Manual action expectation and biomechanical ability in three species of New World monkey
Elias Garcia-Pelegrin1,2,4*, Rachael Miller 2,3, Clive Wilkins 2, Nicola S Clayton2
1Department of Psychology, National University of Singapore, 117572, Singapore.
2Department of Psychology, University of Cambridge, CB2 3EB, United Kingdom.
3School of Life Sciences, Anglia Ruskin University, CB1 1PT, United Kingdom.
4Lead Contact.
*Corresponding author. Elias Garcia-Pelegrin: [email protected]
Twitter: @EGarciaPelegrin
Summary: Being able to anticipate another’s actions is a crucial ability for social animals because it allows for coordinated reactions. However, little is known regarding how hand morphology and biomechanical ability influences such predictions. Sleight of hand magic capitalises on the observer’s expectations of specific manual movements 1,2, making it an optimal model to investigate the intersection between the ability to manually produce an action, and the ability to predict the actions of others. The French drop effect involves mimicking a hand-to-hand object transfer by pantomiming a partially occluded precision grip. Therefore, to be misled by it, the observer ought to infer the opposing movement of the magician’s thumb 4. Here, we report how three species of platyrrhine with inherently distinct biomechanical ability 5–7 – common marmosets (Callithrix jacchus), Humboldt’s squirrel monkeys (Saimiri cassiquiarensis), and yellow-breasted capuchins (Sapajus xanthosternos) – experienced this effect. Additionally, we included an adapted version of the trick using a grip that all primates can perform (power grip), thus removing the opposing thumb as the causal agent of the effect. When observing the French drop, only the species with full, or partial, opposable thumbs were misled by it, just like humans. Conversely, the adapted version of the trick misled all three monkey species, regardless of their manual anatomy. The results provide evidence of a strong interaction between the physical ability to approximate a manual movement and the predictions primates make when observing the actions of others, highlighting the importance of physical factors in shaping the perception of actions.
Results and Discussion
Observing others making actions can activate one’s own motor system 8, and the common-coding theory 9,10 provides a tenet for these phenomena. According to this principle, actions are coded in reference to the causative effects that they should produce. Alongside this, it is theorised that representations (i.e., common-codes) of the causal effects of actions determine the perception and production of those actions. When we produce actions, our common-codes instruct our movement, whilst when we perceive actions, our common-codes allow us to detect the goal of the action being perceived. Consequently, perceiving and performing a particular action should activate the same common-codes, and the more similar an observed action is to the way the observer would perform it, the stronger the activation of these common-codes. Research in neuroscience has produced considerable evidence of common-codes occurring at the level of single neurons in the brain, the so-called mirror system 11–13. For example capoeira dancers have a stronger activation of mirror neurons when they watch videos of capoeira dances than when observing videos of ballet 14,15. However, both the nature of the mirror system and the relevance of mirror neurons in action understanding are still topic of heated debate 16,17. Whilst some attest to the key importance of this system for the development of socio-cognitive ability, communication, and culture 11,18, others propose it to be part of a more domain general system such an associative learning process between both the visual and motor inputs of an action 19,20, thus making action experience a necessary step in the development of these common-codes 21,22. Although there is a strong correlation between mirror neuron activity and action recognition, it is hard to determine whether this activity is a direct cause of the ability to understand actions or just a by-product of it 17,23. Moreover, in autistic participants, who often struggle with imitation tasks 24, no correlation has been found between activation of mirror neurons and the ability to understand the actions or intentions of others 25,26.
In conjunction to the discrepancies surrounding the impact of the mirror system in the formation of common-codes, even less is known regarding how hand morphology and biomechanical ability influences both the activation and creation of them. If experience producing an action moderates the predictions that one will make when observing a similar action in others, a physiognomic inability to perform said action (such as not having evolved the causal limbs that perform it) should result in different predictions than the ones embodied by observers with similar physiological features. In this light, the magic tricks that magicians use to deceive and amaze their audience known as sleight of hand offer an optimal model for such an investigation because they capitalise on the spectator’s intrinsic expectations of the outcome of making certain hand movements 2,4. Indeed, in the last decades some psychologists have endeavoured to investigate how humans experience these intricate techniques of deception 27,28, and whilst investigations into the psychology of magic have primarily been confined to researching human perception 29, this line of inquiry has recently permeated into the field of comparative psychology, where some have started to examine how other species experience these deceptive movements and both the similarities and differences in human and non-human expectations 2,30–32. We focused on three species of New World Monkey, with inherently different manual anatomy and biomechanical ability – yellow-breasted capuchin monkeys (Sapajus xanthosternos), Humboldt’s squirrel monkeys (Saimiri cassiquiarensis), and common marmosets (Callithrix jacchus) (Figure 2).
