Age Differences in Beliefs About Emotion Regulation Strategies

Age Differences in Beliefs About Emotion Regulation Strategies Abstract Objectives: Age shifts in emotion regulation may be rooted in beliefs about different strategies. We test whether there are age differences in the beliefs people hold about specific emotion regulation strategies derived from the process model of emotion regulation and whether profiles of emotion beliefs vary by age. Method An adult life-span sample (N = 557) sorted 13 emotion regulation strategies either by (a) how effective the strategies would be or (b) how likely they would be to use them, in 15 negative emotion-eliciting situations. Results Younger adults ranked attentional and cognitive distraction more effective than older adults, and preferred avoidance, distraction, and rumination more (and attentional deployment less) than middle-aged and older adults. Latent profile analysis on preferences identified three distinct strategy profiles: Classically adaptive regulators preferred a variety of strategies; situation modifiers showed strong preferences for changing situations; a small percentage of people preferred avoidance and rumination. Middle-aged and older adults were more likely than younger adults to be classically adaptive regulators (as opposed to situation modifiers or avoiders/ruminators). Discussion These findings provide insight into the reasons people of different ages may select and implement different emotion regulation strategies, which may influence their emotional well-being. Beliefs, Emotion, Emotion regulation, Implicit theories Theories of life-span emotional development propose that because aging is associated with both gains and losses in resources and functioning, older adults must shift their emotion regulation efforts to maintain high levels of well-being (Charles, 2010; Urry & Gross, 2010). Recent decades have seen a dramatic increase in research on emotion regulation in general. Studies have shown that (a) there are stable individual differences, including age differences, in the use of emotion regulation strategies (e.g., John & Gross, 2004), (b) strategies vary in their effectiveness (e.g., Webb, Miles, & Sheeran, 2012), and (c) effectiveness varies by age (e.g., Shiota & Levenson, 2009). Though research has investigated why such individual differences exist, few studies have examined the beliefs people hold about the effectiveness of the strategies themselves. We propose that due to age-related changes in motivation and resources, there may be systematic age differences in beliefs about the effectiveness and use of specific emotion regulation strategies, and these beliefs guide moment-to-moment choices about which regulation strategy is best for them. Socioemotional selectivity theory (SST) suggests that older adults are especially motivated to pursue emotional well-being goals due to a more limited future time perspective (Carstensen, Isaacowitz, & Charles, 1999). Older adults may therefore be more likely than younger adults to prefer emotion regulation strategies that are most effective in minimizing negative and maximizing positive emotions. The strength and vulnerability integration (SAVI) model proposes that greater life experience provides knowledge about which emotion regulation strategies are likely to support these goals, even as declining cognitive and physical resources may diminish the effectiveness of some strategies (Charles, 2010; see also Urry & Gross, 2010). Given their experience, older adults may better recognize which strategies are likely to be effective compared to younger adults. Meanwhile, older adults may also shift their use of such strategies away from those that rely on cognitive or physical resources that diminish with age and toward those that avoid the emotional experience altogether (Urry & Gross, 2010). Emotion Regulation in the Process Model The process model of emotion regulation proposes five families of emotion regulation strategies people can use to influence emotions, based on which part of the emotion-generative process they target (Gross, 1998). People can select or modify a situation based on their expectations for emotional change (situation selection