[Skip to Content]


Rod and cone mechanisms of mesopic vision

Our visual system has an incredible capacity to detect small changes in light intensity (contrast sensitivity) over a remarkably broad range of light conditions. This is made possible by rods and cones which operate in dim and bright light conditions respectively, and by a process known as adaptation that controls the sensitivity and speed of the visual responses in rods, cones, and downstream neurons. In dim lights, when rods drive the visual responses (scotopic vision), contrast sensitivity is poor and limited to slow variations in light levels (see figure). In bright lights when cones mediate the visual responses (photopic vision), contrast sensitivity increases and extends to fast variations in light levels. At mesopic light levels both rods and cones contribute to vision. Rod photoreceptors are highly specialized for discrete detection of single photons when dark adapted and have slow responses under these conditions. Cones are less sensitive to light and possess intrinsically faster responses, especially in daylight. Thus, for decades it has been accepted that rods subserve vision in dim scenes, detecting only slow contrast changes, and cones subserve vision in mesopic and bright scenes when contrast changes rapidly.

Figure 1

Figure 1. Visual sensitivity depends on mean light levels, speed of the light variations and photoreceptor type. Rods drive vision to slow variations in scotopic (dim) lights (purple sensitivity function) while cones drive responses in photopic (bright) lights (green sensitivity function). x-axis represents frequency of temporal variations in light.

However, Americans spend up to 86% of their time indoors, exposed to artificial lighting, including that of our many video screens. At indoor luminance levels (100-200 cd/m2) rods are producing ~1000 to 10000 R*/s. While earlier studies suggested that mammalian rods are saturated and unresponsive at these light levels, we and others have shown that, in fact, rods respond to light variations. Hence, the mesopic range extends well beyond twilight conditions, into the range of indoor lighting where we spend most of our time. The goal of our lab is to understand how rod and cone- driven vision operates at the intermediate (mesopic) lights, the major mode of vision for people in indoor environments (Fig 1). We test the cellular mechanisms that underlie vision in mesopic lights using an integrative approach that includes retinal electrophysiological techniques, transgenic mouse models, animal visual behavior, and mathematical modeling.

Neural mechanisms underlying mesopic visual behavior in mouse:

To effectively probe the complexities of retinal mechanisms more broadly underlying mesopic visual behavior, we developed an operant conditioning method to measure temporal contrast sensitivity in mice. The optomotor reflex in mice is used as a proxy for the retinal temporal processing. However, subcortical reflexive pathways drive the optomotor response rather than cortical and decision-making areas. In addition, the use of the optomotor response assay is sub-optimal for the study of temporal contrast sensitivity (TCS) because of a) the potential for interactions in the processing of speed vs temporal frequency properties of the drifting grating stimulus, and b) the dynamic response range is substantially less than the retinal responses. To solve this problem, we have developed an operant conditioning assay that allows determination of TCS functions in behaving mice. Indeed, the temporal resolution of the optomotor response is limited to ~12 Hz, while visual responses measured with our new operant approach extend to ~50 Hz.

Temporal contrast sensitivity in the behaving mouse shares fundamental properties with human psychophysics. Umino et al., 2018, eNeuro.

(A) Our operant behavior assay measures TCS in behaving mice using a forced-choice visual task. We applied the theory of signal detection to estimate the discriminability factor (d´), a measure of performance that is independent of response bias and motivation. (B) With this approach, we established in the mouse a model of human vision that shares fundamental properties of human temporal psychophysics such as Weber adaptation in response to low temporal frequency flicker and illumination dependent increases in critical flicker frequency as predicted by the Ferry–Porter law. (C) Human TCS functions adapted from Kelly (1961).

Figure 2

Pupillary light response to steady lights of freely behaving mice. Bushnell et al., 2016, J. Neurosci. Methods .

Because gentle restraint or light anesthesia tends to bias pupillary responses, the ability to determine retinal illumination levels in freely behaving mice is critical to our goal of relating retinal neuronal activity to behavior. We implemented a system to measure the pupillary light response to steady lights of freely behaving mice using a custom-built, portable device that automatically acquires close-up images of their eyes.

In our system, a freely behaving mouse inside a custom-built chamber explores an opening in the side wall of the chamber. Outside the chamber and suspended from the lid of the chamber is a video camera that captures infrared images of the lateral view of the mouse every time its nose crosses the aperture. LEDs provide IR illumination.  Average pupil areas are measured offline and plotted as a function of luminance.

Rod– and cone–driven visual responses interchange roles in mesopic lights. Pasquale et al., 2020, JNeurosci.