Capuchin monkeys have gained a reputation for their manual ability and the varied actions that they employ in object manipulation 33. This is likely due in part to their hand physiognomy, which allows them to individually control their finger digits, as evidenced in their single digit probing actions, and their scissor grip capability i.e., getting hold of an object by holding it in between the sides of two fingers 34–38. Alongside this, capuchin monkeys are the only member of the New World Monkeys to be able to perform a precision grip by bringing the thumb towards the index or middle finger in a pad-to-pad motion 39, which has independently evolved in this species. Moreover, while exploring objects with their hands, capuchin monkeys will employ similar exploratory actions as humans such as probing, pinching, enclosing the item with both hands, and following the contour of the object 40. Capuchins are also renown for being the only platyrrhines to systematically operate tools with their dexterous hands 41. For instance, yellow-breasted capuchins and wild bearded capuchin monkeys (Sapajus libidinosus) will routinely use stone tools to crack nuts in the wild 42, and captive tufted capuchins demonstrate a wide array of tool use ability 33,43–45. While squirrel monkeys are comparably less dexterous than capuchin monkeys 39, these monkeys have been observed rudimentarily using tools in some rare occasions 46. Their hinge like carpometacarpal joint limits the rotation of their thumb, thus restricting their range of motion with their thumb in relation to the pads of the index and middle finger and making full opposition impossible. However, these primates can still oppose their thumb so to touch the side of their index or middle finger in a pad-to-side movement 7, and thus still be able to experience, to a degree, the occlusion of the thumb when hidden behind the index and middle fingers, and the interaction that the thumb has with these digits when occluded. Contrary to both capuchins and squirrel monkeys, marmoset’s hands have evolved mainly for vertical locomotion i.e., vertically climbing tree trunks 34. Consequently, as any type of opposable thumb would still not allow the marmoset to grasp the entirety of a trunk with the hands, evolving one would not be of much use 47,48. Instead, marmosets spread their five digits, which are equidistant form each other, as widely as possible so to increase the amount of suffice area and dig in with their claws by flexing all their digits at unison 5,49. To allow for this, their thumb is substantially shifted distally so to align with the rest of the fingers. The different hand morphology in marmosets makes them unable to operate a precision grip (in contrast to capuchin monkeys), perform a pad-to side precision-like grip (like squirrel monkeys can), or even to occlude their thumb behind the rest of their fingers. Instead, to manipulate objects these primates use a combination of power grips 7,49, and scissor grips 50.
We tested how these three species of platyrrhine experienced the French drop magic trick, a modified version of it called the Power drop, and their respective real transfer counterparts (see Figure 1). In each of the two trials per condition presented to the subjects, a food reward was first shown to the subject with one hand and then either transferred to the opposite hand (if a real transfer) or retained in the same hand (if a sleight of hand magic trick). Following this, the subject was allowed to choose which hand contained the reward, and the hand was opened upon selection. If chosen correctly, the subject was allowed to consume the reward within (see Video S1 for a video of the conditions). If the expectations of a particular movement are embodied by the manual ability of the observer, we hypothesise that only the yellow breasted capuchins, which have experienced enacting a precision grip should infer the causal effect of another opposable thumb and expect the French drop transfer to be completed. By contrast, the primates that cannot perform a precision grip (i.e., squirrel monkeys and marmosets), should not have the necessary expectations to recognise the French drop action as a transfer of objects between hands, and thus not be misled by the French drop. Furthermore, given that the Power drop effect emulates the manoeuvres of a power grip, a movement that all primates have experience enacting 5,34, all subjects should mistake the pantomimed motion for a real transfer regardless of their hand anatomy and biomechanical ability.
