Functionally and anatomically segregated visual pathways in the lobula complex of a calliphorid fly

Moving object detection based on bioinspired background subtraction

Flying insects rely mainly upon visual motion to detect and track objects. There has been a lot of research on fly inspired algorithms for object detection, but few have been developed based on visual motion alone. One of the daunting difficulties is that the neural and circuit mechanisms underlying the foreground-background segmentation are still unclear. Our previous modeling study proposed that the lobula held parallel pathways with distinct directional selectivity, each of which could retinotopically discriminate figures moving in its own preferred direction based on relative motion cues. The previous model, however, did not address how the multiple parallel pathways gave the only detection output at their common downstream. Since the preferred directions of the pathways along either horizontal or vertical axis were opposite to each other, the background moving in the opposite direction to an object also activated the corresponding lobula pathway. Indiscriminate or ungated projection from all the pathways to their downstream would mix objects with the moving background, making the previous model fail with non-stationary background. Here, we extend the previous model by proposing that the background motion-dependent gating of individual lobula projections is the key to object detection. Large-field lobula plate tangential cells are hypothesized to perform the gating to realize bioinspired background subtraction. The model is shown to be capable of implementing a robust detection of moving objects in video sequences with either a moving camera that induces translational optic flow or a static camera. The model sheds light on the potential of the concise fly algorithm in real-world applications.

Motion perception based on ON/OFF channels: A survey

Motion perception is an essential ability for animals and artificially intelligent systems interacting effectively, safely with surrounding objects and environments. Biological visual systems, that have naturally evolved over hundreds-million years, are quite efficient and robust for motion perception, whereas artificial vision systems are far from such capability. This paper argues that the gap can be significantly reduced by formulation of ON/OFF channels in motion perception models encoding luminance increment (ON) and decrement (OFF) responses within receptive field, separately. Such signal-bifurcating structure has been found in neural systems of many animal species articulating early motion is split and processed in segregated pathways. However, the corresponding biological substrates, and the necessity for artificial vision systems have never been elucidated together, leaving concerns on uniqueness and advantages of ON/OFF channels upon building dynamic vision systems to address real world challenges. This paper highlights the importance of ON/OFF channels in motion perception through surveying current progress covering both neuroscience and computationally modelling works with applications. Compared to related literature, this paper for the first time provides insights into implementation of different selectivity to directional motion of looming, translating, and small-sized target movement based on ON/OFF channels in keeping with soundness and robustness of biological principles. Existing challenges and future trends of such bio-plausible computational structure for visual perception in connection with hotspots of machine learning, advanced vision sensors like event-driven camera finally are discussed.

Bioinspired figure-ground discrimination via visual motion smoothing

Flies detect and track moving targets among visual clutter, and this process mainly relies on visual motion. Visual motion is analyzed or computed with the pathway from the retina to T4/T5 cells. The computation of local directional motion was formulated as an elementary movement detector (EMD) model more than half a century ago. Solving target detection or figure-ground discrimination problems can be equivalent to extracting boundaries between a target and the background based on the motion discontinuities in the output of a retinotopic array of EMDs. Individual EMDs cannot measure true velocities, however, due to their sensitivity to pattern properties such as luminance contrast and spatial frequency content. It remains unclear how local directional motion signals are further integrated to enable figure-ground discrimination. Here, we present a computational model inspired by fly motion vision. Simulations suggest that the heavily fluctuating output of an EMD array is naturally surmounted by a lobula network, which is hypothesized to be downstream of the local motion detectors and have parallel pathways with distinct directional selectivity. The lobula network carries out a spatiotemporal smoothing operation for visual motion, especially across time, enabling the segmentation of moving figures from the background. The model qualitatively reproduces experimental observations in the visually evoked response characteristics of one type of lobula columnar (LC) cell. The model is further shown to be robust to natural scene variability. Our results suggest that the lobula is involved in local motion-based target detection.

Visual projection neurons in the Drosophila lobula link feature detection to distinct behavioral programs

Visual projection neurons (VPNs) provide an anatomical connection between early visual processing and higher brain regions. Here we characterize lobula columnar (LC) cells, a class of Drosophila VPNs that project to distinct central brain structures called optic glomeruli. We anatomically describe 22 different LC types and show that, for several types, optogenetic activation in freely moving flies evokes specific behaviors. The activation phenotypes of two LC types closely resemble natural avoidance behaviors triggered by a visual loom. In vivo two-photon calcium imaging reveals that these LC types respond to looming stimuli, while another type does not, but instead responds to the motion of a small object. Activation of LC neurons on only one side of the brain can result in attractive or aversive turning behaviors depending on the cell type. Our results indicate that LC neurons convey information on the presence and location of visual features relevant for specific behaviors.

