Allison M. Sweeney1and Anne Moyer2
1Department of Psychology, University of South Carolina, Columbia, SC, USA
2Department of Psychology, Stony Brook University, Stony Brook, NY, USA
People, especially those at risk, tend to resist persuasive appeals to change their behavior. When people are uninterested in changing a health behavior, they may try to discredit unwanted health information by forming counterarguments or alternative explanations (Ditto & Lopez, 1992). To the extent that people tend to prefer positive information that reflects well on the self, it is not surprising, then, that people scrutinize unfavorable information more than favorable information (Giner‐Sorolila & Chaiken, 1997). To the extent that health information reminds people of their shortcomings, it can be a potent threat to their positive self‐views. Ironically, the more personally relevant a health message is, the less likely an individual is to process it objectively (Liberman & Chaiken, 1992).
One promising approach to overcoming resistance to health information involves leading people to think about their most important values or actions, a process known as self‐affirmation (Steele, 1988). Self‐affirmation theory proposes that highlighting important sources of self‐worth restores or reinforces an individual's overall sense of self‐integrity. By making salient aspects of the self that are important to one's identity, but are unrelated to the threat at hand, self‐affirmation helps to offset any potential self‐threat elicited by health information (Sherman & Cohen, 2006). As a result, self‐affirmation should reduce any motivation to respond defensively to health information. Supporting this prediction, a number of studies have found that when people are self‐affirmed, they are more likely to process health information in an objective manner (Reed & Aspinwall, 1998; Sherman, Nelson, & Steele, 2000).
Much of the early research on self‐affirmation and health focused on the ability ofself‐affirmation to change people's responses to health information, including message acceptance (Sherman et al., 2000). Further efforts have been made to clarify whether increasing people's openness toward health information translates into a change in future health‐related behaviors. Several studies indicate that self‐affirmation facilitates behavior change across a diverse range of behaviors, including physical activity, fruit and vegetable consumption (Epton & Harris, 2008), alcohol consumption (Armitage, Harris, & Arden, 2011), and adherence to medication (Wileman et al., 2014). Such studies typically examine the influence of self‐affirmation over the course of a week or longer, with some studies finding effects extending over several months (Harris et al., 2014; Wileman et al., 2014).
Self‐affirmation has a small‐sized effect on health behaviors (Epton, Harris, Kane, van Koningsbruggen, & Sheeran, 2015; Sweeney & Moyer, 2015). Although small, the effect size of self‐affirmation is similar to those obtained in meta‐analyses of other health behavior change interventions (e.g., Gallagher & Updegraff, 2012). Thus, self‐affirmation holds considerable promise as a useful framework for explaining when people accept health information and, in turn, make changes in their behavior. In this chapter we will examine some of the recent developments and discoveries among studies of self‐affirmation, including (a) the incorporation of diverse research methods, (b) the identification of moderating variables, and (c) efforts to combine self‐affirmation with other intervention strategies. We will then highlight two major questions for future research: (a) how does self‐affirmation influence health behaviors and (b) what is the trajectory of self‐affirmation?
Many early studies on self‐affirmation and health‐related outcomes consisted of laboratory‐based studies that measured immediate responses to health information. After completing a self‐affirmation or a control task, people were presented with health information and reported on their responses to it, such as behavioral intentions and message processing (Reed & Aspinwall, 1998; Sherman et al., 2000). Within the last decade, however, researchers have begun to examine whether self‐affirmation impacts outcomes beyond people's self‐reported experiences. An emerging literature encompassing a diverse array of methods suggests that self‐affirmation impacts the body's immediate and physiological response to threat. For example, relative to non‐affirmed individuals, people who are self‐affirmed have a lower cortisol response when exposed to a stressful situation (Creswell et al., 2005) and take less time for their mean arterial blood pressure to return to baseline after receiving threatening information (Schmeichel & Tang, 2015).
To date, two studies have examined the influence of self‐affirmation on neural activity.
