Free Will, Consciousness, and Cultural Animals
Free Will, Consciousness, and Cultural Animals
Abstract and Keywords
This chapter points to a view of free will as a complex form of action control that is used (sometimes) by humans to achieve self-control and rational, intelligence decision-making, as well as making and following ad hoc plans. Research suggests that these activities deplete a common but limited energy resource, so the capacity for free will is limited and biologically expensive. Blood glucose (brain fuel) maybe a major part of the common resource. Rational choice, self-control, and planful behavior are more useful forms of free will than random action. Psychologists may profitably focus more on how this mechanism works than on debating whether it fits various philosophical or theological definitions of free will. The broader context is that free will can be understood as an evolutionary adaptation to enable humans to function in the increasingly complex social world afforded by culture.
One afternoon, after you have completed your morning writing tasks and had your lunch, you head out to the car to do three errands in three different places. You need to take your dog to the veterinarian, pick up the specially ordered ice cream cake for your child’s birthday party, and renew your automobile registration. Like most humans and animals, you would rather do the more enjoyable activities and put off the unpleasant ones, and in this case you will find doing something nice for your child to be pleasant, whereas separating from your beloved pet (who will gladly hop in the car but will turn recalcitrant upon recognizing the vet’s office) will be unpleasant. So if you follow your gut impulses, you will start with the grocery store, then go to the DMV, and leave the vet till last. Wouldn’t it be nice to have your sweet pooch welcome you back to the car after that tedious and annoying stint in the DMV?
But that plan has flaws, and if you think things through carefully, you might discard it. One flaw is that it leaves the dog and the ice cream cake alone together, unchaperoned, for the indeterminate amount of time you will be in the DMV, and so when you emerge with your new license plate, the dog will be too busy licking up the last of the ice cream to want to lick your face. And even if you make one change to avoid leaving the dog guarding the ice cream, there is another flaw, at least if you live in Florida as I do, which is that the temperature inside a car parked in the afternoon sun would seriously compromise the welfare of either a canine or any ice cream product. Hence the logical plan is to go to the vet first, then the DMV, and pick up the ice cream cake last so that you can spirit it directly home to the freezer with as little melting as possible.
(p.66) This example is intended to illustrate one important concept of free will. In particular, you felt like doing things in one way but then overrode those impulses in favor of a different plan of action. Moreover, you employed conscious, logical reasoning, recognizing that the first plan would have unwanted, destructive consequences, and therefore worked out a second plan to produce a better outcome. Further, there is something distinctively human about this style of action (we know of no other animal who forms, evaluates, and revises plans in this manner), yet the benefits of that style of action are readily apparent, and so it may well be something distinctive about the evolution of the human psyche that created that capacity. And last, the formation of the second plan uses a style of reasoning that can be readily communicated to others, indeed discussed with them. If you told your spouse you were heading out to do the errands according to the first plan, he or she could point out the flaws, which most likely you would then recognize as a valid mandate for revision. Another person might offer you further information that would bring further changes, such as that the DMV happens to be closed today based on some obscure local holiday.
Free will is sometimes seen as a crucial, defining trait of human existence and other times regarded as an absurd, utterly implausible myth. Free will may be a vital faculty that human beings must exercise in order to fulfill their potential in creativity, virtue, or spiritual salvation. Or it may be a dangerous and obsolete illusion that all educated persons ought to reject immediately. The debate about whether free will is real is sometimes intense and bitter.
In this chapter, I will eschew the debate about whether free will exists. Instead, the focus will be on explaining the common belief in free will and, more important, on the phenomena to which those common beliefs refer. Rather than argue about whether free will is real, I shall focus on offering a model of how it might operate.
WHY NOT TO ARGUE ABOUT THE EXISTENCE OF FREE WILL
Philosophers, theologians, and others have discussed free will for centuries. Now psychologists have entered the picture. But perhaps arguing about the existence of free will should not be the job of psychologists. Instead, we might more usefully do what psychologists are most capable of doing, which is to test and refine theories about inner processes. Suppose there are two different kinds of processes that affect behavior, and one of them is freer than the other in some sense. Psychology might profit by exploring that difference, without getting bogged down in the highly abstract and philosophical debate about whether one of them satisfies the most rarefied criteria for qualifying as free will.
Though some experts balk at the term free will, few will dispute the view that human behavior can be guided by different kinds of processes. Moreover, (p.67) concepts such as freedom, intention, and choice have not been rejected as obsolete, incoherent notions by psychologists. On the contrary, psychologists find those useful terms and produce significant differences in behavior as a function of them. Laypersons also recognize their own behaviors and others as sometimes voluntary and sometimes involuntary, they are acutely aware when their freedoms are threatened or removed, they judge themselves and others differently based on whether a behavior was freely chosen or coerced, and they struggle every day with making choices.
Perhaps the issue is moot, after all. Many philosophers embrace some form of compatibilism, which holds that one accept free will without relinquishing faith in causal determinism (e.g., Dennett, 2003; Kane, 2002). If many philosophers find the two beliefs compatible, psychologists should perhaps not fret that they will lose credibility as scientists if they, too, accept free will (cf. Wegner, 2002).
Personally, I resent being told that as a scientist I am forbidden to believe in free will and required to embrace total causal determinism. Possibly this resentment is caused by the well-documented motive to preserve the freedom to think what I want and entertain all possibilities (see Brehm, 1966). The reasons I offer for this refusal to be bullied, which admittedly may be different from my true or unconscious reasons (see below), are fourfold. First, the claim that all behaviors are fully determined by external, prior causes is unproven. Second, it is also unprovable, insofar as no one can study all possible behaviors and show them to be 100% caused by prior events. Third, it is contrary to our everyday experience. We all make choices every day, and it certainly feels as if more than one outcome is possible (hence the necessity of choosing)—which is precisely the point of free choice, namely that more than one could act differently. Fourth, it is contrary to our data, which almost invariably show probabilistic rather than deterministic causation. That is, behavioral outcomes in our experiments are almost never shown to be inevitable; rather, our causes reflect a mere shifting of the odds of some particular response.