Color processing in the medulla of the bumblebee (Apidae: Bombus impatiens)

The mechanisms of processing a visual scene involve segregating features (such as color) into separate information channels at different stages within the brain, processing these features, and then integrating this information at higher levels in the brain. To examine how this process takes place in the insect brain, we focused on the medulla, an area within the optic lobe through which all of the visual information from the retina must pass before it proceeds to central brain areas. We used histological and immunocytochemical techniques to examine the bumblebee medulla and found that the medulla is divided into eight layers. We then recorded and morphologically identified 27 neurons with processes in the medulla. During our recordings we presented color cues to determine whether response types correlated with locations of the neural branching patterns of the filled neurons among the medulla layers. Neurons in the outer medulla layers had less complex color responses compared to neurons in the inner medulla layers and there were differences in the temporal dynamics of the responses among the layers. Progressing from the outer to the inner medulla, neurons in the different layers appear to process increasingly complex aspects of the natural visual scene.

Organization of local interneurons in optic glomeruli of the dipterous visual system and comparisons with the antennal lobes

The lateral protocerebrum of the fly's brain is composed of a system of optic glomeruli, the organization of which compares to that of antennal lobe glomeruli. Each optic glomerulus receives converging axon terminals from a unique ensemble of optic lobe output neurons. Glomeruli are interconnected by systems of spiking and nonspiking local interneurons that are morphologically similar to diffuse and polarized local interneurons in the antennal lobes. GABA-like immunoreactive processes richly supply optic glomeruli, which are also invaded by processes originating from the midbrain and subesophageal ganglia. These arrangements support the suggestion that circuits amongst optic glomeruli refine and elaborate visual information carried by optic lobe outputs, relaying data to long-axoned neurons that extend to other parts of the central nervous system including thoracic ganglia. The representation in optic glomeruli of other modalities suggests that gating of visual information by other sensory inputs, a phenomenon documented from the recordings of descending neurons, could occur before the descending neuron dendrites. The present results demonstrate that future studies must consider the roles of other senses in visual processing.

Visual system of calliphorid flies: motion- and orientation-sensitive visual interneurons supplying dorsal optic glomeruli

Intracellular recordings accompanied by dye fills were made from neurons associated with optic glomeruli in the lateral protocerebrum of the brain of the blowfly Phaenicia sericata. The present account describes the morphology of these cells and their electrophysiological responses to oriented bar motion. The most dorsal glomeruli are each supplied by retinotopic efferent neurons that have restricted dendritic fields in the lobula and lobula plate of the optic lobes. Each of these lobula complex cells represents a morphologically identified type of neuron arranged as an ensemble that subtends the entire monocular visual field. Of the four recorded and filled efferent types, three were broadly tuned to the orientation of bar stimuli. At the level of optic glomeruli a relay neuron extending centrally from optic foci and a local interneuron that arborizes among glomeruli showed narrow tuning to oriented bar motion. The present results are discussed with respect to the behavioral significance of oriented motion discrimination by flies and other insects, and with respect to neuroanatomical data demonstrating the organization of deep visual neuropils.

Diverse speed response properties of motion sensitive neurons in the fly's optic lobe

Speed and acceleration are fundamental components of visual motion that animals can use to interpret the world. Behavioral studies have established that insects discriminate speed largely independently of contrast and spatial frequency, and physiological recordings suggest that a subset of premotor descending neurons is in this sense speed-selective. Neural substrates and mechanisms of speed selectivity in insects, however, are unknown. Using blow flies Phaenicia sericata, intracellular recordings and dye-fills were obtained from medulla and lobula complex neurons which, though not necessarily speed-selective themselves, are positioned to participate in circuits that produce speed-selectivity in descending neurons. Stimulation with sinusoidally varied grating motion (0-200 degrees /s) provided a range of instantaneous velocities and accelerations. The resulting speed response profiles are indicative of four distinct speed ranges, supporting the hypothesis that the spatiotemporal tuning of mid-level neurons contains sufficient diversity to account for the emergence of speed selectivity at the descending neuron level. This type of mechanism has been proposed to explain speed discrimination in both insects and mammals, but has seemed less likely for insects due to possible constraints on small brains. Two additional recordings are suggestive of acceleration-selectivity, a potentially useful visual capability that is of uncertain functional significance for arthropods.

