Rhesus monkeys similarly act as if they know what another sees. When given the option between a grape in front of a person who looks at it and a grape in front of a person who looks away, or whose sight is blocked, they consistently prefer the latter—in apparent recognition that “stealing” from someone who is not looking is safer. Chimpanzees in earlier experiments by Povinelli and others might have failed because the tasks involved cooperation, such as a human informing the ape about where food is. That is not a particularly natural situation. Unlike human toddlers, who constantly want to point out things to their parents and others, primates do not seem to have much inclination to inform others. They tend to compete for rewards instead, and so it may not be surprising that we can observe their competence better when tested in competitive situations.
Unfortunately, these newer results, in spite of their substantial publicity, do not actually prove that the primates reasoned about what the human could see. A much leaner explanation is available. It remains possible that the monkeys simply learned that a human facing the grape is more likely to interfere with their attempt at obtaining the treat than one that is not. The same is true of the chimpanzee example. The lower-ranked individual may have learned purely behavioral rules, such as: if a dominant faces the food, it is not safe to approach.
But other results suggest that chimpanzees appreciate not only what another sees but also what another has previously seen. In an extension of Hare and colleagues’ previous experiments, the chimpanzees took into account whether a dominant competitor had observed the hiding of a food item or not. When the placement of food behind a particular screen was witnessed by the dominant, the subordinate was subsequently less likely to approach the food than when it was not witnessed. In another experiment, the dominant saw the hiding but was then replaced with a different dominant animal that did not see the hiding. Again, the subordinate chimpanzee was more likely to approach the treat when the competitor was ignorant than when he was knowledgeable about the food location. So it remains possible that chimpanzees, after all, have some capacity to reason about the minds of others.
Indeed, other studies support the possibility of competence in great apes. Like two-year-old children, some great apes appear to recognize what someone else is trying to achieve, even if the attempt failed. Some results suggest that they can distinguish accidental from purposeful actions. Other experiments suggest that they can discriminate between someone who is unwilling to do something and someone who is unable to do it. One study suggests that they may be able to distinguish appearance from reality, and they seem to be able to take advantage of when a competitor cannot see them.
There is some indication, though scarce, for such capacities in other species. Grey squirrels, for instance, space their caches farther apart when observed by other squirrels, presumably to avoid pilfering. They even preferentially cache while oriented with their backs to other squirrels. Similarly, scrub jays, when in the presence of a potential competitor, preferentially cache food further away, in darker and more occluded areas, than they do otherwise. Hence other species may also take into account what another can or cannot see. Then again, they may not.
Unfortunately, none of these behaviors need imply any reasoning about the mind of the competitor. Squirrels, jays, and apes in all of these cases may behave as they do simply on the basis of observable behaviors. Acting one way gets a reward; acting another way leads to punishment. Recall that to demonstrate an animal is taking the mind of another into account, one has to show attribution of false beliefs. In spite of several clever attempts at this holy grail of theory of mind research, so far no nonhuman animal has passed false-belief tasks. Even chimpanzees that have demonstrated some impressive competence at other components of the task fail when false beliefs need to be taken into account. It is therefore possible to maintain a lean interpretation, as many comparative psychologists do, and conclude that no other animal has anything like a theory of mind.
MY OWN HUNCH, ACTUALLY, IS that the truth lies between the extreme romantic and killjoy positions. The wealth of recent positive data from great apes suggest, though they certainly do not prove, that they have limited understanding of basic mental states. It will be interesting to see whether apes demonstrate any sign of the early (implicit) false-belief understanding that has recently been documented in infants. Researchers are trying to use eye-tracking devices to test chimpanzees on such tasks, but it is not easy. Perhaps results on such tests will become available by the time you read this book.
Given that great apes perform like two-year-old children on a variety of tasks that involve considering more than meets the eye (as discussed in Chapter 3), it would not be surprising if they did so here. They may have a limited, perhaps implicit, understanding of what another sees, believes, knows, attends, desires, and intends, but this possibility should not detract from the remaining gap between the mind reading in humans and the limited reasoning that on current evidence maximally exists in our closest animal relatives. Although Povinelli’s group and Tomasello’s group, the two most influential research laboratories in this field, have been at loggerheads about lean and rich interpretations of data on ape theory of mind, they both agree that there is no sign yet of any false-belief understanding. Thus we have consensus that there is something unique about human theory of mind. Povinelli argues that only humans have a theory of mind, period. Tomasello, Call, and colleagues believe that their findings are more parsimoniously explained by granting great apes some basic mind-reading capacities. Yet they, too, maintain that apes lack the most fundamental human socio-cognitive skills.
