No abstract

No abstract

When Kosslyn, Ball, and Reiser asked subjects to scan a memorized picture, they found a strong positive linear relationship between distance scanned and reaction time. However, more recent research has suggested that this result may be as much a function of the demand characteristics of the experiment as a reflection of any structural properties of the image. To further test this possibility with complex stimuli, college subjects were either presented with Kosslyn's pictorial stimuli or verbal descriptions of same in a "nonexperiment" and were asked to predict their scanning times. The subjects were able to produce high linear correlations between scanning distance and predicted scanning time. This result is consistent with a demand characteristics explanation of the correlation between distance and reaction time that is obtained in actual image-scanning studies.

Two experiments were conducted to assess the extent of potential experimental demand characteristics inherent in the image-scanning paradigm. The results of the first "mental travel" experiment that pitted verbal versus imagery coding showed that (a) the positive correlation between physical distance and reaction time was replicated, and (b) when given a choice, subjects' reaction times varied as a function of verbal codes rather than imagery. To isolate the effects due to demand constraints from those produced by mode dominance, a nonexperiment in which subjects received only a description of the image-scanning procedure was conducted. Results demonstrated that subjects were capable of predicting the reaction time results for both verbal and imagery codes. The presence of experimental demand in the image-scanning paradigm necessitates caution when structural interpretations of visual images are considered.

The loci of imagery effects in several domains are clarified by separating issues related to the storage of information in memory and its use following retrieval. Empirical findings from studies of memory for word and sentence lists, language comprehension and memory, and symbolic comparisons are discussed. These consistently indicate a functional role for imagery in human cognition but provide no data necessitating the storage of perceptual information related to verbal materials in an analog form. Instead, coucreleness effects in memory appear to result from differential processing of relational (shared) and item-specific (distinctive) information for high-and low-imagery materials. The available evidence suggests that verbal and imaginal processing systems may operate in conjunction with a more generic semantic memory, the form of which is not an issue here, yielding apparently contradictory findings in support of both dual-code and common-code theories.This article concerns the functional role of imagery in several domains of human cognition and memory. We argue, in particular, that although there have been ample demonstrations indicating separate but interconnected imaginal and verbal processing systems, essentially no empirical evidence requires either verbal or imaginal codes as a long-term representational medium. The abundant evidence of a role for mental imagery in on-line processing (e.g., symbolic comparisons and mental rotation and scanning) cannot be taken as proof for theories of analog and imaginal representations in long-term memory. As we will show, concreteness effects in memory for high-and lowimagery verbal materials similarly can be accounted for in terms of functional processing differences at the time of encoding and so also cannot be taken as proof of analog and imaginal storage in memory.The theoretical distinction we wish to make between imaginal processing and analog representation is not intended as an attempt to resolve the continuing debate on whether perceptual information in memory is represented in an analog or prepositional form. We will not consider that issue in any detail here because we believe it is founded on a false dichotomy. Pylyshyn (1973) noted: Cognitive psychology is concerned with two types of questions: What do we know? and How do we acquire and use this knowledge? The origins of this article lie in the Mental Imagery Symposium (C. L. Richman, chair), held at the annual meeting of the Southern Society for Philosophy and Psychology in Fort Worth, Texas, in May 1982.