Purpose. To examine how threshold varies with the polarity (increment or decrement) of the probe in the probed-sinewave paradigm.
Methods. On each trial, a small probe (~1.5deg) was presented on a large (~15deg) background. The background varied sinusoidally over time at a frequency between 0.9-18.8 Hz around a mean luminance of 52 cd/m2. The probe was either an increment or decrement and was presented at one of eight different phases with respect to the flickering background (see left panel). Probe threshold was determined using a QUEST routine. At the end of each trial, the subject pressed a key to indicate that she had or had not seen the probe; in a subset of the experiments, the subject also pressed a key to record her perception of the polarity of the probe.
Results. (1) Regardless of the frequency of the flickering background, increment and decrement probe-threshold-versus-phase functions were very similar (see upper panels); however, there was a small, systematic difference between increment and decrement probe thresholds (see lower panels). (2) The DC-level of the probe-threshold-versus-phase function increased as frequency increased while the shape of the function remained fairly constant across the frequencies tested. (3) Increment probes were almost always perceived as increments, but decrement probes, particularly at low frequencies, were sometimes perceived as increments. (4) WilsonUs model of light adaptation [Wilson (1997) Vis Neurosci, 15, 403-423; Hood & Graham (1998) Vis Neurosci, 15, 957-967] accounted for some aspects of the data.
Conclusions. In the probed-sinewave paradigm, probe threshold varies dramatically as a function of phase and frequency while the polarity of the probe matters relatively little.