To examine the effect of self‐affirmation on immediate sensitivity to threat, participants completed a speeded response time task while their neural activity was recorded with electroencephalography (EEG) (Legault, Al‐Khindi, & Inzlicht, 2012). The response time task was a go/no‐go task that involved making or withholding a response depending on the type of stimuli presented. Participants were instructed to respond as quickly as possible and, thus, were prone to errors. The researchers tested whether self‐affirmation would impact people's sensitivity to making errors by examining their error‐related negativity (ERN) response, an event‐related potential found in past research to index motivational significance. Participants who self‐affirmed prior to the go/no‐go task displayed a greater ERN response, suggesting that they were more sensitive toward making mistakes during the response time task. To the extent that errors are experienced as aversive, the authors interpreted this finding as suggesting that self‐affirmation increases people's openness to processing threat.
Adopting a different approach, one recent study used functional magnetic resonance imaging (fMRI) to examine neural activity during the processing of health information (Falk et al., 2015). The results indicated that people who were self‐affirmed showed greater activation in the ventromedial prefrontal cortex (VMPFC), an area of the brain associated with self‐related processing when reading the health message. Furthermore, increased activation in the VMPFC was associated with greater health behavior change, suggesting that self‐affirmation may engender health behavior change by allowing people to process otherwise threatening health information as self‐relevant. Taken together, research by Legault et al. (2012) and Falk et al. (2015) provide initial evidence that applying a neurophysiological approach to self‐affirmation may be a useful framework for further understanding the cognitive and motivational processes that underlie self‐affirmation.
Another notable change among studies of self‐affirmation has been an increased emphasis on measuring behavioral changes arising from self‐affirmation. Attempts to examine the impact of self‐affirmation on health behaviors have relied primarily on self‐reports (e.g., Armitage et al., 2011; Epton & Harris, 2008). Recently, however, a few studies have begun to incorporate more objective measures of health behavior change. For example, one study found that self‐affirmation led to a reduction in sedentary behavior, as assessed with accelerometers worn by participants for 1 month (Falk et al., 2015). Another study examined whether self‐affirmation would encourage hemodialysis patients to adhere to recommendations (Wileman et al., 2014). By measuring patients' serum phosphate levels across 1 year, the researchers were able to show a positive effect of self‐affirmation on medication adherence.
As research on self‐affirmation has accumulated, researchers have identified several factors that moderate the effectiveness of self‐affirmation. For example, it may be worthwhile to consider the content of health messages and the way that this interacts with self‐affirmation. One study examined the strength of a message related to the risk of fibrocystic disease (FBD) in those who consumed caffeinated beverages (Klein, Harris, Ferrer, & Zajac, 2011). The rationale was that if those who are self‐affirmed are more likely to attend to messages and become sensitive to their strength, then self‐affirmation should result in stronger inclinations to change behavior in response to strong messages. Messages of high strength about the link between caffeine consumption and fibrocystic disease cited reputable organizations and sources that supported the link; messages of low strength cited less well‐known organizations and mentioned sources that did not support the link. Significant interactions between message strength and self‐affirmation were found for feelings of vulnerability (i.e., worry) and intentions to reduce caffeine intake, such that these were higher in those who were self‐affirmed and were exposed to the strong message. Furthermore, feelings of vulnerability were found to mediate the effect on behavior change intentions, suggesting that attention to message strength may be facilitated by self‐affirmation's effect on worry.