Figure 3 shows the choices of all three species for every condition. Squirrel monkeys and capuchin monkeys mostly chose the incorrect hand that did not contain the reward when the French drop sleight of hand was used (binomial test: capuchins – p = 0.02; squirrel monkeys – p < 0.001) but chose the correct hand when observing its real transfer counterpart (binomial test: capuchins – p = 0.04; squirrel monkeys – p = 0.02). This finding shows a clear bias towards believing that a transfer of objects involving a precision grip has been completed even if said transfer has been pantomimed. This pattern of choices has also been seen in humans who, like these other two primate species, are typically misled by the French drop magic effect, but not by its real transfer 3,4. The Generalised Linear Mixed Model (GLMM) revealed that there was a significant effect of condition (χ2 = 39.56; df = 3; p < 0.001), and a significant interaction between condition and species (χ2 = 30.29; df = 6; p < 0.001), but no significant effect of species (χ2 = 2.23; df = 2; p = 0.32). Post-hoc pairwise comparisons further revealed that there were no significant differences between the choices of capuchins and squirrel monkeys in the French drop nor the French drop real transfer (p = 1). By contrast, marmosets performed significantly differently than capuchins (French drop: p = 0.02, real transfer: p = 0.04) and squirrel monkeys (French drop: p = 0.01, real transfer: p = 0.08). Specifically, marmosets chose mostly correctly when observing a French drop effect (binomial test: p < 0.001), but mostly incorrectly when observing its real transfer counterpart (binomial test: p= 0.07).
The French drop sleight of hand occludes the thumb when pretending to grab the focal object. This is a key component of the illusion: instead of performing a normal grabbing motion of the object, the thumb allows the object to fall to the opposite hand whilst simultaneously pretending that an object has been pinched between the thumb, the index finger, and the middle finger, also known as a precision grip 7. The results presented here suggest that marmoset monkeys, which have a rigid thumb and thus cannot oppose it 47,49,51, perceived this sleight of hand effect differently to that of capuchin monkeys, which have an opposable thumb, and squirrel monkeys, which have a partially opposable thumb 33,39. Indeed, it appears that, in this case, the marmosets were most likely using a simple heuristic to choose the hand that contained the reward initially regardless of the pantomime action performed by the experimenter. This pattern of choices has also been observed in corvids when faced with the same French drop effect 2, which of course lack all of the appendages required to produce the precision grip involved in the magic trick.
The choices of marmosets in the French drop and real transfer conditions were moderated by their lack of the necessary expectations about thumb occlusion, opposition, and prehensility. This is supported by the key control conditions – Power drop trick and its real transfer – where the GLMM revealed that the choices of all three species of monkey did not significantly differ from each other neither in the Power drop condition nor in its real transfer version (p = 1 between all three species and conditions). When observing the Power drop, all three species were more likely to choose the incorrecthand (binomial test: capuchins – p = 0.02; squirrel monkeys – p < 0.001; marmosets – p < 0.001). By contrast, all three species chose the correct hand (capuchins – p < 0.001; squirrel monkeys – p = 0.02 for; marmosets – p = 0.004) when observing its real transfer.
This finding reinforces the notion that our results are not a by-product of other species differences, such as the overall size of the subject, or the amount of experience in observing human hands. These two transfers were purposely devised for this study and consisted of the same premise of a French drop sleight of hand effect and a real transfer but removed the necessity to infer the opposable movements of the thumb. This was achieved by performing a variation of a power grip in which the reward is grabbed (or pretend grabbed) with all the digits of the hand that apprehend the reward at unison by pressing it against the palm (Figure 1). This type of grabbing motion was specially chosen because it is regularly performed by marmosets to grasp objects or food and can also be performed by both capuchins and squirrel monkeys in a similar manner 5,7,34,39,51. Therefore, the pattern exhibited by the marmoset monkeys in the Power drop conditions, which stands in complete contrast to their pattern in the French drop conditions, clearly suggests that this sleight of hand movement capitalised on their inherent expectations of hand biomechanics, which fell prey to the same trickery that magicians routinely use to fool their human audiences.
