Signal detection theory cannot distinguish perceptual and response-based biases: Evidence from the Muller-Lyer illusion and application for action-specific effects

摘要

??According to the action-specific account of perception, perception is influenced by the perceivera??s ability to act. For example, softball players who are hitting better than others see the ball as bigger. These claims have been challenged on the grounds that the apparent effects in perception may be due to influences on the post-perceptual processes that generate the responses. Signal detection theory (SDT) seems poised to resolve this issue because it produces separate measures that are often linked to perceptual and decision-based processes. However, we will illustrate that SDT cannot distinguish perceptual biases from decision-based biases, and thus cannot be used to determine if action-specific effects are perceptual. Our illustration involves both an empirical demonstration and a graphical presentation of how a more advanced understanding of SDT would lead to the same conclusion. For the empirical demonstration, we used the Muller-Lyer illusion because it is analogous to the purported effects reported by the action-specific approach in that it leads to a perceptual bias. We found that the Muller-Lyer illusion did not influence da?? and rather influenced the criterion measure (c). In many contexts, this pattern of results would lead to the conclusion that the illusion is not perceptual. Instead, we claim that the illusion is perceptual but that SDT cannot distinguish between perceptual and response-based biases. We then dissect the measures of SDT and illustrate how the measure of c would reveal an influence of both a criterion shift (as in the case of a decision-based process) and a perceptual shift (as in the case of a perceptual bias). Therefore, even though action-specific effects influence c and not da?? (as shown in Experiment 2), this pattern of results cannot be interpreted as evidence that action-specific effects are not perceptual.