by James Geary
Studies of adults show that even when presented with random stimuli, and explicitly informed that the stimuli are random, people still claim to be able to find patterns in the sequences. The brain is so fanatical about pattern that it will gladly generate patterns even where none exist. Look at the figure below.
There are no overlapping triangles in this image; the brain’s pattern recognition circuits create them. Image courtesy of Adam Somlai-Fischer, Prezi.com.
Three wayward Pac men and three pointy brackets are all that is actually printed on the page. What we see, however, are two overlapping triangles. Where patterns are incomplete, or even nonexistent, the brain is happy to fill in the blanks. Some researchers suggest that there is even a specific brain module, called the “interpreter66,” that is tasked with sifting out patterns from the slurry of information continuously flowing through our skulls.
From an evolutionary perspective, pattern recognition is essential. The brain’s pattern recognition circuits take raw data from the senses, sort through it for apparent patterns, and use those patterns to determine a response. The ability to accurately predict the future based on recurring patterns is crucial to everything from hunting and gathering (snakes with cylindrical heads are mostly harmless; snakes with triangular heads could kill me) to mate selection (animals with the most symmetrical, or highly patterned, features tend to be the fittest, and hence the most popular sexual partners).
Once a pattern has repeated itself long enough, it starts to influence behavior. Because lush meadows were found to reliably surround lakes and streams, for example, our brains came to associate green grass with fresh water. Using this type of analogical reasoning, we also began comparing situations to decide whether a new object or environment was sufficiently like a previous object or environment to ensure a steady supply of food and water. Survival depends on having the fittest pattern recognition circuits.
The brain contains some 100 billion neurons, each of which is linked to tens of thousands of other neurons through a vast network of synapses. A fleck of brain tissue roughly the size of a match head67 sports about a billion connections, according to neurobiologist and Nobel Prize–winner Gerald Edelman. He estimates that the number of possible connections in the typical human brain totals somewhere in the region of ten followed by millions of zeroes, a number far in excess of the number of particles in the known universe.
In Second Nature: Brain Science and Human Knowledge, Edelman theorizes that the human brain’s astonishing interconnectivity produces consciousness and, because of the astronomical number of associations our brains are capable of making, pattern recognition is the basis not just for metaphorical thinking but for all thinking:
Brains operate68 . . . not by logic but by pattern recognition. This process is not precise, as is logic and mathematics. Instead, it trades off specificity and precision, if necessary, to increase its range. It is likely, for example, that early human thought proceeded by metaphor, which, even with the late acquisition of precise means such as logic and mathematical thought, continues to be a major source of imagination and creativity in adult life.
Pattern recognition is so basic that the brain’s pattern detection modules and its reward circuitry became inextricably linked. Whenever we successfully detect a pattern—or think we detect a pattern—the neurotransmitters responsible for sensations of pleasure squirt through our brains. If a pattern has repeated often enough and successfully enough in the past, the neurotransmitter release occurs in response to the mere presence of suggestive cues, long before the expected outcome of that pattern actually occurs. Like the study participants who reported seeing regular sequences in random stimuli, we will use almost any pretext to get our pattern recognition kicks.
Pattern recognition is the most primitive form of analogical reasoning, part of the neural circuitry for metaphor. Monkeys, rodents, and birds recognize patterns, too. What distinguishes humans from other species, though, is that we have elevated pattern recognition to an art. “To understand,” the philosopher Isaiah Berlin observed, “is to perceive patterns.”
Metaphor, however, is not the mere detection of patterns; it is the creation of patterns, too. When Robert Frost wrote,
A bank is a place where they lend you an umbrella in fair weather and ask for it back when it begins to rain
his brain created a pattern connecting umbrellas to banks, a pattern retraced every time someone else reads this sentence. Frost believed passionately that an understanding of metaphor was essential not just to survival in university literature courses but also to survival in daily life.
In “Education by Poetry69,” a lecture delivered at Amherst College in 1930, Frost said, “I have wanted in late years70 to go further and further in making metaphor the whole of thinking.” He argued that, without a proper education in metaphor, students could not examine and evaluate the claims made by historians or scientists, newspaper editorialists or political campaigners. People “don’t know when they are being fooled by a metaphor,” he warned:
Unless you are at home in the metaphor, unless you have had your proper poetical education in the metaphor, you are not safe anywhere. Because you are not at ease with figurative values: You don’t know the metaphor in its strength and its weakness. You don’t know how far you may expect to ride it and when it may break down.
Knowing how far to ride a metaphor—and getting off before it breaks down—is fundamental when dealing with the figurative language of finance. Evolution may have made our pattern recognition abilities instinctive and involuntary, but it did not make them infallible.
The trouble starts when we detect patterns that are not really there, as in the stock market. The brain evolved to detect patterns of immediate significance in do-or-die, fight-or-flight situations. The financial markets generate oceans of data: tens of thousands of stocks, each traded thousands of times a day by tens of millions of investors in dozens of markets around the world. By random chance alone, apparent trends will appear everywhere.
