Long-term memory of real-world episodes is independent of recency effects: magic tricks as ecological tasks.

Abstract

How episodic memories decay is an unresolved question in cognitive neuroscience. The role of short-term mechanisms regarding the decay of episodic memories is circumscribed to set the maximum recall from which a monotonic decay occurs. However, this sequential view from the short to the long-term is not compulsory, as short-term dependent memory gains (like recency effects when memorizing a list of elements; serial-position effects) may not be translated into long-term memory differences. Moreover, producing memorable events in the laboratory faces important challenges, such as recreating realistic conditions with elevated recall, or avoiding spontaneous retrievals during memory retention (sociocultural hooks). Here we propose the use of magic to enhance the study of memory. We designed a sequence of magic tricks performed live on stage to evaluate the interaction between memory decay and serial-position effects of those tricks. The audience was asked to freely recall the tricks at four different timepoints: just after the show, 10 days, 1.5 months and 4.5 months. We discovered serial-position differences after the show that were no longer present later on, suggesting that short-term memory gains do not translate into the long-term. Illustrating the power of naturalistic stimuli to study long-term memory while interrogating the interaction between short-term and long-term mechanisms, this work is, to our knowledge, the first scientific study of the memorability of magic tricks.

On the cognitive bases of illusionism

“…We have recently made a first attempt in that direction and showed that the memory of a magic trick decays over time as does that of other episodic memories. However, the serial-position differences in memorability, recency effects, that were evident after the show were no longer present later on, suggesting that short-term memory gains do not translate into the long-term (Bestue et al, 2020). This preliminary work is, to our knowledge, the first scientific study of the memorability of magic tricks; and it illustrates the power that magic holds to study memory, and cognition in general, in real-world environments.…”