An additional consideration in the effectiveness of self‐affirmation is the characteristics (i.e., disposition, life history, or affect) of those who are exposed to health messages. Importantly, if individual differences do interact with self‐affirmation's effects, examining only main effects may result in some important effects being obscured, if moderating effects are not examined (Düring & Jessop, 2015). For instance, people are continually exposed to conflicting health information in their everyday lives, and this can result in confusion and feelings of fatalism about such things as the causes of cancer (Niederdeppe & Gurmankin Levy, 2007). For individuals who perceive high levels of ambiguity about a health topic, health messages may be especially threatening because, in addition to the disconcerting nature of the information, any additional information may only increase this sense of ambiguity (Klein, Hamilton, Harris, & Han, 2015). Self‐affirmation may be particularly useful in reducing defensiveness and facilitating receptivity to health messages in people who feel confused by health recommendations. This idea was tested in a study that exposed women with varying levels of risk and perceptions of ambiguity about the causes of cancer to a clear, unambiguous, and authoritative article describing the connection between high alcohol consumption and increased breast cancer risk (Klein et al., 2015). Self‐affirmation increased message acceptance in those who perceived high levels of ambiguity in extant cancer prevention recommendations.
Although self‐affirmation can be experimentally induced, people are also thought to exhibit tendencies to spontaneously self‐affirm (Pietersma & Dijkstra, 2012). These tendencies are assessed by items such as “When I feel threatened or anxious, I find myself thinking about my strengths” (Harris, Napper, Griffin, Schuz, & Stride, in press, cited in Ferrer et al., 2015). Such tendencies have been found to moderate the relationship between anticipated (but not current) negative affect and intentions to obtain genetic test results for diseases that did not have any medically actionable precautionary measures, such as Huntington disease (Ferrer et al., 2015). Similarly, spontaneous self‐affirmation and optimism moderated the tendencies of those who tend to avoid information to have lowered intentions to learn results for such diseases (Taber et al., 2015).
In addition to individual differences in the tendency to spontaneously self‐affirm, another potential moderating variable of self‐affirmation's effects is self‐esteem. To the extent that those with low self‐esteem have fewer resources to draw upon when presented with messages that threaten the sense of self, self‐affirmation may prove especially beneficial. Accordingly, those with low self‐esteem who were affirmed had more positive attitudes toward behavioral change (increasing exercise), stronger endorsement of intentions to exercise, and lower levels of message derogation, but were no different in terms of behavioral control or actual behavior change (Düring & Jessop, 2015).
So, considering moderating characteristics of individuals or health messages is promising, and looking beyond main effects of self‐affirmation can prove illuminating. Some further considerations include whether self‐affirmation's effects for those who perceive high levels of ambiguity extend to receptivity to, for instance, more ambiguous messages (Klein et al., 2015). Weaker messages may also be more ambiguous; for example, in the study cited above, the strong message read: “The link between FBD and breast cancer is well‐established and FBD has been recognized as a breast cancer risk factor by the Centers for Disease Control and Prevention,” whereas the weak message read: “At a recent meeting of the American Academy of Osteopathic Medicine, a small group of dieticians with links to the fast food industry issued a formal statement supporting the link between FBD and breast cancer as well as the link between caffeine and FBD” (p. 1239). Given the potential effects of perceptions of ambiguity, it could be important to disentangle the effects of message strength and level of ambiguity.
Another approach to expanding research on self‐affirmation has been to combine self‐affirmation with other health behavior change techniques. For example, implementation intentions(“if‐then” plans that concretely specify how to reach a particular goal; Gollwitzer, 1999) are thought to be useful in channeling the motivation generated via self‐affirmation into volition and actual behavior change (Jessop, Sparks, Buckland, Harris, & Churchill, 2014; van Dijk & Dijkstra, 2014). Accordingly, some research has begun investigating the effectiveness of incorporating this technique to enhance the effectiveness of self‐affirmation.
Armitage and colleagues (Armitage et al., 2011; Armitage, Rowe, Arden, & Harris, 2014) developed a self‐affirmation implementation intention intervention to reduce alcohol consumption in adolescents. Forming implementation intentions for alcohol consumption may help individuals plan appropriate behavioral responses when critical situations are encountered, such as being offered a drink. A goal in the design of this intervention was to affirm the self without essay writing or questionnaire elaboration, which depends on good verbal fluency, in order to make it more broadly implementable. Accordingly, participants completed the stem, “If I feel threatened or anxious, then I will…” with one of four options that involved thinking about things they valued in themselves, remembering things they have succeeded in, or thinking about things they stood for, before viewing a diagram depicting the body parts and conditions affected by alcohol consumption. In a first study, this briefer intervention was as effective as a more traditional self‐affirmation manipulation and more effective than a control condition in reducing alcohol consumption 1 month later (Armitage et al., 2011), and, in a second study, more effective than a control condition 2 months later (Armitage et al., 2014).