In conclusion, capuchins – that are capable of thumb opposability and precision grip grasping, but not marmosets – that are not capable of thumb opposability, appear to have similar expectations to humans when observing sleight of handeffects involving the use of an opposable thumb as the causal agent of the manoeuvre. Furthermore, in contrast with our hypothesis, squirrel monkeys – that can oppose their thumb but not operate a precision grip, performed alike to capuchin monkeys in all conditions. This finding prompts several hypothesis concerning how squirrel monkeys acquired biomechanical expectations of precision grip movements without personal experience with the manoeuvre.
It is possible that expectations of others’ manual transfers are not based on the experience of performing the movement at all but rather on the cognitive ability to understand the causal properties of a fully opposable thumb. However, this explanation would require squirrel monkeys to possess a qualitatively different causal understanding of human hand biomechanics than marmoset monkeys, which, if true, would most likely still be attributed to their inherent manual differences with marmoset hands, and similarities with human hands.
Another potential explanation may concern differential levels of exposure to humans’ hands between species – higher in capuchin and squirrel monkeys than marmosets – thus leading to more accurate expectations regarding human hand biomechanics in the former two species. This is unlikely, as both the capuchin and squirrel monkeys were part of zoological collections where animal handling is discouraged, and feeding tends to happen by scattering the food in the enclosure (as a form of enrichment) rather than given by hand 52. Whereas the marmoset monkeys belonged to an animal research facility and were regularly handled and hand fed. Therefore, in this case, it is more likely that the marmosets had richer experiences involving human hands than either of the other two species.
Finally, a more plausible hypothesis is that the requisite experiences to anticipate movement need not be precisely accurate to form common-codes that serve as rough approximations of the potential range of movement. In this case, squirrel monkeys’ expectations may arise because of their ability to both physically occlude the thumb with both the index and middle finger, and to perform a side-to-pad pseudo-precision grip 5. This is an ability that marmoset monkeys do not share, which might lead squirrel monkeys to, when witnessing someone occlude their thumb with their own fingers, infer a grasp of the object without having to deduce a precision grip. This would suggest that the causative cues informing the necessary common-codes allowing the squirrel monkey to recognize the maneuver as a transfer would be formed by the intersection between manual anatomy, observational input, and associative processes 19,20 instructed by the likelihood of an event to occur given some specific manual cues (i.e., a similar hand structure grabbing the object in a similar, yet not the same, maneuver). Overall, although future research is necessary to fully comprehend the neurological and behavioural mechanisms underlying the observed differences in action expectation, this study’s results provide compelling evidence that an observer’s inherent manual capability heavily influences their perception and prediction of others’ manual movements.
Acknowledgments
We would like to thank all the zookeepers at Shepreth Wildlife Park, Cotswolds Wildlife Park and Gardens, and Shaldon Wildlife Trust and the technicians at the University of Cambridge’s Biomedical Centre for their time. Special thanks to Yvonne Morrin (West Section Curator at Shepreth Wildlife Park), Helen Hitchman (Education and Activities Manager at Cotswolds Wildlife Park and Gardens) and Zak Showell (Director of Shaldon Wildlife Trust), for liaising and assisting in this study. Many thanks to the BIAZA committee for their letter of support.
Author Contributions
EG-P contributed to the conceptualisation, methodology, investigation, and original draft writing. RM, CW, and NS contributed to the review and editing of the manuscript. NS supervised the project.
Declaration of Interests
The authors declare no competing interests.
Figure legends
Figure 1. Movements required to perform the French drop, Power drop, and their real transfer counterparts. Note. the use of a precision grip in the French drop involving the thumb, in contrast with the use of the entire hand in the Power grip, in which the thumb is neither occluded by the rest of the fingers, nor it is opposed. See also Video S1.
Figure 2. Three species of New World Monkey with inherently different hand anatomy and biomechanical abilityFrom left to right: yellow-breasted capuchins (Sapajus xanthosternos), squirrel monkeys (Saimiri sciureus), and common marmosets (Callithrix jacchus). (images included to illustrate these species differences, see also Video S1). All pictures are under a creative commons licence.
Figure 3. Percentage of incorrect hand choices in the French Drop and Power Drop conditions (Magic Effect and Real Transfer trials) for marmosets, capuchins, and squirrel monkeys.
STAR★Methods
Resource availability
Lead contact
Further information and requests for resources should be directed to and will be fulfilled by the lead contact, Elias Garcia-Pelegrin ([email protected]).