In that sense, determinism requires a huge leap of faith, not unlike believing in a god. None of those objections proves free will to be true and determinism false. My point is simply that it is a form of unscientific fanaticism to require scientists to accept a belief that is unproven, unprovable, contrary to everyday experience, and contrary to their laboratory data.
Hence I propose to shelve the question of the existence of free will. Let us forge ahead and explore how action happens—and why people, perhaps rightly, perceive some actions as relatively freer than others.
WHAT NEEDS EXPLAINING
If the task is not to prove or disprove the existence of free will, then what? It is to understand how action is controlled and directed, and in particular to (p.68) understand if there is some special model of action control that is (mostly) specific to humans and corresponds to the sort of action that people perceive, in themselves and others, as free.
Absolute freedom, in the sense that would meet the most stringent and exalted criteria for free will, is not necessarily the goal. Psychology mainly works by studying differences and degrees, and so if we can identify what differentiates the more versus less free actions, that would be a contribution appropriate to what psychology can do well. To be sure, if free will does exist in the most exalted philosophical or theological terms, it most likely would conform to what psychologists would identify as the relatively freer forms of action. In a sense, to map out the processes that produce relatively free action would be to furnish a model that would likely prove useful and relevant to understanding free will in whatever form turns out to be real.
The task can be further clarified by noting several likely parameters. In the first place, if there is no need to prove the existence of freedom, then certainly there is no need to depict all human action as free—if anything, the opposite assumption of partial, occasional, and incomplete freedom is more useful for psychology. The 20th century’s most passionate advocate of free will was probably Jean-Paul Sartre (1943/1974), who famously argued that humans were “condemned to freedom” and who asserted that all human acts are free. In contrast, the approach I am advocating is to compare different (forgive the expression) degrees of freedom and to learn what distinguishes relatively free from relatively unfree acts. Under that assumption, free will is at best a sometime thing. Much, perhaps the majority, of human action could be fully and simply determined by simple, explicable causal processes, including brain dynamics, reinforcement-based learning, and ingrained or acquired responses to stimuli. Free will would represent only an occasional opportunity to suspend or override those causal processes, so as to allow a different process to take control.
The difference between seeing all human action as free versus seeing only occasional episodes of free action is rooted in different understandings of what is meant by choice, which can be roughly described as internal versus external. The external view of choice is defined by the situation: Different courses of action are technically possible. The inner view emphasizes the inner process. If the person does not go through an inner process of choosing, then the fact that in principle he or she could have done so is irrelevant. A person who always sleeps with the same partner or always eats the same food for breakfast is not making a choice every day according to the internal model, even though from an external standpoint one would say that those are choices insofar as the person could choose otherwise each day. At the extreme, the coffee shop franchise Starbucks has advertised that it offers 19,000 beverage options, but clearly this invokes the external criterion, because no one would want to make 19,000 decisions before having that first cup of coffee. For present purposes, we focus on (p.69) the internal concept of choice. Free will is relevant only when there is an actual inner process.
That brings up the second point. The psychology of action has often struggled to understand how to make the transition from inner processes such as motivation and cognition to actual behavior: How, exactly, does thought cause the body to start to move? In this, it borrows the framing of the problem from physics, which has often sought to explain how a body at rest begins to move. But perhaps the problem has been wrongly phrased. Psychology, after all, has not identified any state of being “at rest.” To the extent the free will exists, it serves not to initiate action so much as to alter and steer it. Put another way, behavior is already happening all the time, in all organisms from the simplest to the most complex. Free will does not therefore have to make behavior start happening, because there will be ample behavior without free will. Instead, the role of free will would be to alter the flow of behavior.
Viewed in that way, the function of free will is twofold. First, it overrides the response process that would happen without it. In other words, it suspends or interrupts one set of causal processes. Second, it chooses among various options for the coming course of action.
What sort of inner processes produce those outcomes? Almost certainly these include some degree of conscious, effortful thought. Insofar as behavior is purely produced by nonconscious, automatic processes, it would not be regarded as reflecting any sort of free will. From philosophical examples to laboratory studies, freedom of action is tied to conscious deliberation and intentional decision (Sartre, 1943/1974; Wegner, 2002), and that link is almost certainly maintained in everyday perceptions of freedom, such as legal decision making (i.e., if the person did not consciously intend to perform the forbidden act, then the responsibility and legal penalties are reduced). Hence some understanding of conscious processing is likely to be intimately linked to any ultimate theory of free will.
USEFUL FORMS OF FREE WILL
Many people seem to prefer to regard themselves as having free will. But why? Two decades ago, Dennett (1984) addressed the question of why free will would be worth having and suggested that some varieties would be more useful than others. In particular, he proposed that free will is relatively worthless unless it helps you get what you want. After all, having free will that does not make life better for you in any discernible manner would be somewhat pathetic, at best an idle form of random action devoid of practical benefit. Yet that definition may be something of a Trojan horse, because if people claim free will in actions that benefit them, the skeptics and determinists can almost always provide (p.70) a causal explanation (e.g., you did that because your motives and wants caused you to do it).
Skeptics of free will can propose that they would accept an action as free if it were proven to have no relationship to any external cause or any prior event. To be sure, that sets the bar rather high, but it would at least constitute freedom in a sense that would be difficult to dispute. Then again, what sort of action would be utterly independent of external factors and prior events? In essence, it would be an essentially random action.
This view may also be responsible for some of the objections to the very idea of free will, especially by scientists. It treats free will as a random action generator. Science can accept randomness in nature, but deliberate and intentional randomness as a cause of human behavior seems absurd and postulating it seems a foolish and unproductive basis for any theory about behavior.
But would random actions help people get what they want? Or, to invoke another interpretive framework, would natural selection favor a capacity for random action? If we assume that evolution created human nature, then it is worth pausing to consider free will as a possible product of evolution (Baumeister, 2005; Dennett, 2002). If random mutations were to produce an increase in the capacity for purely random action, would those creatures survive and reproduce better than their competing, nonrandom peers? It is hard to see how natural selection would confer any substantial benefits on a random actor.