Harris et al. (2014) examined the independent and synergistic effects of self‐affirmation and implementation intentions with a 2 × 2 study design. There was a significant interaction between the two techniques at the 7‐day but not the 3‐month follow‐up, such that self‐affirmed participants who formed implementation intentions prior to being exposed to a message emphasizing the benefits of eating fruits and vegetables consumed significantly more. Conversely, however, in a series of two studies with a similar 2 × 2 design, combining self‐affirmation and implementation intentions resulted in a negative effect on exercise behavior at 1 week (Jessop et al., 2014). The authors suggested that the counterintuitive finding may have stemmed from the combined techniques producing conflicting construal levels of the behavior or counteracting types of information processing.
Other research has examined whether self‐affirmation interacts with aspects of the health message, such as the framing of health information in terms of gains or losses. One study tested the combined effects of self‐affirmation and message framing on intentions to engage in indoor tanning (Mays & Zhao, 2016). Because indoor tanning, despite being a significant risk factor for skin cancer and melanoma, is believed to be related to the sense of self and feelings of attractiveness, self‐affirmation was thought to be a particularly appropriate technique to prevent defensiveness to loss‐framed messages. Loss‐framed messages (that emphasize the costs of tanning) may be more threatening than gain‐framed messages (that emphasize the benefits of not tanning). After completing a self‐affirmation or a control task, participants viewed an image with an accompanying message that emphasized either the risks of indoor tanning or the benefits of avoiding indoor tanning.
For both intentions to indoor tan and intentions to quit indoor tanning, contrary to expectations, there was no interaction between the framing of the message and self‐affirmation. In addition, although there were main effects for message framing, with loss‐framed messages being more effective, the main effect for self‐affirmation was only significant for intentions to tan, with those in the self‐affirmation conditions reporting stronger intentions to tan. An explanation for this puzzling finding was suggested by the fact that self‐affirmation led to perceptions of lower argument strength. The authors speculated that because indoor tanning is so tied to the self‐concept, the self‐affirmation manipulation could have increased the salience of the importance of indoor tanning and, thus, exacerbated defensiveness to the messages. This finding also brings attention to the notion that the type of target behavior, and its relevance to sense of self, may be an important consideration in self‐affirmation research.
In a rare example of using self‐affirmation techniques in clinical samples, patients with cardiopulmonary disease (angioplasty, hypertension, and asthma) were exposed to a positive affect and self‐affirmation intervention, in addition to a patient education workbook that the control groups also received, in three randomized trials (Mancuso et al., 2012; Ogedegbe et al., 2012; Peterson et al., 2012). The self‐affirmation induction involved thinking of something participants had done that they were proud of and the suggestion that reflecting on these could help them overcome challenges in engaging in a new health behavior. In this instance, the self‐affirmation intervention was intended, in combination with techniques aimed to increase positive affect, such as thinking about small things that made them feel good, to overcome barriers to behavioral change rather than to reduce defensiveness to a message. Related to the point made earlier, the trials with different populations had somewhat different target behaviors, which also yielded different effects: in the angioplasty patients the intervention resulted in increased physical activity relative to the control group; in the hypertensive patients the intervention resulted in increased medication adherence relative to the control group; and in the asthmatic patients the intervention and control groups were no different in terms of physical activity. It is also important to note that, where the interventions were effective, the extent to which the effects can be attributed to the self‐affirmation manipulation versus the components intended to create positive affect is not clear.