and modification); direct their attention toward or away from aspects of the situation (attentional deployment); change the way they are thinking (cognitive change), such as by changing their appraisal of the situation or their own feelings (reappraisal); or target the emotional response—expression, behavior, or physiology (response modulation). Within each strategy family are specific tactics (e.g., using detached vs positive reappraisal; Shiota & Levenson, 2009). An expanded version of the process model proposes that in each case of emotion regulation, a person must decide whether to regulate, which strategy to use (i.e., which aspect of the emotion-generative cycle to target), and which specific tactic within that strategy to implement (Gross, 2015). In each case, a person evaluates the costs and benefits of regulating and of using each tactic. Survey and laboratory studies have identified age differences in both the use and effectiveness of emotion regulation strategies. Although studies generally do not find age differences in use or effectiveness of situation selection (Sands, Livingstone, & Isaacowitz, in press), older adults may use situation modification more (Livingstone & Isaacowitz, 2015). Eye tracking studies of spontaneous attentional deployment show older adults shift attention away from negative and toward positive stimuli, though this is not always associated with decreases in negative affect (Isaacowitz, 2012). Some self-report studies suggest older adults use more reappraisal and less suppression (John & Gross, 2004), whereas others find the opposite (Nolen-Hoeksema & Aldao, 2011). Studies that instruct reappraisal show mixed effects on emotion, with older adults generally successful at using positive reappraisal (Nowlan, Wythrich, & Rapee, 2015) but not detached reappraisal (e.g., Shiota & Levenson, 2009). Several studies have found no age differences in use or effectiveness of expressive suppression (e.g., Shiota & Levenson, 2009). In general, these studies assess either one or a small number of regulation strategies or tactics; to date, there has been no systematic study of age differences in strategy use and effectiveness across the process model. As a starting point, we propose to investigate how beliefs about these strategies and tactics vary by age. Emotion Regulation Beliefs Research on beliefs in other domains shows they have strong effects on motivation, affect, and behavior (Dweck, 1999). In the domain of emotion, negative beliefs about emotion (e.g., that emotions are irrational or damaging) have been linked to emotion regulation difficulties, less use of adaptive regulation strategies (e.g., cognitive reappraisal, problem-solving), and more use of maladaptive strategies (e.g., rumination, avoidance; Trincas, Bilotta, & Mancini, 2016). Beliefs about the effectiveness of regulation strategies can be identified early in development (Dennis & Kelemen, 2009), and greater recognition of effective strategies in children is linked to greater use of constructive strategies in response to frustration (Cole, Dennis, Smith-Simon, & Cohen, 2009). Adults are likely to engage in specific strategies when they believe they will be effective. For example, people engage in aggressive behaviors primarily when they believe it will make them feel better (Bushman, Baumeister, & Phillips, 2001). Likewise, positive beliefs about rumination (e.g., “I need to ruminate about the bad things that have happened in the past to make sense of them”) are linked with increased use (Papageorgiou & Wells, 2001). However, there are also individual differences in which strategies people find effective (Ortner, Briner, & Marjanovic, 2017). Age Differences in Emotion Regulation Beliefs The first goal of this research was to characterize age differences in beliefs regarding the effectiveness and use of a variety of emotion regulation strategies. To investigate this question, we collected data from younger, middle-aged, and older adults. Increasing attention to the role of context has highlighted that the effectiveness of strategies varies by situation (e.g., Troy, Shallcross, & Mauss, 2013). We therefore developed a scenario-based measure of emotion regulation beliefs