In Talking to Our Selves, John Doris claims that attributions of responsible agency may be warranted when the best explanation for certain patterns of behavior across iterated conditions involve the expression of the actor's values. In some cases, however, attributions of responsibility will be unwarranted because we cannot rule out the presence defeaters. For any actor A, behavior (or cognition) X, and practically significant causal factor(s) F, defeaters obtain when the causes of A's X are such that, had A become aware of F, A would not recognize F as reason(s) justifying A's X. In cases where defeaters obtain, exercises of responsible agency do not obtain. If we cannot rule out the presence of defeaters, then our attributions of responsibility are unwarranted.
I want to expand Doris's discussion by further focusing on cases where we are unable to rule out the presence of defeaters. In particular, I am interested in the practical issue of enhancing our responsible agency so that defeaters no longer have a significant impact on behavior and cognition. I maintain that any practically relevant view of responsible agency should have something to say about circumventing the influence of defeaters. Doris does provide an account of how responsible agency may be realized via an extended process of interpersonal negotiation wherein people identify and develop their values. However, it is not entirely clear that these dialogic processes will circumvent the influence of defeaters. Even if I identify and develop my values, defeaters may still bypass those values. In that case, what is needed is an account of how we might become more resistant to the influence of defeaters. Ideally, this account would employ empirically informed self-regulative strategies that enjoy wide applicability. In what follows, I provide an initial sketch of such an account.
Imagine that we have observed a subject, Amy, across various iterated conditions. Suppose that Amy values helping others, but we are unable to infer from her behavioral pattern that defeaters do not obtain. She does not engage in helping behavior in the following iterated conditions: when she is in a neutral mood (O'Malley & Andrews Reference O'Malley and Andrews1983), when she is in the presence of passive bystanders (Darley & Latané Reference Darley and Latané1968), when her willpower has been depleted (DeWall et al. Reference DeWall, Baumeister, Gailliot and Maner2008), and when the potential recipients of her assistance are members of a specific race (Stepanikova et al. Reference Stepanikova, Triplett and Simpson2011). However, Amy does help when her body posture has been mimicked by an experimental confederate (van Baaren et al. Reference van Baaren, Holland, Kawakami and van Knippenberg2004), and when she is in various other defeater-suspect conditions. Given Amy's extended behavioral pattern, it seems plausible to think that we are unable to rule out the presence of defeaters. She would not recognize mimicry effects, group effects, ego-depletion effects, mood effects, and implicit bias as influences that justify her conduct. However, it may be the case that her behavior was influenced by all of these rationally and ethically arbitrary influences.
Imagine that Amy learns that there are various defeater influences on her conduct. She is deeply troubled by this information. Amy thinks of herself as someone who explicitly devalues racist categories. She also thinks of herself as someone who is not easily influenced by numerous other strange factors. Amy decides to visit a therapist for practical advice about how to enhance her responsible agency so that defeaters no longer have a significant impact on her behavior. Is there any practical advice that can be given to Amy? Here's one possible response. We reply that, unfortunately, she cannot enhance her agency so that defeaters no longer have a significant impact on her behavior and cognition. We might support this claim with the following observation: Behavior is overwhelmingly governed by automatic processing, and our ability to override our automatic processing is severely limited (Bargh & Chartrand Reference Bargh and Chartrand1999). Some social psychologists estimate that our ability to override automatic processing is limited to about 5% of the time (Baumeister et al. Reference Baumeister, Bratslavsky, Muraven and Tice1998; Bargh and Chartrand Reference Bargh and Chartrand1999). In light of this evidence, someone might respond to Amy in the following way: Most of us often lack the ability to enhance ourselves in such a way that defeaters no longer have a significant impact on behavior and cognition.
Amy does not find this response very appealing. Amy isn't just concerned about attributing responsible agency to others, she is just as concerned with enhancing her own agency so that attributions of responsibility apply to her own conduct. She has practical aspirations to circumvent the influence of defeaters across various conditions. Furthermore, I take it that most of us share Amy's practical aspirations: We aspire to actively express our values rather than being passive victims of defeaters. A practically relevant account of responsible agency should have something to say about these important practical aspirations. I maintain that we can provide an initial sketch of such an account. This sketch should provide some optimism for thinking that we can give Amy (and others) widely applicable (and empirically informed) practical advice.
