Elementary Motion Detection in<i>Drosophila</i>: Algorithms and Mechanisms

Hh, Yang; Clandinin, Thomas R.

doi:10.1146/annurev-vision-091517-034153

Cited by 53 publications

(55 citation statements)

References 129 publications

(248 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This motion directionality is detected when both the delayed signal from the left location ( 1 ) and the signal from the right location ( 2 ) arrive at the motion detector layer nearly simultaneously, Fig 1 (d), and excite a motion detector cell, Fig 1 (d) to identify a rightward motion (solid green lines, Fig 1). Our model, thus predicts the correct phi-motion direction in a similar way to many other variants of the HRC model (Clark et al 2011;Mo and Koch 2003;Leonhardt et al 2017Leonhardt et al , 2016Salazar-Gatzimas et al 2018;Tuthill, Chiappe, and Reiser 2011;Yang and Clandinin 2018).…”

Section: Results Modelsupporting

confidence: 61%

“…Our suggested model provides a novel approach to the elucidation of motion detection mechanisms, even though the basic structure of the model's architecture is similar to previous models (besides accounting for the cardinal neuronal response). Our model, as well as the previous motion models, are small variants of the classical HRC model of Hassenstein and Reichardt (Von Hassenstein and Reichardt 1956;Mo and Koch 2003;Leonhardt et al 2017Leonhardt et al , 2016Salazar-Gatzimas et al 2018;Adelson et al 1985) and share a common architecture (Clark et al 2011;Mo and Koch 2003;Leonhardt et al 2017Leonhardt et al , 2016Salazar-Gatzimas et al 2018;Tuthill, Chiappe, and Reiser 2011;Yang and Clandinin 2018), consisting of the following layers: 1) the photoreceptor layer, Fig 1 (a). 2) a rectification layer (Reiff et al 2010), which converts the signal to ON and OFF pathways, (e.g., L1/L2 in fly's lamina layers and retinal bipolar cells in mammalians), Fig 1(b).…”

Section: Results Modelmentioning

confidence: 99%

“…The ability to detect motion is an essential quality for an animal's survival (Clark et al 2011;Arenz et al 2017;Yang and Clandinin 2018), since it provides crucial information needed for exploring the external world. This information enables animals to perform tracking, navigation stabilization, and etc.…”

Section: Introductionmentioning

confidence: 99%

“…The low-level neuronal mechanisms and structures of motion detectors that can detect both these types of motions, have been the focus of numerous studies (Salazar-Gatzimas et al 2018;Clark et al 2011;Tuthill, Chiappe, and Reiser 2011;Theobald 2018;Luo-Li et al 2018;Mo and Koch 2003;Leonhardt et al 2017;Yang and Clandinin 2018). The pioneer motion study of Hassenstein and Reichardt (Von Hassenstein and Reichardt 1956), which used insects and has also been adapted to other animals, suggested a simple computational model (Alexander Borst 2000), the Hassenstein and Reichard Correlator (HRC), based on a delay and correlation mechanisms.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