and generated situations eliciting a range of negative emotions. Additionally, whereas most studies on emotion regulation investigate a small number of strategies, we included 13 emotion regulation tactics across the five strategy families proposed by the process model (Gross, 1998). Although emotion regulation strategies vary in their effectiveness depending on context (e.g., Troy et al., 2013), some strategies have been repeatedly linked with better or poorer outcomes. In particular, meta-analysis has linked problem-solving (akin to situation modification), reappraisal, and acceptance with more positive mental health outcomes, and rumination, suppression, and avoidance with negative ones (Aldao, Nolen-Hoeksema, & Schweizer, 2010). One possibility is that older adults, having more experience with emotions (Charles, 2010) and more motivation to maximize well-being (Carstensen et al., 1999), have stronger beliefs about the effectiveness of emotion regulation than younger adults, and are better able to identify putatively effective strategies. On the other hand, age-related declines may make certain strategies more demanding and less efficient (Urry & Gross, 2010), especially those that occur late in the emotion-generative process, once physiological arousal has occurred (Charles, 2010). SAVI proposes that reduced physiological flexibility makes older adults less effective at down-regulating from highly arousing emotional experiences once they are underway. Other strategies, such as reappraisal, may require the flexible use of cognitive resources, which decline with age (Urry & Gross, 2010). Consequently, some strategies, such as cognitive change and response modulation, may be less effective and less often used in old age. If this is the case, older adults’ beliefs are more likely to favor strategies that intervene earlier in the emotion-generative process: situation selection, situation modification, and attentional deployment. When examining age differences in emotion regulation, individual differences in preferences for strategies must be disentangled from differences in the effectiveness of those strategies (Isaacowitz, 2012). These two aspects of emotion regulation have different conceptual and real-life implications. For example, though a person may prefer to use certain strategies, they may not be especially effective for them. Conversely, a person may be able to effectively utilize a strategy in the lab, but be unlikely to employ the strategy in real life. Our focus in the current research was to characterize these beliefs, which we examined in two separate samples to avoid beliefs in one domain influencing beliefs in the other. Though this means that we cannot test the relationship between the two sets of beliefs, we can compare patterns of age differences in beliefs about both effectiveness and use. The second goal of this research was to identify profiles of emotion regulation beliefs and examine whether they vary with age. Profiles may emerge based on the families of strategies derived from the process model. For example, some people may believe that focusing emotion regulation efforts on the situation is most effective, others focusing on cognition, and others focusing on the response. On the other hand, age-related belief patterns could reflect a generalized positivity preference in line with positivity effects identified by SST (Mather & Carstensen, 2005). Age-related positivity effects refer to findings that older adults favor processing of positive over negative information in both attention and memory (Reed, Chan, & Mikels, 2014). This preference may also manifest in choice of emotion regulation strategies (see Livingstone & Isaacowitz, 2016): Older adults may prefer specific tactics like attending toward positive aspects of the situation and away from negative (Isaacowitz, 2012) and positive reappraisal (Shiota & Levenson, 2009). An exploratory third goal was to examine individual differences in beliefs about emotion more broadly as a possible correlate of beliefs about specific emotion regulation strategies. We focus on the degree to which a person believes emotions are malleable (and can be controlled) or fixed (and must play out on their