Memory Explanations (Part 1)

An extensive literature documents that an individual’s response to a particular stimulus can be altered by a prior encounter in the absence of explicit (episodic) recollection of this previous experi-
ence (cf. Roediger & McDermott, 1993; Schacter, 1987). Speculation concerning implicit familiarity as a foundation for a deja vu was originally proposed more than a century ago by Osborn (1884), who suggested that individuals process a considerable amount of information without paying full conscious attention to it and that the subsequent reprocessing of this information may occasionally give rise to a sensation of subjective familiarity in the absence of recollection. What sets the deja vu experience apart from other implicit memory responses is an inordinately strong impression of familiarity in the absence of explicit recollection.

Not all models in this section provide a clear explanation of what causes this intense familiarity, but they do present reasonable frameworks within which the deja vu experience can be interpreted.

Conflict in source monitoring processes

According to the source monitoring framework (SMF; Johnson, Hashtroudi, & Lindsay, 1993; Mitchell & Johnson, 2000), as events are experienced, various features are encoded (perceptual, spatial, emotional, semantic details, temporal information, ongoing cognitive operations), depending on the processing that occurs. Remembering is a mental activity in which current mental experiences are attributed to past events on the basis of their qualitative characteristics and on the individual’s general knowledge and beliefs. Although deja vu has not been specifically interpreted within the SMF, it is relatively straightforward to do so. For example, suppose that you are excited about a first trip to Seattle, enter the airport terminal on your arrival, and have a strong deja vu experience. You know that you never have been in Seattle, but the terminal seems incredibly familiar.

According to the SMF, a deja vu experience could arise from the conflict in two types of source monitoring processes. That is, your attribution based on your general knowledge (of never having been in Seattle) is in conflict with the heuristic attribution that is most natural as based on qualities of the mental experience, which imply it is familiar from past perception. Such familiarity could arise from a number of different types of past events encoded in memory. Perhaps you have seen this airport setting in a movie, television documentary, or picture in a magazine, but your current mental experience does not include this additional information that would allow you to identify the source of the familiarity. If you had a sense of familiarity but could identify its source (e.g., a magazine article on the plane), you would not have a deja vu experience. If you could not identify the source of the familiarity but also thought it was likely that you might have been in Seattle before after all, you also might not have a deja vu experience.

Duplication of processing

Osborn (1884) suggested that it may not be the content of the memory that is duplicated in a deja vu but rather the particular cognitive processing that occurred on a prior occasion:

If at any time in our past lives we passed in actual experience or in imagination over a mental track, say a b c d e , and if to-day [sic] this track is again traversed, although the former experience itself may have been long forgotten, we have a sense that it has been through the mind before. . . . If the mind passes over only part of the former track, say b c d,we sometimes, in the dim recognition which arises, believe we have been over the whole before. (pp. 480–481)

This interpretation can account for why the strong familiarity essential to deja vu can occur in a completely novel setting.  Osborn’s position is similar to the theory of transfer appropriate processing (Kolers, 1973; Kolers & Roediger, 1984; Morris, Bransford, & Franks, 1977; Roediger, Weldon, & Challis, 1989), in which retrieval success depends on the correspondence between the way the information is processed during input and retrieval. If similar mental procedures occur on both occasions, then recollection probability is high. If the processes are dissimilar, the likelihood of remembering the information is low. Thus, if the processing of the new information duplicates the mental procedures that occurred with a prior but unrelated experience, an unexpected sense of familiarity and deja vu may result, even though the stimulus elements in both situations are different.