Interaction of efferent and afferent signals in visual perception A history of ideas and experimental paradigms

A historical review of the development of interaction theories in visual perception is presented. The concept that efferent signals generated in the eye or central nervous system interact with afferent visual signal flow dates back to the circles of pre-Socratic philosophers. They believed, however, that the interaction between observer-generated ‘pneuma’ and visual objects takes place at the site of the object or in the extrapersonal space between objects and eye. This idea was elaborated by Plato, the Stoic philosophers, Galen and some church fathers, but rejected by Aristotle and his school. The interaction theory was modified by Arabian medieval scientists (e.g., Alhazen, Avicenna), who believed the interaction of afferent and efferent signal flow to occur within the eye at the site of the pupil. The interaction theory finally disappeared during the first half of the 18th century when Alkmaion's age-old idea of ‘efferent light’ generated in the eye was experimentally refuted. With the rediscovery of Aristotle's observation of eye-movement-related afterimage movement, however, interaction theory reappeared towards the beginning of the 19th century, and sensory physiologists were asking why the world is perceived as stable despite the fact that its image shifts continuously across the retina (Erasmus Darwin, Stembuch, Purkyneě, Bell). The idea of ‘cancellation’ between afferent visual movement signals and corollary signals evoked by the motor compounds of gaze movement (now called efference copy signals) was first proposed by Purkyně. It was further developed during the 19th century by leading sensory physiologists such as Hering, Helmholtz, Mach and their pupils. The first block diagrams of this idea were presented by Mach (1906) and Von Uexküll (1920/1928). These concepts led to the ‘reafference principles’ of Von Holst and Mittelstaedt (1950) and Sperry (1950).