10 Page Paper 2 Electric Boogaloo
10 Page Paper 2 Electric Boogaloo
10 Page Paper 2 Electric Boogaloo
Eli Traxler
December 7, 2021
Theory of Mind in Macaques; What We Can Say So Far
The many divides between humans and non-human animals have been the subject of
much debate for about as long as humans have been studying animal behavior. It's been long
agreed that there is something that separates humans from the rest of the animal kingdom,
some sort of difference in cognition. However, there has always been the question of how
exactly to define or quantify this difference. Science has brought us far since the days when it
was commonly accepted knowledge that animals were not conscious and did not feel pain, but
more knowledge tends to open up more questions, and this topic is no exception. In 1978,
Premack and Woodruff published a paper titled "Does the chimpanzee have a theory of mind?"
and that question continues to be a hot topic of research and discussion to this day. But what
exactly is a theory of mind? How can it be defined and measured? And perhaps the biggest
question that so many researchers have been struggling to answer, which species exactly can
Theory of mind is, naturally, something that arises from our categorization of different
an organism with zero consciousness may respond to stimuli, but it does not know that it is
responding. It is simply responding in preprogrammed ways, much like a computer. The first
order of consciousness involves being able to think and know and hold beliefs in your own mind,
but being unable to perceive that there are minds different from your own. Most animals were
held to have zero consciousness at some point, but more recently many scientists consider it
plausible if not likely that many animals do have first order consciousness. This is also the level
of consciousness that human infants have until they reach around the age of 4 (Wellman et al.,
2003). However, beyond that point, most humans display what we call second order
consciousness. Second order consciousness involves everything that first order does, with the
addition of being able to understand that other organisms may also have knowledge and beliefs
of their own, differing from yours. It also involves some level of being able to predict what
knowledge and beliefs others may have, based on their own perception of the world that is
fundamentally different from yours. This is what we call theory of mind; if an animal has a theory
of mind, it is able to theorize about the minds of others as entirely different entities from its own
mind.
This is quite easy to measure in humans, as we have language to express such abstract
concepts, and at the age of 4 human children can tell you that another entity might have a false
belief about something, despite the child knowing full well what the truth is (Wellman et al.,
2003). When it comes to non-human animals though, testing for theory of mind is quite difficult
and has spawned a number of different iterations of experiments. One of the most influential of
these is based on that capacity of a human child to discern that others might have false beliefs.
This experiment has been produced in nearly every form one could possibly think of, but the
an item, often food. A participant (usually a human, but not always) will have seen the item in
one location, but isn't aware of its final location. Then, some method such as recording gaze
location or looking time is used to evaluate if the observing animal expects the participant to
search for the object in its actual location, or the alternate location which the observing animal
knows is the final location, but the participant doesn't. If the animal expects the participant to
search in the incorrect location, this is considered a demonstration of the fact that the animal
understands that the participant has a false belief, thus demonstrating theory of mind.
This sort of testing has been done on many species, but non-human primates are of
particular interest when it comes to research on animal consciousness. Due to their evolutionary
closeness to humans, great apes have received a lot of attention in this field of experimentation,
but some researchers have turned their attention to monkeys as well, and many macaque
species in particular. A small island named Cayo Santiago, near Puerto Rico, is home to a
sizable colony of rhesus macaques that are thoroughly habituated to humans, and frequent test
subjects (Drayton et al., 2014). Much experimentation has been conducted to test the cognition
of these macaques, including experiments similar to the false belief task. One such variation is
held to be more relevant socially to the macaques, as it uses the pressures of dominance and
food guarding to motivate them. Two human participants present a grape to the subject, leaving
it on the ground in front of them, and one turns away while the other keeps the grape in their
view. Due to the human's presumed dominance, to demonstrate theory of mind the subject
ought to only take the grape that isn't being watched over by a human, because it knows the
human will be able to see it. The rhesus macaques on Cayo Santiago have displayed this
mind (Flombaum et al., 2005). One concern that has been raised about this experimental format
is that it's possible the macaques are more likely to take from the less threatening human,
without actually representing what the human can perceive visually. However, this was also
tested on Cayo Santiago, in an experiment very similar to the first except for the fact that both
humans turned their side to the subject, and one placed the grape in front of themself while the
other placed the grape behind themself. Under this setup, the macaques continued to prefer the
grape that wasn't directly in the human's visual field (Drayton et al., 2014).
