The main results presented by the authors are that flicker-fusion frequency increased in some conditions. However, these claims are not well-justified because the results presented are vulnerable to observer bias. That is, the apparent improvements could have resulted from an increase in sensitivity like the authors suggest, but they also could have resulted from a change in observer bias or criterion for responding to respond "I perceive no flicker". Indeed, the "method of limits" used to measure the flicker-fusion frequency appears to be entirely subjective, as it is described as being determined by 1) the "frequency at which the participant reported that the flicker recommenced" (an entirely subjective measure) and 2) the frequency at which the participant could no longer detect flicker (this might have been measured objectively measure, but from the context it seems it was subjective). Granted it is difficult to imagine that the pattern of CFFT changes across the several conditions could be entirely due to criterion shifts, esp. since some of the conditions supposedly appeared identical from the point of view of the participants. But with the CFFT measured subjectively, we cannot know which changes are due to sensitivity improvements and which due to criterion shifts (changes in bias to report flicker).