This is a proposal to support a biennial international meeting on the topic of Sensory Coding and the Natural Environment, along with a web resource that will provide a directory of people and publications in the field, as well as a medium for exchanging data and algorithms. The theme of the meeting is highly interdisciplinary, drawing upon expertise in systems and cognitive neuroscience, perceptual psychology, statistics, signal processing, and computer science.
The aim i s to model and understand sensory processes in relation to the statistical structure of the natural environment. This approach is broadly applicable to any sensory modality of any organism. A number of studies over the past decade have shown that the sensory coding strategies of many animals may be understood in terms of efficient coding strategies applied to natural scenes especially in the visual and auditory domains of both vertebrates and invertebrates. This approach is thought to have great potential for shedding light on neural information processing strategies, as well as advancing the development of neural prostheses capable of transforming natural images and sound into a format interpretable by the brain. Two previous meetings have been held on this topic, in 1997 and 2000, and the number of investigators now working in this field, not to mention those entering it, has outgrown these small, informal meetings. More importantly, there is a need to educate both students and current investigators about the techniques, methodologies, and types of results emerging from this field. Funding from this conference grant will enable us to invite experts in the field to a biennial Gordon Research Conference, as well as to provide travel grants and registration fee subsidies to students and post-docs interested in attending the meeting and learning about the field. The web site will complement this effort by providing continuity between the meetings as well as bringing work in the field to the attention of a wider audience.