Materials availability
The study did not generate new unique reagents.
Data and code availability
Experimental model and subject details
We tested three species of monkey with inherently different hand anatomy and biomechanical ability residing in several zoos and laboratories in the UK.
Subjects were: 8 yellow-breasted capuchins (Sapajus xanthosternos) (6 female) at Shepreth Wildlife Park and Shaldon Wildlife Trust, 8 Humboldt’s squirrel monkeys (Saimiri cassiquiarensis) (4 female) at Cotswolds Wildlife Park and Shaldon Wildlife Trust, and 8 common marmosets (Callithrix jacchus) (4 female) from the University of Cambridge’s Marmoset Colony. The marmosets were housed in a conventional barrier facility and belonged to a breeding colony that was not used in research. The main food reward was selected as highly preferred by each species and a regular dietary item: capuchins – peanuts, squirrel monkeys – dried mealworms, marmosets – marshmallows. In all settings, the experimenter interacted with the monkeys through the mesh, which were large enough for subjects to reach through to touch the experimenter’s hands. There was one experimenter (EG-P) that collected all the data, who is a trained magician with 12 years of experience.
Subjects naturally reached for, and pried open, the experimenter’s closed fist through the holes in their meshed enclosures to reach an enclosed food reward without any need of initial training. Subjects were then taught that in the presence of two closed fists, they would only have access to the first fist they tried to pry open. This was accomplished by showing the reward in one hand (which was pseudo-randomised per trial) alongside the empty palm of the other hand, and then simultaneously closing both hands into fists. Once the subjects successfully retrieved 8/10 consecutive rewards (10 trials per session, average nº of sessions = 1 (capuchins); 1 (squirrel monkeys); 2 (marmosets)) they moved to the next stage of training. The final stage of training consisted in the subjects learning to determine which hand contained the reward after observing it being transferred between hands. To do so, the subjects observed the experimenter visibly transfer the reward from one hand to the other and then close their hands into tight fists. Once the subject successfully retrieved 8/10 rewards in two consecutive sessions (10 trials per session, average nº of sessions = 1 (capuchins); 1 (squirrel monkeys); 1 (marmosets)), the subject moved to the testing phase.
Testing
During each session, the food reward was first shown to the subject with one hand and then either transferred to the opposite hand or retained in the same hand (as per condition). Following this, the subject was allowed to choose which hand contained the reward, and the hand was opened upon selection. If chosen correctly, the monkey was allowed to consume the reward within. The starting hand and conditions were pseudo-randomised across trials thus, all subjects experienced fake and real transfers from both hands and not in any specific pattern. If, after the transfer was demonstrated, the subject chose the hand that contained the reward, the subject scored “1” on the trial and was allowed to eat the reward, otherwise the monkey scored “0”. We performed two trials per condition, one transferring (or fake transferring) the reward from left to right and vice versa, with eight trials in total. Fig 1 outlines the movements required to perform the 4 conditions: French drop, French drop real transfer, Power drop and Power drop real transfer.
The data were recorded while being coded in situ and subsequently cross-referenced with the recordings. Inter-rater reliability was measured by a blinded coder scoring a random selection of 20% of the trials, with a balanced quantity of all conditions. Reliability was excellent for all experiments (Cohen’s Kappa = 1). Further, to ensure that the sleight of hand performed to the subjects was free from any possible biases or inadequacies that could, otherwise, compromise the data obtained (i.e. “Clever Hans” effect 53), we also coded for experimental rigor for all trials. Two new blinded coders were shown what an ideal, unbiased, trial would consist of for each condition, then they were asked to observe all trials performed, and were given instructions to flag any trial for possible inconsistencies. Only one trial was flagged by one of the coders: trial 2 of the French drop Real transfer condition for one individual marmoset (individual no 8) – the marmoset responded correctly. As there was disagreement between these two coders regarding the validity of the trial, a third blinded coder was asked to assess this specific trial. The third coder did not identify any inconsistency and therefore the trial was deemed to be valid for inclusion. It is important to note that, even if this trial was excluded, given the significantly larger incorrect responses of the marmosets when observing the French drop Real transfer, this trial would not invalidate the results obtained, but would instead emphasise the significant disparity between correct and incorrect responses found for this species.
