The temporal integration ofluminous energy was compared in the fovea and at 7°eccentricity using a white-appearing stimulus provided by a glow modulator tube transilluminating a 21.8' opal glass target in a free-viewing situation. In Experiment 1, a three-interval temporal forcedchoice adaptive staircase procedure was used to estimate thresholds for brief (3-msec) and long (102-msec) two-pulse stimuli. The data offive observers showed that in the fovea the long stimulus required significantly more luminous energy at threshold than the briefstimulus [t(14) = 5.60, p < .011; however, at 7°, threshold luminous energy was not significantly different for the brief and long stimulus. Thus, Bloch's law held in the periphery but not in the fovea, and the critical duration was longer (> 102 msec) in the periphery than in the fovea « 102 msec), In Experiment 2, the detectability of a brief (3 msec) and a long (102 msec) two-pulse stimulus, both equal in total luminous energy, was compared at the same two retinal locations using a yes-no signal-detection procedure. The data of three observers showed that in the fovea the detectability of the long stimulus was 30% less than that of the brief stimulus, whereas at 7°brief and long stimulus detectability was approximately the same. Thus, both experiments demonstrated a foveal-peripheral difference in two-pulse visual temporal integration, with the fovea evidencing less integration.In visual psychophysics, Bloch's law of temporal integration describes the phenomenon whereby stimuli equal in luminous energy produce the same visual effect regardless of how that energy is distributed in time (Graham, 1965). The longest duration stimulus for which Bloch's law holds is known as the critical duration (Hartline, 1934). In the case of rectangular light pulses, stimulus intensity (I) and duration (T) are interchangeable and reciprocally related such that I x T = K for a constant response such as 50% frequency of seeing. IT reciprocity suggests that luminous energy is completely integrated by the visual system'.Several studies have compared temporal integration in the fovea and peripheral retina. Although most of these investigations indicate longer critical durations or more integration for peripheral than for foveal stimulation (Sperling & Jolliffe, 1965;Dannheim & Drance, 1971;Ronchi & Novakova, 1971), one study (Van den Brink & Bouman, 1954) has shown that temporal integration is independent of retinal locus. The latter study differed from the former in two major respects. First, Van den Brink and Bouman used two-pulse stimuli, whereas the others all used single pulses. Second, Van den Brink and
135Bouman used an equal-energy design, whereas the others used an equal-performance design. In the equal-energy paradigm, the total luminous energy of the stimulus was kept constant and response frequency was measured as a function of the time between two equal luminance pulses. The response frequency for two-pulse stimuli equal in luminous energy but of different durations was the same w...