Theory of Mind - one's ability to understand what others are thinking
Theory of mind is the ability to attribute and reason about the mental states of others. It is often thought of as the pinnacle of social cognition. This paper will describe the Social Brain Hypothesis and the modern anatomy of the prefrontal cortex; as well as an overview of the Theory of Mind literature and studies examining Theory of Mind in adult humans, children, autistic individuals, chimpanzees, elephants and crows. Finally, drawing on the perspective of multiple disciplines including physical anthropology, behavioral ecology and evolutionary psychology, I will investigation of the validity of these theories and their application to non-human animals.
Since Premack and Woodruff first used the term in 1978, research on the subject has mushroomed into a vast literature. Theory of Mind, also referred to as ToM, is a hotly debated topic, especially when it is being used to describe non-human animals. Throughout the literature, the numbers of definitions of ToM have multiplied and popular culture has turned into an umbrella term used to describe many complex mental states, most of which are considered uniquely human, such as empathy, self-consciousness and Machiavellian deceit.
The Social Brain Hypothesis
The social brain hypothesis was proposed by Robin Dunbar and argues that intelligence did not evolve to solve ecological problems, but rather was a means of surviving and reproducing in large, complex social groups. There is a positive correlation between brain size and social group size throughout the animal kingdom, especially within human’s phylogenetic history. Using archaeological evidence from the last 2.5 million years, Gamble and Dunbar usde social brain hypothesis to test the theory that our social lives drove human encephalization. They constructed a map of our evolving social complexity, which concentrates on two resources, materials and emotions. These two resources lie at the basis of all social interaction. They conclude that three temporal movements, or brain size expansions, occurred during our evolution in the Paleolithic 2.6-1.6 Mya, 1.5-0.4 Mya and 300-25 kya (Gamble and Dunbar, 2007).
Behaviors that are adaptive for living in complex societies are reciprocal altruism, deception, and coalition formation. Altruism is behavior that benefits the recipient at the cost of the actor, in term of reproductive fitness. Deception in apes’ tends to be Machiavellian in nature, or used for political gain. These group dynamics relate to ToM because it is imperative to understand the thoughts, emotions and intentions of others, especially when trying to navigate a hierarchical political system (Byrne and Whiten, 1988).
Many human complex cognitive behaviors and ToM abilities are made possible by our huge prefrontal cortex. This area of the brain controls executive functioning, personality expression, planning, decision-making and moderating correct social behavior. Ten years ago there was little knowledge on the functions of the prefrontal cortex, but today much more is known due to functional imaging technology (Shallice, 2001). Functional imaging technology has enabling the processing of information by different centers in the brain to be visualized directly.
Many studies were conducted on a variety of different types of tasks demonstrating that there are significant differences in activation of complex patterns across prefrontal regions. These showed that a number of processes, not previously thought to be strongly related to frontal functions, now seems to involve five, if not six, anterior regions. This suggests that a large number of different types of subprocesses are frontally localized. However, this information has not led to much closure on the nature of the individual processes involved. There are two problems involved with interpreting this information; first is that the subprocesses involved may be too abstract to map simply onto the perceptual input or motor output. Secondly, the tasks that activate the prefrontal cortex can often involve a number of subprocesses and there may be no simple observable manifestations of the successful completion of any one stage in normal performance (Shallice, 2002).
ToM in Children
Human behavior has an enormous range, however, children with normal cognitive abilities begin to exhibit ToM at about age 4. At this point, children demonstrate the understanding that the mind is a representational system, which does not simply reflect reality (Schlinger, 2009). This is proven by many mental tests such as the false-belief task and Piagets A-not-B test. The false-belief task is one of the classic methods in the study of cognitive development, first described by Wimmer and Perner in 1983. The test presents a child with 2 characters, Sally and Anne. First, Sally puts a ball in her basket before departing the scene. In her absence, another character Anne removes the ball and places it in a box. Children are asked to predict, upon Sally’s return, where Sally will look for the object. Normally children over the age of 4 correctly answer IN THE BASKET, but children without a false-belief ability will guess the box because they do not understand that they possess information that Sally does not. One critique of this measure is in a 2000 article by Bloom and German. They argue that the false-belief task should be abandoned as a test of ToM for two reasons. First, passing it requires abilities other than ToM and second, ToM does not necessarily entail the ability to reason about false beliefs.
In general, children who do well in Piaget’s A-not-B problem and false-belief task, do so probably because they have observed adults and other children correctly respond to to such problems and have been reinforced for responding correctly in similar situations (Schlinger 2009). One study found that a child’s conversational experience was directly correlated with success on a variety of ToM tasks, including the false-belief task. Also teaching children to explain other’s behavior in terms of their mental states may account for the children’s ability to predict the behavior of others (Schlinger 2009).
