Mechanisms of rod-cone interaction: evidence from congenital stationary nightblindness

Vision under mesopic and scotopic illumination

Evidence has accumulated that rod activation under mesopic and scotopic light levels alters visual perception and performance. Here we review the most recent developments in the measurement of rod and cone contributions to mesopic color perception and temporal processing, with a focus on data measured using a four-primary photostimulator method that independently controls rod and cone excitations. We discuss the findings in the context of rod inputs to the three primary retinogeniculate pathways to understand rod contributions to mesopic vision. Additionally, we present evidence that hue perception is possible under scotopic, pure rod-mediated conditions that involves cortical mechanisms.

Isolated mesopic rod and cone electroretinograms realized with a four-primary method

The purpose of this study was to evaluate the feasibility of measuring rod and cone electroretinograms (ERGs) at a single mesopic adaptation level. To accomplish this, a four-primary photostimulator was implemented using a commercially available ERG system (Diagnosys ColorDome) to generate three types of stimuli that temporally modulated rods alone, cones alone, and rods and cones simultaneously. For each stimulus type, ERGs were recorded as a function of temporal frequency (2, 4, 8, or 16 Hz) and mesopic light levels (0.02, 0.16, or 1.26 cd/m(2)) in normal observers and patients with retinitis pigmentosa (RP) or cone-rod degeneration. The normal observers ERG waveforms showed a clear periodic pattern, mirroring the sinusoidal stimuli. At all light levels, rod responses were always higher than cone responses for temporal frequencies between 2 and 8 Hz, suggesting that rods dominated the responses. Cone responses were minimal at the lowest light level and increased with increases in light level. The amplitude of the response to the combined stimuli was intermediate between that of the isolated cone and the isolated rod stimuli for all light levels. Good receptoral isolation was confirmed by the results showing (1) minimal or no rod ERGs but recordable cone ERGs in the patients and (2) high correlation between the ERG amplitudes obtained from the four-primary method and those from the ISCEV standard clinical protocol in normal observers.

Effects of light and dark adaptation of rods on specific-hue threshold

Specific-hue threshold as a function of absolute rod threshold was measured with long-, middle-, and short-wavelength monochromatic test lights presented 17 deg extrafoveally. The measurements were obtained both during the rod phase of long-term dark adaptation and under conditions where the rod receptor system was gradually light adapted from a dark-adapted state by a scotopic background field of increasing retinal illumination. The results show that change in specific-hue threshold with change in absolute rod threshold is not, in general, identical for light and dark adaptation of the rod receptor system. Thus, in the long- and middle-wavelength test regions, the specific-hue threshold could be obtained at higher intensities under the light- as compared to the dark-adaptation condition when absolute rod thresholds were the same. Just the opposite was found for the short-wavelength tests. It is concluded that change in specific-hue threshold with light and dark adaptation of the rod receptor system is not, in general, controlled by the same mechanism.

Mesopic state: cellular mechanisms involved in pre- and post-synaptic mixing of rod and cone signals

At least twice daily our retinas move between a light adapted, cone-dominated (photopic) state and a dark-adapted, color-blind and highly light-sensitive rod-dominated (scotopic) state. In between is a rather ill-defined transitional state called the mesopic state in which retinal circuits express both rod and cone signals. The mesopic state is characterized by its dynamic and fluid nature: the rod and cone signals flowing through retinal networks are continually changing. Consequently, in the mesopic state the retinal output to the brain contained in the firing patterns of the ganglion cells consists of information derived from both rod and cone signals. Morphology, physiology, and psychophysics all contributed to an understanding that the two systems are not independent but interact extensively via both pooling and mutual inhibition. This review lays down a rationale for such rod-cone interactions in the vertebrate retinas. It suggests that the important functional role of rod-cone interactions is that they shorten the duration of the mesopic state. As a result, the retina is maintained in either in the (rod-dominated) high sensitivity photon counting mode or in the second mode, which emphasizes temporal transients and spatial resolution (the cone-dominated photopic state). Experimental evidence for pre- and postsynaptic mixing of rod and cone signals in the retina of the clawed frog, Xenopus, is shown together with the preeminent neuromodulatory role of both light and dopamine in controlling interactions between rod and cone signals. Dopamine is shown to be both necessary and sufficient to mediate light adaptation in the amphibian retina.

