Computations in the vertebrate retina: gain enhancement, differentiation and motion discrimination

Koch, Christof; Poggio, Tomaso; Torre, Vincent

doi:10.1016/0166-2236(86)90060-3

Cited by 45 publications

(20 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Does the circuitry of G14a fit any of the current models for directional selectivity? Several models have been postulated for the genesis of directional selectivity in ganglion cells (Barlow and Levick, 1965;Torre and Poggio, 1978;Masland and Mills, 1979;Koch et al, 1982Koch et al, , 1986Werblin et al, 1988;Vaney, 1990;Famiglietti, 1991). Our evidence on circuitry of the G15 and G14a satisfies some of the predictions of these various models.…”

supporting

confidence: 80%

Ultrastructural and immunocytochemical analysis of the circuitry of two putative directionally selective ganglion cells in turtle retina

Guiloff

Kolb

1994

J of Comparative Neurology

View full text Add to dashboard Cite

Two well-stained, horseradish peroxidase-filled varieties of putative ON-OFF directionally selective ganglion cells, G14a and G15, that project to the dorsolateral optic tectum (Guiloff and Kolb [1992a] Vis. Neurosci. 8:295-313) were studied qualitatively and quantitatively. Both were bistratified ganglion cells with one tier of dendrites in the OFF sublamina and the other in the ON sublamina of the inner plexiform layer (IPL). The cells were serially sectioned and examined for synaptic inputs by electron microscopy. Portions of the dendritic trees were also analyzed after postembedding immunocytochemistry for neurotransmitter candidates gamma aminobutyric acid (GABA), glycine, choline acetyltransferase (ChAT), and glutamate in presynaptic neurons. Both G14a and G15 are dominated by amacrine cell inputs and have only minor bipolar cell involvement. Probably at least two different types of bipolar cell are presynaptic. Both ganglion cells receive some GABA-positive (GABA+) amacrine inputs and G14a receives ChAT+ amacrine inputs. Glycine+ and glutamate+ inputs could not be detected in either cell. The GABA+ inputs appeared to be regionally arranged in the dendritic trees. The general distribution of amacrine and bipolar inputs to the two tiers of dendrites in both cell types appeared to be asymmetrical, both along the radial extent of the dendritic trees and within the depth of the IPL. Our data support some aspects of the current models for directional selectivity. We suggest candidate bipolar and amacrine cells that could have input to these ganglion cells. Since many of the putative presynaptic amacrine cells coincide with directionally selective types recorded and stained by other authors, we propose that in turtle retina directional selectivity arises in neurons presynaptic to the ganglion cells.

show abstract

supporting

confidence: 80%

Ultrastructural and immunocytochemical analysis of the circuitry of two putative directionally selective ganglion cells in turtle retina

Guiloff

Kolb

1994

J of Comparative Neurology

View full text Add to dashboard Cite

show abstract

“…The missing preferred-direction enhancement at high contrasts may be simply from saturation nonlinearities at some stage in the motion pathway (16). Hence, the finding of only a null-direction suppression cannot be readily interpreted in terms of an inhibitory one-stage process of motion detection as has been done in several studies (4,21,41,43). Interestingly, although null-direction suppression may often be more pronounced, enhancement of the responses to apparent motion in the preferred direction is frequently also seen (4,22,41,(44)(45)(46).…”

Section: Discussionmentioning

confidence: 70%

Direction selectivity of blowfly motion-sensitive neurons is computed in a two-stage process.

Borst

Egelhaaf

1990

Proc. Natl. Acad. Sci. U.S.A.

View full text Add to dashboard Cite

Direction selectivity of motion-sensitive neurons is generally thought to result from the nonlinear interaction between the signals derived from adjacent image points. Modeling of motion-sensitive networks, however, reveals that such elements may still respond to motion in a rather poor directionally selective way. Direction selectivity can be significantly enhanced if the nonlinear interaction is followed by another processing stage in which the signals of elements with opposite preferred directions are subtracted from each other. Our electrophysiological experiments in the fly visual system suggest that here direction selectivity is acquired in such a two-stage process.Neurons with directionally selective responses to motion are found at different levels of the nervous system in animals as phylogenetically distant as insects and primates. It is an essential requirement for any mechanism underlying direction selectivity that it has at least two input channels subserving neighboring points in visual space, which, after being processed in an asymmetric way, interact nonlinearly (1-3) (Fig. 1 A). Various formal operations have been proposed for this nonlinear interaction, such as a logical gate (4), a multiplication (5-7), and a summation followed by a squaring (8) or a threshold operation (9, 10). However, such a singlestage mechanism may also respond to nonmotion stimuli as, for instance, changes in the mean light intensity (11,12). Because these direction-independent response components are identical in two single-stage mechanisms with the same receptive field but opposite polarity, they can be eliminated by subtracting the output of two such units (Fig. 1B). If such a two-stage process is perfectly mirror-symmetrical, it responds with the same amplitude but an opposite sign to motion in opposite directions (5,6,12). However, even if the symmetry is not exact, direction selectivity is enhanced by a subtraction stage as compared with the corresponding singlestage model (12).The first model of motion detection worked out in formal terms (5), the so-called correlation-type movement detector, makes use of this simple computational principle. This mechanism was initially proposed on the basis of experimental studies on insects (3, 5, 6), where it can account for motion detection surprisingly well under both steady-state and transient stimulus conditions (2, 6, 13-16). Although much effort has been made to characterize the nonlinear interaction, which could be shown to be well approximated by a multiplication (12, 17), there is so far not much direct evidence for the second processing step of direction selectivity, the subtraction stage.In contrast to insects, the responses of directionally selective cells in vertebrates are usually interpreted without taking the possibility of this second processing stage into account (see, however, ref. 18). Instead, most studies are based on a particular single-stage mechanism that goes back to the seminal study of Barlow and Levick (4) on directionally selective retinal g...