Are there any self-regulatory strategies that are widely applicable with respect to mediating the impact of defeaters? A growing amount of evidence suggests that the development of expertise is a widely applicable self-regulative strategy. Cramer et al. (Reference Cramer, McMaster, Bartell and Dragna1988) measured the helping behavior of registered nurses and nursing students in the presence of passive bystanders. An experimental confederate disguised as a maintenance worker loudly fell from a ladder in a nearby room. The helping behavior measured in this experiment involved offering assistance to the confederate. Experimenters found that the nursing students were significantly less likely to help in the presence of bystanders. However, registered nurses were just as likely to help in the presence of passive bystanders as control subjects who were alone.
Cramer and colleagues explain that “the nurses' professional training and experience led to a consistent level of emergency responding whether or not a bystander was present” (Cramer et al. Reference Cramer, McMaster, Bartell and Dragna1988, p. 1142). In addition to this study on group effects, practice and expertise has been found to mediate the influence of implicit bias (Plant & Peruche Reference Plant and Peruche2005; Plant et al. Reference Plant, Peruche and Butz2005), mood effects (Forgas et al. Reference Forgas, Dunn and Granland2008), ego depletion (Hui et al. Reference Hui, Wright, Stewart, Simmons, Eaton and Nolte2009; Muraven et al.Reference Muraven, Baumeister and Tice1999), and various other defeaters (Morewedge et al. Reference Morewedge, Yoon, Scopelliti, Symborski, Korris and Kassam2015). Crucially, some of the evidence on expertise suggests that practiced subjects need not detect defeaters nor override automatic processing. Instead, the evidence suggests that practice modifies the actor's automatic processing so that it is less sensitive to defeaters prior to encountering those influences (Forgas et al. Reference Forgas, Dunn and Granland2008). Thus, there is a growing amount of evidence suggesting a widely applicable self-regulatory strategy: People should develop expertise in certain domains if they want to consistently express their values.
The empirical evidence in support of enhancing responsible agency is not exhausted by the studies on expertise. Peter Gollwitzer and other social psychologists have found that adopting situation-specific plans can mediate the influence of defeaters. They propose that people adopt implementation intentions, “if-then” plans that specify a specific context and desired response (Gollwitzer Reference Gollwitzer1999). For example, suppose I'm trying to avoid buying beer at a local restaurant. I can adopt the implementation intention: “If I'm handed a beer menu, then I will order water instead!” There have been a large number of studies suggesting that implementation intentions can facilitate goal pursuit across various domains (Gollwitzer & Sheeran Reference Gollwitzer and Sheeran2006).
Mendoza et al. (Reference Mendoza, Gollwitzer and Amodio2010) measured whether implementation intentions mediate the influence of implicit bias in a computer shooting task. The shooting program displayed either a Black or White face containing either a gun or a non-threatening object. The subjects must respond as quickly as possible (less than 630 ms) to each trial by choosing “shoot” or “don't shoot” in response to various images. They were instructed to only shoot at armed criminals. Subjects in the experimental group were told to adopt the implementation intention: “If I see a person, then I will ignore his race!” (Mendoza et al. Reference Mendoza, Gollwitzer and Amodio2010, p. 515). Subjects in the control group were not given the instruction to adopt implementation intentions. Experimenters found that control subjects were more likely to shoot unarmed Black targets than unarmed Whites. However, experimental subjects were significantly less likely to make performance errors (i.e., selecting “shoot”) in response to both unarmed Black and White bystanders. This evidence suggests that people can mediate the influence of implicit bias via the adoption of implementation intentions.
The effectiveness of implementation intentions is not restricted to mediating implicit bias. Implementation intentions have also been found to mediate framing effects (Trötschel & Gollwitzer Reference Trötschel and Gollwitzer2007), mimicry effects (Wieber et al. Reference Wieber, Gollwitzer and Sheeran2014), ego depletion (Webb & Sheeran Reference Webb and Sheeran2002), and mood effects (Webb et al. Reference Webb, Schweiger Gallo, Miles, Gollwitzer and Sheeran2012). It is important to point out that the evidence on implementation intentions suggests that subjects can regulate defeaters without needing to override their automatic processing. This is because the ‘if’ component of the plan modifies the actor's automatic processing prior to encountering the specific circumstance, thereby shifting attention away from potential defeaters and toward goal-relevant situational cues (Gallo et al. Reference Gallo, Keil, McCulloch, Rockstroh and Gollwitzer2009). Thus, there is a growing amount of evidence suggesting that implementation intentions can mediate the influence of defeaters.