A very similar variation of this experiment has also been conducted on long-tailed
macaques in captivity, in which another macaque is used as the participant (Overduin-de Vries
et al., 2014). Again, two food items are made visible to a macaque subject who is contained on
one side of the enclosure with the food, but only one of them is visible to another, more
dominant macaque contained on the other side. Then, both macaques are released into the
central enclosure. In consistency with previous experiments, it was found that the subject, the
more submissive macaque, preferred the food item that was not visible to the more dominant
macaque. In some trials, the subject was able to obtain both food items, and in these the
subject obtained the visible food first. The subject's success rate at obtaining both food items
was related to how fast they approached the first, visible item. These results suggest that not
only do macaques prefer to take food that is hidden to others, but that they may be aware that
the dominant macaque will only try to take the visible food, since some subjects were able to
use this to their advantage by rushing for the visible food first. Once the visible food was gone,
they were free to take the other, invisible food unhindered (Overduin-de Vries et al., 2014).
The highly visual nature of many primates, including macaques, has been heavily
influential in these studies, but researchers have also adapted this format to an audio-based
form for further testing. One such experiment, conducted on the Cayo Santiago rhesus
macaques, involves different means to get at food that create different noise levels (Santos et
al., 2006). Two grapes are placed into two plastic containers, both with bells attached to the
outside to make them visually identical. One container has the ringers removed from its bells,
rendering its operation silent, while the other container's bells are intact, making any handling of
it impossible without a significant amount of noise. The grapes are placed into the containers
while the subject is paying attention to the participant, so they can hear which container is
which. Then the participant turns away from both containers, and the macaque may choose
one. In the experiments conducted in this manner on Cayo Santiago, the macaques preferred to
handle the silent container over the noisy one. Additionally, to confirm that this behavior was
based solely on the subject's knowledge of how auditory cues would affect a participant, a follow
up experiment was conducted in which all components were repeated, except the participant
remained looking at the containers. Under these conditions, subjects did not show any
preference towards the silent container, and in fact more of them approached the noisy
container instead of the silent one. This evidence suggests that the macaques are aware of the
human participant's ability to hear the containers while turned away, while simultaneously
knowing that if the participant sees both containers, it doesn't matter if they are heard because
These experiments provide strong evidence that rhesus macaques are able to represent
what others can see, at least to the extent that they are able to use it for their own ends. This is
a very important component of theory of mind, however, it does not establish theory of mind in
these macaques all by itself. One of the biggest difficulties in theory of mind research is,
ironically, determining what concepts and thoughts the subject is actually producing in its mind.
While rhesus macaques have been demonstrated to predict the perceptions and behaviors of
others, and understand that others perceive the world differently, that doesn't necessarily mean
they are consciously understanding that the other is experiencing a state of mind just as they
are. Without reliable language, it's impossible to determine exactly what a monkey is
representing in its mind, no matter how inherently good human theory of mind might be.
However, as discussed previously, showing that a subject knows that another entity can have a
know that another entity can have a false belief, theoretically, one must be able to understand
False belief tasks have been conducted on Cayo Santiago with the rhesus macaque
population as well, using looking times to record whether an event is expected or unexpected to
a subject. A two part experiment was conducted, the first part simply testing the subject's ability
to predict what a presenter will do based on a true belief, and the second part bringing in false
beliefs (Marticorena et al., 2011). In the first part of the experiment, both the presenter and the
subject watch an item move from one box to another, and then the presenter reaches into one
of the two boxes. After this action, the amount of time that the subject looks at the presentation
is recorded. It was found that when the presenter reached into the incorrect box, the macaques
consistently looked longer than when they reached into the correct box. This is consistent with
previous evidence that macaques are able to predict the behavior of others based on their
perceptions. However, the second part of this experiment takes it a step further, by including a
barrier to block the presenter's vision as the item moves from one box to the other. Once again
the presenter reaches into one of the two boxes, and the looking time of the subject is recorded.