own; Tamir, John, Srivastava, & Gross, 2007). Stronger malleability beliefs may be associated with beliefs that favor more putatively effective strategies, especially antecedent-focused (e.g., situation modification, attentional deployment) and emotion-engaging strategies that target cognition (e.g., cognitive change). In contrast, weaker malleability beliefs may be associated with beliefs that favor more response-focused strategies (e.g., expressive suppression) and emotion-disengaging strategies (e.g., avoidance). One study has found that older adults report weaker malleability beliefs than younger adults (Cabello & Fernández-Berrocal, 2015), but we know very little about how age and emotion malleability beliefs together predict beliefs about emotion regulation. We therefore assessed trait-level emotion malleability beliefs along with beliefs about the effectiveness of and preferences for emotion regulation strategies and examined how these beliefs vary with age. Method Participants Power analyses using G*Power (version 3.1.9) for a MANOVA with 13 strategies and three age groups showed that a total sample size of 171 would have 80% power to detect a small effect (ηp2 = .02) at an α level of .05. Therefore, for each version of the study, we collected at least 60 participants in each age group for each version before stopping data collection. The sample consisted of 557 adults (57% female, 73% White, 3% Asian, 6% Black, 7% Hispanic) ages 18–87 (Mage = 42.57, SD = 17.85). Most participants (N = 402) were recruited from www.mturk.com, an online platform for recruiting workers to perform tasks, and were paid $1.05. Additional participants were recruited from a subject pool at a large private university (N = 106) and were given partial course credit, and from an existing database of adults over 60 (N = 49) who had previously been recruited from the Boston community through print and online advertisements; they were entered into a raffle to win a $100 gift card. On MTurk, we used the quota feature in Qualtrics to limit enrolment by age group to ensure adequate numbers of middle-aged (ages 40–59) and older adults (ages 60+), who may be underrepresented on the platform. Participants were randomly assigned to complete either the effectiveness (Ns = 123 YA, 79 MA, 77 OA) or the likelihood version (Ns = 129 YA, 84 MA, 65 OA). Materials Emotion scenarios Rather than ask participants to report on global beliefs about the effectiveness and likelihood of use of emotion regulation strategies, we specified a range of emotion-eliciting situations and asked participants to classify the strategies for each scenario. This procedure was designed to encourage participants to visualize emotion regulation in specific situations, and allowed participants to acknowledge variability in effectiveness and use across contexts (Aldao, 2013; Bonanno & Burton, 2013). Fifteen emotion-eliciting scenarios (see Supplementary Materials) were generated to address a range of negative emotions (e.g., sadness, anger, disgust, guilt) based on core relational themes outlined by Lazarus (1991). The scenarios were phrased broadly to be applicable to people of differing ages and life situations (e.g., “Someone has said something hurtful to you”). An independent sample (N = 66, ages 20–73) rated scenarios for negativity, familiarity, recency, incidence, and likelihood of future occurrence. Descriptive statistics for scenarios are available in Supplementary Materials. Emotion regulation items Thirteen emotion regulation strategies (see Supplementary Materials) were designed to represent specific tactics within the five stages of emotion regulation as outlined by the process model (Gross, 1998, 2015). Beliefs about emotion malleability Participants completed the four-item implicit theories of emotion questionnaire (Tamir et al., 2007). Internal consistency was good (α = .797). Scores were rescaled to a range from 0 to 100 (Cohen, Cohen, Aiken, & West, 2010), with higher scores indicating stronger belief that emotions are malleable (e.g., agrees more strongly with items such as “If they want to, people can change the emotions that they have.”) This scale was used to examine how beliefs about emotion relate to