Newell & Shanks (N&S) perform a valuable service in critically reviewing some of the burgeoning literature on unconscious decision making, and exposing some of the methodological weaknesses of these studies. Nevertheless, there are three related aspects of their thesis that we would like to question: their conceptualization of the unconscious, their dichotomy between proximal and distal cues, and their lack of consideration of any developmental evidence.
First, N&S's treatment of “the unconscious” is conceptually very limited, in that they appear to restrict unconscious decision making to processes that are “cognitively impenetrable” (sect. 6.1.1) – that is, that are not accessible to conscious awareness – without foregrounding this limitation on their definition. This is far from the everyday and classical (e.g., Freudian) meanings of the unconscious as something that is potentially accessible to awareness, but which usually occurs automatically and beneath the level of conscious processing (Bargh & Morsella Reference Bargh and Morsella2008). Indeed, the goal of psychoanalysis is to make hidden or suppressed (temporarily inaccessible) thoughts available once again to conscious awareness.
There are many examples from recent studies of moral reasoning and moral intuition that fit better with this classical understanding of the unconscious. For Haidt (Reference Haidt2007, p. 998), for instance, “Moral intuition refers to fast, automatic, and (usually) affect-laden processes in which an evaluative feeling of good-bad or like-dislike (about the actions or character of a person) appears in consciousness without any awareness of having gone through steps of search, weighing evidence, or inferring a conclusion.” Note that this does not imply that someone cannot, on reflection, come to an awareness of some of the steps that have led him or her to this evaluative feeling. This paints a much more dynamic picture of the unconscious than that presented by N&S. And it is a picture that is shared with a researcher whom the authors criticize, John Bargh, who has explicitly stated that he does not view the unconscious in terms of a lack of awareness of certain (subliminal) stimuli, but rather in terms of a lack of awareness of the effects that certain stimuli have on decision making (Bargh & Morsella Reference Bargh and Morsella2008). So we may have a familiar case of authors from two different, but related, disciplines (cognitive and social psychology) arguing across one another because they are using the same word (“unconscious”) to refer to two different but related phenomena (what is subliminal and inaccessible to consciousness vs. what is merely overlooked or forgotten by consciousness).
This suspicion is reinforced by N&S's reliance on a dichotomy between “proximal” and “distal” cues (when assessing the relevance of a piece of information for decision making; sect. 1.2). They claim an agent might be unaware of a distal cue (e.g., Mother always told me spinach was good for me) and its influence on a current decision (to eat spinach), but still may be able to justify her decision in terms of a proximal cue (the belief spinach is healthy). It is hard to see the difference between a neglected distal cue and an unconscious influence on behavior, in the naturalistic social psychological sense of Bargh and Morsella (Reference Bargh and Morsella2008). Nor do N&S elaborate on what makes a distal cue “irrelevant” to decision making. Even long-past and long-forgotten cues can be highly relevant: for example, a jury selection panel is unlikely to look favorably on any history of interaction between a potential juror and a defendant, whether the juror can recall the interaction or not. Social psychologists of like mind to Haidt and Bargh would claim the authors confuse an agent's ability to give reasons for her own decision with her awareness of all the factors that significantly influenced her decision. That can be thrown into sharp relief by consideration of Haidt's famous cases of “moral dumbfounding” (e.g., when participants are asked why they condemn a case of incest that has no harmful consequences): In such cases, individuals are unable to give either a distal or a proximal cue for their behavior (Haidt Reference Haidt2001; cf. Pizarro & Bloom Reference Pizarro and Bloom2003).
N&S's other example of the relation between proximal and distal cues (in an adjacent endnote) is even less satisfactory. We are asked to consider a judge who sentences people with Afrocentric features more severely (without being aware of this bias) because he perceives them as less intelligent. Yet this time, any conscious justification given by judges cannot be the same as the “proximal” cue, since the judges in these cases would presumably not rationalize their decisions in terms of any aspect of the suspects' appearance whatsoever: They would simply not be aware that the perception of a face as either Afrocentric or unintelligent would influence their judgment. In such cases there is clearly a disconnect between the conscious justification made by an agent and causes of their behavior. In this context, it is interesting that a recent mock-jury study found that when given a chance to deliberate, jurors were more likely to find an attractive defendant guilty, whereas without deliberation they were more likely to find a non-attractive defendant guilty (Patry Reference Patry2008). This finding is in accord with a more dynamic picture of the unconscious than the one that N&S paint.
Finally, N&S omit any discussion of the extensive developmental literature on the ontogeny of children's decision making. Yet that literature is highly relevant to the conceptual limitations of their article, in that there are many experimental tasks (reviewed by Karmiloff-Smith Reference Karmiloff-Smith1992) for which younger children are unable to articulate why they have a preference for a particular behavioral choice, but for which older children are able to explain the conscious reasoning behind their choices. Because of this type of transition, developmental research currently represents a growing point in dual-process accounts of decision making (e.g., Stanovich et al. Reference Stanovich, West and Toplak2011). Taking a developmental perspective leads us again to emphasize the dynamic relationship between conscious and unconscious influences on decision making, because observing how children solve certain tasks makes it clear that conscious thought processes can come to modulate decision making that was previously performed automatically.
