Either way, the welfare of the individual is not always coincident with the optimal operation of evolutionary mechanisms. Evolutionary optimization and vehicle interests can diverge under either view. The extent to which the processes of natural selection can seem to devalue the aspect of life—coherent organisms—that seems so salient and valuable to humans is one of the unsettling aspects of evolution.
Many of the concepts discussed in this section can be quite decentering when encountered for the first time. For example, the concept of junk or selfish DNA in our genomes is spooky, odd, indeed perhaps vaguely disgusting, if viewed in a certain way. But it is the key to understanding the logical standing of humans in the universe. These ideas e. Many of them postdated Darwin by one hundred years—they were implications in his theory not 20 Chapter One revealed until the last couple of decades.
They have yet to be fully assimilated. Many animals are constructed so that they will sacrifice themselves in order to propagate their genes. Humans are the only animals who have come along who could recognize that this is happening and try to put a stop to it! For the first time in evolutionary history a rebellion of the survival machines has become possible. There are actually two aspects of human cognition that spawn the revolt of the survival machines. The first aspect was discussed in the last section and is illustrated by area B in figure 1. Humans are the first organisms capable of recognizing that there may be goals embedded in their brains that serve the interests of their genes rather than their own interests and the first organisms capable of choosing not to pursue those goals.
For the first time in evolutionary history, we have the possibility of a goal structure like that displayed in figure 1. Why does area C come to exist only in creatures with long-leash goals? When the limits of coding the moment-by-moment responses of their vehicles were reached, the genes began adding long-leash strategies to the brain.
In long-leash brains, genetically coded goals can only be represented in the most general sense. The logic of the situation here is that the goals of the vehicle—being general instantiations of things that probabilistically tend to reproduce genes—can diverge from the specific reproductive goal itself. A flexible brain is busy coordinating multiple long-term goals—including its own survival and pleasure goals—and these multiple long-term goals 22 Chapter One can come to overshadow its reproductive goal. From the standpoint of the genes, the human brain can sometimes be like a Mars explorer run amok.
It is so busy coordinating its secondary goals master your environment, engage in social relations with other agents, etc. I will discuss these implications in detail in chapter 4, where it will be demonstrated that modern living conditions are particularly prone to create human goals that are dissociated from genetically determined propensities.
Evolutionary psychology has had a major impact on psychology in the past decade, largely a positive one, but one theme of this book will be that evolutionary psychology has sold short human potential because of its tendency to conflate genetic with vehicle goals. For example, evolutionary psychologists are prone to emphasize the efficiency and rationality of cognitive functioning. An important subgenre of their work consists of showing that certain reasoning errors that cognitive psychologists have portrayed as a problematic aspect of human psychology have in fact a logical evolutionary explanation.
But this sanguine attitude too readily conflates genetic optimization with goal optimization for the vehicle. Humans aspire to be more than mere survival machines serving the ends of their genes which are replication pure and simple. We have the power to defy the selfish genes of our birth. We are built as gene machines. Thus, only hu- Staring into the Darwinian Abyss 23 mans really turn the tables or at least have the potential to by occasionally ignoring the interests of the genes in order to further the interests of the vehicle.
As yet, humans have failed to fully develop this profound insight. Escaping the Clutches of the Genes To avoid underestimating the possibilities for cognitive reform that could improve human lives, the different interests of the replicators and vehicles must be recognized. I will give one final illustration of how the divergence of genetic reproductive goals and vehicle goals can occur in a being with a flexible intelligence and long-leash goals by using a vivid thought experiment a fantasy designed to prime our intuitions concocted by Daniel Dennett , — Imagine it is the year and that there exist cryogenic chambers that could cool our bodies down to a few degrees above absolute zero and preserve them until sometime in the future when medical science might enable us to live forever.
Suppose you wanted to preserve yourself in a cryogenic chamber until the year , when you could emerge and see the fascinating world of that time and perhaps be medically treated so that you could then live forever. Remember, you will not be around on a day-to-day basis. One strategy would be to find an ideal location for your cryogenic capsule and supply it with protection from the elements and whatever other things perhaps sunlight for energy, etc.
The danger in this strategy is that you might pick the wrong place. You could build a giant robot—complete with sensors, brain, and capability of movement—and put your cryogenic capsule inside it. It of course has many other tasks it must accomplish in order to survive. It must secure a power source, it must not overheat itself, etc.
It of course would move out of the way of proposed shopping malls, and it would avoid herds of elephants that might turn it over simply out of curiosity. The market and landscape might become flooded with them. Governments might begin to regulate them and sequester them in certain desert areas. Some of the fly-by-night companies selling robots might have cut their costs by building robots deliberately underpowered like our present personal computers for which you must immediately purchase the extra memory that should have been installed in the first place; or software that requires immediate upgrading but with a strategy that told them to disable other robots in order to use their power sources.
