Steven Sloman and Philip Fernbach, The Knowledge Illusion, Riverhead Books, 2017.
I provided a critique of our book club selection, The Knowledge Illusion, in a previous post. One of my criticisms was that the book did not provide an adequate overall theoretical perspective within which the seemingly disparate topics discussed could be integrated. One perspective that I think could have done so is Evolutionary Psychology. In this essay I will try to highlight a few of the ways Evolutionary Psychology approaches issues pertaining to acquisition and use of knowledge.
Every one of us alive today is embodied with copies of genes that have been passed down from our ancestors. We received those genes at the moment of our conception in the form of a fertilized egg, with half of the genes coming from our mother’s line and half from our father’s. Those genes, via interactions with the environment, guided the pre- and post-natal development of our physical bodies, including sense organs that are capable of picking up information from the environment, and nervous systems that can process that information in order to guide our thoughts and actions.
Psychologists characterize the functioning of our sense organs and nervous systems at molar levels of description, referred to as mental faculties. Examples of mental faculties include perceiving, thinking, and remembering. The ways those faculties operate are determined to a large degree by the genes that guided their development.
The fact that properties of our mental faculties are highly influenced by our genes gives us clues we can use to try to understand how we use those faculties to obtain and use knowledge. One fact that we know for certain is that the genes present in the current population of humans were all passed down to us from a line of individuals who survived (at least to an age where they were able to pass their genes on to offspring). So, we can expect our mental faculties to provide us with the kinds of knowledge that was critical for our ancestors’ survival. At the same time, an evolutionary perspective does not lead to any expectation that other kinds of knowledge obtained via these faculties will be complete or even accurate.
In the remainder of this essay, I will provide some examples of scientific findings regarding the kinds of knowledge provided by our mental faculties.
Our sense organs are designed to extract specific kinds of information from our immediate surroundings. For example, our eyes can detect a narrow band of electromagnetic radiation referred to as visible light. The neural circuitry that collectively functions as the mental faculty of visual perception is then able to extract information from this light to gain knowledge about certain properties of the external environment such as the shapes of objects.
The knowledge we can gain about the world by using our eyes is severely limited compared to what can be extracted using instruments. Radio waves, x-rays, and infrared light are examples of electromagnetic energy that is invisible to the eye but detectable using instruments. Knowledge about properties of the world that we gain using these instruments would be unavailable to us if we had to rely solely on our visual perception faculty. Similar limitations apply to our other sense organs.
Psychologists have identified numerous examples of illusions, examples where the “knowledge” gained from our sense organs is inaccurate. Here is an example.
Most human observers report that the two tabletops in this image do not appear to be identical in size and shape. In actual fact they are identical, as skeptics can demonstrate for themselves by measuring them, or by making a hardcopy of this image on a printer, using a pair of scissors to cut out the two tabletop images, and holding the two cutouts one over the top of the other. Perceptual illusions such as this one demonstrate that even though our eyes provide us with knowledge about features of objects such as size and shape that are correct (close enough for survival purposes) much of the time, our perceptions cannot be trusted to always be correct.
The situations where perception is in error are most commonly ones where the scene being examined by the eye is an artificial one. For example, in the image shown above, the eye is not really looking at 3-dimensional tables in the world but is instead looking at a two-dimensional image of tables. In this artificial situation the eye has been tricked in ways that would be less likely to work if the eye were viewing actual 3-dimensional tables. This is understandable when one takes the perspective that properties of perceptual faculties are governed in large part by genes passed down from ancestors who had to survive in actual 3-dimensional environments.
More complicated examples of the eye being tricked by artificial situations occur in virtual reality environments. Consider the situation most of us have experienced in a relatively primitive, but effective, virtual environment created by a television set. Suppose you are watching a football game on television. While doing so you gain all kinds of knowledge about what is happening – the quarterback has thrown the ball, the receiver has caught it and is running into the end zone, scoring a touchdown that puts the home team ahead! But what did your eye actually see? The “knowledge” you gained about the events taking place during the football game did not come from light waves entering your eye after having been reflected off 3-dimensional objects on the football field. The light that entered your eye was produced by a huge number of very small pixels that turned on and off rapidly on the television screen. The electromagnetic radiation emitted by those pixels was structured very precisely by engineers such that when it entered your eye it had spatial, temporal, and spectral properties that mimicked in certain essential ways properties of the light that would have entered your eye if you had been watching the game in the stadium. Or to state this in Evolutionary Psychology terms, your perceptual faculty interpreted the light entering the eye according to what would likely have been present (football players in a stadium) if this pattern of light stimulation had been encountered in environments in which your ancestors lived.