Statistical analyses were accomplished using JASP (v.0.10.3, http://jasp-stats.org) and RStudio for Mac (version 1.2.1335). To determine the subjects´ choices per condition, we used binomial tests (against value: 0.5). To determine whether the subjects´ choices were influenced by the conditions, and to compare the choices between species, we used a series of Generalised Linear Mixed Models (GLMM) with subject as a random effect, condition, and species as main effects, as well as a condition * species interaction. Significant differences between treatments were further explored using post-hoc pairwise comparisons and were adjusted using the Holm-Bonferroni method to maintain the overall alpha level at the nominated value of 0.05 for multiple pairwise comparisons.
Ethics statement
This study did not require food restriction or any dietary changes, the subjects were fed their regular daily diet by the animal care team, with constant access to water. The experiments were reviewed and approved by the University of Cambridge and conducted under a non-regulated license (zoo 64/19). This study was reviewed and supported by the British and Irish Association of Zoos and Aquariums (BIAZA).
Supplementary Materials
Video S1: Recorded Methodological Examples. Related to Figure 1 & 2.
Video examples of the sleight of hand effects (French drop, Power drop, and real transfer counterparts), and subject species (common marmosets (Callithrix jacchus), Humboldt’s squirrel monkeys (Saimiri cassiquiarensis), and yellow-breasted capuchins (Sapajus xanthosternos))
References
1. Kuhn, G., Amlani, A.A., and Rensink, R.A. (2008). Towards a science of magic. Trends Cogn. Sci. 12, 349–354.
2. Garcia-Pelegrin, E., Schnell, A.K., Wilkins, C., and Clayton, N.S. (2021). Exploring the perceptual inabilities of Eurasian jays (Garrulus glandarius) using magic effects. Proc. Natl. Acad. Sci. U. S. A. 118. 10.1073/pnas.2026106118.
3. Phillips, F., Natter, M.B., and Egan, E.J.L. (2015). Magically deceptive biological motion—the French Drop Sleight. Front. Psychol. 6, 371.
4. Garcia-Pelegrin, E., Wilkins, C., and Clayton, N.S. (2022). Investigating expert performance when observing magic effects. Sci. Rep. 12, 1–10.
5. Fragaszy, D.M., and Crast, J. (2016). Functions of the Hand in Primates. In 10.1007/978-1-4939-3646-5_12.
6. Napier, J. (1962). The evolution of the hand. Sci. Am. 10.1038/scientificamerican1262-56.
7. Napier, J.R. (1960). Studies of the hands of living primates. In Proceedings of the Zoological Society of London (Wiley Online Library), pp. 647–657.
8. Press, C., Cook, J., Blakemore, S.-J., and Kilner, J. (2011). Dynamic modulation of human motor activity when observing actions. J. Neurosci. 31, 2792–2800.
9. Prinz, W. (1997). Perception and action planning. Eur. J. Cogn. Psychol. 9, 129–154.
10. Hommel, B., Müsseler, J., Aschersleben, G., and Prinz, W. (2001). The theory of event coding (TEC): A framework for perception and action planning. Behav. Brain Sci. 24, 849–878.
11. Gallese, V., and Goldman, A. (1998). Mirror neurons and the mind-reading. Trens Cogn. Sci. 10.1016/S1364-6613(98)01262-5.
12. Gallese, V., Fadiga, L., Fogassi, L., and Rizzolatti, G. (1996). Action recognition in the premotor cortex. Brain 119, 593–609.
13. Rizzolatti, G., and Arbib, M.A. (1998). Language within our grasp. Trends Neurosci. 21, 188–194.
14. Calvo-Merino, B., Glaser, D.E., Grèzes, J., Passingham, R.E., and Haggard, P. (2005). Action observation and acquired motor skills: An fMRI study with expert dancers. Cereb. Cortex. 10.1093/cercor/bhi007.
15. Calvo-Merino, B., Grèzes, J., Glaser, D.E., Passingham, R.E., and Haggard, P. (2006). Seeing or Doing? Influence of Visual and Motor Familiarity in Action Observation. Curr. Biol. 10.1016/j.cub.2006.07.065.