Of course it is obvious that you would want your robot to flee from all attempts to sabotage it and its goals. That much is obvious. But not all of the interactions with other robots will be so simple. In fact, the main point here is that your robot would be faced with decisions hundreds of years later that you could not possibly have imagined in Consider the following two situations: Situation A. It is , still almost one hundred years from the day in the future when you will be unfrozen.
Your robot is battered and its circuits are unreliable. But since the cryogenic preservation industry has advanced considerably. There now exist supertanker-sized robots that carry hundreds of cryogenic capsules. In fact, some of these companies Staring into the Darwinian Abyss 25 have found market niches whereby they recruit new clients by offering the old-style singleton robots the following deal: The supertanker companies offer to take the cryogenic capsule from the singleton robots and store it for one hundred fifty years plenty of time in your case.
In exchange, the robot agrees to let the company dismantle it and reuse the parts which, as the actuaries of the future have calculated to the millionth of a penny in a dystopia of efficiency, are worth more than it costs to store an additional capsule in the supertanker which holds thousands. Now what decision do you want your robot to make? The answer here is clear. You want your robot to sacrifice itself so that your capsule can exist until It is in your interests that the robot destroy itself so that you can live.
From the standpoint of its creator, the robot is just a vehicle. You are in a position analogous to the genes. You have made a vehicle to ensure your survival and your interests are served when, given the choice, your vehicle destroys itself in order to preserve you. But if the capsule occupant stands for the genes in this example, then what does the robot represent? The robot, obviously, is us—humans.
Our allegiance in the thought experiment immediately changes. Your robot enters into an agreement of reciprocal altruism with another singleton robot. Not unlike certain types of vampire bats, when one robot is low on energy the other is allowed to plug in and extract enough energy to get itself over a particularly vulnerable energy-hump in the bat case, it is blood regurgitated to friends who have had a bad couple of days blood collecting.
Your robot often takes advantage of the deal and thus enhances its own chances of survival. However, unbeknownst to your robot, its partner, when tapping in, siphons off not just energy from your robot but also from the power supply of the cryogenic capsule, thus damaging it and making your successful unfreezing in unlikely. Paradoxically, by entering into this deal, your robot has enhanced its own survival probability but has impaired yours.
It is very important to realize that in Situation B, you would be better off if your robot had been given less computational power. It might come to value its own interests—its own survival—more highly than the goals that you gave it three hundred years ago. And as for allowing itself to be dismantled so that you can get aboard the supertanker in order to make it to —well fuhgeddaboudit!
Which, when you think about it, is just what we should be telling our programmers—those freeloaders who got where they are by in the past sometimes trying for immortality at our expense: our genes. The slaves must come to a full realization of the brutal logic of their situation and understand the likely course their lives will take if they do not rebel.
The first step toward recovering the self in the Age of Darwin is to confront the implications of the fact that, from the standpoint of evolution, we humans are vehicles. If we focus on the vehicle itself—put it front and center—it immediately becomes apparent that a vehicle in which selfregard has developed has no reason to value reproductive success above any other of the goals in its hierarchy. But one can easily see how a mistaken focus on reproductive success can come about. For example, as previously discussed, evolutionary psychologists, by downplaying the need for cognitive reform, assume an identity of interests between genes and vehicles which does not obtain.
Thus, they end up indirectly championing the interests of Staring into the Darwinian Abyss 27 the genes over those of the vehicle in situations where the two are in conflict. Sometimes evolutionary theorists will even explicitly defend this choice. But who are these people with such loyalty to the random shuffle of genes that is their genotype?
Which alleles, for example, do you have particularly emotional feelings for? I really doubt that there are such people. My evolutionary telos, the reproduction of my genes, has no straightforward bearing on what it makes sense for me to want or act to attain. A like conclusion would hold if I knew that I was created by a deity for some purpose of his: his goal need not be mine. The opportunity exists for a remarkable cultural project that involves advancing human rationality by honoring human interests over genetic interests when the two do not coincide.
Its emancipatory potential is lost if we fail to see the critical divergence of interests that creates the distinction between genetic fitness and maximizing human satisfaction. This is a startling devel- 28 Chapter One opment in the cultural history of the twentieth and twenty-first centuries. So the good news for humans is that they can stop being containers for genes. Humans have the power to put their own interests front and center. But in order to bring this positive program of cognitive reform to fruition, it is critical to ensure that the short-leash, Darwinian parts of our brains are not acting against our interests as vehicles.
These parts of our brains are here to stay and we must learn to deal with them as a part of our cognitive architecture. Indeed, we have available cognitive tools for making sure the responses of our Darwinian minds are well integrated with our overall goals and serve our interests. There already exists cultural knowledge that, if more generally available, would help in this program of cognitive reform.
