RT and perceived brightness against log luminance, covering the range between 3.4 and - 1.6 log cd/m2 in steps of 1 log unit, were measured at the fovea, 20 degrees, and 40 degrees in the periphery. The RT and brightness data were fitted to the RT and brightness power function, respectively. It was found that (a) asymptotic RT increased as eccentricity increased and that (b) the RT exponent decreased as retinal eccentricity increased: Estimated mean RT exponent was found to be approximately - .31 in the fovea, but - .25 and - .17 in the 20 degrees, and 40 degrees periphery, respectively. Further, (c) the brightness exponent tended to increase as eccentricity increased. The RT exponent in the fovea was in agreement with the exponent for perceived brightness (.36); however, the RT exponent in the periphery was about 1/2 to 2/3 as small as that in the fovea. This implies that as luminance changes equal RT contour changes as a function of eccentricity. That is, RT in the periphery does not primarily depend on the target's brightness but its luminance.
The present study investigated the extent to which fatigue influenced the usage of feedback by Ss to make corrections during coincident timing responses. Sixteen college males watched a .01-sec timer and attempted to knock over a barrier at the moment that the sweep hand reached 500 msec. All Ss were given 50 trials of practice on each of four consecutive days. The fatigue group's performance was immediately preceded by a response-specific local fatigue (dynamic exercise). To insure the maintenance of a high level of fatigue throughout practice, the fatiguing exercise was interspersed with the coincident timing task. The control group performed the same exercise as the fatigue group but without any resistance. Schmidt's (1972) index of preprogramming (IP) was calculated for each day in order to determine the level of feedback involvement in controlling the movement. Fatigue did not influence the accuracy with which Ss responded or the rate that they learned. However, the results indicated a relatively high and stable IP throughout all trials with the IP for the fatigue group higher than that of the control, suggesting a preference by these Ss for a programming mode of control.