show abstract

“…In any event, our finding of DS amacrine cells in this retina is a strong indication that at least some processing of directional information takes place presynaptically to ganglion cells in the turtle (Koch et al, 1986;DeVoe et al, 1989;Borg-Graham and Grzcywacz, 1992) rather than as having arisen at the ganglion cell membrane from nondirectional inputs (Torre and Poggio, 1978;Werblin, 19911.…”

Section: Directionally Selective Amacrine and Ganglion Cellsmentioning

confidence: 87%

The organization of the turtle inner retina. II. Analysis of color‐coded and directionally selective cells

Ammermüller

Kolb

1995

J of Comparative Neurology

View full text Add to dashboard Cite

Color coding and directional selectivity (DS) of retinal neurons were studied in the Pseudemys turtle by using similar intracellular recording and staining techniques as in the preceding paper (J. Ammermüller and H. Kolb, 1995, J. Comp. Neuronal. 358:1-34). Color-coded responses were elicited by red (621 or 694 nm), green (525 or 514 nm), and blue (455 nm) light flashes. In addition to red/green and yellow/blue types of chromaticity horizontal cells, in our sample of 305 identified cells we found that 17% of bipolar cells, 6.5% of amacrine cells, and 18% of ganglion cells exhibit color-coded responses. DS responses were found in 37% of the tested ganglion cells and 41% of the tested amacrine cells. Two morphologically identified bipolar cell types, B10 and B11, were red-ON/blue-OFF and red-OFF/green, blue-ON, respectively. Of five identified amacrine cell types, three were red-OFF/blue-ON center (A1, A3, A23b), one was red-OFF/green-ON center (A32), and one (A33) was double color-opponent of red-ON/blue-OFF center:red-OFF/blue-ON surround. Five ganglion cell types had variously color-coded centers (G14 and G24) or surrounds (G3 and G18), including one type, G6, that was double color-opponent (red-OFF/green-ON center:red-ON/green-OFF surround). Responses to colors were found primarily in sustained responses of bipolar and ganglion cells. However, in amacrine cells, transient components of the response also showed color dependence. Red-OFF-center responses were found in ganglion cells that were in a position to make connections at the strata 2/3 border with the red-OFF bipolar cell (B11); red-ON-center responses occurred in ganglion cells with branches in stratum 4 of the IPL where the red-ON-center bipolar (B10) ended. Blue-ON-center signals appeared to be processed mainly in strata 1-2/3, and blue-OFF-center signals in strata 3-5 of the IPL, with contributions of amacrine cells and bipolar cells. Labeled DS amacrine cells could be identified as A9, A20, and A22, and ganglion cells as G19, G20, and G24. The latter type (G24) showed DS and color coding. All response types (ON-center, OFF-center, ON-OFF) were encountered. DS amacrine cells were monostratified near the middle of the IPL, whereas DS ganglion cells were mono-, bi-, and multistratified, although all DS ganglion cells had one feature in common: they had dendrites in stratum 1 of the IPL.(ABSTRACT TRUNCATED AT 400 WORDS)

show abstract

Computations in the vertebrate retina: gain enhancement, differentiation and motion discrimination

Cited by 45 publications

References 46 publications

Ultrastructural and immunocytochemical analysis of the circuitry of two putative directionally selective ganglion cells in turtle retina

Ultrastructural and immunocytochemical analysis of the circuitry of two putative directionally selective ganglion cells in turtle retina

Direction selectivity of blowfly motion-sensitive neurons is computed in a two-stage process.

The organization of the turtle inner retina. II. Analysis of color‐coded and directionally selective cells

Contact Info

Product

Resources

About