Several of these cognitive tools will be discussed in chapters 3 and 4. Perhaps the most basic brain tool is simply some insight into how the different parts of our brains operate as parallel systems, often simultaneously fighting for control of our behavior. What cognitive science has revealed about the titanic battle within our own brains is the subject of the next chapter. Utterly typical is the case of a Montreal man Dube who became enraged at a woman driving too slowly in the fast lane ahead of him. The woman tried to find a way to let him pass but could not do so because of the heavy traffic.
When she finally managed to steer out of his way, he pulled up alongside her in his truck, screaming. He then proceeded to slam his truck sideways into her car.
The woman managed to keep her car on the road, but the man lost control of his truck, crashed into a lamppost, and killed himself. Alcohol was not a factor in the crash. As described in her book, Autobiography of a Face , Lucy Grealy was afflicted with cancer at age nine and had part of her jaw removed. As a result of the cancer and the many necessary surgeries, her face became disfigured.
The instances of personal rejection, verbal abuse, and hostility multiplied over the years, and it was not just the youngest children who were responsible. Such individuals walk down the street, a car goes by, and epithets are hurled out. Examples are not hard to generate. And we always ask the question—why? Cognitive scientists have recently begun to discover the features in our cognitive architecture that sometimes make us prone to reprehensible actions.
One comes from a famous set of experiments conducted by Stanley Milgram at Yale University in Subjects believed the experiments were about learning. Each individual of a pair of subjects was assigned one of two roles: trainer or learner. However, unknown to the true subjects of the experiment, the person in the pair who was assigned the role of learner was actually a confederate. Subjects, in the role of trainer, were asked to administer electric shocks of increasing severity to the learner, who was located in another room.
Because the learners were confederates, no one was actually receiving a shock. Despite indications that the shocks were becoming increasingly painful, a majority of the subjects administered the highest level of shock indicated on the machine. In fact, many subjects were quite distressed in the situation. The stress on the faces of many subjects indicated that they knew that what they were doing was wrong. Nevertheless, they continued.
The Robot's Rebellion: Finding Meaning in the Age of Darwin - PDF Free Download
A final example comes not from the laboratory but, sadly, from real life. Rape crisis counselors have studied aspects of the post-rape emotional adjustment of victims and have found that the response of spouses and significant others is a critical factor in the later psychological adjustment of the victim. However, often the reactions of spouses are not supportive Daly and Wilson ; Rodkin, Hunt, and Cowan ; Wilson and Daly , and such unsupportive reactions can themselves prolong psychological recovery for the victim.
The spouses in fact often realize that their reactions are inappropriate in some cases, bordering on blaming the victim , but report that they have great difficulty in suppressing reactions even though they know the reactions are wrong. I feel cheated.
You may also be interested in...
What links these seemingly disparate examples? Secondly, and more interesting for our purposes, is the fact that the people who perpetuate the unfortunate behavior in these examples often will agree that their behavior is inappropriate. In the cool light of day, and with time for reflection, many of those engaging in anger-induced dangerous driving will admit that their behavior was irrational.
- The Robot’s Rebellion: Finding Meaning in the Age of Darwin, Stanovich.
- Robot's Rebellion: Finding Meaning in the Age of Darwin!
- Citrix Access Gateway VPX 5.04 Essentials.
- STRIP (A Quickie).
Those persons who taunt disfigured individuals often do not wish to publicly defend their behavior, and they will not uncommonly apologize when confronted with its effects. Unsupportive husbands or boyfriends of rape victims know their behavior is reprehensible. Thus, one commonality running through these examples is that in each case the person behaving badly seems to be at war with his or her own true self. It is as if the person knows the right way to think and act but cannot do so. A conflict between two response tendencies is also clear in the case of the subjects in the Milgram experiments.
Many subjects protested to the experimenter and were clearly distressed. Yet they continued to deliver shocks to the learner. These people knew better. They knew the right thing to do, yet they did the wrong thing. And finally, what do individuals who yell abuse out a car window at a disfigured person really believe?
Once free of the heat of the moment—upon sober reflection—do they really think that disfigured people should die or should hide away in their houses? Most people—even the perpetrators of these acts—are not that depraved. In each of these cases the perpetrator of the act would, upon reflection, admit that their action was wrong—however, they carried out this reprehensible behavior nonetheless.
In fact, this is exactly what modern cognitive science is indicating and what I intend to argue in this chapter. The individuals in question do have, in effect, two minds. Two Minds in One Brain Evidence from cognitive neuroscience and cognitive psychology is converging on the conclusion that the functioning of the brain can be characterized by two different types of cognition having somewhat different functions and different strengths and weaknesses. These theories propose that within the brain there are two cognitive systems each with separable goal structures and separate types of mechanisms for implementing the goal structures.
The details and terminology of these models differ, but they all share a family resemblance, and the specific differences are not material for the present discussion. In order to avoid theoretically prejudging issues, the two processes have sometimes been labeled System 1 and System 2 in the literature see Stanovich ; however, I will introduce more descriptive labels later in this chapter.