When this particular version of the experiment was conducted on the Cayo Santiago macaques,
however, they did not display a complete understanding of false beliefs. As before, the
macaques expected the presenter to look in the correct location when the presenter had seen
the item move. But when the presenter had a false belief, the macaques looked for similar
amounts of time regardless of where they searched. This seems to signify that they do
understand that the presenter will behave differently when they have a false belief, but they are
unable to predict exactly what that false belief is (Marticorena et al., 2011). After this study was
conducted, some concerns were raised that the task might simply be too complex and long to
keep the subject's attention, so a similar but slightly streamlined version of the task was tested
on the Cayo Santiago macaques. However, this experiment led to similar results as the first
The positive evidence that has been collected thus far on the representation of the
knowledge of others in various macaque species is too great to ignore; it's certainly far from
conclusive, but also denotes at least some ability to understand that others perceive the world
differently. This is logically sound from an evolutionary view as well, as the ability to predict the
behavior of others in their group could lend a significant advantage in a competitive context.
This is upheld by the overwhelming success with testing macaques and other primates in ways
that mimic natural competitive environments, compared to prior, less ecologically refined
research (Lyons et al., 2006). But thus far, their performance on tasks specifically evaluating
their understanding of false beliefs doesn't hold up quite as well. This is an important piece of
information when trying to determine what level of theory of mind, if any, these primates have,
as false belief tasks are currently something of a gold standard in theory of mind research. It's
also important to consider the fact that this is a relatively young area of research, and much
progress has been made in just the past decade. It's entirely possible that someone will come
up with a new method of evaluating theory of mind without language that renders false belief
tasks obsolete. However, at this time, there is simply not enough positive evidence to say that
aspect has been very strongly emphasized. The neurological and biochemical mechanisms,
however, are a potential source of comparisons that have been overlooked with this recent
same way that is possible with experimentation that revolves around biological mechanisms.
Research on theory of mind is often highly controversial for this reason, because it is so difficult
to establish beyond a reasonable doubt that an animal is thinking in a certain way simply by
watching its behavior. By first studying the mechanisms that allow certain cognitive functions, it
is theoretically possible to bypass the controversy over incomplete evidence and gain a better
overall picture of what cognitive concepts like theory of mind even are (de Waal et al., 2010).
While as yet not very well explored, there is some research in monkeys on exactly what parts of
the brain could be involved in this type of function, and how comparable they are to humans.
One experiment, that may establish a precursor to theory of mind, examines the medial
frontal cortex and how parts of it are activated while observing others. In the experiment, two
monkeys take turns watching each other complete a small task for a reward, while the activity of
neurons in each of their medial frontal cortices are monitored. It was found that for each subject,
many of these neurons were activated while watching their partner perform the task but not
while performing it themself (Yoshida et al., 2011). While a small step, this does suggest the
plausibility of specific regions of the brain dedicated to representing the actions of others.
Additionally, this is consistent with findings in humans that the medial frontal cortex is critical for
social cognition and tasks requiring theory of mind, showing that there may be an analogous
comparison to be made between humans and other animals (Amodio et al., 2006). Another
experiment on this brain region specifically looks at how inhibitory drugs meant to silence the
medial prefrontal cortex affect macaques (Hayashi et al., 2020). Typical false belief task videos
are shown to the subject while their gaze location is recorded, first without inhibitory drugs and
secondly while under the effects of inhibitory drugs. In this particular study it was found that
without the drugs, the subjects did anticipate where the presenter would look when they had a
false belief. This is a very interesting inconsistency with earlier studies on macaques and raises
the question of whether gaze location or some other factor is a better way to demonstrate theory
of mind behaviorally, but more research is required to fully answer that. When the experiment
was repeated while the subjects were affected by the medial prefrontal cortex silencing drug,
however, it was found that they no longer anticipated the presenter's false belief. They
continued to be able to track objects through their own visual perceptions and memory, but were
not able to represent the knowledge of the presenter (Hayashi et al., 2020). This provides
further evidence that the medial frontal and prefrontal cortices are key structures, at the very
least, to the ability to represent the knowledge and actions of others, if not a full theory of mind.
