DYNAMIC CONTRAST PERCEPTION ASSESSED BY PATTERN MASKING

Citation
Rw. Bowen et H. Derudder, DYNAMIC CONTRAST PERCEPTION ASSESSED BY PATTERN MASKING, Journal of the Optical Society of America. A, Optics, image science,and vision., 15(3), 1998, pp. 570-578
Citations number
24
Categorie Soggetti
Optics
ISSN journal
10847529
Volume
15
Issue
3
Year of publication
1998
Pages
570 - 578
Database
ISI
SICI code
1084-7529(1998)15:3<570:DCPABP>2.0.ZU;2-B
Abstract
The perceived contrast of a pulsed grating varies markedly with the ex posure duration and spatial frequency of the grating. We studied dynam ic changes in perceived grating contrast with a pattern-masking paradi gm. We measured masking of a brief, localized test pattern (a D6 stimu lus, 30 ms in duration) by fixed-contrast cosine grating patterns of v arying duration (50-500 ms). The cosine mask pattern had spatial frequ ency of either 1 or 6 cycles per degree (cpd) at a contrast of 0.3. Th e D6 test pattern was centered on a light bar of the mask and was eith er positive peak contrast (same-polarity test and mask) or negative pe ak contrast (opposite-polarity test and mask). In Experiment I, the te st and mask had simultaneous onset. With a 6-cpd mask, the same-polari ty test-threshold elevation versus mask-duration function increases mo notonically. For a 1-cpd mask, the same-polarity threshold-mask-durati on function is nonmonotonic, with peak masking effect produced by a gr ating pulse of 80-100 ms. These masking effects are closely congruent with known dynamic contrast effects. With negative tests, masking-dura tion functions are elevated from same-polarity functions and are essen tially similar in shape for 1- and 6-cpd masks. The elevated threshold s suggest inhibitory interaction between ON and OFF pathways, with a s imilar time course across spatial frequency. In Experiment 2, the D6 t est was delayed from mask onset by 33 ms. Positive contrasts only were employed. For 1-cpd stimuli, the delay of test greatly reduced maskin g at all mask durations and eliminated the nonmonotonic function. This suggests that for low-spatial-frequency patterns, perceived contrast is determined by an early peak component of the neural response. But f or 6-cpd stimuli, masking of the delayed test was somewhat greater at all mask durations, consistent with a gradually increasing underlying neural response to the grating. Finally, in Experiment 3, same-polarit y masking effects at both spatial frequencies were replicated with neg ative-contrast test and mask (OFF pathway mediation). This indicates t hat the ON and OFF pathways have similar response dynamics. (C) 1998 O ptical Society of America.