Systematic analysis of the visual projection neurons ofDrosophila melanogaster. I. Lobula-specific pathways

In insects, visual information is processed in the optic lobe and conveyed to the central brain. Although neural circuits within the optic lobe have been studied extensively, relatively little is known about the connection between the optic lobe and the central brain. To understand how visual information is read by the neurons of the central brain, and what kind of centrifugal neurons send the control signal from the central brain to the optic lobe, we performed a systematic analysis of the visual projection neurons that connect the optic lobe and the central brain of Drosophila melanogaster. By screening approximately 4,000 GAL4 enhancer-trap strains we identified 44 pathways. The overall morphology and the direction of information of each pathway were investigated by expressing cytoplasmic and presynapsis-targeted fluorescent reporters. A canonical nomenclature system was introduced to describe the area of projection in the central brain. As the first part of a series of articles, we here describe 14 visual projection neurons arising specifically from the lobula. Eight pathways form columnar arborization in the lobula, whereas the remaining six form tangential or tree-like arborization. Eleven are centripetal pathways, among which nine terminate in the ventrolateral protocerebrum. Terminals of each columnar pathway form glomerulus-like structures in different areas of the ventrolateral protocerebrum. The posterior lateral protocerebrum and the optic tubercle were each contributed by a single centripetal pathway. Another pathway connects the lobula on each side of the brain. Two centrifugal pathways convey signals from the posterior lateral protocerebrum to the lobula.

Sign-conserving amacrine neurons in the fly's external plexiform layer

Amacrine cells in the external plexiform layer of the fly's lamina have been intracellulary recorded and dye-filled for the first time. The recordings demonstrate that like the lamina's short photoreceptors R1-R6, type 1 lamina amacrine neurons exhibit nonspiking, "sign-conserving" sustained depolarizations in response to illumination. This contrasts with the sign-inverting responses that typify first-order retinotopic relay neurons: monopolar cells L1-L5 and the T1 efferent neuron. The contrast frequency tuning of amacrine neurons is similar to that of photoreceptors and large lamina monopolar cells. Initial observations indicate that lamina amacrine receptive fields are also photoreceptor-like, suggesting either that their inputs originate from a small number of neighboring visual sampling units (VSUs), or that locally generated potentials decay rapidly with displacement. Lamina amacrines also respond to motion, and in one recording these responses were selective for the orientation of moving edges. This functional organization corresponds to the anatomy of amacrine cells, in which postsynaptic inputs from several neighboring photoreceptor endings are linked by a network of very thin distal processes. In this way, each VSU can receive convergent inputs from a surround of amacrine processes. This arrangement is well suited for relaying responses to local intensity fluctuations from neighboring VSUs to a central VSU where amacrines are known to be presynaptic to the dendrites of the T1 efferent. The T1 terminal converges at a deeper level with that of the L2 monopolar cell relaying from the same optic cartridge. Thus, the localized spatial responses and receptor-like temporal response properties of amacrines are consistent with possible roles in lateral inhibition, motion processing, or orientation processing.

The computational basis of an identified neuronal circuit for elementary motion detection in dipterous insects

Based on comparative anatomical studies and electrophysiological experiments, we have identified a conserved subset of neurons in the lamina, medulla, and lobula of dipterous insects that are involved in retinotopic visual motion direction selectivity. Working from the photoreceptors inward, this neuronal subset includes lamina amacrine (α) cells, lamina monopolar (L2) cells, the basket T-cell (T1 or β), the transmedullary cell Tm1, and the T5 bushy T-cell. Two GABA-immunoreactive neurons, the transmedullary cell Tm9 and a local interneuron at the level of T5 dendrites, are also implicated in the motion computation. We suggest that these neurons comprise the small-field elementary motion detector circuits the outputs of which are integrated by wide-field lobula plate tangential cells. We show that a computational model based on the available data about these neurons is consistent with existing models of biological elementary motion detection, and present a comparable version of the Hassenstein-Reichardt (HR) correlation model. Further, by using the model to synthesize a generic tangential cell, we show that it can account for the responses of lobula plate tangential cells to a wide range of transient stimuli, including responses which cannot be predicted using the HR model. This computational model of elementary motion detection is the first which derives specifically from the functional organization of a subset of retinotopic neurons supplying the lobula plate. A key prediction of this model is that elementary motion detector circuits respond quite differently to small-field transient stimulation than do spatially integrated motion processing neurons as observed in the lobula plate. In addition, this model suggests that the retinotopic motion information provided to wide-field motion-sensitive cells in the lobula is derived from a less refined stage of processing than motion inputs to the lobula plate.