While this more biological evidence is hardly conclusive, neither is the behavioral
evidence, and both open up the possibility for plenty of further questions and experimentation.
Theory of mind research has begun to advance rapidly in recent years, and the next few
decades of research will likely open up even more knowledge. For macaques, as of yet, it is
impossible to say for certain that they do possess a human-like theory of mind. However, the
evidence that does exist strongly points to at least some cognitive functioning that can be
And the evidence gathered on the involvement of brain regions in theory of mind shows that a
structural similarity exists here between humans and macaques. So while macaques may not
be proven to have a theory of mind like us, it would not be inaccurate to say that they
Wellman, H. M., Cross, D., & Watson, J. (2003). Meta-Analysis of Theory-of-Mind Development:
The Truth about False Belief. The Society for Research in Child Development, 72(3),
655-684. https://doi-org.proxy.lib.umich.edu/10.1111/1467-8624.00304
Drayton, L. A, & Santos, L. R. (2014). A decade of theory of mind research on cayo santiago:
Insights into rhesus macaque social cognition. American Journal of Primatology, 78(1),
106-116. https://doi-org.proxy.lib.umich.edu/10.1002/ajp.22362
Flombaum, J. I., & Santos, L. R. (2005). Rhesus Monkeys Attribute Perceptions to Others.
https://doi-org.proxy.lib.umich.edu/10.1016/j.cub.2004.12.076
Overduin-de Vries, A. M., Spruijt, B. M., & Sterck, E. H. M. (2014). Long-tailed macaques
https://doi-org.proxy.lib.umich.edu/10.1007/s10071-013-0639-6
Santos, L. R., Nissen, A. G., & Ferrugia, J. A. (2006). Rhesus monkeys, Macaca mulatta, know
what others can and cannot hear. Animal Behavior, 71(5), 1175-1181.
https://doi-org.proxy.lib.umich.edu/10.1016/j.anbehav.2005.10.007
Marticorena, D. C. W., Ruiz, A. M., Mukerji, C., Goddu, A., & Santos, L. R. (2011). Monkeys
represent others’ knowledge but not their beliefs. Developmental Science, 14(6),
1406-1416. https://doi-org.proxy.lib.umich.edu/10.1111/j.1467-7687.2011.01085.x
Martin, A., & Santos, L. R. (2014). The origins of belief representation: Monkeys fail to
https://doi-org.proxy.lib.umich.edu/10.1016/j.cognition.2013.11.016
Lyons, D. E., & Santos, L. R. (2006). Ecology, Domain Specificity, and the Origins of Theory of
https://doi-org.proxy.lib.umich.edu/10.1111/j.1747-9991.2006.00032.x
de Waal, F. B. M., & Ferrari, P. F. (2010). Towards a bottom-up perspective on animal and
https://doi-org.proxy.lib.umich.edu/10.1016/j.tics.2010.03.003
Yoshida, K., Saito, N., Iriki, A., & Isoda, M. (2011). Representation of Others' Action by Neurons
https://doi-org.proxy.lib.umich.edu/10.1016/j.cub.2011.01.004
Amodio, D. M., & Frith, C. D. (2006). Meeting of minds: the medial frontal cortex and social
https://doi-org.proxy.lib.umich.edu/10.1038/nrn1884
Hayashi, T., Akikawa, R., Kawasaki, K., Egawa, J., Minamimoto, T., Kobayashi, K., Kato, S.,
Hori, Y., Nagai, Y., Iijima, A., Someya, T., & Hasegawa, I. (2020). Macaques Exhibit
Implicit Gaze Bias Anticipating Others’ False-Belief-Driven Actions via Medial Prefrontal
https://doi-org.proxy.lib.umich.edu/10.1016/j.celrep.2020.03.013