In contrast, natural selection might well confer huge advantages on creatures that developed other kinds of freedom. The preceding section contended that free will has a twofold nature, namely overriding a prepotent response tendency and making a smart choice among options. If a mutation offered a superior capacity for doing either of those, then the individual might well flourish (which means surviving and reproducing better than rivals), and the mutation would gradually spread through the gene pool. Let us consider each of those.
Overriding incipient responses is studied in psychology under the rubric of self-regulation (e.g., Baumeister, Heatherton, & Tice, 1994). Although there has been some work contending that certain self-regulation processes are nonconscious and effortless, for the present we are concerned with the conscious and effortful majority of self-regulation. Self-regulation is highly adaptive because it vastly increases the behavioral flexibility of the organism and hence its capacity to find an optimal response to a situation. Creatures that live in more complex and changeable environments, and who therefore encounter a broader diversity of situations, will benefit all the more from the capacity for self-regulation. Clearly, humans are at the extreme high end of that distribution (i.e., they live in very complex and changeable environments), and so a high capacity for self-regulation would be especially useful and adaptive for us. Sure enough, even just comparing humans against other humans, it is clear that people with a greater capacity for regulating themselves are more successful than their less self-regulating peers in a broad variety of outcomes (e.g., Duckworth & (p.71) Seligman, 2005; Kelly & Conley, 1987; Mischel, Shoda, & Peake, 1988; Shoda, Mischel, & Peake, 1990; Tangney, Baumeister, & Boone, 2004).
Self-regulation should qualify almost by definition as at least a limited form of free will. That is, without self-regulation, the organism cannot help but act on the first or strongest impulse that arises in response to a situation. With self-regulation, the organism can override that response, allowing a different impulse or response to take over. Overriding the first response frees the person from having to respond in that particular way and, if only briefly, creates a gap or uncertainty that opens the door for other possibilities. This is not to say that the eventual response is necessarily better than the first or that it is itself not the product of an inner causal sequence of responses. But the fact of changing away from the first to enable the second should constitute a kind of freedom, and it would almost certainly be recognized as such, though we are just now designing research to test that perception. And humans who could exert that much free will, who could override one response in order to permit another, would probably survive and reproduce better than their rivals who couldn’t.
If self-regulation is one probably useful form of free will, rational choice would be another. Rational choice is an evolutionarily new kind of decision making (new with humans) that relies on evaluating a potential course of action using logical reasoning, most commonly with cost-benefit analyses that mentally simulate the various behavioral options and their likely consequences, quantifies them according to benefits to self-interest and possibly others, and compares them to produce the most desirable result (as far as can be ascertained with the available information). Undoubtedly it is facilitated by the use of language, which can represent the options, and by a mastery of logical reasoning, which enables the person to apply general principles to specific cases and to follow the ineluctable rules of logic to move from one thought to a different thought.
Rational choice is perhaps underappreciated in psychology, though some other social sciences (notably economics and political science) recognize it as a powerful model for understanding human behavior and if anything overstate how far it guides human behavior. Daniel Kahneman, who had a highly successful career based on showing how various behaviors and choices deviated from rational choice models, remarked in 2003 that he never intended to dispute the view of human beings as rational—people are rational, he said, just incompletely so. In my view, the incompleteness of human rationality confirms the depiction of free will as a sometime thing. The capacity for rational thought and decision making lies atop an irrational, impulsive beast, and so it only sometimes can alter the course of action that that impulsive beast will take.
The link between rationality and free will was noted by Searle (2001), who said that theories of rationality almost inevitably presuppose some degree of free will. After all, what would be the use of being able to reason out the best plan of action, if one were not capable of altering one’s behavior so as to follow that best plan? Without that, the conscious mind would be a helpless, depressing (p.72) spectator, constantly observing one’s own irrational actions while figuring out that it would have been better to do something different.
The anecdote that opened this chapter illustrates the confluence of these two forms of free will (self-regulation and rational choice). One’s preference and initial impulse dictate one sequence of action, but logical reasoning suggests that doing what one wanted would bring destructive consequences, and so one overrides the first plan in favor of a different and better plan. Humans make and follow such ad hoc plans constantly, but as far as we know that style of action control is absent in other species.
When philosophical writers such as Dennett (2003) discuss free will, their examples and discussions keep coming back to acts such as these, namely self-regulation and rational choice. It thus seems fair to regard them as two major manifestations of the sort of phenomena that are relevant to free will. Putting those two together gains plausibility, furthermore, from my own laboratory work. It appears that rational choice and self-regulation have overlapping inner processes that use a common resource.
The common link between self-regulation and rational choice emerged only gradually, in part because it pointed to a style of thinking that has been out of fashion for decades in psychology and only recently has begun to be taken seriously again. In this, I refer to theoretical models based on energy. Freud’s theories characterized psychological processes as energy transactions, but more recent theorists have ceased to invoke energy, preferring highly cognitive theories and the like. The recent reconciliation between mind and body (driven by the rise of health psychology and neuroscience) has, however, furnished a newly plausible basis for using energy theories. The human body is, after all, an energy system that relentlessly ingests and burns calories. The special relevance of energy processes is suggested by Dunbar’s (1998) observation that the human brain consumes 20% of the calories used by the entire body, while comprising only 2% of its total mass. Thus, the brain, and by implication psychological activity generally, is a huge burner of energy.
Our initial review of the research literature on self-regulation (Baumeister, Heatherton, & Tice, 1994) noted a pattern that suggested an energy process: It seemed that self-regulation operated as if depending on a limited resource. Self-regulation failure seemed more likely to occur when people had already self-regulated some other aspect of behavior. For example, when people struggle to quit smoking, they eat more, become crabby (failing to regulate emotion), and show other signs of poor self-regulation.