profiles of emotion regulation. Procedure Surveys were hosted on Qualtrics.com (Qualtrics, Provo, UT). MTurk participants first completed a demographic screening questionnaire, for which they were paid $0.05. Participants who qualified based on age were given the option to complete an additional 20-min survey for $1.00. After providing informed consent, participants read instructions and completed a practice sorting task followed by the main sorting task. Participants then completed a second demographic questionnaire and the implicit theories of emotion questionnaire. Final demographic statistics were based on the second demographic questionnaire, which was also used to verify age; those with mismatched age reports were excluded. Participants recruited from campus and the local community simply saw the consent form, tasks, and demographic survey. Effectiveness sorting task Instructions read, “We are interested in the many ways that people can manage their emotions or feelings. For the following situations, imagine yourself in that situation and think about what would be the most effective way for you to deal with your emotion. Assume that you want to feel less negative emotion.” Participants were presented with one scenario at a time, written in large font at the top of the web page, in a randomized order. Strategies were presented by description (not name) in a fixed order (see Supplemental Materials) on the left side of the screen. On the right side of the screen were three boxes, labeled Most effective (“These strategies would be most effective in helping you feel less negative emotion”), Somewhat effective (“These strategies may be effective in helping you feel less negative emotion”), and Least effective (“These strategies would not be effective in helping you feel less negative emotion, or may even make things worse”). Three categories were chosen rather than rating scales to reduce participant fatigue across the 15 scenarios and to allow participants to visually compare strategies using different boxes. Participants sorted the strategies by dragging each strategy into the box that best matched their rated effectiveness in that scenario. To ensure adequate variability, participants were asked to include at least three strategies in each box and to sort all strategies into a box. Likelihood of use sorting task Instructions were the same as for the effectiveness version except for the third sentence: “We are not necessarily interested in what you would do to deal with the situation, but rather what you would do to deal with emotions that arise in that situation.” Materials and procedures were the same as for the effectiveness task, except that participants were asked to sort the strategies into those that they would be Most likely to use (“These are strategies that you would most likely use if faced with the situation”), Somewhat likely to use (“These strategies are the ones you may use if you were faced with the situation”), or Least likely to use (“These strategies are the ones you would be least likely to use, or would definitely not use, if faced with the situation”). Results Effectiveness Beliefs Responses were scored 1 (Most effective), 0 (Somewhat effective) or −1 (Least effective). Emotion regulation strategies generally had acceptable internal consistency across scenarios (average Cronbach’s α = .819), so were averaged across scenarios to create a continuous score of for each strategy for each participant ranging from −1 to 1. Multivariate analysis of variance (MANOVA) tested age differences in effectiveness beliefs across 13 strategies. Because of known gender differences in emotion regulation (e.g., Nolen-Hoeksema & Aldao, 2011), gender was included as a covariate. The multivariate test was marginally significant for age, Wilks Λ = .848, F(26,458) = 1.51, p = .054, ηp2 = .079. There was a significant age difference in cognitive change—distraction and a marginal age difference for attentional deployment—away from negative (see Table 1). Younger adults rated attentional deployment and cognitive distraction as more effective than older adults; middle-aged adults did not differ from either group, ps > .05. There was also an age difference for cognitive