Purpose. We have used a visual masking paradigm (Pelli, 1990) to estimate limits on contrast detection which arise early in the visual system. In this paradigm, we measure contrast detection thresholds for gratings in 2-dimensional dynamic white noise. The masking effectiveness of the noise indicates visual efficiency, or the magnitude of (internal) visual noise limiting threshold. Our previous visual evoked potential (VEP) measures of visual efficiency appeared to fit Pelli's additive model less well than psychophysical data do (Skoczenski et al., ARVO 1995). We made extensive VEP measures of visual efficiency to investigate this discrepancy, and have found conditions under which the VEP correlates well with psychophysical data. Methods. We used a swept-contrast VEP technique and psychophysics to measure contrast detection thresholds in five adult human observers with normal vision. To estimate visual efficiency, we used 3 c/deg sinewave gratings in the presence of 2-dimensional dynamic broadband noise. The masking noise contrast that elevates contrast threshold by √2 is the estimate of efficiency. We measured VEP and psychophysical thresholds for 5.5 Hz on/off and contrast-reversal gratings. Results. VEP contrast thresholds measured with reversal stimuli were more variable, and provided a systematically poorer fit to the masking paradigm model, compared to VEP thresholds for on/off gratings. On/off VEP data matched psychophysical data quite well. Conclusion. VEPs recorded for on/off stimuli provide more reliable estimates of early limits on contrast detection than do grating-reversal VEPs. This is likely because VEPs to on/off stimuli are more robust than those to reversal stimuli in the context of the contrast dynamics of a swept-contrast stimulus.
|Original language||English (US)|
|Journal||Investigative Ophthalmology and Visual Science|
|State||Published - Feb 15 1996|
ASJC Scopus subject areas
- Sensory Systems
- Cellular and Molecular Neuroscience