THEOBALD, J.C.*; O’CARROLL, D.: Neural strategies that optimize motion detection in the crepuscular hawkmoth Manduca sexta
Perhaps the greatest demand on visual system design occurs with the transition from day to night, since illuminance can differ by a factor of 108. For an animal to function in darker conditions, its visual system must deal with an inherently ‘noisier’ image. This is a special problem for hovering hawkmoths, as the detection of low image velocities required for hovering is visually demanding. We tested the hypothesis that hawkmoths use not just optical, but neural strategies to adapt their vision and motion detection to low light levels. We compared the responses of neurons in the lobula plate (a retinotopic neuropil rich in motion-sensitive cells) and photoreceptors to a moving sinusoidal stimulus of varying stripe density (spatial frequency), rate (temporal frequency), and light intensity. Our results support the hypothesis that hawkmoths use temporal neural filtering at low light levels, but not spatial filtering. Because the velocity of an image is the quotient of its temporal and spatial frequencies, low velocity information is found in both low temporal and high spatial frequencies. By sacrificing only high temporal frequency information, hawkmoths improve the signal to noise ratio of a dark image, but retain the high spatial frequency information required for hovering in spatially complex habitats.