change—rumination, F(2,240) = 3.34, p = .037, ηp2 = .027; middle-aged adults rated rumination as less effective than older adults; younger adults did not differ from either group. Table 1. Descriptive Statistics and Inferential Tests of Age Differences in Emotion Regulation Beliefs   Mean Belief (SD)        Younger  Middle-Aged  Older  F  p  ηp2  Effectiveness        F(2,240)       SS Avoid  −.31 (.43)  −.38 (.36)  −.44 (.45)  1.42  .243  .012   SS Leave  −.05 (.33)  −.08 (.32)  −.15 (.45)  1.45  .237  .012   SM Modify  .56 (.33)  .61 (.29)  .59 (.35)  0.60  .551  .005   AD Negative  .41 (.30)  .37 (.31)  .30 (.28)  3.01  .051  .024   AD Positive  .35 (.35)  .28 (.39)  .42 (.41)  2.12  .123  .017   CC Detached  .03 (.42)  .09 (.44)  .13 (.44)  1.18  .310  .010   CC Distract  .09 (.34)  −.02 (.43)  −.11(.40)  5.24**  .006  .042   CC Positive  .25 (.37)  .16 (.40)  .27 (.43)  1.44  .238  .012   CC Ruminate  −.78 (.28)  −.87 (.24)  −.74 (.37)  3.33*  .037  .027   CC Accept  .02 (.49)  .03 (.49)  .14 (.49)  1.05  .350  .009   RM Suppress  −.18 (.34)  −.23 (.42)  −.21 (.39)  0.45  .640  .004   RM Mask  −.11 (.47)  −.09 (.46)  −.12 (.50)  0.07  .928  .001   RM Physiology  .12 (.43)  .12 (.49)  .03 (.48)  1.17  .313  .010  Preferences        F(2,238)       SS Avoid  −.12 (.38)  −.27 (.42)  −.37 (.43)  7.77**  .001  .061   SS Leave  .04 (.33)  −.06 (.34)  −.08 (.38)  3.11*  .046  .025   SM Modify  .51 (.29)  .55 (.27)  .54 (.30)  0.67  .515  .006   AD Negative  .16 (.33)  .32 (.38)  .31 (.35)  6.05**  .003  .048   AD Positive  .04 (.41)  .24 (.45)  .23 (.48)  5.87  .003  .047   CC Detached  .01 (.37)  .07 (.40)  −.01 (.44)  0.78  .462  .006   CC Distract  .02 (.37)  −.13 (.42)  −.14 (.43)  4.76**  .009  .038   CC Positive  −.13 (.43)  .06 (.46)  .05 (.53)  4.84**  .009  .039   CC Ruminate  −.03 (.49)  −.30 (.48)  −.34 (.51)  10.29***  <.001  .080   CC Accept  −.01 (.41)  .09 (.49)  .05 (.48)  1.23  .295  .010   RM Suppress  .02 (.39)  −.04 (.47)  −.01 (.35)  0.65  .523  .005   RM Mask  −.12 (.43)  −.16 (.45)  −.09 (.46)  0.41  .664  .003   RM Physiology  −.08 (.45)  −.02 (.55)  .07 (.38)  1.87  .157  .015    Mean Belief (SD)        Younger  Middle-Aged  Older  F  p  ηp2  Effectiveness        F(2,240)       SS Avoid  −.31 (.43)  −.38 (.36)  −.44 (.45)  1.42  .243  .012   SS Leave  −.05 (.33)  −.08 (.32)  −.15 (.45)  1.45  .237  .012   SM Modify  .56 (.33)  .61 (.29)  .59 (.35)  0.60  .551  .005   AD Negative  .41 (.30)  .37 (.31)  .30 (.28)  3.01  .051  .024   AD Positive  .35 (.35)  .28 (.39)  .42 (.41)  2.12  .123  .017   CC Detached  .03 (.42)  .09 (.44)  .13 (.44)  1.18  .310  .010   CC Distract  .09 (.34)  −.02 (.43)  −.11(.40)  5.24**  .006  .042   CC Positive  .25 (.37)  .16 (.40)  .27 (.43)  1.44  .238  .012   CC Ruminate  −.78 (.28)  −.87 (.24)  −.74 (.37)  3.33*  .037  .027   CC Accept  .02 (.49)  .03 (.49)  .14 (.49)  1.05  .350  .009   RM Suppress  −.18 (.34)  −.23 (.42)  −.21 (.39)  0.45  .640  .004   RM Mask  −.11 (.47)  −.09 (.46)  −.12 (.50)  0.07  .928  .001   RM Physiology  .12 (.43)  .12 (.49)  .03 (.48)  1.17  .313  .010  Preferences        F(2,238)       SS Avoid  −.12 (.38)  −.27 (.42)  −.37 (.43)  7.77**  .001  .061   SS Leave  .04 (.33)  −.06 (.34)  −.08 (.38)  3.11*  .046  .025   SM Modify  .51 (.29)  .55 (.27)  .54 (.30)  0.67  .515  .006   AD Negative  .16 (.33)  .32 (.38)  .31 (.35)  6.05**  .003  .048   AD Positive  .04 (.41)  .24 (.45)  .23 (.48)  5.87  .003  .047   CC Detached  .01 (.37)  .07 (.40)  −.01 (.44)  0.78  .462  .006   CC Distract  .02 (.37)  −.13 (.42)  −.14 (.43)  4.76**  .009  .038   CC Positive  −.13 (.43)  .06 (.46)  .05 (.53)  4.84**  .009  .039   CC Ruminate  −.03 (.49)  −.30 (.48)  −.34 (.51)  10.29***  <.001  .080   CC Accept  −.01 (.41)  .09 (.49)  .05 (.48)  1.23  .295  .010   RM Suppress  .02 (.39)  −.04 (.47)  −.01 (.35)  0.65  .523  .005   RM Mask  −.12 (.43)  −.16 (.45)  −.09 (.46)  0.41  .664  .003   RM Physiology  −.08 (.45)  −.02 (.55)  .07 (.38)  1.87  .157  .015  Note: Possible scores ranged from −1 to 1. SS = situation selection. AD = Attentional deployment; CC = Cognitive change; RM = Response modulation; SM = Situation modification. *p < .05, **p < .01, ***p < .001. View Large Table 1. Descriptive Statistics