In sum, the emerging evidence regarding the combination of self‐affirmation with other intervention techniques is mixed, such that the research on even the most commonly studied added technique, implementation intentions, shows both increases and decreases in effectiveness. However, the literature is diverse, with different target behaviors, study populations, follow‐up time points, and even types of self‐affirmation inductions, which may contribute to this lack of uniformity. Considering aspects of the message itself (e.g., framing) may be especially worthwhile, as these vary in the literature also, and, as most messages are framed either positively or negatively, giving explicit attention to this is worthwhile. Finally, some authors have pointed out that self‐affirmation may play only a small role in the larger behavior change process, among other techniques, such as self‐monitoring and identifying barriers, so continuing to consider its role in combined interventions is useful, despite the mixed evidence that has emerged thus far (van Dijk & Dijkstra, 2014).
Having highlighted some of the recent developments among studies of self‐affirmation and health‐related outcomes, we will now turn to discussing some of the major research questions that remain for future research.
Although numerous studies have indicated that self‐affirmation increases acceptance of health information and facilitates behavior change, surprisingly little is known about how it impacts health outcomes. To date, researchers have focused primarily on people's self‐reported responses to health information as a source of potential mediators. For example, one possibility is that self‐affirmation changes people's attitudes toward health information (e.g., by increasing perceptions of risk), which, in turn, leads to changes in health behavior. Although there is some preliminary evidence that an increase in perceived vulnerability mediates the effect of self‐affirmation on behavioral intentions (Klein et al., 2011), other studies have found that neither attitudes toward the health behavior nor perceived threat (Armitage et al., 2014) mediates the effect of self‐affirmation on health behavior.
Another possibility is that self‐affirmation engenders a change in intentions, which, in turn, leads to a change in behavior. Supporting this prediction, one study found that among people who are at a higher risk for developing type 2 diabetes, intentions to take an online diabetes risk test mediated the effect of self‐affirmation on risk test participation (van Koningsbruggen & Das, 2009). Other studies of self‐affirmation, however, have failed to find a mediating effect of intentions on behavior change (Armitage et al., 2014). Further casting doubt on the viability of behavioral intentions as a mediator, several studies have found that self‐affirmation does not always elicit a positive change in behavioral intentions (Harris et al., 2014; Reed & Aspinwall, 1998). Among studies of self‐affirmation measuring both health intentions and behavior, one meta‐analytic review found that intention effect sizes did not predict behavior effect sizes (Sweeney & Moyer, 2015), suggesting that a change in behavioral intentions does not translate to a comparable change in behavior. Taken together, limited research supports a causal intention–behavior relation among self‐affirmation studies.
Whereas much of the extant research has focused on the role of single potential mediating variables, future research may examine whether self‐affirmation influences health outcomes through multiple variables. Few self‐affirmation studies have adopted a path analysis approach to examine whether a set of variables helps to explain the impact of self‐affirmation on health behavior. One exception is by Armitage, Harris, Hepton, and Napper (2008) who found that health message acceptance mediated the relation between self‐affirmation and intentions and intentions mediated the relation between message acceptance and behavior. Self‐affirmation has been applied to health behaviors that vary across a number of dimensions, including the type of behavior (i.e., health promoting vs. health damaging) and the temporal impact of the behavior (i.e., immediate vs. distant consequences). Given this variability, in addition to the inherent complexities associated with long‐term health behavior change, future research may consider adopting an analytic approach that examines a set of variables and takes into account variables specific to the targeted health behavior.
Another avenue for future research may involve testing mediating variables that extend beyond people's immediate self‐reported responses to health information. Several studies have suggested that self‐affirmation facilitates a broader perspective from which to view information (Critcher & Dunning, 2015; Sherman et al., 2013). For example, when people are self‐affirmed, they are more likely to think about actions in terms of their abstract, superordinate aspects (e.g., why to improve one's health), rather than in terms of their concrete, subordinate aspects (e.g., how to improve one's health; Sherman et al., 2013).