Newell & Shanks (N&S) perform a valuable service in critically reviewing some of the burgeoning literature on unconscious decision making, and exposing some of the methodological weaknesses of these studies. Nevertheless, there are three related aspects of their thesis that we would like to question: their conceptualization of the unconscious, their dichotomy between proximal and distal cues, and their lack of consideration of any developmental evidence.
First, N&S's treatment of “the unconscious” is conceptually very limited, in that they appear to restrict unconscious decision making to processes that are “cognitively impenetrable” (sect. 6.1.1) – that is, that are not accessible to conscious awareness – without foregrounding this limitation on their definition. This is far from the everyday and classical (e.g., Freudian) meanings of the unconscious as something that is potentially accessible to awareness, but which usually occurs automatically and beneath the level of conscious processing (Bargh & Morsella Reference Bargh and Morsella2008). Indeed, the goal of psychoanalysis is to make hidden or suppressed (temporarily inaccessible) thoughts available once again to conscious awareness.
There are many examples from recent studies of moral reasoning and moral intuition that fit better with this classical understanding of the unconscious. For Haidt (Reference Haidt2007, p. 998), for instance, “Moral intuition refers to fast, automatic, and (usually) affect-laden processes in which an evaluative feeling of good-bad or like-dislike (about the actions or character of a person) appears in consciousness without any awareness of having gone through steps of search, weighing evidence, or inferring a conclusion.” Note that this does not imply that someone cannot, on reflection, come to an awareness of some of the steps that have led him or her to this evaluative feeling. This paints a much more dynamic picture of the unconscious than that presented by N&S. And it is a picture that is shared with a researcher whom the authors criticize, John Bargh, who has explicitly stated that he does not view the unconscious in terms of a lack of awareness of certain (subliminal) stimuli, but rather in terms of a lack of awareness of the effects that certain stimuli have on decision making (Bargh & Morsella Reference Bargh and Morsella2008). So we may have a familiar case of authors from two different, but related, disciplines (cognitive and social psychology) arguing across one another because they are using the same word (“unconscious”) to refer to two different but related phenomena (what is subliminal and inaccessible to consciousness vs. what is merely overlooked or forgotten by consciousness).
This suspicion is reinforced by N&S's reliance on a dichotomy between “proximal” and “distal” cues (when assessing the relevance of a piece of information for decision making; sect. 1.2). They claim an agent might be unaware of a distal cue (e.g., Mother always told me spinach was good for me) and its influence on a current decision (to eat spinach), but still may be able to justify her decision in terms of a proximal cue (the belief spinach is healthy). It is hard to see the difference between a neglected distal cue and an unconscious influence on behavior, in the naturalistic social psychological sense of Bargh and Morsella (Reference Bargh and Morsella2008). Nor do N&S elaborate on what makes a distal cue “irrelevant” to decision making. Even long-past and long-forgotten cues can be highly relevant: for example, a jury selection panel is unlikely to look favorably on any history of interaction between a potential juror and a defendant, whether the juror can recall the interaction or not. Social psychologists of like mind to Haidt and Bargh would claim the authors confuse an agent's ability to give reasons for her own decision with her awareness of all the factors that significantly influenced her decision. That can be thrown into sharp relief by consideration of Haidt's famous cases of “moral dumbfounding” (e.g., when participants are asked why they condemn a case of incest that has no harmful consequences): In such cases, individuals are unable to give either a distal or a proximal cue for their behavior (Haidt Reference Haidt2001; cf. Pizarro & Bloom Reference Pizarro and Bloom2003).
N&S's other example of the relation between proximal and distal cues (in an adjacent endnote) is even less satisfactory. We are asked to consider a judge who sentences people with Afrocentric features more severely (without being aware of this bias) because he perceives them as less intelligent. Yet this time, any conscious justification given by judges cannot be the same as the “proximal” cue, since the judges in these cases would presumably not rationalize their decisions in terms of any aspect of the suspects' appearance whatsoever: They would simply not be aware that the perception of a face as either Afrocentric or unintelligent would influence their judgment. In such cases there is clearly a disconnect between the conscious justification made by an agent and causes of their behavior. In this context, it is interesting that a recent mock-jury study found that when given a chance to deliberate, jurors were more likely to find an attractive defendant guilty, whereas without deliberation they were more likely to find a non-attractive defendant guilty (Patry Reference Patry2008). This finding is in accord with a more dynamic picture of the unconscious than the one that N&S paint.
Finally, N&S omit any discussion of the extensive developmental literature on the ontogeny of children's decision making. Yet that literature is highly relevant to the conceptual limitations of their article, in that there are many experimental tasks (reviewed by Karmiloff-Smith Reference Karmiloff-Smith1992) for which younger children are unable to articulate why they have a preference for a particular behavioral choice, but for which older children are able to explain the conscious reasoning behind their choices. Because of this type of transition, developmental research currently represents a growing point in dual-process accounts of decision making (e.g., Stanovich et al. Reference Stanovich, West and Toplak2011). Taking a developmental perspective leads us again to emphasize the dynamic relationship between conscious and unconscious influences on decision making, because observing how children solve certain tasks makes it clear that conscious thought processes can come to modulate decision making that was previously performed automatically.