Science proceeds by approaching the truth gradually through successive approximations. A generation ago, psychologists began to realize that the conscious control of action had been overestimated, and many began to search for unconscious processes. The pendulum has now swung the other way, as Newell & Shanks (N&S) have shown in their valuable corrective to the excesses of that view. Now it is the unconscious effects that have been overestimated.
We think the way forward for psychological theory is to stop pitting conscious against unconscious and instead figure out how the two work together. Plainly, there is plenty of processing of information that occurs outside of awareness. Likewise as plainly, consciousness has advantages and can accomplish things that the unconscious cannot.
N&S propose their lens model with multiple stages of processing. They rightly criticize the tendency to claim that some outcome is unconscious based on showing that only one of the five steps is unconscious. Still, the fact that some steps are unconscious is a genuine contribution that will need to be included in the eventual, correct account of human thought and action.
Recently, some theorists have become increasingly bold in asserting that consciousness is an epiphenomenon with no impact on behavior. We have been skeptical that such a complex, advanced phenomenon as human conscious thought would evolve without conferring vital advantages. Inspired by this skepticism, we conducted an extensive review of experimental evidence for the conscious causation of behavior (Baumeister et al. Reference Baumeister, Masicampo and Vohs2011). We searched for experiments in which the independent variable was a conscious thought or other conscious event and the dependent variable was overt behavior. By the logic of experimental design, such studies prove causation. We found a wide assortment, leading to our conclusion that the evidence for conscious causation was extensive, diverse, and undeniable.
Still, none of the evidence we found ruled out important, even essential contributions by unconscious processes. We speculated that there may well be no human behaviors that are produced entirely by consciousness (and likewise few produced entirely by unconscious processes). If all behavior is indeed produced by both conscious and unconscious processes, then it is imperative to understand both types and how they interact.
In fact, we think that conscious processes are themselves produced in part by unconscious processes. Baumeister and Masicampo (Reference Baumeister and Masicampo2010) concluded that consciousness is best considered a place where the unconscious constructs meaningful sequences of thought. This is linked to evidence that, for example, the unconscious processes single words but not sentences and paragraphs (see Baars Reference Baars2002). Likewise, logical reasoning deteriorates sharply when consciousness is preoccupied and improves when engaged (DeWall et al. Reference DeWall, Baumeister and Masicampo2008). Logical reasoning requires putting together complex sequences of ideas while ruling out other possible sequences, and that may too big a job to do effectively without consciousness.
What else is consciousness good for? We are wary of making assertions that something absolutely cannot be done unconsciously – but perhaps that is not necessary. After all, the capacity for conscious thought would have been favored by natural selection simply on the basis of doing something better or more thoroughly than unconscious processes, even if the unconscious could occasionally do something along the same lines well enough to produce an experimental finding. For example, one recent paper has proposed that the unconscious can do some arithmetic (Sklar et al. Reference Sklar, Levy, Goldstein, Mandel, Maril and Hassin2012). Even if this finding could measure up to the methodological standards proposed by N&S, we think that is hardly a reason to dismiss the usefulness of conscious thought for mathematical work. Does anyone seriously think that a student could pass a college math test without conscious thought?
Key themes from our survey of experimental findings on conscious causation (Baumeister et al. Reference Baumeister, Masicampo and Vohs2011) included the following: Conscious thoughts integrate across time. That is, conscious thought permits the deliberate combining of past and future into causing present behavior, as well as helping present cognitions to cause future behavior, and probably other combinations. Most animals live largely in the present, whereas the stupendous success of the human race has benefited immensely from integrating across time – for which conscious thought probably deserves much of the credit.
Conscious thought also helps translate abstract principles into specific behaviors. Humankind has benefited from moral principles, legal rules, economic calculations, application of scientific and mathematical principles, and other sorts of general, abstract understandings. The unconscious may be effective at processing highly specific stimuli and responses, but without conscious thought, it may lose most of the benefits of abstract principles for guiding behavior.
We also found that conscious causation of behavior was typically found in situations involving multiple possibilities, such as for deliberating among multiple possible courses of action by considering their likely consequences. Consciousness enables the person to mentally simulate nonpresent realities, including possible sequences of future events. In our view, the jury is still out on whether the act of choosing is conscious or unconscious – but most decisions will be considerably more effective insofar as one uses conscious thought to ponder what each possible action will produce and the desirability of downstream consequences.
Last, and perhaps most important, consciousness is highly useful for communication. We have not heard even the most assertive critics of conscious thought claim that a person could carry on a conversation unconsciously. Indeed, we think that the evolutionarily decisive advantages of conscious thought are not to be found in private, solipsistic ratiocination but rather in its contribution to communication (Baumeister & Masicampo Reference Baumeister and Masicampo2010). Humankind's biological strategy for surviving and reproducing has been centrally based on sharing information and coordinating joint performance. Although the principle that thinking is for doing has been widely considered sacrosanct since first asserted by William James (Reference James1890), we propose that a viable partial alternative is that conscious thinking is for talking.
In sum, we applaud N&S for pushing the field forward. Conscious thought is a vital part of human life. We think the view that humans could operate effectively without conscious thought will soon be regarded as quaint and naïve.
