The production effect in memory: Multiple species of distinctiveness

The production effect is the difference in memory favoring words read aloud relative to words read silently during study. According to a currently popular explanation, the distinctiveness of aloud words relative to silent words at the time of encoding underlies the better memory for the former. This distinctiveness is attributable to the additional dimension(s) of encoding for the aloud items that can be subsequently used during retrieval. In this study we argue that encoding distinctiveness is not the sole source of distinctiveness and that, in fact, there is an independent source of distinctiveness, statistical distinctiveness, which may or may not work in harmony with encoding distinctiveness in influencing memory. Statistical distinctiveness refers to the relative size of a subset of items marked by a(ny) unique property. Silently read words can carry statistical distinctiveness if they form a salient minority on the background of a majority of vocalized words. We show that, when the two sources are placed in opposition, statistical distinctiveness modifies the PE in a profound way.