False Recognition

Memory & Learning

Roediger and McDermott (1995)

Can automatic associations lead to false memories in the form of false recognition?


The original article refers to Bartlett’s (1932) Schema Theory, stating that “false recall and false recognition are produced by means of activation of implicit associative responses” (Roediger and McDermott, 1995). As defined by Vernon (1955), schemas are: “persistent, deep-rooted, and well-organized classifications of ways of perceiving, thinking, and behaving”. In relation to thinking, schemas are thus networks of ideas about a certain type of event, object, person, or group, that have been clustered together based on prior experience. One of their functions is to organize the knowledge stored in memory in a way that is meaningful and relevant. Another one of their functions is to produce expectations. In both cases, schematic processing speeds up cognition, making it more efficient. For instance, if the words “desk”, “teacher” and “homework” are mentioned in a certain context, they will immediately evoke the idea (schema) of “school”, which unifies them. However, schematic processing also lends cognition to error, as expectations can prove to be false assumptions. Because of the immediate association between “desk, teacher, homework” and “school”, a person asked to memorize and recall the former might also falsely memorize (at encoding) and recall (at retrieval) the latter. This is due to the fact that, as Bartlett’s (1932) explained, memory is not a passive copy and paste, but an active reconstruction based on schematic assumptions.

Participants & Procedures

36 undergraduates from Rice University (Houston, TX, USA)

The researchers chose 6 target words  using the most “successful” ones from the first experiment on this topic (Deese, 1959), and then created 6 lists by adding the first 12 words schematically associated with these targets according to Russell and Jenkins’ (1954) word association norms. Participants were asked to listen to the lists and take an immediate recall test after each one. The instructions specified that they were to write the last few items they could remember first, and then as many as they could,  but only if they were highly confident—not just guesses. After a 2mn filler conversation, the subjects were given a recognition test. The recognition test consisted of 6 blocks of 7 words (one for each list), including: 2 words from the corresponding list, 2 words weakly related with the words from the list, 2 words unrelated to any from the list, and 1 target word (the “critical lure”, which was always last). Participants were asked to determine whether the words had been previously studied by rating them on a 4-point scale: 1=”Sure new”, 2=”Probably new”, 3=”Probably old”, 4=”Sure old”. Subjects took the recognition test at their own pace.


As can be seen in the table above, the mean ratings were very similar for previously studied words and critical lures. This suggests that schematic associations led to false recollections. The low mean rating for weakly related lures also indicates that the results were not due to the participants using wild guesses.

IA Tip

This study is very easy to replicate, including online, and meets all IA requirements without any modification. Students should simply focus on the individual and group ratings rather than on the percentages, which will not be useful for the statistical analysis. They should also be meticulous in their creation of the word lists, using proper instruments and/or a pilot study to ensure that one and only one contains schematic associations and that the lures are properly selected.


Roediger, H. L., & McDermott, K. B. (1995). Creating false memories: Remembering words not presented in lists. Journal of Experimental Psychology: Learning, Memory, and Cognition, 21, 803–814.