Science proceeds by approaching the truth gradually through successive approximations. A generation ago, psychologists began to realize that the conscious control of action had been overestimated, and many began to search for unconscious processes. The pendulum has now swung the other way, as Newell & Shanks (N&S) have shown in their valuable corrective to the excesses of that view. Now it is the unconscious effects that have been overestimated.
We think the way forward for psychological theory is to stop pitting conscious against unconscious and instead figure out how the two work together. Plainly, there is plenty of processing of information that occurs outside of awareness. Likewise as plainly, consciousness has advantages and can accomplish things that the unconscious cannot.
N&S propose their lens model with multiple stages of processing. They rightly criticize the tendency to claim that some outcome is unconscious based on showing that only one of the five steps is unconscious. Still, the fact that some steps are unconscious is a genuine contribution that will need to be included in the eventual, correct account of human thought and action.
Recently, some theorists have become increasingly bold in asserting that consciousness is an epiphenomenon with no impact on behavior. We have been skeptical that such a complex, advanced phenomenon as human conscious thought would evolve without conferring vital advantages. Inspired by this skepticism, we conducted an extensive review of experimental evidence for the conscious causation of behavior (Baumeister et al. Reference Baumeister, Masicampo and Vohs2011). We searched for experiments in which the independent variable was a conscious thought or other conscious event and the dependent variable was overt behavior. By the logic of experimental design, such studies prove causation. We found a wide assortment, leading to our conclusion that the evidence for conscious causation was extensive, diverse, and undeniable.
Still, none of the evidence we found ruled out important, even essential contributions by unconscious processes. We speculated that there may well be no human behaviors that are produced entirely by consciousness (and likewise few produced entirely by unconscious processes). If all behavior is indeed produced by both conscious and unconscious processes, then it is imperative to understand both types and how they interact.
In fact, we think that conscious processes are themselves produced in part by unconscious processes. Baumeister and Masicampo (Reference Baumeister and Masicampo2010) concluded that consciousness is best considered a place where the unconscious constructs meaningful sequences of thought. This is linked to evidence that, for example, the unconscious processes single words but not sentences and paragraphs (see Baars Reference Baars2002). Likewise, logical reasoning deteriorates sharply when consciousness is preoccupied and improves when engaged (DeWall et al. Reference DeWall, Baumeister and Masicampo2008). Logical reasoning requires putting together complex sequences of ideas while ruling out other possible sequences, and that may too big a job to do effectively without consciousness.
What else is consciousness good for? We are wary of making assertions that something absolutely cannot be done unconsciously – but perhaps that is not necessary. After all, the capacity for conscious thought would have been favored by natural selection simply on the basis of doing something better or more thoroughly than unconscious processes, even if the unconscious could occasionally do something along the same lines well enough to produce an experimental finding. For example, one recent paper has proposed that the unconscious can do some arithmetic (Sklar et al. Reference Sklar, Levy, Goldstein, Mandel, Maril and Hassin2012). Even if this finding could measure up to the methodological standards proposed by N&S, we think that is hardly a reason to dismiss the usefulness of conscious thought for mathematical work. Does anyone seriously think that a student could pass a college math test without conscious thought?
Key themes from our survey of experimental findings on conscious causation (Baumeister et al. Reference Baumeister, Masicampo and Vohs2011) included the following: Conscious thoughts integrate across time. That is, conscious thought permits the deliberate combining of past and future into causing present behavior, as well as helping present cognitions to cause future behavior, and probably other combinations. Most animals live largely in the present, whereas the stupendous success of the human race has benefited immensely from integrating across time – for which conscious thought probably deserves much of the credit.
Conscious thought also helps translate abstract principles into specific behaviors. Humankind has benefited from moral principles, legal rules, economic calculations, application of scientific and mathematical principles, and other sorts of general, abstract understandings. The unconscious may be effective at processing highly specific stimuli and responses, but without conscious thought, it may lose most of the benefits of abstract principles for guiding behavior.
We also found that conscious causation of behavior was typically found in situations involving multiple possibilities, such as for deliberating among multiple possible courses of action by considering their likely consequences. Consciousness enables the person to mentally simulate nonpresent realities, including possible sequences of future events. In our view, the jury is still out on whether the act of choosing is conscious or unconscious – but most decisions will be considerably more effective insofar as one uses conscious thought to ponder what each possible action will produce and the desirability of downstream consequences.
Last, and perhaps most important, consciousness is highly useful for communication. We have not heard even the most assertive critics of conscious thought claim that a person could carry on a conversation unconsciously. Indeed, we think that the evolutionarily decisive advantages of conscious thought are not to be found in private, solipsistic ratiocination but rather in its contribution to communication (Baumeister & Masicampo Reference Baumeister and Masicampo2010). Humankind's biological strategy for surviving and reproducing has been centrally based on sharing information and coordinating joint performance. Although the principle that thinking is for doing has been widely considered sacrosanct since first asserted by William James (Reference James1890), we propose that a viable partial alternative is that conscious thinking is for talking.
In sum, we applaud N&S for pushing the field forward. Conscious thought is a vital part of human life. We think the view that humans could operate effectively without conscious thought will soon be regarded as quaint and naïve.