Rod-cone interactions and the temporal impulse response of the cone pathway

Abstract

Dark-adapted rods suppress cone-mediated flicker detection. This study evaluates the effect that rod activity has on cone temporal processing by investigating whether rod mediated suppression changes the cone pathway impulse response function, regardless of the form of the temporal signal. Stimuli were generated with a 2-channel photostimulator that has four primaries for the central field and four primaries for the surround. Cone pathway temporal impulse response functions were derived from temporal contrast sensitivity data with periodic stimuli, and from two-pulse discrimination data in which pairs of briefly pulsed stimuli were presented successively at a series of stimulus onset asynchronies. Dark-adapted rods altered the amplitude and timing of cone pathway temporal impulse response functions, irrespective of whether they were derived from measurements with temporally periodic stimuli or in a brief presentation temporal resolution task with pulsed stimuli. Rod-cone interactions are a fundamental operation in visual temporal processing under mesopic light levels, acting to decrease the temporal bandwidth of the visual system.

Figure 1

Temporal profiles of the stimuli used for the temporal contrast sensitivity (Panel A) and 2-pulse measurements (Panels B and C). Panel A gives an example of a 9 Hz periodic sinusoidal cone signal modulated within a 1 sec raised cosine envelope (constant time-averaged luminance) that alternated with a 1 sec steady field. Panel B shows a pair of 4 ms incremental pulses separated by a 70 ms stimulus onset asynchrony (SOA). The SOA was defined as the time (ms) between pulse onsets. Panel C shows an increment-decrement pulse combination separated by a 70 ms SOA.

Figure 2

The effect of the state of rod adaptation on cone pathway sensitivity to periodic stimuli. Left and right panels show the data for two experienced psychophysical observers. Open circles show contrast sensitivity as a function of temporal frequency (Hz) for luminance modulated stimuli (L+M+S) during the cone plateau and the closed squares show contrast sensitivity after dark adaptation. The average standard error of measurement (SEM) was 1.65 for AJZ and 1.86 for DC. Panel inserts show the difference in log contrast sensitivity between the cone plateau (CP) and dark-adapted viewing conditions (CP/DA) for each observer. The solid lines show the Fourier transformation of the estimated IRF with the periodic stimulus (Fig 4, upper panels).

Figure 3

The effect of the state of rod adaptation on cone pathway sensitivity to pulsed stimuli. Left and right panels show the data for two experienced psychophysical observers. Circular symbols show the two-pulse discrimination data as a function of the stimulus onset asynchrony (SOA) of the two 4 ms pulses measured during the cone plateau, the square symbols after dark adaptation. Measurement conditions included two incremental pulses (open symbols) and one incremental and one decremental pulse (closed symbols). The average SEM was 0.67 for AJZ and 0.26 for DC. The solid lines show the predicted two pulse discrimination sensitivity based on the IRF with the pulsed stimuli (Fig 4, lower panels).

Figure 4

The effect of dark-adapted rods on the cone pathway temporal impulse response functions. Upper panels: Impulse response functions derived from the temporal contrast sensitivity data in Figure 2. Lower panels: Impulse response functions derived from the 2-pulse data in Figure 3. The grey lines show the IRFs during the cone plateau, the black lines show the IRFs after dark adaptation. The state of rod adaptation alters the amplitude and time time-to-peak of the cone pathway impulse response.