Recognition of motion-defined shapes in patients with multiple sclerosis and optic neuritis.

We have developed a simple procedure for assessing the ability of the visual pathway to extract a two-dimensional shape from motion. The test requires a patient to read motion-defined (MD) letters. These letters differ physically from the familiar contrast-defined (CD) letters that are dimmer or brighter than their surroundings in that the boundaries of MD letters are rendered visible exclusively by a step in velocity while the boundaries of CD letters are rendered visible by a step in luminance. Subjects viewed a random pattern of bright dots containing a perfectly camouflaged letter. Then the letter was revealed by moving dots within and outside the letter at equal speeds in opposite directions. Letter reading scores for 50 eyes of 25 patients with multiple sclerosis (MS) or optic neuritis were compared with norms based on 50 control subjects. When tested with large (50 arc min, i.e., 6/60) MD letters, 34/50 eyes of patients required abnormally high dot speeds to read letters, visual loss being sufficiently selective in 10 eyes that contrast sensitivity, Snellen acuity, 11%-contrast and 4%-contrast acuity were all spared. Four eyes were effectively motion blind in the sense that they could not read large letters even at our highest relative speed of 0.9 deg/s and the failure could not be attributed to reduced Snellen acuity. Our normal limit was 2.5 SD from the control mean and there were 1/50 false positives. Of the 34/50 eyes with elevated speed thresholds, 23 had normal Snellen acuities. The number of eyes abnormal for intermediate (11%) contrast CD letters, was 19/50 of which 8 had normal Snellen acuity, confirming our previous finding that MS can degrade the ability to see low-contrast objects while sparing Snellen acuity. We conclude that MD test letters can detect lesions that are not picked up by testing with CD test letters of high or low contrast. We suggest that the MD letter test can detect dysfunction in the human equivalent of a pathway in monkey brain that originates in large retinal ganglion cells, passes through the magnocellular layers of the lateral geniculate body, includes cortical area MT, and is involved in processing motion.