Academic journal article Memory & Cognition

The Role of Test Structure in Creating False Memories

Academic journal article Memory & Cognition

The Role of Test Structure in Creating False Memories

Article excerpt

In the Deese/Roediger-McDermott (DRM) paradigm, studying lists of semantic associates results in high rates of false recognition of a nonpresented critical word. The present set of experiments was designed to measure the contribution of additional processing of list items at test to this false memory effect The participants studied sets of lists and then performed a recognition task for each set. In three experiments, using this paradigm, we investigated false recognition when the number of studied list items presented at test (0, 6, or 12) was manipulated. In Experiments 2 and 3, false recognition of critical lures associated to both studied and nonstudied lists increased significantly as the number of list items included in the test increased. These results indicate that processes occurring at retrieval contribute to false memory effects found with the DRM paradigm.

In recent years, the study of experimentally induced false memory phenomena has increased substantially. One method with which false memories are studied involves using word list paradigms that are based on properties of semantic association and that reliably reveal robust false memory effects. Many recent studies have utilized the Deese/Roediger-McDermott (DRM) paradigm (Roediger & McDermott, 1995), in which participants study lists of semantic associates (e.g., bed, rest, awake, drowsy, etc.) of a nonpresented critical lure (e.g., sleep). During subsequent free recall or recognition tasks, participants tend to remember the critical lure at rates comparable to the studied items (Roediger & McDermott, 1995). Furthermore, participants often report similar phenomenological experiences (e.g., a vivid memory of having studied the item) for both the list items and the critical lures (Roediger & McDermott, 1995). In other words, participants are able to "retrieve" the experience of encoding the critical item, when in fact it was not studied.

The false memory effect appears to be quite robust, and it occurs reliably across a variety of experimental manipulations. Indeed, some factors have been identified that appear to increase the strength of the illusion. Roediger and McDermott (1995) found higher rates of false recognition of critical lures associated to lists that had been previously recalled than of lures associated to lists that had not been recalled, indicating that repeated retrieval attempts might enhance the false memory effect. In addition, several studies that examined the effects of longer retention intervals on accurate and false memory have found that false memories actually increase over time, whereas accurate memories decrease (McDermott, 1996; Thapar & McDermott, 2001 ; Toglia, Neuschatz, & Goodwin, 1999). Seamon et al. (2002), however, found that false memories did not increase over time; rather, they were more resistant to decay than accurate memories. Thus, it appears that false memory for the critical lure in the DRM paradigm can be enhanced or at least maintained by testing factors such as the number of tests.

Evidence also suggests that retrieval processes may play a significant role in creating false memories for the critical lure. In the original Roediger and McDermott (1995) study, an output serial position analysis of the free recall data indicated that participants tended to falsely recall the critical lure toward the end of the recall session. This may indicate that prior recall of list items served as a cue for the lure, or that recalling the list items increased the probability that participants would recall the lure as an item highly associated to all studied items. In recognition tasks, one or more list items are often presented before the lure, which may have thus contributed to additional priming of the lure (e.g., Roediger & McDermott, 1995).

One explanation of the effectiveness of the DRM paradigm in creating false memories is semantic activation. According to activation theories, words are linked to one another in a network, and the activation of one lexical concept results in the spread of activation to surrounding concepts (Collins & Loftus, 1975). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.