Masked transposition effects for simple versus complex nonalphanumeric objects

When two letters/digits/symbols are switched in a string (e.g., jugde-judge; 1492-1942; *?$&-*$?&), the resulting strings are perceptually similar to each other and produce a sizable masked transposition priming effect with the masked priming same-different matching task. However, a parallel effect does not occur for strings of pseudoletters (e.g., (sic); Garcia-Orza, Perea, & Muoz, Quarterly Journal of Experimental Psychology, 63, 1603-1618, 2010). In the present study, we examined whether masked transposition priming is specific to alphanumeric stimuli or whether it also occurs with strings composed of other objectsaEuronamely, line drawings of common objects (Experiment 1) and geometrical shapes (Experiment 2). Results showed a significant masked transposition priming effect for geometrical shapes (e.g., (sic)), but not for line drawings of common objects (e.g., (sic)). These findings suggest that the mechanism involved in the coding of position in masked priming works only with perceptually simple, familiar objects (i.e., letters, numbers, symbols, or geometrical shapes), once their identities have been well ascertained.