But while the stock market is filled with transitory patterns, the vast majority of them are meaningless, at least in the short term. The hourly variance of a stock price, for example, is far less significant than its annual variance. If you’re in the habit of checking your portfolio every hour on the hour, the noise in those statistics drowns out any real patterns. Still, people insist on making decisions based on these supposed patterns, even when explicitly told the patterns do not exist.
Oh, how quickly we’re deceived when we a bogus pattern do perceive.
Perhaps no other creature has had its pattern recognition abilities plumbed in such depth as the humble frog. And the frog’s gullibility in going after false patterns is a sobering analogy for the way we can be fooled by misleading financial metaphors.
The frog does not have a very discerning palate. In fact, a frog will try to eat anything you put in front of it, as long as the object is about the size of an insect and moves around in jerky, staccato bursts. If it looks like a fly and acts like a fly, frogs think, it must be a fly. Warren S. McCulloch, a neurophysiologist and early contributor to the field of cybernetics, and a group of colleagues proved this back in the late 1950s when they performed a meticulous study of the frog’s visual apparatus.
McCulloch and his collaborators placed an aluminum hemisphere before some prostrate frogs. Objects were attached to this metal plate by magnets and, like puppeteers at a children’s show, the researchers moved the objects across the frog’s visual field by moving the magnets on the back of the plate. As they did so, they recorded the electrical traffic along the frogs’ optic nerve, enabling them to detect which nerve fibers fired in response to which visual stimuli. In this way, they learned a lot about what and how the frog sees.
One of the team’s most important discoveries was that, by the time a visual image reaches a frog’s brain, it is already to a large extent classified and interpreted. So, for example, when a frog sees an object about the size of an insect movin
g around in jerky, staccato bursts, it does not delay, debate, or deliberate. The frog immediately shoots out its tongue to grab it. The researchers called the fibers that respond in this way “bug perceivers71.”
The scientists put on quite a show for their captive amphibian audience. They displayed “not only spots of light72 but things [the frog] would be disposed to eat, other things from which he would flee, sundry geometrical figures, stationary and moving about.” No matter what was put before them, though, the frogs always focused on the zooming, buzzing confusion of the presumed fly’s flight path. The frogs weren’t interested if the object didn’t move, or if it moved only in a straight line. So, the authors observed:
The frog does not seem to see73 or, at any rate, is not concerned with the detail of stationary parts of the world around him. He will starve to death surrounded by food if it is not moving. His choice of food is determined only by size and movement. He will leap to capture any object the size of an insect or worm, providing it moves like one. He can be fooled easily not only by a bit of dangled meat but by any moving small object.
The experiments showed, McCulloch concluded, “that the eye speaks to the brain74 in a language already highly organized and interpreted, instead of transmitting some more or less accurate copy of the distribution of light on the receptors.”
This way of seeing makes a lot of evolutionary sense. Flies typically trace erratic patterns in the frog’s natural habitat. They buzz about, alight for a moment, then zigzag off again on their tipsy trajectories. The frog’s chances of snaring a meal are vastly increased if it simply snaps at any object that moves in this way, rather than investing a lot of time and brain power in trying to analyze the object’s shape and color, too.
Among pond life, he who hesitates starves to death.
When it comes to stock price movements, we, too, take a frog’s eye view. We rely on implied trajectories, suggested in part by agent and object metaphors. If it looks like a pattern and acts like a pattern, we think, it must be a pattern. This passion for pattern clouds our judgment whenever numbers or probabilities are involved, as when picking stocks—or playing basketball.
A team of researchers, including Amos Tversky who, along with Daniel Kahneman, founded the field of behavioral economics, which uses insights from the social sciences to inform theories of how people make financial decisions, investigated the notion of the “hot hand” in basketball. The “hot hand” theory is the belief that a player’s chances of hitting a shot are greater following a basket than following a miss on the previous shot.
The researchers studied detailed shooting records of the Philadelphia 76ers and the Boston Celtics—and even conducted a controlled experiment with the men and women of Cornell University’s varsity teams—and concluded that the outcomes of previous shots influenced predictions about players’ chances but not their actual performance.
Tversky’s group found that more than 90 percent of fans believed that a player has “a better chance of making a shot75 after having just made his last two or three shots than he does after having just missed his last two or three shots.” Yet the probability of a hit was actually lower following a basket—51 percent—than following a miss—54 percent.
The researchers attributed the widespread belief in the “hot hand” to “a general misconception of chance76 . . . If random sequences are perceived as streak shooting77, then no amount of exposure to such sequences will convince the player, the coach, or the fan that the sequences are in fact random. The more basketball one watches and plays, the more opportunities one has to observe what appears to be streak shooting.”
As on the basketball court, so in the stock market, where the same confusion between pattern and chance makes people think they’re on a winning streak.
Our brains greedily seek patterns in everything, even in the chaos of stock prices and other financial statistics. As soon as we spot anything that looks like a pattern, we latch onto it as quickly as frogs snatch at anything that looks like a fly. Investors regularly chase hot stocks and hot funds, trying to invest in them before they go cold. This is especially true when agent metaphors are at work, because agents pursue goals—and that makes them special.
