Stimuli presented with a low frequency elicit larger pupillary responses than those presented with a high frequency. Similar effects are known for motor responses observed in reaction time experiments. Utilizing this stimulus probability effect, we conducted a Go/NoGo reaction time experiment and measured pupillary dilation to evaluate categorization. We found no probability effect when participants used distinctive features to classify stimuli with different frequencies into the same category. If no distinctive features could be applied, however, the effect was observed in both pupillary and reaction time data. The stimulus probability effect can serve as a tool for the evaluation of categorization and is discussed in the context of orienting.