It should be admitted here that there may be some limits to the self-regulatory strategies described in my initial sketch. Lauren Olin and John Doris have noted some limits on expertise (Olin & Doris Reference Olin and Doris2014). They discuss some studies suggesting that physicians become worse at diagnosing heart sounds after years of practice (Choudhry et al. Reference Choudhry, Fletcher and Soumerai2005; Ericsson et al. Reference Ericsson, Whyte and Ward2007), and experts are not better than non-experts at making economic predictions (Tetlock Reference Tetlock2009). Thus, it seems that practice sometimes results in worse performance, and it may well be that some of these failures involve defeaters.
Doris also claims that implementation intentions may be limited because “countervailing pressures may perturb the goal relevant situation” (Doris Reference Doris2015b, p. 128). Someone might also worry that implementation intentions may collaborate or produce defeaters in specific circumstances. The increased salience of situational cues contained in the “if” component of one's plan may automatically attract attention even when these cues are not relevant to one's current goal (Parks-Stamm & Gollwitzer Reference Parks-Stamm, Gollwitzer, Moskowitz and Grant2009). There is evidence that the human capacity for attention is limited (Shiffrin & Schneider Reference Shiffrin and Schneider1977), and so implementation intentions may influence us in such a way that we are sometimes unable to attend to relevant contextual cues during the pursuit of multiple goals. Perhaps even some of these failures of attention may be construed as defeaters, where the causes of the actor's behavior (i.e., the ‘if’ component of the plan together with the situational cue) would not be recognized as reason(s) justifying their behavior.
Admittedly, the strategies suggested here may be limited in various ways. However, the purpose of my discussion was only to present an initial sketch that enjoys some empirical support. I have not attempted to provide a fully developed account of enhancement that can adequately address all of these potential limitations. Obviously, the details of this account need to be developed further so that these limitations are adequately addressed. But I do not see sufficient evidence for thinking that this project is unlikely to produce a compelling response to Amy (and others). In conclusion, accounts of responsible agency face an important practical challenge: Specify conditions and strategies wherein actors can circumvent the influence of defeaters. I hope to have provided an initial sketch of such an account.
In Talking to Our Selves, John Doris claims that attributions of responsible agency may be warranted when the best explanation for certain patterns of behavior across iterated conditions involve the expression of the actor's values. In some cases, however, attributions of responsibility will be unwarranted because we cannot rule out the presence defeaters. For any actor A, behavior (or cognition) X, and practically significant causal factor(s) F, defeaters obtain when the causes of A's X are such that, had A become aware of F, A would not recognize F as reason(s) justifying A's X. In cases where defeaters obtain, exercises of responsible agency do not obtain. If we cannot rule out the presence of defeaters, then our attributions of responsibility are unwarranted.
I want to expand Doris's discussion by further focusing on cases where we are unable to rule out the presence of defeaters. In particular, I am interested in the practical issue of enhancing our responsible agency so that defeaters no longer have a significant impact on behavior and cognition. I maintain that any practically relevant view of responsible agency should have something to say about circumventing the influence of defeaters. Doris does provide an account of how responsible agency may be realized via an extended process of interpersonal negotiation wherein people identify and develop their values. However, it is not entirely clear that these dialogic processes will circumvent the influence of defeaters. Even if I identify and develop my values, defeaters may still bypass those values. In that case, what is needed is an account of how we might become more resistant to the influence of defeaters. Ideally, this account would employ empirically informed self-regulative strategies that enjoy wide applicability. In what follows, I provide an initial sketch of such an account.
Imagine that we have observed a subject, Amy, across various iterated conditions. Suppose that Amy values helping others, but we are unable to infer from her behavioral pattern that defeaters do not obtain. She does not engage in helping behavior in the following iterated conditions: when she is in a neutral mood (O'Malley & Andrews Reference O'Malley and Andrews1983), when she is in the presence of passive bystanders (Darley & Latané Reference Darley and Latané1968), when her willpower has been depleted (DeWall et al. Reference DeWall, Baumeister, Gailliot and Maner2008), and when the potential recipients of her assistance are members of a specific race (Stepanikova et al. Reference Stepanikova, Triplett and Simpson2011). However, Amy does help when her body posture has been mimicked by an experimental confederate (van Baaren et al. Reference van Baaren, Holland, Kawakami and van Knippenberg2004), and when she is in various other defeater-suspect conditions. Given Amy's extended behavioral pattern, it seems plausible to think that we are unable to rule out the presence of defeaters. She would not recognize mimicry effects, group effects, ego-depletion effects, mood effects, and implicit bias as influences that justify her conduct. However, it may be the case that her behavior was influenced by all of these rationally and ethically arbitrary influences.