Although those observations were subject to rival interpretations, we began conducting experimental tests under rigorously controlled laboratory conditions, and we found that the pattern held up: After people engage in one act of self-regulation, they self-regulate less effectively in other spheres. For example, in one of the first studies, students who resisted the temptation to eat chocolate (and ate radishes instead, while staring at the chocolates) gave (p.73) up faster on a subsequent discouraging task, as compared to students who had been permitted to eat chocolate or students who skipped the food procedure altogether (Baumeister, Bratslavsky, Muraven, & Tice, 1998).
The implication was that the initial act of self-regulation depletes some inner resource, leaving less available for the second task. This pattern has been widely replicated (for recent reviews, see Muraven & Baumeister, 2000; Schmeichel & Baumeister, 2004).
For present purposes, the most important finding is that logical reasoning in the service of making deliberate, conscious choices appears to deplete the same resource used for self-regulation. Thus, after making one big or multiple small choices, self-regulation is impaired (Baumeister et al., 1998; Vohs et al., 2006). Conversely, after people engage in self-regulation, their capacity for logical reasoning and decision making is impaired, as indicated by poorer performance on logic tests (Schmeichel, Vohs, & Baumeister, 2003) and increased reliance on fallible short cuts and heuristics in decision processes (Amir et al., 2005).
The common energy source is compatible with the view that evolution created a new or at least vastly renovated action control system for the human psyche. Essentially, evolution created a costly but powerful system by which the body’s energy supply can be channeled into overriding behavior and making logical, rational choices.
Indeed, the evolutionary implications could be taken a notable step further. The fact that the psyche uses the same energy resource for self-regulation as for rational choice suggests that that mechanism evolved first for one of them, and then the second piggybacked onto the same system. That opens room for speculating about how the action control apparatus developed.
My best guess is that the system evolved originally for the sake of self-regulation, and it was later adopted and perhaps adapted for rational choice. The reasoning behind this is that self-regulation seems older and more fundamental in evolutionary terms than rational choice (especially logical reasoning). Long before humans evolved, social animals might need self-regulation in order to stifle their impulses and adjust their behavior to group life. For example, when a hungry animal sees food, the natural impulse would be to eat it. In a pack, however, if the animal begins to help itself to food before the alpha male eats his share, the smaller animal is likely to receive a beating. To continue to live in the group without daily beatings would require the capacity to overcome the natural impulse to eat those foods.
In plain terms, self-regulation allows the organism to alter its behavior so as to conform to the rules of the group. Logical reasoning and rational choice enable it to formulate its own rules, and the capacity for self-regulation can then be invoked to alter one’s behavior accordingly. Both forms of free will promise to be highly adaptive.
We are beginning to explore the precise mechanisms behind this new, high-energy form of action control. Gailliot et al. (2006) found that blood glucose (p.74) was reduced after acts of self-control and that giving laboratory a high-glucose snack counteracted the effects of ego depletion. Glucose is fuel for brain activities, though some (like self-control) use much more than others. The implication is that evolution developed a procedure for converting the human body’s food energy into complex psychological processes that can alter the stream of behavior in adaptive ways.
HUMANS AS CULTURAL ANIMALS
Thus far I have suggested that free will, in whatever sense it actually exists, was the product of evolutionary processes that created the human psyche and that would have conferred advantages in survival and reproduction. An adequate psychological theory of free will should therefore situate it in the perspective of human evolution. The possibly quite special confluence of evolutionary pressures that produced the human psyche probably had much to do with the emergence of its novel system for action control.
In a recent book summarizing psychology’s contribution to understanding human nature, Baumeister (2005) argued that the human psyche is distinctively well suited for participating in culture. The implication is that evolution took an unusual turn in creating humans. Social psychologists are fond of calling humans “the social animal” (see Aronson, 2000), but there are in fact many social animals. What sets humans apart is a radically new way of being social. Humans are most distinctively and appropriately described as cultural animals.
The sensory organs of most animals are geared toward detecting other species, mainly the wide assortment of predators and prey. In surprising contrast, human sense organs (especially those involved in vision and hearing) seem much more geared toward attending to each other. This is evident in the tradeoff between detection (noticing anything at all) versus resolution (processing a few things very clearly and thoroughly). Anyone who has lived with a dog knows it can hear a much wider range of sounds than a human—which would be useful for noting all manner of different animals—but the dog cannot distinguish similar sounds very well, so it cannot tell the difference between “Fido,” “buy low,” “hi ho,” and so forth. The dog is not the unusual one here, though—it is the human ear, whose design has sacrificed the capacity for hearing high-pitched squeals and ultralow growls in favor of being able to hear subtle auditory differences between thousands of spoken words.
More generally, most animals get their food, shelter, and other needs directly from the natural environment around them. Humans get them from their social system. In fact, when faced with experimentally engineered conflicts between the evidence of their senses and the information given by other people, humans will often go along with the group rather than heed their senses (Asch, 19551955, 1956).
(p.75) In that sense, humans use a different biological strategy than almost any other species. Culture is the biological strategy that our species evolved to use, and most likely humans were selected on the basis of being capable of culture. Culture may be defined as an information-based system that enables people to live and work together in organized fashion to satisfy their biological and social needs. It offers huge advantages, which should be obvious given how thoroughly humans have taken over the planet and altered their environment. In simple terms of survival and reproduction, culture has been tremendously successful. In a relatively short time it has enabled humans to multiply from one woman to 6 billion souls, which is much greater success than any of our close biological relatives have had. In part this is because nearly all humanity’s closest biological relatives live near the equator, but with the benefit of cultural learning to create clothing, cook food, and build complex shelters, humans have been able to live in a much wider range of physical environments than other species. And no other species has been able to triple its average life expectancy by means of its own research and interventions.