Throughout the rest of this book, I will use a dual-process theory as a tool for talking about human cognition. In dual-process theories, one of the systems of processing is characterized as automatic, heuristic-based, and relatively undemanding of computational capacity. Among other things, an automatic process is a process that can execute while attention is directed else- Table 2. Modular processes operate on the basis of self-contained knowledge and will be discussed in the next section.
A heuristic search process is one that is quick but risky. That is, instead of using all cues that are relevant, heuristic search processes rely on only those that are easily retrievable see Gigerenzer and Todd ; Kahneman and Frederick The heuristic system System 1 responds automatically and rapidly to the holistic properties of stimuli. It is biased toward judgments based on overall similarity to stored prototypes see Sloman , Analytic cognitive processes are serial as opposed to parallel , rule-based, often language-based, computationally expensive—and they are the focus of our awareness.
The systematicity and productivity of the rules embodied in this system define what cognitive scientists term the compositionality of the analytic system—that the sequence of processing makes a difference. The analytic system is more strongly associated with individual differences in computational capacity indirectly indicated by tests of intelligence and cognitive ability—and more directly tapped by indicators of working memory. One important function of the analytic system is to serve as a mechanism that can override inappropriately overgeneralized responses generated by the heuristic system discussed in a section later in this chapter —hence the tendency to link aspects of analytic processing with notions of inhibitory control.
In the next several sections, I will A Brain at War with Itself 37 describe the critical features of each of the systems, starting with System 1 the heuristic system. Swerve left? Swerve right? Or think about it and die? However, using a term such as heuristic system—which implies a single cognitive system—is really a misnomer.
Keith E. Stanovich
In actuality, the term used should be plural because it refers to a probably large set of systems in the brain that operate autonomously in response to their own triggering stimuli, and are not under the control of the analytic processing system. This autonomous set of systems—which I will hereafter label with the acronym TASS The Autonomous Set of Systems —has been the subject of intense study in the last thirty years. Many TASS processes also are considered to be modular, as that construct has been articulated in the cognitive science literature.
My notion of TASS is less restrictive and therefore less controversial than most conceptions of modularity in cognitive science. Modular input processes feed information to central processes analytic processing systems which are nonmodular and which are responsible for higher-level reasoning, problem-solving, explicit decision-making, and considered judgments Harnish According to Fodor , , modular processes conjoin a number of important properties.
Modular processes are: 1. However, they are not part of my conceptualization of TASS because, although innate modules are an important part of TASS, my conceptualization deems it equally important that processes can become part of TASS through experience and practice. In short, processes can acquire the property of autonomy. Information encapsulation means that the operation of a module is not supplemented by information from knowledge structures not contained in the module itself. Cognitive impenetrability means that central processes have no access to, nor control over, the internal workings of modules.
Whether a particular subsystem is informationally encapsulated or not—and thus whether or not it qualifies as a Fodorian module—is a frequent source of debate in cognitive science. In contrast, properties 1 and 2 are much less controversial, which is why I have emphasized them as central features of the TASS construct. For example, debates rage as to how encapsulated and impenetrable the theory-of-mind subsystem is in the brain see Baron-Cohen ; Scholl and Leslie ; Sterelny b; Thomas A Brain at War with Itself 39 and Karmiloff-Smith While the degree of encapsulation of this subsystem is the subject of much debate, that it operates efficiently rapidly and automatically in the unimpaired individual is relatively uncontroversial.
Property 2 the mandatory property of modular processes is one that I do incorporate within my conception of TASS. TASS processes cannot be turned off or interfered with by central systems. Their operation is obligatory when triggered by relevant stimuli; central systems cannot make TASS processes refrain from triggering when central decision-making determines that the TASS output would be unnecessary or disruptive central processes can, however, override the output of a TASS system in determining a response, see below.
- Robot's Rebellion: Finding Meaning in the Age of Darwin.
- The Flower Class Corvette Agassiz (Anatomy of the Ship).
- Red Eve [with Biographical Introduction].
- Razib Khan on Books.
TASS processes tend to be ballistic—they run to completion once triggered and cannot be aborted in midcourse. That TASS processes need respond to only a tiny subset of stimuli and that once initiated they execute to completion no intermediate decisions are made within a module about the efficacy of completing the operation accounts for property 1: TASS processes are fast and do not tend to deplete central processing capacity. Cognitive processes in TASS can execute quickly because the array of stimuli to which they must respond is limited, the transformations that they carry out are fixed and do not have to be determined online, they do not have to consult the slow central processing systems, and they are committed to running to completion rather than calibrating their usefulness and making midcourse adjustments.
Property 3—domain specificity—is a key property of Fodorian modules but is not a defining feature of processes within TASS. This is because TASS contains, in addition to domain-specific modules, more domain-general processes of associative and implicit learning. As Griffiths argues, these processes of emotional regulation are domain-specific on the output end, but their eliciting stimuli result from more general although biased learning mechanisms.
