Abstract
Contrast polarity effects have been ascribed to the negative polarity system being more sensitive and having higher spatial resolution. Higher cell density for negative polarity ganglion cells has been found in Macaque peripheral parasol cells. We repeated the letter identification task we presented here last year, adding a condition in which the letters were presented in one of four peripheral locations instead of in the central fovea. Single letters of positive or negative polarity were either presented normally or as background gray letters surrounded by a small positive or negative pedestal. The pedestal polarity determined the channel polarity, opposite the apparent polarity of the pedestal-defined letter. Response latencies were collected from 16 observers for 3 repetitions of 12 letters presented randomly at 6 letter or pedestal contrasts (+/−0.4, +/−0.2, +/−0.1) and at two eccentricities (0 and 3 degrees). For the foveal presentations, only the absolute amount of contrast significantly affected the latencies. Peripheral presentations led to slower latencies overall, and for the highest contrast peripheral conditions neither contrast polarity nor pedestal vs. letter mattered. At the lowest contrast, there again was no polarity effect, but the pedestal condition was much faster than the letter condition, the increased stimulus area of the pedestal apparently guiding fixations better. At the middle contrast level, the positive contrast letters were slowest to be identified. This could be the result of poorer fixation guidance or smaller letter span by the positive polarity system.