In the context of health behaviors, a broader perspective may be helpful for several reasons. For example, threatening information may be experienced as less aversive when viewed from a distance (Sherman, 2013). Offering preliminary support for this prediction, one study tested whether thinking in broad (vs. narrow) terms would influence tanners' receptivity toward health information on the risks of tanning (Belding, Naufel, & Fujita, 2015). Tanners led to think that broad, superordinate categories were more motivated to reduce their risk of skin cancer than tanners led to think in narrow, subordinate categories. These authors suggest that thinking in broad terms promotes long‐term self‐change motivation, rather than short‐term self‐protection motivation. Relatedly, when a threat or temptation does not loom quite so largely, a broadened perspective may make it easier for people to exert self‐control. Several studies have indicated that people are better at practicing self‐control when led to think in broad (vs. narrow) terms, as evidenced, for example, by differences in preferences for healthy versus unhealthy foods (Fujita & Han, 2009).
Another possibility is that a broadened perspective helps people to connect their current actions to their long‐term aspirations. Supporting this prediction, past research has found that construing information in broad (vs. narrow) terms facilitates attention toward one's long‐term goals when faced with reminders of goal‐related temptations (Fujita & Sasota, 2011). Such research suggests, for example, that a dieter who adopts a broader perspective may find it easier to bring to mind his or her dieting goal when faced with tempting food. Furthermore, other research has found that a broad (vs. narrow) perspective helps people to recognize the commonalities across their various life goals (Clark & Freitas, 2013). As a result, a broadened perspective may make it easier to connect a targeted health behavior (e.g., eating more fruits and vegetables) with one's long‐term goals (e.g., living a long and happy life).
Having established that self‐affirmation has the potential to be a useful tool for promoting health behavior change, an important next step for researchers is to develop an empirically supported mechanistic account of self‐affirmation. Given that people's immediate self‐reported responses to health information do not appear to mediate the effect of self‐affirmation, future research may consider other ways in which self‐affirmation impacts the self and the consequences this has for evaluation, motivation, and goal‐directed action. Identifying the mechanisms that underlie self‐affirmation effects may help to further refine self‐affirmation theory as a whole and also offers important practical benefits, such as increasing understanding of the specific conditions under which self‐affirmation is most effective.
Much of the research on self‐affirmation and health has focused on the immediate changes that occur after self‐affirming, such as changes in message acceptance. However, measures of people's deliberate responses to health information after self‐affirming have yielded some inconsistent findings. As noted previously, for example, self‐affirmation does not always lead to a positive change in behavioral intentions (e.g., Harris et al., 2014; Jessop et al., 2014). Similarly, other studies have failed to find a significant effect of self‐affirmation on outcomes such as perceived behavioral control, perceived threat, message derogation, or self‐efficacy (Armitage et al., 2008, 2014; Jessop et al., 2014). Although most studies have assessed whether self‐affirmation elicits immediate changes in health‐related cognitions, one recent study measured intentions and attitudes immediately after participants read a health message and again during a 1‐week follow‐up. There were no immediate differences in intentions and attitudes between the self‐affirmation and control group; however, the self‐affirmation group did express greater intentions and attitudes 1 week later.
Such findings may lead one to wonder whether self‐affirmation elicits an immediate change or is the effect gradual? In a meta‐analytic review Epton et al. (2015) found that self‐affirmation exerts a smaller effect on outcomes that are measured immediately after encountering health information (i.e., message acceptance) than measures that typically involve some delay (i.e., behavior). Importantly, however, the time of measurement of a health behavior (e.g., days vs. weeks) has not been found to moderate the effect of self‐affirmation on behavior (Epton et al., 2015; Sweeney & Moyer, 2015). Furthermore, as reviewed previously, a recent study using fMRI found that when people were self‐affirmed, they engaged in greater self‐related processing, as reflected through activity in the VMPFC, while reading health information (Falk et al., 2015). Additionally, neural activity during message processing predicted changes in health behavior that were distinct from changes predicted from participants' self‐reports of behavioral intentions and attitudes. These findings suggest that self‐affirmation enacts immediate effects on the self (as reflected through participant's neural activity) that relate to subsequent behavior change.
