic principle leads to the simultaneous development of both smooth pursuit behavior and model neurons whose properties are similar to those of primary visual cortical neurons selective for different directions of visual motion. We suggest that this general principle may form the basis for a unified and integrated explanation of many perception/action loops.