Suppressive rod-cone interactions: evidence for separate retinal (temporal) and extraretinal (spatial) mechanisms in achromatic vision

We investigated the influence of selective rod light and dark adaptation on cone-mediated sensitivity to monocular displays modulated sinusoidally in both spatial and temporal domains. Rod light adaptation (1) increased sensitivity to high spatial frequencies [> or = 8 cycles per degree (cpd)] flickered slowly (< or = 2 Hz), an effect that we refer to as grating suppressive rod-cone interaction (gSRCI); (2) increased sensitivity to low spatial frequencies (< or = 2 cpd) flickered rapidly (> or = 8 Hz), an effect that we refer to as flicker suppressive rod-cone interaction (fSRCI); and (3) had relatively little influence on intermediate temporal-spatial-frequency combinations. The magnitudes of both gSRCI and fSRCI increased as the retinal position of the test display was increasingly displaced parafoveally. In parafoveal retina, both forms of suppressive rod-cone interaction increased as the overall dimension of the test stimulus decreased. However, sensitivity to high spatial frequencies is equally well influenced by adaptation of the viewing and the contralateral eye, while the adapted state of the nonviewing eye negligibly influences sensitivity to rapid flicker. Moreover, gSRCI cannot be observed with a small (30-arcmin) grating restricted to the fovea, while fSRCI is a prominent effect with small foveal test stimuli. Collectively, these results and neurobiological evidence suggest that fSRCI reflects a mechanism restricted to distal retinal, while gSRCI involves extraretinal neural circuitry.

Functional consequences of oncogene-induced horizontal cell degeneration in the retinas of transgenic mice

Visual function was evaluated in transgenic mice expressing the simian virus 40 early region under the control of the promoter for phenylethanolamine-N-methyltransferase. These transgenic mice undergo a degeneration of the retinal horizontal cells and the outer plexiform layer. Electroretinograms (ERGs) were recorded under stimulus conditions chosen to elicit both receptoral and postreceptoral responses. The dark-adapted a-waves obtained from transgenic mice were not different from control recordings, indicating that the degenerative process does not interfere with function of the rod photoreceptors. In comparison, the ERG b-wave was markedly reduced in transgenic mice under both dark- and light-adapted conditions. Reproducible visual evoked potentials (VEPs) were recorded from transgenic mice in response to both low luminance stimuli that isolate rod function, and to higher luminance stimuli, indicating that retinal activity is transmitted centrally to the visual cortex. However, VEPs were delayed at all stimulus luminances compared to controls. Analysis of luminance-response functions suggests that the VEP delays could reflect the combination of a decrease in synaptic efficacy and an overall loss in visual sensitivity. These functional abnormalities correlate well with the anatomical abnormalities that have been previously observed in the transgenic retina (Hammang et al., 1993), namely a reduced number of synapses between photoreceptors and second-order neurons.

Peripheral colour vision: effects of rod intrusion at different eccentricities

Chromaticities of monochromatic lights from different parts of the spectrum were measured both during the cone-plateau period of the long-term dark-adaptation curve and in a completely dark-adapted state. The measurements were obtained at 3, 8, 30 and 65 deg in the temporal field of view and at 1, 2, 3 and 4 log units above the cone-plateau level. The results show that cone-mediated colours obtained during the cone-plateau period in general are desaturated when rod signals intrude during long-term dark adaptation. The desaturation effect of rods obtained at high mesopic illumination levels was found to increase when the test field was moved from 3 to 30 deg but to reduce markedly between 30 and 65 deg. Surprisingly, the desaturation was clearly observable even at a retinal illumination of 20,000 ph td. The desaturation effect of rods is explained by the suggestion that differences in ongoing activity rates of the different types of spectrally opponent cells become levelled out to some extent when light signals from rods intrude during dark adaptation.

Suppression of flicker response with increasing test illuminance: roles of temporal waveform, modulation depth, and frequency

