A recent surge of interest into the empirical measurement of mind-wandering has led to an increase in the use of thought-probing to measure attentional states, which has led to a large degree of variability in methodologies across studies (Weinstein, in press). Three sources of variation in methodology include the frequency of thought probes during a task, the number of response options provided for each probe, and the way in which various attentional states are framed during task instructions. Method variation can potentially affect behavioral performance on the tasks into which thought probes are embedded, the experience of various attentional states within those tasks, and/or response biases to thought probes. Therefore, such variation can be problematic both pragmatically and theoretically. Across three experiments we examined how manipulating probe frequency, response options, and framing affected behavioral performance and responses to thought probes. Probe frequency and framing did not affect behavioral performance or probe responses. But based on the present results we argue that thought probes need at least three response corresponding to on-task, off-task, and task-related interference. When specifically investigating mind-wandering, probe responses should also distinguish between mind-wandering, external distraction, and mind-blanking.