The rebound response plays a role in the motion mechanisms and perception

Cohen-Duwek

Spitzer

2020

Preprint

View full text Add to dashboard Cite

9Motion estimation is an essential ability for sighted animals to survive in their natural 10 environment. Many anatomical and electrophysiological studies on low visual levels have been 11 based on the classic pioneering HRC (Hassenstein & Reichaedt Correlator) computational model. 12 The accumulated experimental findings, which have given rise to a debate in the current 13 computational models regarding the interaction between the On and Off pathways. The previous 14 algorithms were challenged to correctly predict physiological experiment results and the two types 15 of motion: a) Phi motion, also termed apparent motion. b) Reverse-phi motion that is perceived 16 when the image contrast flips during the rapid succession. We have developed a computational 17 model supported by simulations, which for the first time leads to correct predictions of the 18 behavioral motions (phi and reverse-phi), while considering separated On and Off pathways and is 19 also in agreement with the relevant electrophysiological findings. This has been achieved through 20 the well-known neuronal response: the rebound response or "Off response". We suggest that the 21 rebound response, which has not been taken into account in the previous models, is a key player in 22 the motion mechanism, and its existence requires separation between the On and the Off pathways 23 for correct motion interpretation. We furthermore suggest that the criterial reverse-phi effect is 24 only an epiphenomenon of the rebound response for the visual system. The theoretical predictions 25 are confirmed by a psychophysical experiment on human subjects. Our findings shed new light on 26 the comprehensive role of the rebound response as a parsimonious spatiotemporal detector for 27 motion and additional memory tasks, such as for stabilization and navigation. 28 29 32the external world. This information enables animals to perform tracking, navigation stabilization, 33 and etc. A knowledge of the neuronal mechanisms and the structures of the motion detectors is 34 critical to understand how an animal behaves in its complex world. 35In general, animals, from mammals to insects, are sensitive to two types of motion. These were 36 described by pioneering studies, as a common apparent or phi motion, and an illusory motion, the 37 reverse-phi illusion (Anstis 1970). Phi motion requires at least two alternating stimuli, where the 38 second stimulus, is slightly shifted spatially relative to the first stimulus. This configuration yields 39 a perception of motion in the direction of the second stimulus. In contrast, for reverse-phi motion, 40 the second stimulus is displayed with inverse contrast polarity (in addition to the shift in spatial 41 location, as for phi motion). This yields a perception of motion in the opposite direction towards 42 the first stimulus (Anstis 1970). 43 65 into two types, where the first assumes that directionality can be detected by discrete ON and OFF 66 pathways (Leonhardt et al. 2017, 2016), while the second type assumes that there must...

show abstract

Section: Results Modelsupporting

confidence: 61%

Section: Results Modelmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

The rebound response plays a role in the motion mechanisms and perception

Cohen-Duwek

Spitzer

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…These phenomenological models have provided striking insights into neural and behavioral 42 responses in a variety of species, including in flies (Yang and Clandinin, 2018). 43…”

Section: Introduction 33 34mentioning

confidence: 99%

A minimal synaptic model for direction selective neurons inDrosophila

Zavatone-Veth

Badwan

Clark

2019

Preprint

View full text Add to dashboard Cite

14 15 Visual motion estimation is a canonical neural computation. In Drosophila, recent advances have 16 identified anatomical and functional circuitry underlying direction-selective computations. 17Models with varying levels of abstraction have been proposed to explain specific experimental 18 results, but have rarely been compared across experiments. Here we construct a minimal, 19 biophysically inspired synaptic model for Drosophila's first-order direction-selective T4 cells 20 using the wealth of available anatomical and physiological data. We show how this model relates 21 mathematically to classical models of motion detection, including the Hassenstein-Reichardt 22 correlator model. We used numerical simulation to test how well this synaptic model could 23 reproduce measurements of T4 cells across many datasets and stimulus modalities. These 24 comparisons include responses to sinusoid gratings, to apparent motion stimuli, to stochastic 25 stimuli, and to natural scenes. Without fine-tuning this model, it sufficed to reproduce many, but 26 not all, response properties of T4 cells. Since this model is flexible and based on straightforward 27 biophysical properties, it provides an extensible framework for developing a mechanistic 28 understanding of T4 neural response properties. Moreover, it can be used to assess the 29

show abstract

Visual Processing in Free Flight

Egelhaaf

2022

Encyclopedia of Computational Neuroscience

View full text Add to dashboard Cite

Elementary Motion Detection inDrosophila: Algorithms and Mechanisms

Cited by 53 publications

References 129 publications

The rebound response plays a role in the motion mechanisms and perception

The rebound response plays a role in the motion mechanisms and perception

A minimal synaptic model for direction selective neurons inDrosophila

Visual Processing in Free Flight

Contact Info

Product

Resources

About