This study examined the detectability of flicker for small foveal long-wavelength test stimuli centered within surrounding long-wavelength annular adaptation stimuli. Flicker threshold-versus-illuminance (tvi) curves were analyzed for four different test-stimulus waveforms--sine-wave, square-wave, and rapid-on sawtooth and rapid-off sawtooth flicker--at temporal frequencies ranging from 12 to 21 Hz and at temporal modulation depths ranging from approximately 50% to 100%. For all stimulus combinations that were examined involving temporal frequencies above 12 Hz, the resultant flicker tvi curves shared the following characteristic features: First, at operationally dim surround illuminances, there was always a single elevated threshold for detection of flicker. Second, some surround illuminance always could be found for which flicker threshold decreased abruptly, typically by approximately 1.5 log units within 0.1 log unit of surround illuminance increase. Third, when test illuminance was incremented above this lower flicker threshold, flicker always vanished; when test illuminance was incremented still further, flicker reappeared. Finally, at sufficiently bright surround illuminances flicker did not disappear with increasing test illuminance. Although these effects held for all waveforms, the abrupt decrease of flicker threshold occurred at brighter surround illuminances for sawtooth than for sine-wave flicker, and at brighter surround illuminances for sine-wave than for square-wave flicker, at least for fully modulated waveforms (of a given temporal frequency). Moreover, when modulation depth was adjusted so that any two different waveforms had the same first-harmonic contrast, the resultant flicker tvi curves became identical when plotted as first-harmonic amplitude versus surround illuminance. This identity held for any given temporal frequency, even though the flicker tvi curves for 12-Hz fully modulated sine-wave or square-wave flicker did not manifest flicker response suppression, whereas the flicker tvi curves for sawtooth flicker did. These and other results imply that the first-harmonic contrast of the test stimulus fully determines the shape of the entire flicker tvi curve and that the dc component of the test stimulus helps to cause flicker response suppression. The results also demonstrate that first-harmonic equivalence is only a necessary, not a sufficient, condition for linearity.

Nonmonotonic effects of test illuminance on flicker detection: a study of foveal light adaptation with annular surrounds

This study examined the detectability of flicker for small long-wavelength foveal test stimuli centered within larger long-wavelength surround stimuli. Flicker visibility was evaluated as a function of surround and test illuminance and as a function of test wavelength, of the time elapsed following test or surround onset, and of surround dimensions. Consistent with prior flicker threshold-versus-illuminance results [Vision Res. 26, 917 (1986)], flicker threshold decreased abruptly once the surround illuminance became sufficiently great. However, as test illuminance was increased above flicker threshold, flicker again vanished. Flicker reappeared at still higher test illuminances, as middle-wavelength-sensitive (M-) cone-mediated flicker threshold was exceeded. Meanwhile, the time required for the surround to render flicker visible increased at a rapidly accelerating rate with decreasing surround illuminance; it increased at a more sporadic rate with increasing test illuminance. At bright enough surround illuminances, flicker did not vanish with increasing test illuminance. These and other results are compatible with a framework derived from previous dark-adaptation data [Vision Res. 32, 1975 (1992)]. In that framework the test stimulus itself induces losses of flicker sensitivity by sufficiently perturbing retinal response during states or stages of adaptation that fail to cause spectrally antagonistic processes to redress that perturbation adequately. The relevant adaptation processes, which can require minutes, involve an adaptation pool that includes (and is affected by) the test stimulus.

Rod-cone interaction in form detection

Using a Wright colorimeter, absolute threshold, absolute form threshold and specific form threshold were measured during long-term dark adaptation in the extrafoveal retina. The specific form threshold was found to fall markedly at about the cone-rod break but thereafter rose steeply. Furthermore, during the rod phase of the dark adaptation the form percept of the small, slender rectangular test field changed qualitatively from a line or rectangle to a circular field at all mesopic intensities. The results indicate that light signals from rods may both facilitate and suppress cone-mediated information about form, and that the rod system may completely dominate the perception of form several log units above the absolute dark-adapted cone threshold when the eye is dark adapted.

Profound reductions of flicker sensitivity in the elderly: can glaucoma involve the retina distal to ganglion cells?