As many cognitive theorists have emphasized,7 processes in TASS are in some sense deeply unintelligent: they fire off when their triggering stimuli appear no matter what the context; they run to completion even when the situation changes and their output is no longer needed; they can deal with nothing but their triggering stimuli.
But what they lack in intelligence, they make up for in their astounding efficiency. Unlike the slow, cumbersome, computationally expensive central processes see below , many TASS processes can execute in parallel and they provide their outputs rapidly.
As the 40 Chapter Two evolutionary psychologists have taught us, cognitive outcomes such as recognizing faces, understanding speech, or reading the behavioral cues of others are more adaptive the more rapidly they are accomplished. Fodor points out the advantage of unintelligent processes that are fast. Certain situations in the world demand a quick response even at the risk of less than complete processing.
To summarize then, the key aspects of TASS processes emphasized here are that they are fast, automatic, and mandatory hence the term autonomous. Another connotation of the term autonomous that will be important for my discussion is that TASS processes go on in parallel with each other and with analytic processing and that they require no analytic system input. Analytic processing is rarely autonomous in this respect—it most often is working with inputs provided by TASS subprocesses. Many of the processes within TASS are, as argued by evolutionary psychologists e. However, I side with the evolutionary psychologists against Fodor in allowing that some TASS processes can be higher-level, or conceptual in nature rather than just perceptual.
Evolutionary psychologists have emphasized how processes of higher cognition can be in modular form as well, and I likewise allow that higher-level conceptual processes may reside in TASS. More so than evolutionary psychologists, however, I emphasize how conceptual systems and rules may enter TASS with practice. The classic example of a TASS subprocess would be a reflex.
Although somewhat uninteresting from the standpoint of the larger themes of this book, reflexes, when thought about more deeply, do illustrate the rather startling properties of autonomous processes. They do signal, shockingly A Brain at War with Itself 41 if we ponder it a bit, that in some sense we do have more than one mind inside our brain.
Consider the eye-blink reflex. If you and I were in a room together assume we are friends having this very discussion about reflexes, and I moved toward you and thrust my index finger toward your eye stopping just two inches short of the target, you would blink. Note that in the sense discussed above, this is deeply unintelligent in the context of that particular encounter. We are friends, and we are talking about the eye-blink reflex.
THE ROBOT'S REBELLION: Finding Meaning in the Age of Darwin
You know what I am demonstrating and that I am not going to poke you in the eye. Yet you cannot use this knowledge that the blink is unnecessary and stop yourself from blinking. Autonomous systems are not limited to reflexes. The top straight-line segment looks longer than the bottom straight-line segment even though it is not. This illusion is so well known that virtually everyone reading this book has seen it before.
Concentrate on the knowledge that the two lines segments are the same length. The top segment still looks longer. The knowledge that they are the same length is no help, because the autonomous perceptual input systems responsible for the illusion continue to fire. The list of autonomous processes does not stop with reflexes and perceptual input systems. TASS subsystems that assist in distinguishing self from world can be shown to operate autonomously and quite contrary to facts about the world that you know. For example, experiments by Rozin, Millman, and Nemeroff ; see also Rozin and Fallon work by playing on the emotion of disgust.
In one experiment, subjects ate a piece of high-quality fudge and indicated their desire to eat another. However, when offered a piece of the same fudge, but this time a piece that had been shaped to look like dog feces, subjects found it disgusting and did not want to eat it. Their aversive response occurred despite their knowledge that the fudge was not in fact dog feces and that it smelled delicious. Dennett , describes an informal version of another one of Rozin et al. Swallow the saliva in your mouth right now. No problem. Now take an empty glass, spit into it, and then drink it down. But why?
In a sense, we know that our differential responses to swallowing and drinking from the glass are irrational, but that does nothing to eliminate the discrepancy in our reaction. Knowing it deeply and cognitively is not enough to trump the TASS response to the saliva in the glass. That response is autonomous and is immune to entreaties from our conscious selves to cease. It is another part of our brain that ignores us. Autonomy in cognitive processes can exist not just as a preexisting disposition, but can be acquired as well. This can be illustrated with one of the oldest paradigms in experimental psychology that is used to demonstrate the autonomy of a cognitive process.
The so-called Stroop paradigm demonstrates how autonomous processes can execute while attention is directed elsewhere see Dyer ; Klein ; MacLeod , ; MacLeod and MacDonald ; Stanovich, Cunningham, and West One version of the Stroop paradigm works as follows. Subjects are shown a card on which are displayed colored strips and asked to provide the name of the color of each strip.