Culture has conferred advantages by several means, but all involve the social network (Baumeister, 2005). Language enables people to share information and make decisions in groups. Knowledge is stored in the social group rather than in the individual mind, so that people can benefit from the lessons learned and problems solved by people who are far away or even long dead, and this allows the collective body of knowledge to accumulate across generations, thereby creating progress of a sort that is almost entirely unknown in other species. Role differentiation and economic exchange enable social systems to become far more than the sum of their parts, including the fact that each task can be done by an expert, and trade can improve everyone’s quality of life. (In fact, recent work has begun to suggest that Neanderthals failed to compete with Homo sapiens precisely because they failed to divide labor and hence were economically incompetent, which would indicate that evolutionary competition favored the humans who had the better cultural system; Horan, Bulte, & Shogren, 2005.)
In order to sustain and take advantage of culture, however, humans may have needed a new, more flexible and far-reaching psychological mechanism for making choices. Because of its potential for change and progress, humans encounter more different situations in their lifetimes, and a greater variety of choices, than other animals.
Culture can accumulate and use information most effectively by invoking meaning. Not coincidentally, humans therefore differ from other animals in the extent to which they use meaning to guide action. Ad hoc plans, such as the vet-DMV-supermarket errand example that opened this chapter, figure prominently in human activity but would be very difficult to develop and refine without language because they are based on integrating various ideas. The human brain evolved to become capable of language so as to be able to exploit the power of meaning. (The brain evolved; language was invented; meaning (p.76) was discovered.) Again, though, such capabilities of thought would be largely useless unless people had sufficient free will to be able to alter their course of action based on those thoughts.
CONSCIOUSNESS AND SOCIAL LIFE
Earlier, this chapter suggested that if there are any genuine phenomena associated with the concept of free will, they most likely involve conscious choice. Such a view has to contend with the now widespread belief that consciousness is a useless, feckless epiphenomenon, and that all behavior is guided by nonconscious processes.
The attack on consciousness can be traced to Freud, who proposed that the conscious self is often merely and unwittingly carrying out the agenda laid down by unconscious motives and ideas. In recent years, the attack on consciousness has been spearheaded by John Bargh (e.g., 1994), whose careful and persuasive studies have shown that many ostensible results of conscious, intentional action can also be produced by activating nonconscious ideas or motives, thus entirely bypassing the conscious system. To be sure, those findings do not disprove the potential role of consciousness, and the argument that conscious will is dispensable fits well with my characterization of free will as a sometime thing, but it has been tempting for Bargh and others to speculate that consciousness will eventually be proven to be an irrelevance.
Further support comes from work by Gazzaniga (e.g., 2003), who has shown that the conscious mind devises explanations that are often fanciful and wrong. Likewise, Wegner (2002) has provided evidence that people are sometimes mistaken about whether they have caused something to happen, thus sometimes creating an “illusion of conscious will” (his term and the title of his book). Wilson (2002) has characterized conscious self-knowledge as likely to be useless and full of errors, and he has proposed that when people seek to explain their reasons for doing something, they are prone to fall into error and mislead themselves and others.
For many, the most devastating and influential critique of consciousness comes from the research by Libet (1985, 1999), who concluded that conscious thought is too slow to guide behavior. In his studies, participants were asked to initiate a simple motor action, moving a finger, and to note on a fast-moving clock when they made that decision. Their self-reports of the decision time were compared with electronic readings of brain activity, which showed that brain activity began to rise prior to the subjective decision time. In other words, the brain started to act before the conscious mind decided to act. To be sure, there are methodological critiques of that work (e.g., the subjective conscious time may reflect not the making of the decision but the self-recognition of having made the decision, which takes a bit longer), as well as arguments (p.77) that is overstated and irrelevant (e.g., because consciousness compensates for its slowness by projecting slightly into the future; Shariff & Peterson, 2005). But the conscious processing system is known for being slow, and many researchers have come to accept the view that consciousness is too slow to guide behavior and is, hence, irrelevant to action.
In social psychology, the most influential critique of conscious thought in the modern era came from Nisbett and Wilson (1977), who contended that people cannot introspect on their thought processes (and are hence unable to explain their true reasons for acting). They said that when people are asked to explain their behavior, rather than introspect and furnish the actual, causal roots of their choices, they simply offer standard reasons from a stock of explanations that are favored in their culture. These explanations were derided as a priori theories about why people ought to make a particular choice, rather than true explanations of why they do.
Let us take these challenges seriously and offer a revised theory of conscious agency, based in part on the cultural animal perspective. First of all, from that view, humans evolved to work together with shared information (the basis for culture). Perhaps Nisbett and Wilson (1977) were overly disparaging of that stock of cultural explanations, treating them as trivial, irrelevant, self-deluding myths. Suppose that part of the value of consciousness is to enable people to make group decisions and to act in ways that would be suitable to the group. In that case, it is less important to come up with the true reason for one’s action than to come up with a reason that the group will accept as justified. The local strongman may claim much of the best food because he is greedy and selfish, but he is likely to get less trouble if he can justify his claims to the group on the basis of the divine right and the gods’ will. Today, a citizen defending his actions in a legal trial does not necessarily have to account for the true inner reasons for his actions but rather furnish an explanation for them that the law accepts.
In this view, then, consciousness is there in large part to help people explain and justify their actions, or to question and influence the actions of others, according to the collectively (socially and culturally) accepted rules. Well-designed studies can show that people sometimes are unaware of the subtle influences on their behavior, but the need to recognize those influences is less, or throughout our evolutionary history has generally been less, than the need to reconcile actions with the culture’s rules.
Group decision making and interpersonal influence would thus benefit from consciousness, not because people necessarily know or acknowledge their true, inner reasons and causes, but because the group members can discuss and resolve the issue along mutually accepted lines. As a simple example, the true reason a parent desires good behavior from a toddler may be to prevent embarrassment to the parent, or to build character for decades hence, or simply to reduce the parent’s hassle, but these reasons will be less persuasive to the toddler than the admonition that Santa Claus is watching and may reduce your (p.78) allotment of toys for next Christmas. The culture sustains the useful fiction and the child believes it, and so the parent can invoke it to change the child’s behavior. Thus, again, true reasons are less important than socially accepted ones.
