Desplan Lab
New York University. 100 Washington Square East.
1009 Silver Center. New York, NY. 10003.

Post-Doctoral Fellow

Rudy Behnia

(212) 992-9526


  • 2006. Cambridge University. PhD.


  • Mechanisms describing how neurons compute direction-selective signals through interpretation changes in luminance in space and time have been the focus of extensive studies in neuroscience. In the 1950s, Hassentein and Reichardt developed their now famous, eponymous correlator model (HRC). This mathematical model relies on the differential filtering of two spatially separated input channels so that the output of one becomes delayed as compared to the other. This delay allows direction-selective amplification of the output only when the delayed and non-delayed signals coincide in time, signaling motion (Movie 1). Since the inception of this model, a wealth of data on insects including Drosophila, have demonstrated that motion responses display the fundamental signatures predicted by the HRC, but the neurons that implement the different steps of the HRC have remained elusive.

  • Movie 1

  • In an effort to understand the role of specific optic lobe neurons in the processing of visual information, starting with motion detection, I have developed an in-vivo system in which the activity of medulla neurons in response to specific patterns of light is measured using whole-cell patch-clamp recording (Figure 1). I am combining our detailed knowledge of the anatomy of the optic lobe and the availability of specific marker lines to target my recordings to specific neuronal types using GFP expression.

  • Figure 1

  • I am currently routinely recording from Mi1, Tm3, Tm1 and Tm2 neurons, four columnar cell types in the medulla part of the optic lobe, whose connectivity patterns are suggestive of an involvement in motion detection (Figure 2). Indeed, Mi1 and Tm3 are postsynaptic to lamina neuron L1 and presynaptic to T4 neurons while Tm1 and Tm2 are both postsynaptic to lamina neuron L2 and presynaptic to T5 Neurons. L1 and L2 are the first steps in the motion detection pathway in Drosophila. The L1 pathway has been shown to feed into a motion circuit that is dedicated to the detection of moving bright edges whereas the L2 pathway feeds into a motion circuit that responds to dark edges. T4 and T5 are both direction selective, with T4 responding exclusively to moving bright edges and T5 to moving dark edges.

  • Figure 2

  • I present flies with flashes of light and Gaussian noise stimuli while recording from Mi1, Tm3, Tm1 and Tm2, in order to assess their response to brightness increments and decrements as well as their kinetic properties. Using these techniques I have shown that these four neurons in the medulla part of the optic lobe implement processing steps that are fundamental to the detection of the direction of motion in Drosophila. Indeed Mi1 andTm3 respond selectively to brightness increments, with the response of Mi1 delayed relative to Tm3. Conversely, Tm1 and Tm2 respond selectively to brightness decrements, with the response of Tm1 delayed compared with Tm2 (Figures 3 and 4). In an article published in 2014 in Nature, we propose that Mi1 and Tm3 perform critical processing of the delayed and non-delayed input channels of the correlator responsible for the detection of light edges, while Tm1 and Tm2 play analogous roles in the detection of moving dark edges. Our data show that these specific medulla neurons possess response properties that allow them to implement the algorithmic steps that precede the correlative operation in the Hassenstein–Reichardt correlator, revealing elements of the long-sought neural substrates of motion detection in the fly.

  • Figure 3

    Figure 4

  • I will be extending this type of experiments to other cells types and I am modifying the system to study other kinds of visual processing such as color discrimination in the fly optic lobe of Drosophila.


  • Behnia R, Clark DA, Carter AG, Clandinin TR and Desplan C. Processing properties of Drosophila ON and OFF pathway for motion detection. Nature doi:10.1038/nature13427


  • European Molecular Biology Organization (EMBO)
  • Human Fronteir Science Program (HFSP)