Abstract
Purpose: :
Here we describe a study which set out to show parallel processing of chromatic and achromatic information in the retina using the electroretinogram (ERG). We used a novel chromatic-achromatic compound stimulus which was the temporal analogue of the spatial compound described by Lee et al. (J Physiol 589, 59-73, 2011).
Methods: :
ERGs were recorded from 4 colour normal and one deuteranopic male subjects using DTL fibre electrodes. The red and green LEDs in a Diagnosys ColorDome ganzfeld were used to produce a temporal compound stimulus. In each temporal cycle, first red and then green luminance were modulated using raised cosines.Thus there was red-green information at one frequency (which we defined as the fundamental), and luminance information at twice that frequency (2nd harmonic). To determine isoluminance, temporal frequency was set to 18Hz. It was assumed that, at this temporal frequency, there is no response to red-green chromatic stimulation, so that any first harmonic component is the result of luminance imbalance. ERGs were recorded to a range of red:green luminance ratios. The ratio at which luminance intrusion was minimised was then used to measure responses to a series of temporal frequencies. In control studies we also used simple sinusoidal modulation of red, green and yellow.
Results: :
At 18Hz, all subjects ERGs reached a clear minimum in the fundamental (red-green) response as red-green ratio was changed showing that isoluminance could be determined. The dichromat and one colour normal subject with a known high L:M ratio required substantially more green than red at isoluminance, and the other three subjects reached isoluminance when red and green luminances were equal. The fundamental (red-green) response at isoluminance as a function of temporal frequency was different for the normals and the dichromat: all four normals produced low pass functions which descended to noise at 20Hz, whilst the dichromat had no fundamental response. The second harmonic (luminance) response was band-pass for all subjects, peaking at around 30Hz. A similar shape was seen with simple luminance modulation. All subjects showed similar 4th harmonic responses.
Conclusions: :
Here we demonstrate that it is possible to simultaneously record information about parallel chromatic and achromatic pathways in the human retina. We can be confident that the fundamental (red-green) response is colour-specific because it is not seen in the deuteranope. This is also evidence that the fundamental response is not coming from rods or S-cones.
Keywords: electroretinography: non-clinical • color vision • temporal vision