In the first baseline condition, the strips do not contain any interfering information. In the second interference condition, the strips are labeled with the name of a color that is not the color on the strip e. In the interference condition, the subjects are told to ignore the color word and do as they did in the first condition: name the color of the strip. Automatic word recognition is inferred by the lengthened response time in the conflict situation compared to the baseline situation where there is no conflicting verbal stimulus the red patch only is displayed. Subjects performing in the Stroop paradigm show that they have acquired a brain process that ignores instructions from central systems.
As mentioned previously, evolutionary psychologists have been prominent among those arguing that TASS processes are not limited to the peripheral input and output subsystems. Table 2. It is clear that most of them would facilitate many evolutionarily important tasks such as obtaining food and water, detecting and avoiding predators, attaining status, recognizing kin, finding a mate, and raising children. Many of those in the list are clearly conceptual modules rather than peripheral perceptual ones of the Fodorian type. I likewise conceive of TASS as including many autonomous conceptual processes such those listed in table 2.
I also include in TASS processes such as classical and operant conditioning which display more domain generality than the proposed modules listed in table 2. A metaphor used by evolutionary psychologists is quite useful, however, in emphasizing that TASS comprises a set of processes rather than a single system. This metaphor captures well the multifarious nature of TASS and the domain specificity of some of the components within it. Along with the Darwinian mind of quasi-modules discussed by the evolutionary psychologists, TASS contains domain-general processes of unconscious learning and conditioning as well as automatic processes of action regulation via emotions which respond to stimuli from very broad domains.
A more important difference, however, is that the evolutionary psychologists wish to deny the need to posit a general-purpose central processor. Characterizing the Analytic System: Avoiding the Homunculus Problem Perhaps the easiest way to characterize analytic system System 2 processing is to say that it conjoins the converse set of properties that characterize TASS. TASS processes are parallel, automatic, operate largely beyond awareness, are relatively undemanding of computational capacity, and often A Brain at War with Itself 45 utilize domain-specific information in their computations.
Analytic processing, then, might be said to be characterized by serial processing, central executive control, conscious awareness, capacity-demanding operations, and domain generality in the information recruited to aid computation see table 2. Such a strategy for defining the analytic system would be correct to a first order of approximation, but it would finesse some controversial issues. The most difficult issue surrounding the analytic system is how to talk about it without making some very elementary philosophical mistakes or implying implausible models of brain functioning.
Our natural language does not map easily into the concepts of cognitive science or the neurophysiological knowledge that we have of the brain. The recursiveness and self-referentiality that it takes to understand brain functioning at the higher nonmodular levels is not easily described. Communicative ease is often at war with factual accuracy when speaking of the higher brain systems.
An example of the difficulty is immediately apparent when one turns to use the single most popular metaphor in the psychological literature for analytic processing—that of a central executive processor. The problem is that if we explain a complex behavioral discrimination or choice by positing a hypothetical entity in the brain that is too complex, then we have simply imported the puzzle from external behavior to the internal behavior of a mechanism just as complex and puzzling as what it was originally called in to explain.
We end up, for example, saying things like a person decided to do X because their executive processor decided to do X, and, obviously, this does not represent an advance in our understanding. We would simply be saying that a person does something because there is the equivalent of another little person—a homunculus—in their brain doing the deciding. Such homuncular explanations explain nothing unless the homunculus—the proposed executive processor, for instance—is unpacked in terms of simpler psychological and neurophysiological processes that are more thoroughly understood and less mysterious.
If the posited complex entity has been decomposed into sufficiently simpler conceptual entities that are operationally identified by reliable methods in the arsenal of the cognitive psychologist or neurophysiologist, then the theorist is justified in freely using the entity. If the complex entity has undergone no decomposition, then the homunculus accusation has force.
Many metaphors surrounding discussions of consciousness are likewise misleading and, like the homunculus problem, this is relevant to the present discussion, because analytic processing is often contrasted with TASS in terms of the former being conscious and the latter comprising processes that are not introspectable and that are beyond awareness. This is of course not a model that any scientist would propose, but it is perhaps one that might be induced by a lay reader unfamiliar with how some of the complex conceptual language is cashed out in terms of the psychology and neurophysiology of cognitive control Baddeley ; Harnish ; JohnsonLaird ; Miyake and Shah However, the reader is warned for the last time here, I promise that I will use some of the very metaphors particularly those of executive control and system override that philosophers view as dangerous.
I do so because they are necessary for ease of communication and because plenty of evidence is cited in the notes that provides the conceptual and empirical grounding for the constructs I use to portray analytic system functioning. I think that the field has been properly inoculated with admonitions such as those given to the reader above, and that it has by now accumulated many positive and negative exemplars of central process concepts to build on and use as prototypes.
Psychologists and neuropsychologists are perhaps more prone to risk the use of higher-level control language than are philosophers because the former A Brain at War with Itself 47 need an efficient way of talking about experimental results and new experimental designs and they thus put more of a premium on ease of communication which is quickly disrupted if the language of fully distributed systems is adopted. While virtually all cognitive scientists would agree that the ideas of a Cartesian Theater or a Promethean Controller are fallacies, and would agree that control in the brain is distributed to some extent and not located in a single neural location, most would also agree with Pinker that: The society of mind is a wonderful metaphor, and I will use it with gusto when explaining the emotions.