Taken together, self‐affirmation appears to exert some immediate influences, as evidenced, for example, by studies showing that self‐affirmation produces an immediate change in physiological responses to threat (e.g., Creswell et al., 2005) and in neural activity (e.g., Falk et al., 2015). However, the observed heterogeneity in people's self‐reported responses suggests that self‐affirmation may not elicit a change in health‐related cognitions that can be reliably captured at a conscious level. That is, some of the influence of self‐affirmation may occur outside of people's conscious awareness.
Increased attention is being given to understanding how health‐related decisions are shaped by both conscious and unconscious processes (Sheeran, Gollwitzer, & Bargh, 2013). To date, relatively few studies have examined the implicit effects of self‐affirmation in relation to health outcomes. One exception is Klein and Harris (2009) who used a dot probe task to examine whether self‐affirmation enhances people's tendency to direct their attention toward threatening health information. After completing a self‐affirmation or control task and reading a message about the risks of alcohol consumption, female participants completed a task in which a neutral and a threatening word from the health message were presented simultaneously and followed by a dot. Participants were asked to quickly identify whether the dot appeared on the left or right side of the screen. Self‐affirmed participants were faster to identify the location of the dot when it appeared after a threat‐related word, suggesting that self‐affirmed participants displayed a stronger attentional bias for threat‐related words than non‐affirmed participants. Whereas other studies have used self‐report measures to assess differences in perceived threat, this methodology allowed for a novel test of the extent to which self‐affirmation elicits differences in implicit processing of health information.
Another exception is van Koningsbruggen, Das, and Roskos‐Ewoldsen (2009) who used a lexical decision‐making task to examine whether self‐affirmation increases responsiveness toward health information at an implicit level. After completing a self‐affirmation or control task and reading a message about the risks of caffeine, coffee drinkers and non‐coffee drinkers completed a response time task that required them to distinguish between non‐words, neutral words, and threat‐related words (e.g., “heart disease”). Among coffee drinkers, those who self‐affirmed were faster at recognizing threat‐related words than non‐affirmed individuals, suggesting self‐affirmation increased accessibility of threat‐related cognitions among people for whom the health message was most relevant.
These two studies provide initial support for the potential usefulness of implicit measures in studies of self‐affirmation and health. In light of recent research suggesting that people may not always be able to reliably report on conscious changes that arise after self‐affirmation, future research may consider adopting a framework that encompasses both conscious and unconscious processes. Increasingly health behavior change models are adopting a dual‐systems approach that incorporates both reflective and automatic health‐related processes (Sheeran et al., 2013). Whereas much of the extant research has focused on conscious changes arising from self‐affirmation, future research is needed to further elucidate the nonconscious influences of self‐affirmation.
Research on self‐affirmation and health outcomes has grown extensively in recent decades. Building upon studies that highlighted the potential of self‐affirmation as a tool for reducing defensive processing, numerous studies have been devoted to understanding the different types of health‐related cognitions that are influenced by self‐affirmation and any implication this may have for health behavior change. In this chapter, we highlighted three recent developments in this area of research including the adoption of a wide array of methodological approaches, emphasis on identifying moderating variables, and efforts to combine self‐affirmation with other behavior change techniques. By drawing attention to the need for developing a mechanistic account of self‐affirmation and the need to further our understanding the trajectory of self‐affirmation, we hope this chapter will encourage future investigations of self‐affirmation as a tool for health behavior change.
Allison M. Sweeney is a postdoctoral researcher at the University of South Carolina. Integrating theories and methods across health, social, and cognitive psychology, her research focuses on developing theory‐based interventions for promoting health behavior change.
Anne Moyer is an associate professor of social and health psychology at Stony Brook University. Her research focuses on psychosocial factors surrounding cancer and cancer risk, medical decision making, research methodology and research synthesis, and the psychological aspects of human research participation.