Reversing the Picture Superiority Effect

Abstract

Imaginal coding typically enhances retention. Pictures are remembered better than words; words for which subjects imagine referents are better remembered than words studied without such coding; concrete words are better retained than abstract words; and mnemonic devices employing imagery can produce dramatic effects on retention. These facts have long been noted (Paivio, 1971) and many contributions in this volume confirm the efficacy of imagery and imagination in remembering. If we conceive the process of remembering as involving the broad stages of encoding, storage, and retrieval (Kohler, 1947; Melton, 1963), then it is fair to say that theories explaining the superior retention of information encoded imagi-nally have emphasized encoding and storage processes as the loci of the effects. For example, Paivio's (1969, 1986) dual code theory attributes the picture superiority effect to redundant coding, and Nelson's (1979) sensory-semantic model to pictures' superior sensory codes. Interestingly, the possible importance of retrieval factors has been relatively neglected. In this chapter we present evidence that the neglect of the retrieval phase in such effects is unwarranted. In fact, we show that retrieval factors are critical in producing the "standard" imagery effects. We concentrate primarily on the picture superiority effect, the finding that pictures are better remembered than words, but at the end of the chapter we argue that our thesis also holds for other standard effects of imagery and mnemonic devices. First, we briefly review evidence documenting the picture superiority effect. Next, in something of a detour to lay the groundwork of our argument, we review recent evidence showing important differences between certain classes of retention tasks. From this literature we hypothesize that one may in fact find a reversal of the usual picture superiority effect on certain tests, such that words are retained better than pictures. We present several experiments that illustrate superior retention for words than for pictures and then resurrect some older evidence bearing on this point. We conclude with a section summarizing our argument and extending it to other imagery manipulations and mnemonic techniques. M. A. McDaniel et al. (eds.), Imagery and Related Mnemonic Processes