Stereotyped neuropil branching of an identified stomatogastric motor neuron

Anatomical studies of the crab stomatogastric ganglion (STG) have suggested only minimal organization within the neuropil of this structure. Here, we present evidence that, for at least one intrinsic neuron type, the ventricular dilator (VD) neuron, a highly organized and stereotyped branching structure exists within the stomatogastric neuropil. Specifically, we show the morphology of the VD neuron consists of a single primary neurite that projects from the soma into the neuropil and bifurcates into a pair of subprimary neurites, which in turn exit the neuropilar region, one entering the left and the other the right medial ventricular nerve. Nearly all secondary neurite branching of the VD neuron is from the subprimary neurites. There are approximately 22 secondary branches/neuron (range 14-28), with no significant difference between the number of secondary branches off the right vs. the left subprimary neurite, although the ratio of secondary branches between subprimaries varies (range 0.4-1.6). The fine neurites that branch from the secondary processes segregate hemispherically within the neuropil, based on the subprimary neurite of origin. Within this hemispherical organization, another level of fine neurite segregation is present, namely, the fine neurites derived from each secondary branch are restricted to discrete regions of the hemisphere with only minimal overlap with those derived from other secondary branches. Monte Carlo simulations show that this segregation differs significantly from a random distribution. The organization of branching seen in the VD neuron may play a critical role in the electrotonic and local computational organization of this neuron and sets the stage for physiological experimentation addressing these issues.

Anatomical organization of retinotopic motion-sensitive pathways in the optic lobes of flies

Anatomical methods have identified conserved neuronal morphologies and synaptic relationships among small-field retinotopic neurons in insect optic lobes. These conserved cell shapes occur across many species of dipteran insects and are also shared by Lepidoptera and Hymenoptera. The suggestion that such conserved neurons should participate in motion computing circuits finds support from intracellular recordings as well as older studies that used radioactive deoxyglucose labeling to reveal strata with motion-specific activity in an achromatic neuropil called the lobula plate. While intracellular recordings provide detailed information about the motion-sensitive or motion-selective responses of identified neurons, a full understanding of how arrangements of identified neurons compute and integrate information about visual motion will come from a multidisciplinary approach that includes morphological circuit analysis, the use of genetic mutants that exhibit specific deficits in motion processing, and biomimetic models. The latter must be based on the organization and connections of real neurons, yet provide output properties similar to those of more traditional theoretical models based on behavioral observations that date from the 1950s. Microsc. Res. Tech. 62:132-150, 2003.

Organization and neural connections of the anterior optic tubercle in the brain of the locust, Schistocerca gregaria

The anterior optic tubercle is a small neuropil in the insect brain and a major target of visual interneurons from the optic lobe. The functional role of the tubercle is poorly understood, but recent evidence from locusts points to a possible involvement in polarization vision. The present study examines the organization of the anterior optic tubercle in the locust Schistocerca gregaria and its connections with other brain areas. The tubercle of the locust consists of an upper and a lower subunit. Both units are connected in parallel with the medulla and lobula of the optic lobe, with the contralateral tubercle, and with the lateral accessory lobe in the median protocerebrum. Wide-field transmedullary neurons provide input from the medulla. Neurons with processes in the dorsal rim of the medulla, a relay station in the polarization vision pathway, project exclusively to the lower unit of the tubercle. Visual input from the lobula to the upper and lower unit originates from topographically distinct strata. The most prominent output target of the tubercle is the lateral accessory lobe in the median protocerebrum. Neurons from the upper unit project widely in the lateral accessory lobe, whereas neurons from the lower unit have focused projections confined to the median olive and to the lateral triangle. The two subunits of the anterior optic tubercle are, therefore, processing stages in two parallel visual pathways from the optic lobe to the median protocerebrum. Pathways via the lower unit of the tubercle appear to be involved in polarization vision.