16. Glenberg, A.M. (2011). Positions in the mirror are closer than they appear. Perspect. Psychol. Sci. 6, 408–410.
17. Hickok, G. (2009). Eight problems for the mirror neuron theory of action understanding in monkeys and humans. J. Cogn. Neurosci. 21, 1229–1243.
18. Rizzolatti, G., Fogassi, L., and Gallese, V. (2001). Neurophysiological mechanisms underlying the understanding and imitation of action. Nat. Rev. Neurosci. 10.1038/35090060.
19. Heyes, C. (2010). Where do mirror neurons come from? Neurosci. Biobehav. Rev. 34, 575–583.
20. Cooper, R.P., Cook, R., Dickinson, A., and Heyes, C.M. (2013). Associative (not Hebbian) learning and the mirror neuron system. Neurosci. Lett. 540, 28–36.
21. Cook, R., Bird, G., Catmur, C., Press, C., and Heyes, C. (2014). Mirror neurons: from origin to function. Behav. Brain Sci. 37, 177–192.
22. Catmur, C., Press, C., and Heyes, C. (2016). Mirror neurons from associative learning. Wiley Handb. Cogn. Neurosci. Learn., 515–537.
23. Caramazza, A., Anzellotti, S., Strnad, L., and Lingnau, A. (2014). Embodied cognition and mirror neurons: a critical assessment. Annu. Rev. Neurosci. 37, 1–15.
24. Williams, J.H.G., Whiten, A., and Singh, T. (2004). A systematic review of action imitation in autistic spectrum disorder. J. Autism Dev. Disord. 34, 285–299.
25. Bird, G., Leighton, J., Press, C., and Heyes, C. (2007). Intact automatic imitation of human and robot actions in autism spectrum disorders. Proc. R. Soc. B Biol. Sci. 274, 3027–3031.
26. Spengler, S., Bird, G., and Brass, M. (2010). Hyperimitation of actions is related to reduced understanding of others’ minds in autism spectrum conditions. Biol. Psychiatry 68, 1148–1155.
27. Cocchini, G., Galligan, T., Mora, L., and Kuhn, G. (2018). The magic hand: Plasticity of mental hand representation. Q. J. Exp. Psychol. 71, 2314–2324. 10.1177/1747021817741606.
28. Kuhn, G., and Land, M.F. (2006). There’s more to magic than meets the eye. Curr. Biol. 16, 950–951. 10.1016/j.cub.2006.10.012.
29. Kuhn, G. (2019). Experiencing the impossible: The science of magic (MIT Press).
30. Garcia-Pelegrin, E., Schnell, A.K., Wilkins, C., and Clayton, N.S. (2020). An unexpected audience. Science (80-. ). 369, 1424 LP – 1426. 10.1126/science.abc6805.
31. Garcia-Pelegrin, E., Schnell, A.K., Wilkins, C., and Clayton, N.S. (2022). Could it be proto magic? Deceptive tactics in nonhuman animals resemble magician’s misdirection. Psychol. Conscious. Theory, Res. Pract.
32. Schnell, A.K., Loconsole, M., Garcia-Pelegrin, E., Wilkins, C., and Clayton, N.S. (2021). Jays are sensitive to cognitive illusions. R. Soc. Open Sci. 8, 202358. 10.1098/rsos.202358.
33. Fragaszy, D.M., Visalberghi, E., and Fedigan, L.M. (2004). The complete capuchin: the biology of the genus Cebus (Cambridge University Press).
34. Napier, J.R. (1967). Evolutionary aspects of primate locomotion. Am. J. Phys. Anthropol. 10.1002/ajpa.1330270306.
35. Reynolds, V., Napier, J.R., and Napier, P.H. (1969). A Handbook of Living Primates. Man. 10.2307/2799276.
36. Christel, M.I., and Billard, A. (2002). Comparison between macaques’ and humans’ kinematics of prehension: the role of morphological differences and control mechanisms. Behav. Brain Res. 131, 169–184.
37. Spinozzi, G., Truppa, V., and Lagana, T. (2004). Grasping behavior in tufted capuchin monkeys (Cebus apella): grip types and manual laterality for picking up a small food item. Am. J. Phys. Anthropol. Off. Publ. Am. Assoc. Phys. Anthropol. 125, 30–41.
