Evaluating the evidence for nonconscious processes in producing false memories☆
Introduction
The initial study of Seamon, Luo, and Gallo (1998) attracted our attention because they claimed to have obtained a reliable false memory effect even when the subjects “were unable to discriminate studied words from unrelated nonstudied words” (Seamon et al., 1998, p. 20). There were two main reasons why we were surprised by the conclusions that Seamon et al. drew from their studies. First, although they claimed to have obtained false memory effects “even when memory for list items was essentially absent” (p. 25), their data did in fact consistently show higher hit rates than false alarm rates. Second, they claimed to have presented the items for 20 ms per list item. Based on previous, unrelated, experiments that we had performed using masked perceptual identification, we thought it highly unlikely that subjects could have identified any of the list items at such brief presentation times. In view of the fact that performance on the list items was in fact above chance, we thought it likely that their subjects might have occasionally identified one or more of the list items (perhaps because their equipment was not suitable for such brief presentation times). If that was indeed the case, then these identifications might have been responsible for the observed false memory effects rather than the “nonconscious processing” that they believed was responsible.
In the experiments that we reported in the Zeelenberg, Plomp, and Raaijmakers (2003) paper, we tried to replicate the Seamon et al. (1998) results using better calibrated equipment for controlling the presentation durations of the stimuli. We reasoned that if nonconscious processing of the list items could indeed lead to false recognition, we should be able to obtain such an effect using a more carefully controlled stimulus presentation procedure even when such a procedure would not lead to conscious activation of list items. Recognition performance for both the presented list items as well as the nonpresented critical lures was at chance level for stimulus durations of 20 ms and, somewhat to our surprise, even for stimulus durations of 40 ms. However, a veridical recognition memory effect as well as a substantial false memory effect were obtained for the same items when longer stimulus durations (2000 ms) were used.
In order to explain the difference between our results and those of Seamon et al. (1998), we speculated that “in the Seamon et al. studies due to the combination of software and hardware used words were (sometimes) being presented for more than 20 ms and that as a result subjects could occasionally identify the list items.” (Zeelenberg et al., 2003, p. 410), a not unreasonable speculation given the fact that it is very difficult to accurately control presentation times using the kind of operating system that they used.
Gallo and Seamon (2004) now claim that the use of a 20 ms presentation rate was not really crucial for their conclusions: “the goal was not to eliminate the perception of the list words, but to minimize the conscious processing of those list words” (p. 160). They even go as far as to assume that Zeelenberg et al. (2003) did not obtain a false memory effect precisely because the “presentation rates were so rapid that even the necessary minimal amount of processing did not occur” (p. 161). However, in the abstract of their 1998 article, they wrote that they “found that the subjects falsely recognized semantically related nonstudied words in all conditions, even when they were unable to discriminate studied words from unrelated nonstudied words.” (Seamon et al., 1998, p. 20; see also the title of their article). Although their present position comes close to our conclusion that a false memory effect cannot be obtained under conditions where “the list items are presented for such a short time that they cannot be identified” (Zeelenberg et al., 2003, p. 412), it is still remarkable that we also did not find any effect at a presentation rate of 40 ms, i.e., twice as long as the 20 ms presentation rate that Seamon et al. (1998) claimed to have used in their experiment. We agree with Gallo and Seamon (2004, p. 160) that there must have been “imperfections” in the techniques that they used for stimulus presentation.
Section snippets
False memories at fast presentation rates: Evidence for nonconscious processes?
Although Gallo and Seamon (2004) now seem to believe otherwise, it is worth emphasizing that in our view the logic of the evidence for the nonconscious processing conclusion is critically dependent on the assumption that the list items are not consciously identified during study. The finding of a false memory effect when fast presentation rates are used for list items during study in and of itself does not mean that the false memory effect is due to nonconsious processes. If one or more of the
New experiment of Gallo and Seamon (2004)
In their article Gallo and Seamon (2004) report a new experiment (although apparently still using the same equipment) in which DRM lists were briefly presented during study. Immediately after each list, subjects were asked to report any word they had seen. In this immediate recall test subjects rarely generated the related critical lure. At a later two-alternative forced-choice recognition test, however, subjects choose the related lure with a probability of 58% (where the other alternative is
Conclusions
Given the fact that there are many other explanations for the occurrence of false memory effects when some of the list items are in fact consciously perceived that do not rely on the assumption of nonconscious processes, we see no reason to change our conclusion that these experiments fail to provide evidence for such an assumption. Thus, even though Gallo and Seamon now admit that “the evidence is mixed as to whether conscious recollection of the list items is necessary” (p. 166) for false
References (4)
- et al.
Can false memories be created through nonconscious processes?
Consciousness & Cognition
(2003) - Criss, A. H., & Shiffrin, R. M. (in press). Context noise and item noise jointly determine recognition memory: A...
Cited by (6)
Can false memory for critical lures occur without conscious awareness of list words?
2018, Consciousness and CognitionCitation Excerpt :This finding clearly challenges Gallo and Seamon’s (2004) assumption that all consciously activated critical lures will be produced as false recall, allowing them to be excluded from subsequent data analyses to obtain pure measures of false memory for unconsciously activated critical lures. The support for the unconscious activation hypothesis in Gallo and Seamon (2004) also relies on an assumption that studied critical lures in the test pairs were not selected based on their familiarity arising from association with remembered list words (e.g., choosing sleep because you remember bed) (Gallo & Seamon, 2004; Raaijmakers & Zeelenberg, 2004). An alternative explanation is that some, if not all, of a participant’s above chance selection of critical lures from studied rather than nonstudied lists might be based on conscious activation of the critical lures during study or familiarity with recollected test items.
Development of false memories in bilingual children and adults
2008, Journal of Memory and LanguageActivating the critical lure during study is unnecessary for false recognition
2005, Consciousness and CognitionAssociative illusions of memory: False memory research in DRM and related tasks
2013, Associative Illusions of Memory: False Memory Research in DRM and Related TasksChildren (but not adults) can inhibit false memories
2005, Psychological Science
- ☆
Reply to Commentary on Zeelenberg, R., et al. (2003). Can false memories be created through nonconscious processes2. Consciousness and Cognition, 12, 403–412.