Flicker sensitivities were measured for more than 100 people age 60 and older with stimulus-conditions originally designed to obtain estimates of preretinal absorption by the lens and macular pigment. Flicker sensitivities were measured on two chromatic backgrounds: a 1000-td, 480-nm background and a 5800-td, Wratten 33 background (approximately metameric with 633 nm). Testing sessions were administered at 18-month intervals across a 3-yr period. No subject tested had a history of glaucoma or ocular hypertension at the time of entry into the study. For ten subjects, however, flicker sensitivity was sometimes reduced by more than 2.0 log units from the mean norm for at least one of the two backgrounds. For most other subjects, flicker sensitivities were within 0.5 log units of the mean norms. On retrospective analysis, the profound reductions of flicker sensitivity (PRFS) were associated significantly with (a) advanced age (perhaps especially when combined with relatively high intraocular pressure), and (b) the use of cardiovascular medications. The PRFS probably were associated with (c) female sex, and (d) large intraocular pressure fluctuations. In addition, the majority of subjects with PRFS were found to have evidence of glaucomatous cupping or field loss. These results suggest that PRFS result from glaucoma or share etiologies with low-tension glaucoma. The use of cardiovascular medications suggested that PRFS could depend on retinal dysfunction rather than on optic nerve compromise alone. Predicted results from two additional subject populations support this possibility. For young healthy subjects, flicker threshold vs illuminance curves attained very steep slopes for sufficiently short wavelength tests on sufficiently extreme long wavelength backgrounds (655 nm, 50,000 td); the steep slopes coincided with the breakdown of effective M-cone isolation. Reductions of flicker sensitivity on the 5800-td Wratten 33 background depended correspondingly on test wavelength for subjects with well-documented low-tension glaucoma.

Dark adaptation of foveal cones during the cone-plateau period

Following substantial bleaching by "white" light, absolute threshold, relative spectral sensitivity and sensation of hue of monochromatic lights were measured at the central fovea during the cone-plateau period. The absolute-threshold level was found to increase and then decrease markedly, the relative spectral sensitivity remained invariant, while the sensation of hues of monochromatic lights from the long- and middle-wave regions of the spectrum changed toward hues of shorter wavelengths.

Rod suppression of cone-mediated information about colour and form during dark adaptation

Following substantial bleaches, the specific form and hue thresholds were measured during dark adaptation with a test stimulus of 1 x 2 degrees at 40 degrees extrafoveally. The wavelength of the test field was varied between runs. The results show that both thresholds started to rise at about the cone-rod break of the dark-adaptation curve, irrespective of wavelength used in the test. Furthermore, the specific threshold for form was found to rise when a scotopic stimulus was superimposed on a photopic test flash. On the other hand, both thresholds remained at the cone-plateau level when the test flash was confined within the rod-free fovea. In order to explain the rise in the specific thresholds, it is suggested that signals from rods generated directly in response to the test stimulus may suppress both cone-mediated form and colour. It is also suggested that this type of rod-cone interaction represents a general characteristic involved in several kinds of visual information processing.

Spatial properties of rod-cone interactions in flicker and hue detection

Rod-cone interactions in flicker and hue detection were compared to examine the hypothesis that they are mediated by mechanisms with different spatial properties. Flicker and hue thresholds for a 1 deg test stimulus (TS) were measured as a function of background luminance and diameter. Flicker thresholds were reduced from their dark-adapted value by an 11 deg diameter background, but not by a 1 deg background. These results, in agreement with previous work, demonstrate that light adaptation of rods surrounding the TS is necessary to eliminate their effect on cone flicker thresholds. In contrast, hue thresholds were reduced from their dark-adapted value to a comparable degree by 1 and 11 deg backgrounds, indicating that light adaptation of the rods stimulated by the TS is sufficient to abolish the rod-cone hue interaction. Our results support the contention that the rod-cone flicker and hue interactions are mediated by different mechanisms. We also demonstrated that changing the detection task while keeping stimulus parameters constant is sufficient to shift between these two types of rod-cone interactions.

Losses of foveal flicker sensitivity during dark adaptation following extended bleaches

Flicker sensitivity to a small foveal test stimulus can decrease appreciably during the period of dark adaptation that follows extinction of a bleach. For 20 min diameter, long wavelength tests that followed 70% L cone bleaches of various wavelengths, 18 Hz flicker sensitivity decreased precipitously (i.e. within about 45 sec) by about 1.5-2.0 log units beginning at about 100-200 sec following extinction of the bleach. For short wavelength tests that followed long wavelength bleaches, the corresponding precipitous losses of flicker sensitivity were relatively small and early; flicker sensitivity decreased by about 0.6-1.0 log units beginning no later than 60 sec following extinction of the bleach. Whenever flicker sensitivity decreased precipitously, the losses of flicker sensitivity were followed by plateaus of M cone dominated flicker sensitivity. For combinations of test and bleaching wavelengths that did not cause flicker sensitivity to decrease precipitously, flicker sensitivity instead decreased gradually over a prolonged period of time, and incorporated a substantial L cone contribution. The precipitous suppression of flicker sensitivity found for certain combinations of bleaching and test wavelengths appears to depend on the action of a broadly tuned, red-green spectrally opponent process.