Reexamining the “Distinctiveness Effect”: Poorer Recognition of Distinctive Face Silhouettes

A recognition advantage for distinctive faces has been widely reported (e.g., Valentine, 1991). In such studies, distinctive faces produce more hits and fewer false alarms than typical faces. Although the finding is robust, the mechanism for this advantage has not been carefully explored. The choice of distractors in these studies does not guarantee equivalent target-distractor distances for typical and distinctive faces. In fact, because typical faces lie in a denser, more central region of face space (Valentine, 1991), they will be on the whole more similar to the distractor set than distinctive faces. The location of distractors may thus be sufficient to explain the distinctiveness advantage. In fact, theories of perceptual learning would predict a processing disadvantage for distinctive faces that we have less experience with. To control for the effect of unevenly spaced distractors, we constructed a parameterized face space and created equally spaced targets and distractors.