Contrast coding in the electrosensory system: parallels with visual computation

• Published in: Nature reviews. Neuroscience Vol. 16; no. 12; pp. 733 - 744
• Main Authors: Clarke, Stephen E., Longtin, André, Maler, Leonard
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 01.12.2015; Nature Publishing Group
• Subjects: 631/378/116/2395; 631/378/2613/1786; 631/378/3917; 9/30; 9/97; Afferent Pathways - cytology; Afferent Pathways - physiology; Algorithms; Animal Genetics and Genomics; Animals; Behavior, Animal; Behavioral Sciences; Biological Techniques; Biomedicine; Brain mapping; Brain research; Electric fields; Electric instruments; Electric Organ - physiology; Neural circuitry; Neural Networks (Computer); Neurobiology; Neurosciences; Perception; review-article; Senses; Sensory Receptor Cells - physiology; Social interaction; Canada; Ottawa Ontario Canada
• ISSN: 1471-003X; 1471-0048; 1469-3178
• DOI: 10.1038/nrn4037

Key Points First- and second-order contrast stimuli can be described for the visual, auditory and tactile senses. Similarly, electrosensory contrast patterns vary over large spatial and temporal scales. Fast exponential adaptation and power law adaptation in electrosensory afferent neurons partition the range of natural electrosensory stimulus frequencies and enable different signal processing goals. Comparison of the electrosensory and retinal adaptation algorithms reveals many similarities. The electrosensory of ON and OFF ganglion cells can also be compared to retinal ON and OFF ganglion cells. Despite different biophysics and network architectures, these cells share the same algorithmic role for the electrosense and vision. Envelope encoding and decoding mechanisms in the electrosense are related to both locomotion and social behaviours. Both the electrosense and vision combine ON and OFF cell responses to improve the coding efficiency of second-order contrast stimuli. Motion reversal triggers switches in electrosensory ON and OFF cell preferences for spatial contrast (polarity), which has also been noted to occur in salamander and mouse retina. Individual ON and OFF cell firing rates encode scalar quantities of motion such as object distance and speed, whereas sequences of population activity encode vector information (motion direction). There are many benefits of these flexible coding paradigms for spatiotemporal contrast using ON and OFF cells. It will be important to understand how downstream decoding neurons interpret patterns and sequences of activity of ON and OFF cell populations. Sensory systems encode and interpret patterns of contrast in sensory signals to provide an accurate representation of an animal's environment. Maler and colleagues here outline our current understanding of the principles of contrast coding in the electrosensory system and make comparisons with contrast coding in the visual system. To identify and interact with moving objects, including other members of the same species, an animal's nervous system must correctly interpret patterns of contrast in the physical signals (such as light or sound) that it receives from the environment. In weakly electric fish, the motion of objects in the environment and social interactions with other fish create complex patterns of contrast in the electric fields that they produce and detect. These contrast patterns can extend widely over space and time and represent a multitude of relevant features, as is also true for other sensory systems. Mounting evidence suggests that the computational principles underlying contrast coding in electrosensory neural networks are conserved elements of spatiotemporal processing that show strong parallels with the vertebrate visual system.