Imagine that Amy learns that there are various defeater influences on her conduct. She is deeply troubled by this information. Amy thinks of herself as someone who explicitly devalues racist categories. She also thinks of herself as someone who is not easily influenced by numerous other strange factors. Amy decides to visit a therapist for practical advice about how to enhance her responsible agency so that defeaters no longer have a significant impact on her behavior. Is there any practical advice that can be given to Amy? Here's one possible response. We reply that, unfortunately, she cannot enhance her agency so that defeaters no longer have a significant impact on her behavior and cognition. We might support this claim with the following observation: Behavior is overwhelmingly governed by automatic processing, and our ability to override our automatic processing is severely limited (Bargh & Chartrand Reference Bargh and Chartrand1999). Some social psychologists estimate that our ability to override automatic processing is limited to about 5% of the time (Baumeister et al. Reference Baumeister, Bratslavsky, Muraven and Tice1998; Bargh and Chartrand Reference Bargh and Chartrand1999). In light of this evidence, someone might respond to Amy in the following way: Most of us often lack the ability to enhance ourselves in such a way that defeaters no longer have a significant impact on behavior and cognition.
Amy does not find this response very appealing. Amy isn't just concerned about attributing responsible agency to others, she is just as concerned with enhancing her own agency so that attributions of responsibility apply to her own conduct. She has practical aspirations to circumvent the influence of defeaters across various conditions. Furthermore, I take it that most of us share Amy's practical aspirations: We aspire to actively express our values rather than being passive victims of defeaters. A practically relevant account of responsible agency should have something to say about these important practical aspirations. I maintain that we can provide an initial sketch of such an account. This sketch should provide some optimism for thinking that we can give Amy (and others) widely applicable (and empirically informed) practical advice.
Are there any self-regulatory strategies that are widely applicable with respect to mediating the impact of defeaters? A growing amount of evidence suggests that the development of expertise is a widely applicable self-regulative strategy. Cramer et al. (Reference Cramer, McMaster, Bartell and Dragna1988) measured the helping behavior of registered nurses and nursing students in the presence of passive bystanders. An experimental confederate disguised as a maintenance worker loudly fell from a ladder in a nearby room. The helping behavior measured in this experiment involved offering assistance to the confederate. Experimenters found that the nursing students were significantly less likely to help in the presence of bystanders. However, registered nurses were just as likely to help in the presence of passive bystanders as control subjects who were alone.
Cramer and colleagues explain that “the nurses' professional training and experience led to a consistent level of emergency responding whether or not a bystander was present” (Cramer et al. Reference Cramer, McMaster, Bartell and Dragna1988, p. 1142). In addition to this study on group effects, practice and expertise has been found to mediate the influence of implicit bias (Plant & Peruche Reference Plant and Peruche2005; Plant et al. Reference Plant, Peruche and Butz2005), mood effects (Forgas et al. Reference Forgas, Dunn and Granland2008), ego depletion (Hui et al. Reference Hui, Wright, Stewart, Simmons, Eaton and Nolte2009; Muraven et al.Reference Muraven, Baumeister and Tice1999), and various other defeaters (Morewedge et al. Reference Morewedge, Yoon, Scopelliti, Symborski, Korris and Kassam2015). Crucially, some of the evidence on expertise suggests that practiced subjects need not detect defeaters nor override automatic processing. Instead, the evidence suggests that practice modifies the actor's automatic processing so that it is less sensitive to defeaters prior to encountering those influences (Forgas et al. Reference Forgas, Dunn and Granland2008). Thus, there is a growing amount of evidence suggesting a widely applicable self-regulatory strategy: People should develop expertise in certain domains if they want to consistently express their values.
The empirical evidence in support of enhancing responsible agency is not exhausted by the studies on expertise. Peter Gollwitzer and other social psychologists have found that adopting situation-specific plans can mediate the influence of defeaters. They propose that people adopt implementation intentions, “if-then” plans that specify a specific context and desired response (Gollwitzer Reference Gollwitzer1999). For example, suppose I'm trying to avoid buying beer at a local restaurant. I can adopt the implementation intention: “If I'm handed a beer menu, then I will order water instead!” There have been a large number of studies suggesting that implementation intentions can facilitate goal pursuit across various domains (Gollwitzer & Sheeran Reference Gollwitzer and Sheeran2006).