Retinotopic pathways providing motion-selective information to the lobula from peripheral elementary motion-detecting circuits

Recordings from afferent channels from the medulla supplying deep neuropils of the fly's optic lobes reveal different filter properties among the three classes of afferent neurons: transmedullary cells, T2 neurons, and Y cells. Whereas transmedullary cells respond to local flicker stimuli without discriminating these from directional or oriented motion, the T2 afferent neurons show clear motion orientation selectivity, which corresponds closely with a morphological bias in the orientation of their dendrites and could also be influenced by systems of local recurrent neurons in the medulla. A Y cell having a clearly defined terminal in the lobula, but having dendrite-like processes in the medulla and, possibly, the lobula plate, discriminates the direction of motion and its orientation. These results demonstrate unambiguously that the lobula receives information about motion and that the channels carrying it are distinct from those supplying wide-field motion-selective neurons in the lobula plate. Furthermore, recordings from a newly identified recurrent neuron linking the lobula back to the inner medulla demonstrate that the lobula discriminates nondirectional edge motion from flicker, thereby reflecting a property of this neuropil that is comparable with that of primary visual cortex in cats. The present findings support the proposal that elementary motion detecting circuits supply several parallel channels through the medulla, which segregate to, but are not shared by, the lobula and the lobula plate. The results are discussed in the context of other intracellular recordings from retinotopic neurons and with analogous findings from mammalian visual systems.

Chemical neuroanatomy of the fly's movement detection pathway

In Diptera, subsets of small retinotopic neurons provide a discrete channel from achromatic photoreceptors to large motion-sensitive neurons in the lobula complex. This pathway is distinguished by specific affinities of its neurons to antisera raised against glutamate, aspartate, gamma-aminobutyric acid (GABA), choline acetyltransferase (ChAT), and a N-methyl-D-aspartate type 1 receptor protein (NMDAR1). Large type 2 monopolar cells (L2) and type 1 amacrine cells, which in the external plexiform layer are postsynaptic to the achromatic photoreceptors R1-R6, express glutamate immunoreactivity as do directionally selective motion-sensitive tangential neurons of the lobula plate. L2 monopolar cells ending in the medulla are accompanied by terminals of a second efferent neuron T1, the dendrites of which match NMDAR1-immunoreactive profiles in the lamina. L2 and T1 endings visit ChAT and GABA-immunoreactive relays (transmedullary neurons) that terminate from the medulla in a special layer of the lobula containing the dendrites of directionally selective retinotopic T5 cells. T5 cells supply directionally selective wide-field neurons in the lobula plate. The present results suggest a circuit in which initial motion detection relies on interactions among amacrines and T1, and the subsequent convergence of T1 and L2 at transmedullary cell dendrites. Convergence of ChAT-immunoreactive and GABA-immunoreactive transmedullary neurons at T5 dendrites in the lobula, and the presence there of local GABA-immunoreactive interneurons, are suggested to provide excitatory and inhibitory elements for the computation of motion direction. A comparable immunocytological organization of aspartate- and glutamate-immunoreactive neurons in honeybees and cockroaches further suggests that neural arrangements providing directional motion vision in flies may have early evolutionary origins.

Fundamental mechanisms of visual motion detection: models, cells and functions

Taking a comparative approach, data from a range of visual species are discussed in the context of ideas about mechanisms of motion detection. The cellular basis of motion detection in the vertebrate retina, sub-cortical structures and visual cortex is reviewed alongside that of the insect optic lobes. Special care is taken to relate concepts from theoretical models to the neural circuitry in biological systems. Motion detection involves spatiotemporal pre-filters, temporal delay filters and non-linear interactions. A number of different types of non-linear mechanism such as facilitation, inhibition and division have been proposed to underlie direction selectivity. The resulting direction-selective mechanisms can be combined to produce speed-tuned motion detectors. Motion detection is a dynamic process with adaptation as a fundamental property. The behavior of adaptive mechanisms in motion detection is discussed, focusing on the informational basis of motion adaptation, its phenomenology in human vision, and its cellular basis. The question of whether motion adaptation serves a function or is simply the result of neural fatigue is critically addressed.

Neurobiological constraints and fly systematics: how different types of neural characters can contribute to a higher level dipteran phylogeny