It was Baron-Cohen in 1985 that linked Autism with a deficit of ToM. Autism spectrum disorder is often characterized by abnormal behavior and development and is diagnosed in children 18 months to 3 years old. These children typically have difficulties with social interactions, communication, adjusting to change, feeling empathy and imitation of others. Individuals with Asperger's syndrome, a mild form of autism, perform well on simpler ToM tests but show deficits on more developmentally advanced ToM tests (Stone & Baron-Cohen 1998). Only very extreme cases would be characterized as completely lacking a ToM, but these examples point out the continuum that exists within complex social functions, even in humans.
One study regarding prefrontal cortex activity showed that bilateral orbito-frontal lesion patients performed similarly to individuals with Asperger's syndrome, performing well on simpler tests and showing deficits on tasks requiring more subtle social reasoning, such as the ability to recognize a faux pas. In contrast, no specific ToM deficits were evident in the unilateral dorsolateral frontal lesion patients. The dorsolateral lesion patients had difficulty only on versions of the tasks that placed demands on working memory (Stone &Baron-Cohen 1998).
One Baron-Cohen study focused on the manifestations of a basic metarepresentational capacity including a ‘theory of mind’. Believing that autistic children lack such a ‘theory’, would cause them to be unable to impute beliefs to others and to predict their behavior. This hypothesis was tested using Wimmer and Perner's puppet play paradigm. Normal children and those with Down's syndrome were used as controls for a group of autistic children. Even though the mental age of the autistic children was higher than that of the controls, they alone failed to impute beliefs to others. Thus the dysfunction they have postulated and demonstrated is independent of mental retardation and specific to autism (Baron-Cohen 1985).
The first animal to be described as having a ToM was the Chimpanzee (Pan troglodytes). They are our closest living relatives whose evolutionary branch diverged from ours only six to eight million years ago. They have been said to be capably of insight, problem solving, and use of symbols including American Sign Language, deception and a ‘theory of mind’.
Premack and Woodruff’s 1978 study described the behavior of their sign language trained chimpanzee Sarah. Sarah was shown video sequences of an experimenter in various predicaments. After each sequence, Sarah was presented with a number of photograph showing possible solutions to the problems. Sarah was highly accurate in choosing the appropriate photo in which the experimenter solved the problem correctly. Premack and Woodruff argued that therefore she understood the actors’ intentions. Ever since this interesting study, the capacity for non-human animals to attribute mental states to others has been put to the test by the brightest brains of both humans and animals.
Hare and colleagues conducted the most convincing study of ToM in Chimps, which exploited their competitive nature. This experiment was the observation of a subordinate and dominant individual in competition for two food resources. If the subordinate individual knew that she was the only one who knew the location of one of the food items, she always went for that one first. Subordinate individuals also could keep track of which dominant individuals had knowledge about hidden food items. If a subordinate individual knew about a hidden food item that other conspecifics did not, they often pretended nothing exciting is going on when the group first enters the enclosure. Then when the other group members were distracted, she would sneakily procure the reward. These tests display the importance of deception and perspective taking within chimp social behavior. The subordinate individual knew that if she immediately alerts the whole group of the food, a dominant individual would immediately take it, and she would be left with nothing. Hare et all also took this as reason to conclude that chimpanzees are not only aware of what others can see, but also what conspecifics do and do not know. This element of intentionality is a subject that is also observed in other animals such as elephants and corvids. Intentionality is implied to be of high importance for ToM abilities.
Elephants have the largest brain of terrestrial mammals. Century of myth, folklore and observation have led to the general belief that they “never forget” and are highly intelligent and social. They have been known to engage in very “human-like” behavior such as mourning their dead, recognizing themselves in a mirror and tool use abilities. Many of these behaviors make us seriously question their level of self-consciousness and their understanding of other’s intention and desires.
Some tests done on elephants regarding their cognitive capacity try to determine whether they know that people can and cannot see. Two Asian elephants from the Detroit Zoo were trained to use a begging gesture (or pointing with their trunk for 2 seconds) to identify which of two trainers was “looking at them” rather than standing with a bucket on their head, turning around or lying down and turning around. They were trained using positive reinforcement, by rewarding them if they pointed to the correct trainer or ignored if they didn’t, they were correct 9/10 times by the first pre-training session. In the end results, the elephants were correct 69% of the time, compared to Povinelli’s similar study on chimpanzees who were correct 54% of the time.