In fact, Tomasello and colleagues argue that great apes do not even show the basic social awareness that is typical of human infants when they point, show, or offer things because they want to communicate about them. These authors argue that the main difference is that humans have what they call “shared intentionality.” As we have seen, humans have a fundamental motivation to share their own mental states with others. This inclination allows us to construct a sense of “we” that enables us to collaborate on unheralded flexible scales—socially constructing tools, meals, games, and theories (drawing extensively on language and mental time travel). Infants demonstrate an inclination toward this sharing early, well before they pass false-belief tasks. For instance, when one-year-olds are engaged in a collaborative task with an adult and the adult stops, the infants typically try to get the adult to reengage. Infant chimpanzees, on the other hand, simply try to do the task themselves. Chimpanzees may gesture to get someone to do something for them, but humans often gesture (and talk) just to inform.
Chimpanzees do not point to each other in the wild. This may be the case because it is pointless to point (pardon the pun) if the other chimpanzees do not give you what you want. In experiments, they are poor at using and providing social cues in cooperative tasks requiring communication. For a while experiments suggested that great apes, unlike dogs, could not even understand pointing by humans. More recent work shows that they only struggle when the human points to things that are close to each other; they can discriminate when the options are far apart. Great apes can also learn to point to humans but do so virtually only to request, rather than to declare. (Recall that only some 5 percent of utterances in “language-trained” apes could be categorized as resembling statements or declarations.) Children, on the other hand, constantly want to point out things to share information. My own children are quite insistent that I drop everything and join in the excitement. Humans may be uniquely motivated to establish links between their minds, through reading and telling, that allow us to create common mental worlds of goals, ideas, and beliefs.
Great apes may have some capacity to reason about basic states of mind. Even so, they do not seem to have a particularly strong drive to build connections between their minds, which may significantly limit what cooperation they can mount. A recent large-scale examination of over a hundred great apes found that they performed similarly to two-and-a-half-yea
r-old children on a battery of tasks on physical cognition, but much more poorly on tasks of social cognition.8 Of course, such comparisons depend on what exact tasks are used: social tasks in which apes have to interact with human adults may not be comparable to social tasks in which human infants interact with human adults. Nonetheless, the results add to the mounting evidence that humans’ urge to link with other minds is unique. After over three decades of research, there is no strong evidence that apes understand the representational character of beliefs. It remains possible that they only reason about observables, rather than about minds.
Humans evidently are mind readers. Though we may frequently get it wrong, we can read and tell our minds sufficiently well to enable us to cooperate in a multitude of clever ways. We share ideas, advice, and goals. We can develop intricate plans and collaborate in bringing them to fruition. We teach and learn from each other’s experiences. We set out to entertain each other and care about what others might find funny or pleasing. We come together to share attention in celebrations and performances. Our cultural inheritance, as we will discuss later, can be conceived of as an accumulation of cooperative exchanges between minds over many generations. We reflect on the nature of mind reading. We even spend considerable effort trying to overcome the obstacles associated with establishing the nature of animal minds. Much of this scientific mind measuring has focused on animals’ capacity for smart problem solving. And so we turn now to research on intelligence.
1Daniel Dennett calls this the “intentional stance” and contrasts it with explanations and predictions based on function (“design stance”) or on other forces (“physical stance”).
2For this and some other reasons, congenitally blind children are typically delayed in their development of theory of mind.
3There do appear to be cultural differences, however. For instance, in one study Israeli parents showed much more face-to-face interactions with their five-month-old infants than Palestinian parents, who, in turn, created more affiliation through touch.
4For example, some people believe and others do not believe in telepathy or in a vigilant divine observer of one’s mind.
5Chimpanzee mothers and infants sometimes show some brief bouts of gazing in each other’s eyes. For the most part, however, primates, even mother and offspring, do not spend much time in such gazing.
6An alternative possibility is that the need to understand other species drove the evolution of mind reading. Predators, for instance, could benefit from better prediction of prey behavior and prey species from being better able to predict predators. This may have led to an evolutionary arms race of wits. In support of this theory, note that there is evidence that mammalian predator and prey brain sizes have increased in tandem over geological time.
7Such pursuits can certainly look thoughtful, though. Consider the following example of a male chimpanzee aggressively chasing a female. As the female sought cover behind a tree trunk, the male moved to the left, which prompted the female to move to the right. In full motion, the male then threw a brick in the direction of her rightward path while continuing to move to the left himself. To avoid the projectile the female changed directions, only to find herself caught by the male.
8Note that such a coarse distinction between social and physical cognition is far from clear-cut. Social factors often feature in physical tasks (e.g., when presented by a social other), and social tasks often involve reasoning about physical components.
SEVEN
Smarter Apes
Man is most uniquely human when he turns obstacles into opportunities.
—ERIC HOFFER
HUMANS HAVE OVERCOME MANY OBSTACLES on the way to planetary dominance. We have created light where we could not see. We have created warmth where we were cold. Our smarts have given us tools to do what could not be done before, whether it’s hunting from a distance or curing illness. Increasingly, our technologies give us control over what we care about. We keep finding new solutions to problems and call this progress—all while tacitly ignoring the fact that most of our solutions have spawned new problems of their own, from hard labor to pollution.