We make agents out of objects, given even the slightest provocation, by imputing the characteristics of living things to them. These attributions are often based on nothing more than the fact that the behavior patterns of non-living things look like the behavior patterns of living things. Psychologists Fritz Heider and Mary-Ann Simmel memorably demonstrated this back in the mid-1940s. They showed subjects a simple animated film78 involving basic geometric shapes—a big triangle, a smaller triangle, a circle, and a large rectangle that opened and closed on one side—moving around on a plain white background.
Asked to describe what they saw, participants invariably reported dramatic stories in which the circle and the smaller triangle were in love, the bigger triangle was trying to steal away the circle, but the circle and the smaller triangle managed to trap the bigger triangle in the large rectangle and so live happily ever after. Personification like this involves the same kind of metaphorical thinking involved in attributing agency to stock price movements.
This anthropomorphic instinct is the jumping-off point for many a literary flight of fancy. Known in the jargon of cognitive psychology as “physiognomic perception79,” it operates whenever we endow the inanimate with emotional or expressive qualities. Angry thunderclouds, smiling sunbeams, and plaintive melodies are all metaphors of physiognomic perception. A related phenomenon, physiognomic projection, takes place whenever we address the inanimate as animate. You are physiognomically projecting every time you scold your computer for crashing, cajole your car for stalling, or have feelings (of affection or aggression) for your in-car GPS voice.
Brian Scholl of the Yale Perception and Cognition Laboratory has made films starring simple geometric shapes similar to those of Heider and Simmel. In one film, two small squares are situated across from one another. Square A moves in a straight line toward Square B. But as soon as Square A gets close to Square B, Square B moves quickly away from Square A until it is several inches from it. The pattern repeats, and observers inevitably interpret it as a classic chase scene. “You see A cause the motion of B80,” Scholl wrote in a paper describing the experiments. “You see A and B as alive, and perhaps as having certain intentional states, such as A is wanting to catch B, and B trying to escape.”
Of course, there is no chase scene in Scholl’s film. It’s simply some geometric shapes moving around on a screen, just as Heider and Simmel’s circle and smaller triangle are not in love and three wayward Pac men and three pointy brackets do not two overlapping triangles make. Nevertheless, we perceive living physiognomies in these non-living things, inserting patterns where none exist.
Why are we so promiscuous with pattern, so profligate with personification? For the same reason frogs leap at anything that moves like a fly: it is essential for survival.
From an evolutionary perspective, it is far safer to automatically attribute agency to inanimate objects that behave like living things81 than it is to mistake a living thing for a seemingly inanimate object. That swaying in the trees may just be a breeze or it could be a wild beast, coiled and ready to strike. You can misperceive the breeze as a beast or the beast as a breeze. Which mistake would you rather make? And if you were an early hominid, which mistake would be more likely to ensure that you would survive long enough to reproduce?
Non-human primates share some of our pattern detection abilities. Starting in the late 1970s, psychologists David Premack, Guy Woodruff, and their collaborators taught Sarah, an African-born chimpanzee82, to use and comprehend a simplified visual language. Sarah’s vocabulary consisted of colored plastic tokens of various shapes, sizes, and textures. She learned to arrange these into simple sentences, such as “Apple is red.”
Premack and Woodruff taught Sarah how to use tokens for “same” and “different” and then showed her
two sets of objects—geometric shapes (triangles and crescents) that differed in size, color, or markings and ordinary household objects that differed in function (locks and can openers). The first set was designed to test Sarah’s perceptual matching abilities83 (large triangles are “the same” as large crescents) and the second was designed to test her functional matching abilities (can openers have “the same” function as keys). Children are given similar tests to evaluate their metaphoric competence.
Sarah consistently spotted the right analogical patterns, correctly indicating that a large and a small triangle were “the same” as a large and a small crescent. She chose the “different” symbol when presented with a large and small triangle and two small crescents. In the test of her functional matching ability, Sarah chose the “same” symbol for a picture of a lock and a key and a picture of a can and can opener, demonstrating her understanding that unlocking a lock is the same kind of activity as opening a can.
The agent and object metaphors of economics tap into this primal urge for pattern. Agents are special because only agents move of their own volition; only agents move with a purpose. And pattern recognition evolved in large part to predict the purpose of living things. According to Morris and his collaborators, “Uptrend stimulus trajectories84 should automatically trigger schemas for animate action and downtrends should trigger schemas for inanimate motion, regardless of whether the trajectories are encountered on sand dunes or stock charts.”
Decades of statistical analysis suggest the random walk metaphor is still the most accurate way to describe price movements in the stock market. Yet people insist on dragging bulls and bears into it85. When it comes to spotting price trends, we all have amphibian brains. “People making sense of stock charts86 may be in a predicament something like that of the frogs,” the Morris team wrote of McCulloch’s research, “victims of their automatized responses to trajectory.”
