Jan Bielecki:
Vision and visual information processing in cubozoans

Date: 30-06-2013    Supervisor: Jens T. Høeg & Anders Garm



Eyes have been considered support for the divine design hypothesis over evolution because, surely, eyes cannot function with anything less than all the components that comprise a vertebrate camera type eye. Yet, devoted Darwinists have estimated that complex visual systems can evolve from a single light sensitive cell within 400 000 generations and all intermediate stages can be found throughout the Metazoa. Eyes have evolved to accommodate increasingly more complex visual behaviours, from light sensitive tissues involved in circadian entrainment to the complex camera type eyes that can guide animals in a wide range of behaviours. It is intuitive that a complex eye is energetically very costly, not only in components but also in neural involvement. The increasing behavioural demand added pressure on design specifications and eye evolution is considered an optimization of the inverse relationship between acuity and light sensitivity. Animals have evolved a wide variety of solutions to this problem such as folded membranes, to have a larger receptive surfaces, and lenses, to focus light onto the receptive membranes. On the neural capacity side, complex eyes demand huge processing network to analyse the received information, illustrated by the fact that one third of the human brain is devoted to visual information processing. The cost of maintaining such neural network deter most organisms from investing in the camera type option, if possible, and settle for a model that will more precisely fit their need. Visual neuroethology integrates optics, sensory equipment, neural network and motor output to explain how animals can perform behaviour in response to a specific visual stimulus.

In this doctoral thesis, I will elucidate the individual steps in a visual neuroethological pathway by considering the experimentally tractable system of cubozoan jellyfish. The cubomedusae are equipped with rather complex camera type eyes by which they modulate their elaborate behaviours. However, cubozoans were a little short-sheeted in the neural department, and have to accomplish image analysis with about 1000 neurons, which make these stunning animals the perfect model organism to explore basic visual information processing.