On the separability of two mechanisms involved in the detection of grating patterns in humans.

Abstract

1. The detectability of contrast modulation (M) of sinusoidal gratings was explored at the rate of 8 Hz. The luminance profile of a contrast modulated sinusoidal grating is L = L(1 + C cos 2 pi F chi). This stimulus may also be regarded as the sum of a steady grating pattern and counter phase flicker of the same spatial frequency. 2. Contrast modulation sensitivity (1/M) was established in five observers at several levels of constract and over a range of spatial frequencies, where M = delta C/C of delta C is the just detectable contrast change and C is the mean contrast of the grating. The slope of a modulation sensitivity function (C/delta C vs. C) is 1 (i.e. delta C = constant) near threshold contrast at each spatial frequency, but in the suprathreshold contrast range the slope flattens from close to 1 at 1.5 c/deg to almost 0 (delta C/C = constant) at 12 c/deg. 3. Adaptation to a high contrast steady grating of the same spatial frequency as the contrast modulated test gratings shifts each modulation sensitivity function to the right at low contrasts, but not at high. As a result the adapted curves cross their corresponding unadapted ones. At each spatial frequency the modulation sensitivity function is now fitted by a straight line of slope 1. While delta C needs to be higher than half the detection threshold of the same grating at spatial frequencies above 3 c/deg, in the adapted condition the values are nearly equal at each frequency. Thus pattern adaptation unmasks the threshold of the counterphase component of the contrast modulated grating near threshold contrast as well as above it. The phase of the steady adapting grating, relative to the steady component of the test grating, does not make any difference. 4. Apparently contrast modulation reveals differences beyond threshold sensitivity between spatial frequencies adjacent to the peak of the contrast sensitivity curve. For each spatial frequency channel there must be different neural coupling between steady and modulated inputs. Electrophysiological studies using contrast modulated gratings would be useful in the exploration of individual and ensemble properties of neurones of the visual cortex.