We are a long way from creating worldwide utopias with sustainable supplies of what we want. Even basic necessities, such as sufficient food and clean water, remain out of reach for many human communities. But no matter to what embarrassing failures or glorious successes you might point, it is self-evident that humans can be outstandingly resourceful and clever. We survive by our wit. We keep finding ingenious solutions to obstacles and turning them into opportunities.
Other animals also have effective ways to solve problems such as finding food, shelter, and partners. They generally cope well with whatever re-current challenges they face. It could hardly be any other way. Just as their bodies are adapted to their environments, so are their senses, cognitions, and behaviors. On many tasks other species are clearly our superiors. Some sharks, birds, and turtles navigate according to electromagnetic fields we do not even perceive. Bats, dolphins, and some shrews use echolocation to scan their environment. Bees use optic flow to control their flight in such an elegantly simple way that humans are now trying to implement it in aircraft navigation. There is a difference, however, between an individual figuring out a clever solution and mechanisms that are universal to all members of a species. Some clever-looking animal behaviors are hardwired. In these cases new obstacles may remain barriers they cannot turn into opportunities. We saw in Chapter 5 that digger wasps, for instance, always inspect their nest before dragging food inside and are incapable of snapping out of this routine even when an experimenter keeps moving the food away.
Yet some animals demonstrate flexible problem-solving capacities, and this is not restricted to big-brained primates and cetaceans. For example, some Australian crows have figured out a way of eating poisonous cane toads. These toads were introduced to Australia in 1935 and have become a great pest. The birds learned to turn the toads onto their backs and then peck at their harmless bellies. Even invertebrates may have considerable smarts. We have seen that some cephalopods show immense deceptive capacities (and octopuses are invertebrates that enjoy vertebrates for lunch). Whether or not these examples can be explained by lean interpretations, we should not be presumptuous about intelligence only existing in vertebrates, in mammals, or in primates. I suspect that so far comparative psychology has merely scratched the surface in its attempts at documenting animal problem-solving capacities.
Nevertheless, human minds demonstrate intellectual flexibility that appears to be unmatched. What might be unique about our intelligence and creativity? We can start by examining what research has uncovered about the nature of human intelligence.
RESEARCH ON INTELLIGENCE IS IN some sense one of the greatest success stories of the discipline of psychology. Millions of people are given intelligence tests every year. In the Netherlands, for example, for decades virtually all young males have been tested. Intelligence tests originated from two main sources. In England, Sir Francis Galton, a cousin of Charles Darwin, believed that intelligence was a combination of sensory acuity and effort, and he developed various tests to measure these factors. But people became disenchanted with Galton’s approach, as scores on his tests failed to make good predictions about, for example, who would perform well at school. In France, intelligence tests derived from just such pragmatic concerns. The French government asked Alfred Binet to devise an objective test to identify children who would not benefit from normal classroom teaching with their same-aged peers. In response Binet developed a battery of tasks—including measures of memory, general knowledge, problem solving, and so forth—based on his view that intelligence was an aggregate of diverse abilities. He gave his test to many children to establish their competences and calculate the average score for each age group. These averages became his yardstick. If you performed like the average twelve-year-old, your mental age was twelve. Simple.
In 1912 the German psychologist William Stern used mental age to deri
ve the infamous Intelligence Quotient. The IQ was your mental age divided by your actual chronological age times 100. If you are 10 years old and perform like the average 12-year-old, you have an IQ of 120. If you perform like the average 10-year-old, your IQ is 100. This quotient, however, only makes sense for assessing children. If you extend this logic into adulthood, then you would find that everyone would want to make comparisons to the aged. After all, if you perform like the average 90-year-old and are in fact 30 years old, you would have an IQ of 300. Today’s measure of IQ is not really a quotient but your score’s relative position in a standard population distribution.1 Binet’s test was revised in the United States into the Stanford-Binet test, which, together with the Wechsler scale, is still the most widely used intelligence test.
Chances are you have taken one of these tests at some stage in your life. They test your capacity to comprehend and define words; share knowledge of some facts; reason by analogy, deduction, and inference; solve arithmetic problems; repeat strings of digits backwards; put puzzle pieces together; copy a design; identify what is missing from a picture; put picture sequences in order; and translate signs from a novel symbol system. Being good at these things means you are intelligent. At least that is what many intelligence researchers believe, because a boring definition of intelligence is “intelligence is what the tests test” (as E. G. Boring wrote in 1923). This is circular reasoning, of course, but Boring noted strong findings supporting these tests. They unearthed individual differences, and children’s relative rank order tended to be stable even though performance improved with age. Importantly, it has long been found that people who are good at one part of these tests also tend to be good at others. This points to a general intelligence factor, usually called “g,” and this factor has been shown to make reliable predictions about real-life outcomes.
The Gap Page 16