The tests used to determine whether chimpanzees and elephants “know that a person can see” have some faults. Gomez believes that the prolonged training of subjects in artificial tasks is a mistaken approach. Instead research should be focused on observing natural behavior that provides evidence of high cognitive function. A major problem with these studies in animals is the fact that ToM is very hard to test for; often these attributes we observe have multiple factors and evolutionary advantages. Some tests don’t necessarily require ToM rather they focus on human interaction, which is not helpful for understanding their natural behavior.
Surprising to the interdisciplinary study of animal behavior and cognitive function is the similarity between primates and corvids (Corvidae). Emery argues that certain aspects of corvid and parrot sociology, neurobiology, and life history are prerequisites for intelligence in birds, as they appear to be for primates. In line with the social brain hypothesis, commonalities between corvids and primates are their brain size, the manufacture special tools to acquire unobtainable foods, adaptation to extreme ecological environments, and extended developmental period before becoming nutritionally independent (Emery, 2006).
Corvids are among the most social birds. For example, in Florida Scrub-Jays, several closely related family members share the responsibility of raising young. Corvids live from pole to pole, thriving in the freezing tundra of Siberia to the extreme heat of the Sahara Desert. Corvids are omnivorous, generalist foragers. Corvid and Parrot forebrains (the nidopallium) are very large for their body size. It was once thought that the avian cerebrum was derived from the basal ganglia, but it is now known that it is derived from the pallium, which is like the mammalian neocortex (Emery 2006). Another remarkable aspect of the avian brain is the hippocampus (important for special memory). There is a relationship between the hippocampus and food-storing, this relationship is strongest in those species who hide and recover large amounts of food items over long periods of time, such as corvids. The social context of caching behavior may be viewed as an arms race between storers and pilferers, in which storers use counter strategies to minimize the risk of having their catches pilfered (Emery, 2006). Some corvids return to caches they had hidden in the presence of conspecifics, and readily recache them in new places unbeknown to the potential thief. This kind of behavior suggests that Crows also know what conspecifics can see and know; storers may understand the visual perspectives of the observer.
Both corvids and primates not only make a range of tools for different uses, but both species exhibit different tool use in geographically isolated areas, showing cultural variation. Crows actually display social learning from individual to individual and from on generation to the next. Recently, crows shocked us once again when they were found using multiple tools in the correct sequence to achieve a goal with no explicit training. Emery suggests that corvids and apes may represent a case for convergent mental evolution, but with divergent brains.
ToM is extremely complicated to measure and understand in humans, let alone other animals. It is not only dependent on a highly developed prefrontal cortex, but is arguably impossible to achieve fully without language. When modern humans appeared about 200,000 years ago it is believed that they possessed mental faculties similar to our own and that they used language. Earlier than that, one million years ago our ancestors were crafting hand-axes and manipulating fire and it is almost certain that these ancient people used some language. This was the most important derived trait of human evolution that catapulted our rapid development of technology and cultural spread of knowledge.
Language is a behavior that conveys conscious thought 10 times faster than any other mode of communication. It is also believed that the tool that allows us to think abstractly and imagine events that happened in the past or that could possibly happen in the future are dependent on language. Many studies have been done to test the influence of language on Theory of Mind in humans. These tests confirm that narrative discourse, language acquisition and training are important for young children to succeed on theory of mind tests. Autistic children with older siblings often score higher on a ToM tests than only children with autism.
One of the most amazing developments in non-human primate study was teaching chimps to communicate with humans through the use of sign language. When this happened in the sixties there was an excited optimism that this would result in proof that other apes are capable of manipulating symbols to communicate abstract ideas. However, decades later an enormous amount of research has shown that although chimps are able to learn an impressive vocabulary, signing of chimpanzees in interactions with humans is usually acquisitive in motivation, mostly limited to object and action signs, and lacking in semantic or syntactic structure in sequence of signs (Rivas, 2005). The fact that other apes lack the ability to use symbols semantically or syntactically shows that they cannot utilize the rules and principles that govern the sentence structure of any individual language. This prevents them from using symbols and signs in the same ways that humans innately do, by using grammar and syntax to express new concepts and make abstract inferences.
Problems with ToM
We can see that shifting gears from ToM in humans and children to other animals can be very complex because of multiple definitions, apparent lack of it in some human beings and difficulty with testing in other animals. A major problem with asking ToM is the fact that a large majority of studies on the intelligence of non-human animals have focused on how closely the intelligence of that species resembles human intelligence. There is a flaw in measuring non-human characteristics by human standards. We are tempted to anthropomorphize apes because they remind us of ourselves. We are not so inclined to ask whether or not Crows possess a ToM. Even though they display very similar abilities. On top of that, discussions of ToM focus almost exclusively on inferred cognitive structures and processes and shed little light on the actual behaviors involved. Schlinger suggests that the behavioral relations referred to by ToM scholars can be described and explained parsimoniously within theoretical framework of behavior analysis. (Schlinger, 2009).