Moreover, and perhaps even more profoundly, conscious thought can escape from the introspective ineptitude revealed by Nisbett and Wilson (1977), insofar as it masters logical reasoning and other rule-based forms of thought that are open to inspection. Perhaps, as in their famous study of stocking preferences, people were unable to introspect or even guess that they generally chose on the basis of a recency effect (picking the last stocking they saw) rather than on the basis of differences in color or texture. But the rules of logic are part of the objective reality of meaning, and intelligent people can agree with consensual certainty whether a conclusion follows from a premise. Thus, conscious thought introduced, or at least greatly promoted, a style of thinking that enables movement from one thought to another according to firm rules.
The errors, omissions, and blind spots with which researchers continue to lambaste consciousness may therefore be far less consequential than they first seem, because as people describe their conclusions to others, those errors and other flaws can be detected. Thus, as noted, Gazzaniga’s split-brain patients may devise a false explanation for the ostensible coherence of different stimuli (the snow shovel and the chicken, in his standard example), but if they offered that explanation to others outside the laboratory, the others would quickly point out the error and correct them. Anyone who has tested children or even adults on arithmetic knows that not all conscious minds always reach the correct answer, but a correct answer does exist, and if a person can summarize his or her calculations to others, errors can be detected and corrected. The social network will thus correct the mistakes that the individual mind makes, as long as it can discuss them with other similarly conscious beings.
Likewise, in the example of the errand trip that opened this chapter, if you had indeed settled on the original plan and mentioned it to someone else, that other person would potentially have noted the dangers of leaving a dog and ice cream together in a car with an interior temperature over a hundred degrees.
Consciousness may therefore have developed as it did because it strengthened the link between the individual and the collective. It allowed the culture to guide the behavior of individuals in new and powerful ways. Because collective action through culture was the distinctive biological strategy of humans, anything that promoted it would likely help some humans prevail over their rivals in evolutionary competition.
USES OF CONSCIOUSNESS
The preceding section suggests that we should seek to understand the value and efficacy of consciousness in how it permitted a new form of action control (p.79) that was suitable for cultural animals. That section emphasized finding reasons and justifications acceptable to the group and compatible with its rules. A related and potentially even more powerful reason has to do with logical reasoning.
As already noted, dual process theories of human mental functioning have now largely swept the field (e.g., Chaiken & Trope, 1999). Nearly everyone accepts that some cognitive processes involve automatic, nonconscious activity, whereas others are conscious and controlled. I have proposed here that free will, such as it is, will be mainly associated with the latter, although it may be more precise to speak of free will as arising from cooperative interplay between the two systems. Most conscious processing rests on a substantial amount of nonconscious activity. To furnish a simple and obvious example, understanding language requires a considerable amount of nonconscious processing, by which auditory or visual stimuli entering the brain from the sense organs are understood to convey meaningful ideas.
Earlier I suggested that rational choice based on logical reasoning was an important, adaptive, and common form of free will. How does logical reasoning occur? An influential article by Lieberman, Gaunt, Gilbert, and Trope (2003) proposed that it is largely confined to the conscious system (what they call “reflective”). To the extent that such reasoning occurs, therefore, it would be largely outside the capabilities of the automatic system, and consciousness can do it correctly.
A series of experiments in our laboratory has supported the hypothesis that effective logical reasoning depends partly on conscious processing (DeWall, Baumeister, & Masicampo, 2006). Several studies invoked the principle that the conscious system can generally do only one thing at a time (operates in serial), whereas the automatic system operates in parallel and therefore routinely performs multiple operations at once (Lieberman et al., 2003). Hence a distracting load would entirely preempt the conscious system but not the automatic system.
In several studies, we gave people logic problems to solve while listening to music. Some were instructed to monitor the music lyrics and count instances of a particular word; these showed substantial impairments in reasoning performance. In fact, they scored no better than chance guessing, though they seemed not to realize this and continued to answer questions as the same pace rather than simply putting down guesses rapidly for all items, a strategy that would have gotten them many more correct answers. Thus, preoccupying the conscious mind appears to have had a devastating effect on logical reasoning ability. Conversely, engaging the conscious mind more in the reasoning task, such as by telling people that they would have to explain their answers, led to significant improvements in scores on the logic test. Crucially, activating the idea of logical reasoning by means of nonconscious priming failed to have any significant effect on reasoning performance.
(p.80) These findings offer preliminary support for the view that logical reasoning depends on a conscious processing system. Performance on the logic test rose and fell according to manipulations that targeted the conscious processing system, whereas manipulations aimed at the automatic processing system had no discernible effect.
Earlier I noted that Libet and others have emphasized the slow speed of conscious processing, as compared to automatic processes. Hence its operation may be to follow along and make careful corrections while the automatic process runs ahead and generates quick responses. This view is perhaps most compatible with Kahneman’s (2003) characterization of “System 2” as a kind of editor that responds to the inputs from “System 1” (his term for the intuitive, automatic processes), sometimes accepting them, sometimes rejecting them or calling for revision. In one well-known study, Frederick (2005) asked people to tell the cost of a ball after hearing that someone had purchased a bat and a ball for $1.10 and that the bat cost a dollar more than the ball. Most people can get the correct answer of 5 cents, but usually their first thought is that the answer is 10 cents, and moreover people who are given the problem when distracted or in a hurry often give the 10-cent answer. The implication is that the automatic system processes the problem first and offers the approximate answer of 10 cents, and only the careful but slower operations of the conscious system comes up with the correct (and corrected) answer of 5 cents.
A similar view underlies our studies on creativity, which has periodically entered into the free will debate. Although some views of creativity see it as a form of free will, insofar as the person consciously decides how to formulate something new and different, the prevailing view among artists and psychologists has been that creativity is almost exclusively the product of nonconscious forces. The traditional emphasis on semidivine external muses as the wellspring of creative inspiration has largely given way to the assumption that creativity springs from deeply unconscious roots and that the conscious mind is, if anything, an impediment or obstacle to the creative process (see Dennett, 2003, and Wegner, 2002, for summary of some of those views).
