In his doctoral research, Borst studied the olfactory sense of Drosophila melanogaster. He discovered that dedicated brain structures known as mushroom bodies play a crucial role in olfactory learning in flies (Heisenberg, Borst, Wagner, Byers, J. Neurogenetics 1985).[2]
Since then, his work has focused on the question of how nerve cells process information and perform certain calculations. As an example of neuronal computation, he uses motion vision, i.e. the process by which nerve cells in the visual center of flies calculate the direction of motion of an object.
Borst’s research aims to connect a formal, mathematical description of these processes with the underlying biophysical properties of nerve cells. By combining theoretical approaches and computer simulations with various experimental studies, he made the following major discoveries:
The calculation of directional motion in the fly brain formally follows to a large extent the so-called Reichardt model (Single & Borst, Science 1998).[3]
Similar to the vertebrate retina, this computation is performed in two parallel pathways, an ON and an OFF channel (Jösch et al., Nature 2010).[4]
In each of these channels, four subsets of neurons exist (T4 cells in the ON, T5 cells in the OFF channel), each maximally sensitive to one of the four orthogonal directions of motion (right, left, up, down). These cells are connected to four separate layers of neural tissue, where they are interconnected with large heading control neurons (Maisak et al., Nature 2013).[5]
In T4 cells, the signal multiplication postulated in the Reichardt model is biophysically based on disinhibition (Groschner et al., Nature 2022).[6]