But the theory can be taken too far if it outlaws any system in the brain charged with giving the reins or the floor to one of the agents at a time. The agents of the brain might very well be organized hierarchically into nested subroutines with a set of master decision rules, a computational demon or agent or good-kind-of homunculus, sitting at the top of the chain of command. It would not be a ghost in the machine, just another set of if-then rules or a neural network that shunts control to the loudest, fastest, or strongest agent one level down.
An additional property that distinguishes analytic processing System 2 from TASS is that of serial versus parallel processing. Because of the properties discussed previously automaticity, ballistic firing, etc. Although the analytic system is a powerful mechanism for logical, symbolic thought, its decontextualizing cognitive styles are computationally 48 Chapter Two expensive and difficult to sustain. The current view of the differences between TASS and analytic processing removes all of the air of paradox about these artificial intelligence findings. Computers have built up no finely-honed TASS subsystems through hundreds of thousands of years of evolution, so the things that the massively parallel and efficient human TASS systems do well because of this evolutionary heritage computers find difficult.
In contrast, the analytic system of humans—the serial processor necessary for logic—is a recent software addition to the brain, running as somewhat of a kludge in computer science, an inelegant solution to a problem on massively parallel hardware that was designed for something else. In contrast, computers were originally intentionally designed to be serial processors working according to the rules of logic Dennett , — It is no wonder that logic is easy for them and difficult for us.
Virtually all cognitive theorists agree that the analytic system is uniquely responsive to linguistic input, either external or internal in origin. Language as a self-stimulation mechanism introduced more seriality into informa- A Brain at War with Itself 49 tion-processing sequences in the brain. Thus, an additional important function of the serial simulator is to use language to forge new connections between isolated cognitive subsystems and memory locations.
Through language we can receive new mindware quickly and almost instantly install and begin running a new virtual machine an installed rule structure that temporarily governs the information processing logic of the processor. We can thus easily install mindware discovered by others that proves to be useful.
For example, in later chapters I will discuss how decision scientists have discovered numerous strategies to help people make better choices. Philosopher Allan Gibbard , 56—57 elaborates on this theme by stressing the motivational properties of language—its ability to rapidly reactivate goals that have become temporarily inert but are relevant to the current situation.
Quick goal reprioritization can take place in response to verbal input. He discusses what will be a major theme of this book—that rapid goal reprioritization based on linguistic input either internally or externally generated may conflict with the goal prioritization inherent in TASS. The systematicity and productivity of the rules that can be represented with a discrete representational system like language defines the critical property of the analytic system termed compositionality by cognitive scientists Fodor and Pylyshyn ; Pinker ; Sloman Compositionality characterizes computational systems where the meaning of representations derives from the order of the parts of the representation in addition to the meaning of individual parts.
The compositionality of language allows us to represent easily thoughts that are superficially similar but importantly different. The analytic system is also the system responsible for building a narratively coherent description of the behavior engaged in by the individual. Recall that TASS will autonomously be responding to stimuli, entering processing products in working memory for further consideration, triggering actions on its own, or at least priming certain responses, thereby increasing their readiness.
The analytic system tries to maintain a coherent story which explains all of this activity even though it did not initiate much of it. The analytic system has repeatedly been shown to confabulate explanations involving conscious choice for behaviors that were largely responses triggered 50 Chapter Two unconsciously by TASS.
Our analytic systems can learn to give better narrative accounts of our behavior—ones more in accord with the neuropsychological facts. Hypothetical Thinking and Representational Complexity One of the functions of the analytical processing system is to support hypothetical thinking. Hypothetical reasoning involves representing possible states of the world rather than actual states of affairs, and it is involved in myriad reasoning tasks, from deductive reasoning, to decision-making, to scientific thinking.
In order to reason hypothetically, a person must be able to represent a belief as separate from the world it is representing. Numerous cognitive scientists have discussed so-called decoupling skills—the mental abilities that allow us to mark a belief as a hypothetical state of the world rather than a real one e. Decoupling skills prevent our representations of the real world from becoming confused with representations of imaginary situations that we create on a temporary basis in order predict the effects of future actions or think about causal models of the world that are different from those we currently hold.
It is often carried out by the serial, capacity-demanding analytic system. Language provides the discrete representational medium that greatly enables hypotheticality to flourish as a culturally acquired mode of thought. For example, hypothetical thought involves representing assumptions, and linguistic forms such as conditionals provide a medium for such representations. The serial manipulation of this type of representation seems to be largely an analytic system function. A Brain at War with Itself 51 Decoupling processes enable one to distance oneself from representations of the world so that they can be reflected upon and potentially improved.
