Memory for images of concealed objects: A reexamination of Neisser and Kerr

Reexamined U. Neisser and N. Kerr's study in an attempt to resolve an apparent contradiction in the literature concerning whether objects that are concealed in an image are less accessible in memory than objects that are not concealed. Neisser and Kerr found that concealed objects are just as memorable as pictorial objects; it is argued that this may be due to the failure of their Ss to construct images that actually concealed objects. In the present experiment with 36 undergraduates, 1 group of Ss received explicit instructions to conceal; a 2nd group, comparable to the Ss in Neisser and Kerr's study, received no special instructions. In the incidental memory test, concealed objects were recalled significantly less often than pictorial objects for the group receiving instructions to conceal, whereas the group receiving no special instructions replicated the results of Neisser and Kerr. Three explanations of the results show that in all cases it is necessary to assume that an imaginal representation is constructed; whether it is the imaginal representation or a propositional representation that is actually stored, however, cannot be determined from the present results. This is not because issues of internal representation are fundamentally undecidable. It is shown that J. R. Anderson's argument to this effect is wrong. (17 ref) (PsycINFO Database Record (c) 2006 APA, all rights reserved)