These examples demonstrate that our perceptual faculties cannot be trusted to provide us with complete or even veridical knowledge about our surroundings. If they did, we would simply state while watching a football game on television, “All I see are a boring bunch of pixels on the front of the screen turning on and off very rapidly.” And we would not be susceptible to optical illusions. And we would be able to use your eyes to gain knowledge directly that is only available to us by using machines that can detect other forms of electromagnetic radiation such as x-rays.
The bottom line is that “seeing is believing” may often be true, but in the search for truth it can lead the seer astray. From an evolutionary perspective, all we can expect from our perceptual faculty is an ability to gain and use knowledge about our immediate surroundings that was good enough to allow our ancestors to survive.
The cognitive faculty is used for various forms of thinking and reasoning. This faculty also worked well enough to allow our ancestors to solve problems that were important for their survival, but is prone to errors when applied more generally to other domains of problems. Psychologists have discovered a number of cognitive analogues of perceptual illusions, called cognitive fallacies. I will illustrate two examples here.
Humans often have difficulty solving this cognitive problem, and the solutions that are chosen commonly have errors. The most frequent error is to state that the card with 5 showing is one of the cards that should be turned over. Psychologists refer to this error as an example of a confirmation bias. We tend to pay too much attention to data that simply confirms our hypotheses and not enough to data that could potentially refute them. But the solution to the card problem requires that we examine data that could refute the assertion shown in the box, and the correct solution to this problem is that one should turn over two cards, E and 4.
Another cognitive fallacy, The Monte Hall Problem, is one that even individuals with strong backgrounds in logic and statistics often get wrong because the solution goes against the intuitions that are provided to us by the cognitive faculty. Imagine that you are a contestant on a game show hosted by Monty Hall and are shown three doors and told there is an expensive car behind one of the doors and only a goat behind each of the other two. You are asked to choose a door and will receive as a prize whatever is behind that door. For purposes of this example let’s suppose you choose door 1. Monte Hall then surprises you and informs you that before he opens the door 1 you have chosen, he is going to open a different door to let you take a peek. He opens door 3 and here is what you see.
Then Monte Hall gives you a choice. You can stick with your original pick of door 1 or you can switch to door 2. Assuming you want to take home the car instead of the other goat, what should you do?
The vast majority of humans respond that it should not make any difference which option you pick because there is an equal 50% chance of the automobile being behind either of the two remaining doors. In actual fact statistical analyses and simulations demonstrate that you will approximately double your odds of getting the automobile if you take the option of switching to door 2.
Cognitive fallacies such as these demonstrate that our cognitive faculties do not always provide us with correct solutions to solvable problems. Our abilities to use thinking, including various forms of reasoning and intuition, are good enough to solve the kinds of problems that were necessary for our ancestors to survive. However, we cannot trust them to come up with correct solutions to problems in general, even when our intuition leads us to believe that we can do so.
The memory faculty has to deal with different kinds of memories. These include short-term memories ( such as remembering a phone number for a few seconds while you dial it), procedural memories for motor skills you have learned previously (such as remembering how to ride a bicycle), semantic memories (facts you learned earlier, such as that George Washington was the first president), and autobiographical (memories of your own life history). I will limit my examples to autobiographical memories.
Try doing the following exercise that I used to assign to students enrolled in my Introductory Psychology classes.
Take out a piece of paper and write a paragraph or more describing one of your earliest precious childhood memories.
After the assignment was turned in, I would announce to the students in the class that what they described very likely never happened. As you might expect, this met with a high level of resistance from my students, and probably does to you also!
Contrary to popular belief, our memories do not provide us with the kinds of information that would have been recorded if a video camera had been attached to our foreheads throughout our lives. Only sparse snippets, if anything at all, remains in most of our autobiographical memories. Most of our very early memories are completely lost and the ones that remain have the perverse characteristic that the ones that seem most vivid are the ones most likely to be false! This is because the memories we think about the most are those that are most precious to us. And each time we think about a memory a few new details can be added in and a few others dropped. Our memory systems operate like children playing the game of telephone tag, with details being added and dropped at each stage of the retelling until at the end the story has little resemblance to the original. As a result, our autobiographical memories, instead of functioning as an accurate history of our experiences, provide us with stories (highly fictional) that we have edited to create meaning in our lives.
