In this study, magicians were better than non-magicians at differentiating real from fake transfers using both the palm transfer (Experiment 1) and the French drop (Experiment 2). Furthermore, there was a clear relationship between the years of experience practising magic and the ability to detect sleight of hand effects. As expected, spectators of magic effects with no prior engagement in the craft were not able to differentiate a fake transfer from a real one in either experiment.
There was also a clear effect of expertise. The choices made by magicians with a limited amount of experience (i.e., 1 to 4 years of experience practising) did not significantly differ from the choices of non-magicians. By contrast, magicians with more than 5 years of experience practising magic excelled at not being misled by the sleight of hand techniques presented. Remarkably, while both participants with none or little experience and experienced magicians excelled at choosing correctly when observing the real transfer emulating the palming technique, only participants with none or little experience did so when observing a real transfer emulating the French drop movements. In this case, experienced magicians appeared to have difficulties at correctly identifying the hand containing the coin.
While, as a group, magicians in Experiment 1 performed significantly better than non-magicians at identifying a magic effect, further inspections of the magician’s experience revealed that only magicians that had trained in sleight of hand for more than 5 years were able to significantly choose the correct hand when observing a palm transfer technique. This five-year threshold is worth further exploration: one might wonder whether a magician must have mastered the sleight to be able to recognise it in others, or whether it is a case of having had more experience observing other magicians performing it.
These results are in accordance with most evidence substantiating the level of ability that deliberate practice across time in a particular domain can elicit (Farrington-Darby & Wilson, 2006). However, it should be noted that these skills are not necessarily prompted by the number of years practising per se. Indeed, the opposite might be true in some capacity, the lower performance of individuals with years of experience auditing or performing medical diagnostics when compared to less experienced individuals shows that experience might not necessarily be related to expertise (Ericsson, 2008). As such, the expertise skill might be moderated by the type and quality of the practice, rather than the number of years (Ward et al., 2007). For example, in singers, amateurs tend to focus on emotional expression amongst other factors when practising, while expert professional singers tend to focus on precision and technique (Grape et al., 2002). Such differences in the type of practice might be the cause of the effects we see with magicians when identifying false movements. For example, expert magicians might be focusing more on making their movements as similar to real transfers as possible when practising, and this focus might be granting them the skills necessary to perceive the same technique when performed by others. The positive effect that spending extensive time deliberately practising a skill has on the level of expertise of the individual has been discussed in detail (for a review see Ericsson and Lehmann, 1996). Moreover, even for those with already significant experience, evidence suggests that on average, ten years practising a particular domain is necessary for both full expert dominion over it, and success at a professional level (Ericsson et al., 1993; Simon & Chase, 1988; Bryan & Harter, 1899). As such, further experimental evidence should investigate whether the level and quality of the practice that magicians undergo elicit a similar relationship in sleight of hand recognition than the one presented here.
When observing the real transfer of Experiment 1 all participants excelled at identifying the location of the coin regardless of their level of experience practising magic. This was expected as the movement itself reveals the location of the coin. When the experimenter performed the real transfer, the coin was tossed from one hand to the other, thus the observer physically sees the coin move. Moreover, as the movement is a common method of transferring an object from one hand to the other, which is mastered early in child development (Eppler, 1995; Hinojosa et al., 2003; Soska et al., 2010), it is reasonable to assume that any spectator with an unimpaired visual and attentional system would easily locate the hand in which the coin has been transferred to.
In Experiment 2 (French drop) similar patterns were found to those reported for Experiment 1 in terms of the magic effect condition, namely magicians did, as a group, perform significantly better than non-magicians at identifying the sleight of hand. Non-magicians and magicians with less than 5 years of experience were significantly misled by the French drop technique, while more experienced magicians were able to recognise that the motion presented was mimed, and thus not a real transfer of object.
Regarding the real transfer counterpart in Experiment 2, non-magicians and magicians with little experience excelled at significantly choosing the correct hand, whilst there was no significant difference in the choices of experienced magicians. These results stand in contrast to the pattern found in Experiment 1, in which all participants (including experienced magicians) significantly chose the correct hand. To further explain these results, it is important to reiterate that there is a clear difference between the techniques used in Experiment 1 and in Experiment 2, not only methodologically but also regarding the amount of perceptual information that the techniques provide to the observer. In the palming technique used in Experiment 1, whilst there is a simulation of a normal transfer of objects, the movement is not ambiguous (i.e., with attentive observation, the observer can identify that the hand mimicking the transfer is not actually transferring the coin). Similarly, the real transfer technique used in Experiment 1ofers visual clues to the attentive observer indicating that a real transfer has occurred (such as the visible drop of the coin from one hand to the other). This is not the case for either condition in Experiment 2, in which the index and middle finger fully cover the coin (Figure 1). As the coin is fully covered for both the initial grab of the coin and the retention of it, the observer cannot be certain whether the transfer of objects is real or not (as the coin cannot be seen either way).
The performance exhibited by experienced magicians in the real transfer counterpart of Experiment 2 may be explained by the magician’s mindset rather than their discriminatory ability. Given the ambiguity of the movements observed, it is possible that magicians with several years practising how to fool others by utilising sleight of hand have a more pessimistic mindset than non and less experienced magicians when a motion does not provide clear and observable information. When observing an ambiguous movement, experienced magicians might consider the possibility of foul play and choose accordingly, whilst non-magicians and less experience magicians might not suffer from similar pessimistic perceptive biases. The superior ability of magicians to detect mimed movements from real ones has already been documented by Quarona et al., (2020), where magicians performed significantly better when discriminating the mimed motion of an individual reaching to grasp a glass than non-magicians. However, given our results, this superior ability seems to be developed after several years of training in the craft. Moreover, as evidenced by the experienced magicians' poor performance in the real French drop transfer, this prowess does not seem to grant the experienced magicians with a superior ability to discriminate between mimed and non-mimed movements when such movements are perceptually ambiguous.
It is important to note that, in this study, we took a broad-brush approach to magicians’ expertise, and as such the 5-year threshold presented ought to be appraised with caution. Given that the experience level of the participants in this study was measured as four broad groups, the data does not offer insight into the experience variation within the groups (i.e., the exact number of years practising magic), or other variables that might have contributed to such differences in responses (such as the amount of time spent per day practising for example). Consequently, further research is merited to provide insight into the relationship between the number of years of deliberate practice of magicians, and their ability to discriminate deceptive movements.
Nonetheless, our results suggest that whilst, as a group, magicians are better at perceiving deceptive motions in contrast to non-magicians, this ability appears to be moderated by the quantity of experience practising magic: with expert magicians excelling at the recognition of deceptive movements, and non-magicians and magicians with little experience being unable to recognise the fake transfers as deceptive actions. Interestingly the ability to distinguish mimed from real movements does not seem to grant the expert magicians any advantage when these motions are ambiguous in nature. In this case, non-magicians and less experienced magicians seem to be more likely to be correct. We suggest that this might be because expert magicians who have heavily trained in the art of deception, and routinely use their movements to mislead a naïve audience, tend to adopt a pessimistic approach when a movement does not offer clear perceptive clues and consequently, they are more prone to suspect foul play even if the movements are genuine.