Se samtliga publikationer

Neurophysiology of motion vision

Note, that we have a more up to date website.

As an animal moves through the world, its own movement generates optic flow across the retina, which it can use to maintain a straight path or to avoid obstacles. The visual system can also easily disambiguate the motion of objects that move independently of the surround from such self-generated optic flow.

Vision plays a large role for us humans, as well as for many other animals. If you have ever tried to walk in a straight line with your eyes closed, you know how important self-generated optic flow is for maintaining a straight trajectory. Besides such widefield optic flow cues, we can also visualize the motion of objects that move independently of the remaining visual surround. Such targets may represent the motion of a ball during a game of cricket or tennis. Despite the fact that you are moving, and thereby generating optic flow across your visual field, you can still visualize and indentify the independent trajectory of the ball. Motion vision is not only important for human sports stars, but also for insects who use these cues for tasks such as maintaining a straight flight trajectory, avoiding colliding with approaching tree trunks, and importantly, to identify targets, such as the motion of a conspecific or potential prey.

Studying the neurophysiology underlying target detection in human subjects, and other vertebrates, quickly becomes quite complicated. Besides the relative inaccessibility of the vertebrate visual cortex, there is the additional inconvenient complication of the eyes being able to move independently of the head. In insects, however, the eyes are fixed to the head’s exoskeleton, which means that we know what the insect looks at if we know what direction the head was facing. Intriguingly, however, despite being separated by huge evolutionary distances, and being equipped with completely different eyes, motion vision is coded in remarkably similar ways in the vertebrate visual cortex and the insect brain. We can therefore, somewhat surprisingly, use the insect visual system to understand the coding of visual cues in our own brain.

The underlying mechanisms that allow sensory systems to extract salient features from noisy surrounds are still poorly understood, despite being important for several senses. The visual detection of target motion is an interesting example of such feature extraction: Target visualization is computationally challenging, but evolution has solved it beautifully, even in the tiniest of insects, despite carrying eyes with low spatial resolution and small brains. As insects are physiologically accessible for in vivo recordings of single neurons, such as the exquisitely tuned small target motion detectors (STMDs), they provide an excellent model system for investigating the mechanisms underlying sensory selectivity. We are, for example interested in how the small insect nervous system efficiently extracts targets from cluttered backgrounds, with short behavioral delays (ca. 30 ms).

Insects who pursue targets have evolved a range of optical and behavioral adaptations to increase the relative contrast of targets during initial visualization. However, during actual pursuit, targets are inevitably displayed against the pursuer’s self-generated optic flow. For successful capture the pursuer must thus be able to visualize the target’s motion against a moving background. Considering that insects solve this faster than humans, despite carrying tiny brains and low-resolution compound eyes, is astonishing. We currently approach these questions using intracellular electrophysiology of single neurons in the optic lobes of intact insects while they view experimenter-controlled visual stimuli, enabling us to correlate the exact visual input with the neural response on a frame-by-frame basis.

Frank Lee, Olga Dyakova, Jozi Huotari, Malin Thyselius. Photo by Frank Lee.