38. Spinozzi, G., Lagana, T., and Truppa, V. (2007). Hand use by tufted capuchins (Cebus apella) to extract a small food item from a tube: digit movements, hand preference, and performance. Am. J. Primatol. Off. J. Am. Soc. Primatol. 69, 336–352.
39. Costello, M.B., and Fragaszy, D.M. (1988). Prehension in Cebus and Saimiri: I. Grip type and hand preference. Am. J. Primatol. 15, 235–245. 10.1002/ajp.1350150306.
40. Lacreuse, A., and Fragaszy, D.M. (1997). Manual exploratory procedures and asymmetries for a haptic search task: A comparison between capuchins (Cebus apella) and humans. Laterality Asymmetries Body, Brain Cogn. 2, 247–266.
41. Ottoni, E.B., and Izar, P. (2008). Capuchin monkey tool use: overview and implications. Evol. Anthropol. Issues, News, Rev. Issues, News, Rev. 17, 171–178.
42. Fragaszy, D.M., Liu, Q., Wright, B.W., Allen, A., Brown, C.W., and Visalberghi, E. (2013). Wild bearded capuchin monkeys (Sapajus libidinosus) strategically place nuts in a stable position during nut-cracking. PLoS One 8, e56182.
43. De Resende, B.D., Ottoni, E.B., and Fragaszy, D.M. (2008). Ontogeny of manipulative behavior and nut-cracking in young tufted capuchin monkeys (Cebus apella): A Perception-action perspective. Dev. Sci. 11, 828–840. 10.1111/j.1467-7687.2008.00731.x.
44. Fujita, K., Kuroshima, H., and Asai, S. (2003). How do tufted capuchin monkeys (Cebus apella) understand causality involved in tool use? J. Exp. Psychol. Anim. Behav. Process. 29, 233.
45. Ottoni, E.B., and Mannu, M. (2001). Semifree-ranging tufted capuchins (Cebus apella) spontaneously use tools to crack open nuts. Int. J. Primatol. 22, 347–358.
46. Buckmaster, C.L., Hyde, S.A., Parker, K.J., and Lyons, D.M. (2012). Spontaneous tool-use by captive born squirrel monkeys (Saimiri sciureus sciureus). In AMERICAN JOURNAL OF PRIMATOLOGY (WILEY-BLACKWELL 111 RIVER ST, HOBOKEN 07030-5774, NJ USA), p. 39.
47. Schmitt, D. (2003). Evolutionary implications of the unusual walking mechanics of the common marmoset (C. jacchus). Am. J. Phys. Anthropol. Off. Publ. Am. Assoc. Phys. Anthropol. 122, 28–37.
48. Young, J.W., and Chadwell, B.A. (2020). Not all fine-branch locomotion is equal: grasping morphology determines locomotor performance on narrow supports. J. Hum. Evol. 142, 102767.
49. Novikova, M., and Kuznetsov, A. (2017). Palmar flexion creases and finger linkage groups in New World Monkeys — Functional and evolutionary palmistry. Biol. Commun. 62, 181–201. 10.21638/11701/spbu03.2017.304.
50. Tia, B., Takemi, M., Kosugi, A., Castagnola, E., Ansaldo, A., Nakamura, T., Ricci, D., Ushiba, J., Fadiga, L., and Iriki, A. (2017). Cortical control of object‐specific grasp relies on adjustments of both activity and effective connectivity: a common marmoset study. J. Physiol. 595, 7203–7221.
51. Fox, D.M., Mundinano, I.C., and Bourne, J.A. (2019). Prehensile kinematics of the marmoset monkey: Implications for the evolution of visually-guided behaviors. J. Comp. Neurol. 527, 1495–1507. 10.1002/cne.24639.
52. Garcia-Pelegrin, E., Clark, F.E., and Miller, R. (2021). Increasing Animal Cognition Research in Zoos. bioRxiv, 2021.11.24.469897. 10.1101/2021.11.24.469897.
53. Favareau, D., and Favareau, D. (2009). The Clever Hans Phenomenon from an Animal Psychologist’s Point of View. Essent. Readings Biosemiotics Anthol. Comment., 237–255.