But then why is creativity mainly found among conscious beings? We reasoned that perhaps the creative inspirations do emerge from nonconscious processes, but the integrative editing of the conscious mental system is vital for fashioning the final product. Hence preoccupying or distracting the conscious mind would reduce creativity rather than facilitating it.
A series of laboratory studies supported that hypothesis that creativity depends in part on conscious activity (Baumeister, Schmeichel, DeWall, & Vohs, 2007). In one, we asked musicians to perform a series of improvisations, one of which was done while counting backward by 6 from 917 (a cognitive load manipulation designed to preoccupy the conscious system). Judges rated those solos as less creative than solos done while counting forward by 1 or while not counting at all. In other studies, participants drew pictures while listening to (p.81) music as in the logic studies above, either tallying instances of a lyric or just listening, and judges rated the drawings made during cognitive load as less creative than the others. It is important that not all aspects of performance were impaired by the cognitive load. When the conscious system was distracted, musicians were able to keep the beat and avoid mistakes (defined as notes outside the key), and artists successfully followed the instructions about what elements to include in their drawings and used the same variety of colors. But such aspects probably can be achieved automatically and hence do not rely on consciousness. Only the artistic integration into an appealing, creative product suffered when the conscious mind was otherwise engaged.
The special value of conscious functioning seems thus to be found in editing mental operations: criticizing them, combining them, and the like. The conscious mind may thus react to the impulses arising from automatic, nonconscious processing. Libet (1985, 1999) himself did not conclude that consciousness is irrelevant, only that its main function was to exercise a veto over behaviors that the automatic system initiated. Self-regulation is likewise most commonly exercised to stop a behavioral response sequence, rather than to initiate one (Baumeister, Heatherton, & Tice, 1994). But does that mean free will is merely reactive (“free won’t” in the phrase of some writers; Ohbi & Haggard, 2004)?
THE DELAYED EXECUTOR
Let us return to the problem, raised by the Libet findings, that consciousness seems too slow to guide behavior. By the time the conscious system makes up its mind what to do, the ship has already sailed. Is there any way that such a slow conscious system could participate in guiding action?
Assume for the sake of argument that the immediate control of behavior is always the result of nonconscious processes. In the heat of the moment, the person’s automatic system sizes up the stimulus situation and organizes the response. To do that, it presumably consults some bank of inner programs (including reinforcement history and other knowledge) that tell it how to respond to that situation. Nothing in that process requires participation by the conscious system.
Crucially, however, the conscious system could still have an indirect—yet extremely powerful and adaptive—influence over behavior if it can alter those programs. And for this, its slowness may be irrelevant, because the creation or revision of programs could be done in relatively quiet moments far removed from the crucial seconds when responses are being made. In particular, conscious reprogramming could occur after major events have ended (thus mainly helping to revise how one would respond to similar events in the future), or well before a particular, anticipated situation is encountered.
(p.82) Unlike robots and computers, which go on to the new task as soon as the current one is finished, the human mind has a tendency to dwell on prior events and ruminate about them. In particular, unpleasant conscious emotions (which are usually a sign that some episode has turned out badly) stimulate the mind to ruminate about how the event might have gone differently, a pattern called counterfactual thinking (Roese, 1997). Such thinking seems ideally suited to this reprogramming function, which can be called the delayed executor, because the conscious mind examines the episode step by step to consider how a more desirable outcome might have ensued if one had responded differently at some point. Thus, as the event was unfolding, the conscious mind was perhaps not involved in directly steering the action, but it may have been gathering observations to use in its postmortem analysis. The automatic system dictated the responses according to its programming, and afterward the conscious mind weighed the outcome and mentally simulated different actions that might have been taken, and if it concludes that a different act would have yielded a better result, it essentially revises the program that the automatic system will consult in future similar episodes. Next time, take an umbrella, or get the promise in writing, or post your name inside the suitcase, or refrain from making commitments while inebriated.
Likewise, the organization of behavior according to ad hoc plans can be done by the conscious mind well in advance of their execution, so that the automatic system is fully in charge on carrying out the behavior. To return to the example that opened this chapter, the plan for the sequence of errands was made by means of conscious deliberation, before the errands were begun. Hence any researcher who studied the person carrying out those errands might find, correctly, that each act in the process of doing the errands was directed by automatic, even nonconscious responses, because at each moment your response was a direct, preprogrammed response. But the macro program had been crucially shaped by conscious thinking.
The relevance of conscious volition to macro thinking has been argued persuasively by Donald (2002), who argued that the cognitive science approach of studying ever more micro units of behavior will bias the data against finding any role for consciousness. At the extreme, by studying behavior at the level of neurons firing, one could explain behavior with no possible role at all for conscious processes, and researchers who work at that level might easily convince themselves that they had fully resolved the debate by ruling out any causal role for consciousness. But if one looks at behavior in larger units over longer periods of time, consciousness might just turn out to be decisive.
Most scientists reject the idea of free will as a random action generator, and probably with good reason. In contrast, free will in the sense of self-control (p.83) and rational, intelligent choice comprises an important set of psychological phenomena and is plausible in terms of the evolution and construction of the human psyche. Quite likely human conscious processing emerged as a way to facilitate this new form of action control. It may operate less by direct initiation of behavior than by macro and sometimes delayed reflecting on optimal courses of action, possibly setting up and altering response tendencies that guide the automatic responses that are the immediate, proximal causes of behavior.
Amir, O., Dhar, R., Pocheptsaya, A., & Baumeister, R. F. (2005). The fatigued decision maker: Ego depletion changes decision process and outcome. Unpublished findings, Yale and Florida State Universities.
Aronson, E. (2000). Social animal (7th ed.). New York: Freeman.
Asch, S. E. (1955, November). Opinions and social pressure. Scientific American, 31–35.
Asch, S. E. (1956). Studies of independence and conformity: I. A minority of one against a unanimous majority. Psychological Monographs, 70, (No. 416).