We applied our pupilometer and operant behavior assay to measure TCS in WT mice, GNAT1 KO mice (dysfunctional rods) and GNA2 KO (or GNAT2cpfl3) mice (with dysfunctional cones) to investigate the contributions of rods and cones to mesopic vision. (A), In dim lights, rods in WT and G2 mice, relay relatively slow temporal variations. (B), However, in daylight conditions, rod pathways exhibit high sensitivity to fast — but not to slow — temporal variations, whereas cone-driven responses supplement the loss in rod-driven sensitivity to slow temporal variations. (C), Our findings highlight the dynamic interplay of rod- and cone-driven vision as light levels rise from night to daytime levels. The functional map illustrates rod and cone-driven responses as a function of illumination and frequency. For example, cones drive the responses to low frequencies (<6 Hz) at 10000 ph/s/um2 while rods responses to high frequencies (> 6 Hz) (indicated by the grey arrow).

Furthermore, the fast, rod-driven signals do not require the rod-to-cone Cx36 gap junctions as proposed in the past, but rather, can be relayed by alternative Cx36-independent rod pathways (not shown here, but see Pasquale et al 2020 for details).

Figure 6

In summary, using transgenic mice to selectively assess the functional contributions of rods and cones to visual behavior, we found that at indoor (mesopic) light levels, rods drive the visual responses to fastnot slow — temporal variations. Remarkably, cones exhibit poor sensitivity to fast light variations at mesopic intensities. As light levels rise from night to mesopic levels, rod vision shifts from low frequency detection to high frequency and cones take over as the low frequency detectors.

Figure 7

Current projects in the lab build on these interesting findings and focus on the following questions:

1) What neural mechanisms underlie the fast mesopic rod responses?

2) How do rod-driven visual responses to high frequencies in mesopic lights determine behavioral sensitivity in natural environments?

 Keep posted for future results!

Another major line of research in the lab investigates the retinal mechanisms that determine temporal resolution and limit contrast sensitivity. Umino et al., 2019, J Neurosci.

We determined that R9AP95 mice (with fast photoresponse recovery kinetics caused by overexpression of the transducin GAP complex) exhibit increased behavioral TCS to (A) low (6 Hz) but not (B) high (21 Hz) temporal frequencies at retinal irradiance levels ranging from 100 to 4000 ph/s/mm2. These light levels correspond to the mesopic range in mouse vision. This is also the range where mouse TCS exhibits rod-driven Weber adaptation in response to 6 Hz flickering lights (TCS remains constant as irradiance increases). TCS to 21 Hz flicker did not adapt to background light levels and TCS to 21 Hz flicker is independent of the level of R9AP expression. Our results established that rod photoresponse kinetics limit temporal contrast sensitivity to low temporal frequencies in mesopic vision.

Figure 4

The photoresponse recovery time-constant controls the magnitude (and phase) of the pharmacologically isolated ERG flicker responses. Umino et al., 2019, J Neurosci.

 (A), The pharmacologically isolated flicker ERG responses of WT and transgenic mice with fast photoresponse recovery kinetics (R9AP95 line) grew differentially at irradiance levels > 80 ph/s/mm2. (B), Flicker responses of R9AP95 mice had markedly higher amplitudes and faster responses (as inferred from the relative phase advance of the waveforms) than WT mice. Recordings with mice with the GNAT2cpfl3 background confirm that the responses originate from rods. (C), A simple quantitative model satisfactorily explains the increase in the magnitude of the flicker responses in R9AP95 mice in terms of their faster photoresponse kinetics.

Figure 5

Mesopic temporal contrast sensitivity increases despite photoreceptor degeneration in a mouse model of retinitis pigmentosa with fast rod recovery kinetics. Pasquale et al., 2021, eNeuro.

Uncharacteristically fast rod recovery kinetics are facets of both human patients and transgenic animal models with a prevalent cause of retinitis pigmentosa: a P23H rhodopsin mutation (RhoP23H/+). We found that these mice exhibit a 1.2 to 2-fold increase in retinal (A, B) and optomotor (C) TCS despite significant photoreceptor degeneration. A simple linear-non-linear model suggests that the increase in sensitivity can be explained by the change in rod kinetics. Measurement of TCS could be used as a non-invasive early diagnostic tool indicative of rod dysfunction in some forms of retinal degenerative disease.

Figure 8

A current goal of the lab is to identify the phototransduction and membrane mechanisms that increase the magnitude of rod contrast responses when the kinetics speed up, in health and disease conditions.