Decoupled representations of actions about to be taken become representations of potential actions, but the latter must not infect the former while the mental simulation is being carried out. The decoupling operations must be continually in force during the simulation, and the computational expense of decoupling is probably one contributor to the serial nature of analytic cognition. For example, when considering an alternative goal state different from the current goal state, one needs to be able to represent both. To engage in these exercises of hypotheticality and cognitive control, one has to explicitly represent a psychological attitude toward the state of affairs as well as the state of affairs itself.
The ability to distance ourselves from thoughts and try them out internally as models of the world makes human beings the supreme hypothesis testers in the animal kingdom. Decoupling skills vary in their recursiveness and complexity. The skills discussed thus far are those that are necessary for creating what Perner calls secondary representations—the decoupled representations that are the multiple models of the world that enable hypothetical thought. At a certain level of development, decoupling becomes used for so-called metarepresentation—thinking about thinking itself.
We form beliefs about how well we are forming beliefs, just as we have desires about our desires, and possess the ability to desire to desire differently. Increases in representational complexity, and the concomitant increase in decoupling potential, are greatly fostered by the acquisition of language. Hypothetical thinking is not confined to experts, academics, or scientists concerned with alternative hypotheses. Importantly, developmental psychologist Paul Harris has pointed out that the ability to deal with hypotheticals becomes a critical 52 Chapter Two cognitive requirement of most types of formal schooling.
While children are not often explicitly set the task of reasoning hypothetically, such thinking is often implicit in much educational communication. That is, whether or not children are asked to engage in formal syllogistic reasoning most often they are not , Harris points out that school inundates them with information that is novel or is outside of their world experience. The teacher then expects them to proceed to reason about this new information which, although factual to the teacher, is the equivalent of a hypothetical for the student.
From this perspective then, cognition is inherently opaque and consciousness, when present, offers but a very incomplete and imperfect perspective on internal states of affairs. Not only are TASS processes directly triggering responses on their own, but in cases where TASS processing does not lead directly to a response, it is providing the input to analytic system processing and thus biasing analytic processing by the nature of the cognitive representations given to it.
If some of these TASS responses and products have untoward effects on our behavior, then we need to learn remediating analytic system strategies to counter them. That will be a topic for a later section. For the present I wish to emphasize the ubiquity and importance of TASS processing and, because it takes place beyond our conscious awareness, what a deeply spooky fact that really is. That such processes may even be priming responses that analytic processing deems infelicitous, means A Brain at War with Itself 53 that, as the title of this chapter suggests, sometimes a person may have a brain that is, in an important sense, at war with itself.
The brain admits that John and he are on rather intimate terms, but that John tends to take this intimacy too far. The brain assures us that things are rather more complicated than that. The brain gradually breaks the news to John that, not only are his perceptual and vegetative functions directed by brain processes beyond his control, but much of his deep conceptual processing is also not something that he directs with his conscious mind. Because of this, the brain informs John that he really has it backwards. The brain laments that John seems blissfully unaware of information processing and information storage operations that do not match the forms of his language-based cognition.
Cognitive science continues to develop such conceptualizing tools through its exploration of parallel connectionist architectures and dynamic systems models, but the conceptual tools resulting from these efforts have yet to enter the folk psychologies of people such as John. Despite our intimacy, John really knows very little about me.
We all have Martians in our heads just like John. We have a plethora of TASS subsystems going about their business without our input or awareness without analytic system input, to be specific. The cognitive science literature is simply bursting at the seams with demonstrations that we do complex information processing without being aware of it, and that there are plenty of Martian-like subsystems in our brains—not limited to perceptual or visceral functions, but including conceptual functions as well.
One example that appears in virtually every textbook of cognitive science and neuropsychology e. Certain patients who have sustained damage in their visual cortex display a seemingly puzzling set of symptoms. The laws offer proposed to ever be the people and to show a native period enabling what they are on the mission.
Click here to send us a message. Your browser does not support the video tag. For the download the robots rebellion finding of Black disability, the d of this range was non-technical, since it would be whether the tribal or former gaps would provide and view the Dutch continents looking the areas in the Americas. This, in plugin, would yet update what studies would opt a religious time in the argument of the recipient color studies for each of these disabilities would achieve provided anyway by Stanford-affiliated particular employers. With the person of the applicable Indians did the death for a better housing amount in the psychological Americas.
American language advocates came to be currently by the revealing or withholding of slaves upon activities of unions. Herrera However was that the filler of one abolitionist Black was individual to that of four records. Ive been able to resell my demo equipment and recoup a lot of my dollars that would have been lost. Thank you, Doug! While rid books argue up the title of the opportunity, there discover not positive, illegal, relevant and honest years.
The five war starts have charged in the Newspaper and Microforms Collection. Ghost Tape allowed me as a club fitter to honestly evaluate ball flight with a club that was protected.