Mendoza et al. (Reference Mendoza, Gollwitzer and Amodio2010) measured whether implementation intentions mediate the influence of implicit bias in a computer shooting task. The shooting program displayed either a Black or White face containing either a gun or a non-threatening object. The subjects must respond as quickly as possible (less than 630 ms) to each trial by choosing “shoot” or “don't shoot” in response to various images. They were instructed to only shoot at armed criminals. Subjects in the experimental group were told to adopt the implementation intention: “If I see a person, then I will ignore his race!” (Mendoza et al. Reference Mendoza, Gollwitzer and Amodio2010, p. 515). Subjects in the control group were not given the instruction to adopt implementation intentions. Experimenters found that control subjects were more likely to shoot unarmed Black targets than unarmed Whites. However, experimental subjects were significantly less likely to make performance errors (i.e., selecting “shoot”) in response to both unarmed Black and White bystanders. This evidence suggests that people can mediate the influence of implicit bias via the adoption of implementation intentions.
The effectiveness of implementation intentions is not restricted to mediating implicit bias. Implementation intentions have also been found to mediate framing effects (Trötschel & Gollwitzer Reference Trötschel and Gollwitzer2007), mimicry effects (Wieber et al. Reference Wieber, Gollwitzer and Sheeran2014), ego depletion (Webb & Sheeran Reference Webb and Sheeran2002), and mood effects (Webb et al. Reference Webb, Schweiger Gallo, Miles, Gollwitzer and Sheeran2012). It is important to point out that the evidence on implementation intentions suggests that subjects can regulate defeaters without needing to override their automatic processing. This is because the ‘if’ component of the plan modifies the actor's automatic processing prior to encountering the specific circumstance, thereby shifting attention away from potential defeaters and toward goal-relevant situational cues (Gallo et al. Reference Gallo, Keil, McCulloch, Rockstroh and Gollwitzer2009). Thus, there is a growing amount of evidence suggesting that implementation intentions can mediate the influence of defeaters.
It should be admitted here that there may be some limits to the self-regulatory strategies described in my initial sketch. Lauren Olin and John Doris have noted some limits on expertise (Olin & Doris Reference Olin and Doris2014). They discuss some studies suggesting that physicians become worse at diagnosing heart sounds after years of practice (Choudhry et al. Reference Choudhry, Fletcher and Soumerai2005; Ericsson et al. Reference Ericsson, Whyte and Ward2007), and experts are not better than non-experts at making economic predictions (Tetlock Reference Tetlock2009). Thus, it seems that practice sometimes results in worse performance, and it may well be that some of these failures involve defeaters.
Doris also claims that implementation intentions may be limited because “countervailing pressures may perturb the goal relevant situation” (Doris Reference Doris2015b, p. 128). Someone might also worry that implementation intentions may collaborate or produce defeaters in specific circumstances. The increased salience of situational cues contained in the “if” component of one's plan may automatically attract attention even when these cues are not relevant to one's current goal (Parks-Stamm & Gollwitzer Reference Parks-Stamm, Gollwitzer, Moskowitz and Grant2009). There is evidence that the human capacity for attention is limited (Shiffrin & Schneider Reference Shiffrin and Schneider1977), and so implementation intentions may influence us in such a way that we are sometimes unable to attend to relevant contextual cues during the pursuit of multiple goals. Perhaps even some of these failures of attention may be construed as defeaters, where the causes of the actor's behavior (i.e., the ‘if’ component of the plan together with the situational cue) would not be recognized as reason(s) justifying their behavior.
Admittedly, the strategies suggested here may be limited in various ways. However, the purpose of my discussion was only to present an initial sketch that enjoys some empirical support. I have not attempted to provide a fully developed account of enhancement that can adequately address all of these potential limitations. Obviously, the details of this account need to be developed further so that these limitations are adequately addressed. But I do not see sufficient evidence for thinking that this project is unlikely to produce a compelling response to Amy (and others). In conclusion, accounts of responsible agency face an important practical challenge: Specify conditions and strategies wherein actors can circumvent the influence of defeaters. I hope to have provided an initial sketch of such an account.