Report on Research Study: "Why It’s Easier to Remember Seeing a Face We Already Know Than One We Don’t: Preexisting Memory Representations Facilitate Memory Formation" by Reder et. al.
Authors of study: Lynne M. Reder, Lindsay W. Victoria, Anna Manelis, Joyce M. Oates, Janine M. Dutcher, Jordan T. Bates, Shaun Cook, Howard J. Aizenstein, Joseph Quinlan, and Ferenc Gyulai.
Published in: Psychological Science: 2013 24: 363
A team of researchers used two experiments to test a hypothesis that stimuli with already-established memory representations (e.g. faces of celebrities) are easier to relate to the encoding context than stimuli that do not have long-term memory representations (e.g. faces of strangers). The theoretical backboard is the idea that semantic long-term memory of famous people (and factual knowledge about them) assists identification of context by facilitating elaboration of the encoding episode.
In Experiment 1, research participants were exposed to different images. The fan (the number of memories associated with a certain context) of the background of each picture was manipulated with familiar and unfamiliar faces. The researchers hypothesized that these manipulations would influence memory of the context depending on whether the faces were famous or not.
Experiment 2 involved a drug intervention that inhibited the creation of new memories but did not affect the familiarity-based recognitions that were tested in Experiment 1. Furthermore, in both experiments, there was an additional test for the theory that memory enhancement for known faces arises from the ease of relating stimuli and context that have previous memory representations: the participants were asked whether the control response (when they identified contexts without a familiar face in the picture) was based on “contextual information” or on “item familiarity”.
The study excluded trials with the faces of celebrities that were not identified by the participants from the analysis. Nonetheless, the participants correctly identified 90% of the celebrities and were even more accurate at discarding unfamiliar faces. The results of both experiments propose that long term memory representations to faces really do facilitate an association to context. Reaction times and correct identification rates corroborate this. When the participants were given midazolam – a drug that inhibits forming of new memories – the enhanced effects previously observed were diminished.
The research study had a powerful control of the variables. The paper outlines the many factors that were controlled in order to reduce opportunity for confounding. For example, it was confirmed through testing that the “familiar” faces used in the experiments were actually recognizable by all the subjects, and when it was not, these participants were not included in the study.
This is important because the very nature of the independent variable would be invalid if there was no recognition of the celebrity. The research was also very extensive (sample size was not limited, although a call for further research would not be superfluous), and the analysis was conveyed through quantitative depictions rather than simple qualitative reasoning (i.e. a confidence interval was applied to the research values).
I found the results to be particularly compelling. The control of the experiment was having the subjects take the drug that prevented the formation of associations with new faces. The research successfully managed to test a cognitive function through repeated trials. In terms of validity of theory, the concepts put forth by the paper stand in tandem with common psychological canon (i.e. that interference can occur when the unfamiliar is added to the familiar).