Internally represented forces may be cognitively penetrable: comment on Freyd, Pantzer, and Cheng (1988)

Freyd, Pantzer, and Cheng (1988) provided considerable evidence for the proposition that people can represent underlying forces within static scenes. However, they explicitly assumed that their observed memory shifts were the result of perceptually modular information processing. For several reasons, I suggest herein that this assumption of cognitive impenetrability is a dubious one. The assumption is challenged by recent empirical findings, some theoretical considerations, and calculations that show that the observed effects are minute when compared with those expected by means of physical forces. Three explanations for the evidence are proposed, including the alternative hypothesis that although people do represent static physical forces, these representations can be almost completely overridden by the conscious intention to remember an object's precise location. Freyd, Pantzer, and Cheng (1988) clearly demonstrated that when a pictorial element is shown first in equilibrium and then with an applied, nonzero net force, subjects tend to exhibit a memory shift in the direction of that force. In other words, when subjects are asked whether a test display shows the pictured element in the same position as in two previous displays, they are more likely to agree when the element is "moved" (repositioned) in concert with the implied disequilibrium than when it is moved counter to that net force. Freyd et al. cited this memory shift as evidence that people tend to represent the underlying forces of static scenes. They further suggested that such representations may form the basis of our phenomenal, conscious sense of concreteness. This interpretation seems reasonable and is supported by other recent results (see Hubbard & Bharucha, 1988, p. 213, regarding gravity). The foundation for the present discussion, however, stems from a more dubious notion: Freyd et al.'s explicit assumption that the observed effect is the result of encapsulated information processing (see Fodor, 1983), specifically, perceptual modularity.