Much uncertainty still exists regarding higher level phylogenetic relationships in the insect order Diptera, and the need for independent analyses is apparent. In this paper, I present a parsimony analysis that is based on details of the nervous system of flies. Because neural characters have received little attention in modern phylogenetic analyses and the stability of neural traits has been debated, special emphasis is given to testing the robustness of the analysis itself and to evaluating how neurobiological constraints (such as levels of neural processing) influence the phylogenetic information content. The phylogenetic study is based on 14 species in three nematoceran and nine brachyceran families. All characters used in the analysis are based on anatomical details of the neural organization of the fly visual system. For the most part they relate to uniquely identifiable neurons, which are cells or cell types that can be confidently recognized as homologues among different species and thus compared. Parsimony analysis results in a phylogenetic hypothesis that favors specific previously suggested phylogenetic relationships and suggests alternatives regarding other placements. For example, several heterodactylan families (Bombyliidae, Asilidae, and Dolichopodidae) are supported in their placement as suggested by Sinclair et al. (1993), but Tipulidae and Syrphidae are placed differently. Tipulidae are placed at a derived rather than ancestral position within the Nematocera, and Syrphidae are placed within the Schizophora. The analysis suggests that neural characters generally maintain phylogenetic information well. However, by "forcing" neural characters onto conventional phylogenetic analyses it becomes apparent that not all neural centers maintain such information equally well. For example, neurons of the second-order visual neuropil, the medulla, contain stronger phylogenetic "signal" than do characters of the deeper visual center, the lobula plate. These differences may relate to different functional constraints in the two neuropils.

Visual input regulates circuit configuration in courtship conditioning of Drosophila melanogaster

Courtship and courtship conditioning are behaviors that are regulated by multiple sensory inputs, including chemosensation and vision. Globally inhibiting CaMKII activity in Drosophila disrupts courtship plasticity while leaving visual and chemosensory perception intact. Light has been shown to modulate CaMKII-dependent memory formation in this paradigm and the circuitry for the nonvisual version of this behavior has been investigated. In this paradigm, volatile and tactile pheromones provide the primary driving force for courtship, and memory formation is dependent upon intact mushroom bodies and parts of the central complex. In the present study, we use the GAL4/UAS binary expression system to define areas of the brain that require CaMKII for modulation of courtship conditioning in the presence of visual, as well as chemosensory, information. Visual input suppressed the ability of mushroom body- and central complex-specific CaMKII inhibition to disrupt memory formation, indicating that the cellular circuitry underlying this behavior can be remodeled by changing the driving sensory modality. These findings suggest that the potential for plasticity in courtship behavior is distributed among multiple biochemically and anatomically distinct cellular circuits.

In vivo calcium accumulation in presynaptic and postsynaptic dendrites of visual interneurons

In this comparative in vivo study of dendritic calcium accumulation, we describe the time course and spatial integration properties of two classes of visual interneurons in the lobula plate of the blowfly. Calcium accumulation was measured during visual motion stimulation, ensuring synaptic activation of the neurons within their natural spatial and temporal operating range. The compared cell classes, centrifugal horizontal (CH) and horizontal system (HS) cells, are known to receive retinotopic input of similar direction selectivity, but to differ in morphology, biophysics, presence of dendrodendritic synapses, and computational task. 1) The time course of motion-induced calcium accumulation was highly invariant with respect to stimulus parameters such as pattern contrast and size. In HS cells, the rise of [Ca(2+)](i) can be described by a single exponential with a time constant of 5-6 s. The initial rise of [Ca(2+)](i) in CH cells was much faster (tau approximately 1 s). The decay time constant in both cell classes was estimated to be at least 3.5 times longer than the corresponding rise time constant. 2) The voltage-[Ca(2+)](i) relationship was best described by an expansive nonlinearity in HS cells and an approximately linear relationship in CH cells. 3) Both cell classes displayed a size-dependent saturation nonlinearity of the calcium accumulation. Although in CH cells calcium saturation was indistinguishable from saturation of the membrane potential, saturation of the two response parameters differed in HS cells. 4) There was spatial overlap of the calcium signal in response to nonoverlapping visual stimuli. Both the area and the amplitude of the overlap profile was larger in CH cells than in HS cells. Thus calcium accumulation in CH cells is spatially blurred to a greater extent than in HS cells. 5) The described differences between the two cell classes may reflect the following computational tasks of these neurons: CH cells relay retinotopic information within the lobula plate via dendritic synapses with pronounced spatial low-pass filtering. HS cells are output neurons of the lobula plate, in which the slow, local calcium accumulation may be suitable for local modulatory functions.

Encoding of motion in real time by the fly visual system

Direction-selective cells in the fly visual system that have large receptive fields play a decisive role in encoding the time-dependent optic flow the animal encounters during locomotion. Recent experiments on the computations performed by these cells have highlighted the significance of dendritic integration and have addressed the role of spikes versus graded membrane potential changes in encoding optic flow information. It is becoming increasingly clear that the way optic flow is encoded in real time is constrained both by the computational needs of the animal in visually guided behaviour as well as by the specific properties of the underlying neuronal hardware.