When referring to ToM we should, “appeal to parsimony, suggesting that a variety of complex behaviors in non-humans can be explained more in terms of basic learning processes” (Seyfarth and Cheney, 2000). Occam's razor, or the Law of Parsimony, is a principle that suggests when selecting from competing hypotheses the one that makes the fewest new assumptions is usually the “best” until we can trade simplicity for increased explanatory power. This principle is used as a heuristic guide to aid scientists in the development of theoretical models. Occam's razor is not considered an irrefutable principle of logic, however it helps sort between competing explanations and can weed out hypothesis that are too general and make too many assumptions.
A critique by Heyes maintains that empirical studies of imitation, self-recognition, social-relationships, deception and perspective taking, in every case where non-human behavior has been interpreted as ToM could have occurred by chance or processes such as associate learning (Heyes, 1988). Other critiques, including my own, suggest that the search for ToM in animals is misguided in principal. Instead behavioral ecology, evolutionary psychology and physical anthropology should focus on more parsimonious explanations that help us gain a better understanding of other animals’ true cognitive abilities.
In reply to Hayes, Baum sharply wrote: “The problem is not empirical but logical…. “Does the chimpanzee have a theory of mind?” is neither intriguing nor important. It begs the question of whether it is helpful for a science of behavior to use like “theory of mind” at all…. One must consider whether such a concept advances our understanding of behavior at all” (Schlinger, 2009). On the other hand, Tomasello and Call argue that “in a broad construal of the phrase ‘theory of mind’, then, the answer to Premack and Woodruff’s question of 30 years ago is a definite yes; chimpanzees do have a theory of mind. But chimpanzees probably do not under-stand others in terms of a fully human-like belief–desire psychology in which they appreciate that others have mental representations of the world that drive their actions even when those do not correspond to reality” (Call and Tomasello, 1997). I believe the solution to this problem is to stop asking whether other animals have ToM, develop a more comprehensive understanding of the implications of the Social Brain Hypothesis, utilize functional imaging technology on a wider range of species, and conduct observations in natural setting that don’t anthropomorphize our animal subjects. The scholarly literature could benefit from an influence from a wider range of perspectives, especially those from outside the Judeo-Christian, or Western ideology, which have a fresh interpretation of animal minds and capabilities.
Here is an example of a theory of mind experiment done with young children. I found it interesting.
Baron-Cohen S, Leslie AM, Frith U. (1985) Does the autistic child have a "theory of
mind"? Cognition. 21(1):37-46. PubMed PMID: 2934210.
Byrne, R. W. & Whiten, A (1988) Machiavellian Intelligence: social expertise and the evolution of intellect in monkeys, apes and humans. Oxford, UK: Clardon Press.
Emery, N. J. (2005). Cognitive ornithology: the evolution of avian intelligence. Philosophical Transactions of The Royal Society, 361, 23-43. doi: 10.1098/rstb.2005.1736
Dunbar R. (1992) Neocortex size as a constraint on group size in primates. Journal of Human Evolution. 20, 469-493)
Dunbar R. (1998) The Social Brain Hypothseis. Evolutionary Anthropology. 6, 178-190.
Gamble, C., Gowlett, J., & Dunbar, R. (2011). The social brain and the shape of the paleolithic. Cambridge Archaeological Journal, 21, 115-135.
Hare, B, Call, J, & Tomasello, M. (2001). Do chimpanzees know what conspecifics know?. Animal behaviour, 61(1), 139-151.
Heyes, C. M. (1988) Theory of Mind in nonhuman primates. Behav. Brain Science 21, 101-148.
Penn, D. C., & Povinelli, D. J. (2007). On the lack of evidence that non-human animals possess anything remotely resembling a 'theory of mind'. Philosophical Transactions of The Royal Society, 362, 731-744.
Premack, D., & Woodruff, G. (1978) Does the chimpanzee have a “theory of mind”? Behavioral and Brain Sciences, 4, 515-526.
Rivas, E. (2005). Recent use of signs by chimpanzee (pan troglodytes) in interactions with humans. Journal of Comparative Psychology, 119(4), 404-417.
Schlinger, H. D. (2009) Theory of Mind: an iverview and behavioral perspective. The Psychological Record. (59) 435-448.
Seyfarth, R. M., & Cheney, D. L. (2000) Social awareness in the monkey. American Zoologist, (40) 902-909.
Shallice, T. (2001) ‘Theory of Mind’ and the prefrontal cortex. Brain. 101-14124(2): 247-248.
Tomasello, M. &Call, J. (1997) Primate cognition. New York: Oxford University Press.