Bargh, J. A. (1994). The four horsemen of automaticity: Awareness, efficiency, intention, and control in social cognition. In R. S. Wyer, Jr., & T. K. Srull (Eds.), Handbook of social cognition (2nd ed., pp. 1–40). Hillsdale, NJ: Erlbaum.
Baumeister, R. F. (2005). The cultural animal: Human nature, meaning, and social life. New York: Oxford University Press.
Baumeister, R. F., Bratslavsky, E., Muraven, M., & Tice, D. M. (1998). Ego depletion: Is the active self a limited resource? Journal of Personality and Social Psychology, 74, 1252–65.
Baumeister, R. F., Heatherton, T. F., & Tice, D. M. (1994). Losing control: How and why people fail at self-regulation. San Diego, CA: Academic Press.
Baumeister, R. F., Schmeichel, B. J., DeWall, C. N., & Vohs, K. D. (2006). Is the conscious self a help, a hindrance, or an irrelevance to the creative process? Advances in psychology research, 53, 137–52.Brehm, J. (1966). A theory of psychological reactance. New York: Academic Press.
Chaiken, S., & Trope, Y. (Eds.). (1999). Dual-process theories in social psychology. New York: Guilford.
Dennett, D. C. (1984). Elbow room: The varieties of free will worth wanting. Cambridge, MA: MIT Press.
Dennett, D. C. (2003). Freedom evolves. New York: Viking/Penguin.
DeWall, C. N., Baumeister, R. F., & Masicampo, E. J. (2006). Evidence that logical reasoning depends on conscious processing. Manuscript submitted for publication, Florida State University.
Donald, M. (2002). A mind so rare: The evolution of human consciousness. New York: Norton.
Duckworth, A. L., & Seligman, M. E. P. (2005). Self-discipline outdoes IQ in predicting academic performance of adolescents. Psychological Science, 16, 939–44.
Frederick, S. (2005). Cognitive reflection and decision making. Journal of Economic Perspectives, 19, 25–42.
Gailliot, M. T., Baumeister, R. F., DeWall, C. N., Maner, J. K., & Plant, E. A. (2006). Self-control relies on glucose as a limited energy source: Willpower is more than a metaphor. Manuscript in preparation, Florida State University.
Gazzaniga, M. S. (2003, January). The when, where, what, and why of conscious experience. Presented to the National Institute on the Teaching of Psychology, St. Petersburg Beach, FL.
Horan, R. D., Bulte, E., & Shogren, J. F. (2005). How trade saved humanity from biological exclusion: An economic theory of Neanderthal extinction. Journal of Economic Behavior and Organization, 58, 1–29.
Kahneman, D. (2003). A perspective on judgment and choice: Mapping bounded rationality. American Psychologist, 58, 697–720.
Kane, R. (2002). Oxford handbook of free will. New York: Oxford University Press.
Kelly, E. L., & Conley, J. J. (1987). Personality and compatibility: A prospective analysis of marital stability and marital satisfaction. Journal of Personality and Social Psychology, 52, 27–40.
Libet, B. (1985). Unconscious cerebral initiative and the role of conscious will in voluntary action. Behavior and Brain Sciences, 8, 529–66.
Libet, B. (1999). Do we have free will? Journal of Consciousness Studies, 6, 47–57.
Lieberman, M. D., Gaunt, R., Gilbert, D. T., & Trope, Y. (2002). Reflection and reflexion: A social cognitive neuroscience approach to attributional inference. Advances in Experimental Social Psychology, 34, 199–249.
Mischel, W., Shoda, Y., & Peake, P. K. (1988). The nature of adolescent competencies predicted by preschool delay of gratification. Journal of Personality and Social Psychology, 54, 687–96.
Muraven, M. R., & Baumeister, R. F. (2000). Self-regulation and depletion of limited resources: Does self-control resemble a muscle? Psychological Bulletin, 126, 247–59.
Nisbett, R. E., & Wilson, T. D. (1977). Telling more than we can know: Verbal reports on mental processes. Psychological Review, 84, 231–59.
Obhi, S.S., & Haggard, P. (2004). Free will and free won’t. American Scientist, 92, 358–65.
Roese, N. J. (1997). Counterfactual thinking. Psychological Bulletin, 121, 133–48.
Sartre, J.-P. (1974). Being and nothingness. Secaucus, NJ: Citadel. (Original work published 1943)
Schmeichel, B. J., & Baumeister, R.F. (2004). Self-regulatory strength. In R. Baumeister & K. Vohs (Eds.), Handbook of self-regulation: Research, theory, and applications (pp. 84–98). New York: Guilford.
Schmeichel, B. J., Vohs, K. D., & Baumeister, R. F. (2003). Ego depletion and intelligent performance: Role of the self in logical reasoning and other information processing. Journal of Personality and Social Psychology, 85, 33–46.
Searle, J. R. (2001). Rationality in action. Cambridge, MA: MIT Press.
(p.85) Shariff, A., & Peterson, J. B. (2005). Anticipatory consciousness, Libet’s veto, and a close-enough theory of free will. In R. Ellis & N. Newton (Eds.), Consciousness & emotion (pp. 197–215). Amsterdam, Netherlands: J. Benjamins.
Shoda, Y., Mischel, W., & Peake, P. K. (1990). Predicting adolescent cognitive and self-regulatory competencies from preschool delay of gratification: Identifying diagnostic conditions. Developmental Psychology, 26, 978–86.
Tangney, J. P., Baumeister, R. F., & Boone, A. L. (2004). High self-control predicts good adjustment, less pathology, better grades, and interpersonal success. Journal of Personality, 72, 271–322.
Vohs, K. D., Baumeister, R. F., Twenge, J. M., Schmeichel, B. J., & Tice, D. M. (2006). Decision fatigue exhausts self-regulatory resources. Manuscript submitted for publication, University of Minnesota.
Wegner, D. M. (2002). The illusion of conscious will. Cambridge, MA: MIT Press.
Wilson, T. D. (2002). Strangers to ourselves: Discovering the adaptive unconscious. Cambridge, MA: Harvard University Press.