Numerous scientific studies have also demonstrated that false memories can be implanted. Paradigms such as the following are used. A group of subjects are brought into the laboratory and, sitting around in a group, each subject is asked to describe an early childhood memory. What the subjects in the experiment do not know is that one of subjects in the group is a confederate, i.e., a person working with the experimenter. The confederate relates an experience he had as a child such as becoming separated from his mother and seeing a clown.
Later, the same subjects are brought back to the laboratory for a follow up to the study. The experimenter asks each subject a series of questions, one of which is, Did the subject recall ever having had a scary experience involving being separated from their mother? A startling number of subjects (ranging from zero to over 50% in various studies but averaging about 25%) report recalling, AS THEIR OWN EXPERIENCE, an implanted false memory, in this example a memory of being frightened while separated from their mother and seeing a clown.
We are all susceptible to implanted false memories, and not aware of which of our memories are true and which are false. This is because all memories, however originally stored, feel similar while being recalled. Some actually are true in the sense that they are based on snippets of what was stored at the time the remembered experience actually occurred. Others might be accurate in terms of some of the details remembered, but we are remembering those details based on what we learned about the event later from other sources. A common example is a memory of an early birthday party where what is remembered is not based on the actual experience but instead based on looking at photos of the event or having parents tell us about the event. Finally, even though this fact goes against our beliefs and intuitions, some of our memories are most likely implanted and false. They were implanted from stories we heard or read, from movies or television shows we watched, or were fictions created gradually over time as we revisited old memories over and over, changing slightly with each re-remembrance, until what is remembered is mostly if not entirely false.
Our early human ancestors did not survive as individuals. Instead, they survived by living in social groups. This fact provides a potential explanation for why our memory faculty operates in this manner. Being able to survive as part of a social group requires many traits that support the group rather than the individual. These include feelings of strong loyalty to the group and being highly motivated to play roles that aid the group such as cooperating with others and following orders from group leaders. Constructing stories (memories) in which we identify with these roles was very likely adaptive to our ancestors and helps explain why our memories operate in this way.
The bottom line is that properties of the memory faculty cannot be relied on to provide a factual account of the events that have happened to us during our lifetime. Instead, they provide us with stories about how we think about ourselves that are meaningful. The expression, precious memories, is apt.
Psychological Faculties Compared to Scientific Methods
A common theme I have tried to emphasize in this commentary is that our psychological faculties cannot be trusted to always provide us with true knowledge. The examples I have presented illustrate cases where the perceptual, cognitive, and memory faculties have been demonstrated to be faulty. Similar analyses of the other psychological faculties would reveal similar findings. Our faculties provide us with the kinds of knowledge that allowed our ancestors to survive. Nothing more.
When our book club discussed The Knowledge Illusion one of the questions that was raised was why it took humans so long to develop our current technologies. Since the beginning of the industrial revolution technological advances have been happening at an accelerating pace. But why did it take so long for humans to get the industrial revolution started? An evolutionary perspective provides a potential answer to this question. Until the 17th Century when humans began to develop scientific methods, we had to rely solely on our psychological faculties to gain knowledge about how the world operates, and as I hope this essay has demonstrated, those cannot always be trusted.
A useful way of characterizing scientific methods is that they are designed to try to overcome some of the limitations of our psychological faculties. For example, instead of relying exclusively on our perceptual faculty, scientific methods utilize instruments for making observations of the physical world. Instead of relying only on our cognitive faculty to make sense of these observations, scientific methods employ formal tools such as algorithms that can be implemented on computers to construct and evaluate models, statistics that can help minimize cognitive errors such as the confirmation bias, and formal systems of logic and mathematics to evaluate assertions. Rather than relying on a faulty memory faculty, lab notebooks, computer databases, peer reviewed published papers, and other forms of documentation are used for keeping track of what is known. And so on with respect to the other psychological faculties.
Use of scientific methods as an adjunct to our own psychological faculties is what has allowed, and hopefully will continue to allow, rapid advances in our knowledge about properties of the physical world. And that knowledge is in a form that has facilitated accelerating advances in technology.