and Inferential Tests of Age Differences in Emotion Regulation Beliefs   Mean Belief (SD)        Younger  Middle-Aged  Older  F  p  ηp2  Effectiveness        F(2,240)       SS Avoid  −.31 (.43)  −.38 (.36)  −.44 (.45)  1.42  .243  .012   SS Leave  −.05 (.33)  −.08 (.32)  −.15 (.45)  1.45  .237  .012   SM Modify  .56 (.33)  .61 (.29)  .59 (.35)  0.60  .551  .005   AD Negative  .41 (.30)  .37 (.31)  .30 (.28)  3.01  .051  .024   AD Positive  .35 (.35)  .28 (.39)  .42 (.41)  2.12  .123  .017   CC Detached  .03 (.42)  .09 (.44)  .13 (.44)  1.18  .310  .010   CC Distract  .09 (.34)  −.02 (.43)  −.11(.40)  5.24**  .006  .042   CC Positive  .25 (.37)  .16 (.40)  .27 (.43)  1.44  .238  .012   CC Ruminate  −.78 (.28)  −.87 (.24)  −.74 (.37)  3.33*  .037  .027   CC Accept  .02 (.49)  .03 (.49)  .14 (.49)  1.05  .350  .009   RM Suppress  −.18 (.34)  −.23 (.42)  −.21 (.39)  0.45  .640  .004   RM Mask  −.11 (.47)  −.09 (.46)  −.12 (.50)  0.07  .928  .001   RM Physiology  .12 (.43)  .12 (.49)  .03 (.48)  1.17  .313  .010  Preferences        F(2,238)       SS Avoid  −.12 (.38)  −.27 (.42)  −.37 (.43)  7.77**  .001  .061   SS Leave  .04 (.33)  −.06 (.34)  −.08 (.38)  3.11*  .046  .025   SM Modify  .51 (.29)  .55 (.27)  .54 (.30)  0.67  .515  .006   AD Negative  .16 (.33)  .32 (.38)  .31 (.35)  6.05**  .003  .048   AD Positive  .04 (.41)  .24 (.45)  .23 (.48)  5.87  .003  .047   CC Detached  .01 (.37)  .07 (.40)  −.01 (.44)  0.78  .462  .006   CC Distract  .02 (.37)  −.13 (.42)  −.14 (.43)  4.76**  .009  .038   CC Positive  −.13 (.43)  .06 (.46)  .05 (.53)  4.84**  .009  .039   CC Ruminate  −.03 (.49)  −.30 (.48)  −.34 (.51)  10.29***  <.001  .080   CC Accept  −.01 (.41)  .09 (.49)  .05 (.48)  1.23  .295  .010   RM Suppress  .02 (.39)  −.04 (.47)  −.01 (.35)  0.65  .523  .005   RM Mask  −.12 (.43)  −.16 (.45)  −.09 (.46)  0.41  .664  .003   RM Physiology  −.08 (.45)  −.02 (.55)  .07 (.38)  1.87  .157  .015    Mean Belief (SD)        Younger  Middle-Aged  Older  F  p  ηp2  Effectiveness        F(2,240)       SS Avoid  −.31 (.43)  −.38 (.36)  −.44 (.45)  1.42  .243  .012   SS Leave  −.05 (.33)  −.08 (.32)  −.15 (.45)  1.45  .237  .012   SM Modify  .56 (.33)  .61 (.29)  .59 (.35)  0.60  .551  .005   AD Negative  .41 (.30)  .37 (.31)  .30 (.28)  3.01  .051  .024   AD Positive  .35 (.35)  .28 (.39)  .42 (.41)  2.12  .123  .017   CC Detached  .03 (.42)  .09 (.44)  .13 (.44)  1.18  .310  .010   CC Distract  .09 (.34)  −.02 (.43)  −.11(.40)  5.24**  .006  .042   CC Positive  .25 (.37)  .16 (.40)  .27 (.43)  1.44  .238  .012   CC Ruminate  −.78 (.28)  −.87 (.24)  −.74 (.37)  3.33*  .037  .027   CC Accept  .02 (.49)  .03 (.49)  .14 (.49)  1.05  .350  .009   RM Suppress  −.18 (.34)  −.23 (.42)  −.21 (.39)  0.45  .640  .004   RM Mask  −.11 (.47)  −.09 (.46)  −.12 (.50)  0.07  .928  .001   RM Physiology  .12 (.43)  .12 (.49)  .03 (.48)  1.17  .313  .010  Preferences        F(2,238)       SS Avoid  −.12 (.38)  −.27 (.42)  −.37 (.43)  7.77**  .001  .061   SS Leave  .04 (.33)  −.06 (.34)  −.08 (.38)  3.11*  .046  .025   SM Modify  .51 (.29)  .55 (.27)  .54 (.30)  0.67  .515  .006   AD Negative  .16 (.33)  .32 (.38)  .31 (.35)  6.05**  .003  .048   AD Positive  .04 (.41)  .24 (.45)  .23 (.48)  5.87  .003  .047   CC Detached  .01 (.37)  .07 (.40)  −.01 (.44)  0.78  .462  .006   CC Distract  .02 (.37)  −.13 (.42)  −.14 (.43)  4.76**  .009  .038   CC Positive  −.13 (.43)  .06 (.46)  .05 (.53)  4.84**  .009  .039   CC Ruminate  −.03 (.49)  −.30 (.48)  −.34 (.51)  10.29***  <.001  .080   CC Accept  −.01 (.41)  .09 (.49)  .05 (.48)  1.23  .295  .010   RM Suppress  .02 (.39)  −.04 (.47)  −.01 (.35)  0.65  .523  .005   RM Mask  −.12 (.43)  −.16 (.45)  −.09 (.46)  0.41  .664  .003   RM Physiology  −.08 (.45)  −.02 (.55)  .07 (.38)  1.87  .157  .015  Note: Possible scores ranged from −1 to 1. SS = situation selection. AD = Attentional deployment; CC = Cognitive change; RM = Response modulation; SM = Situation modification. *p < .05, **p < .01, ***p < .001. View Large Likelihood of Use Beliefs Responses of Most likely were assigned a value of 1, Somewhat likely a value of 0, and Least likely a value of −1. Emotion regulation strategies generally had acceptable internal consistency across scenarios (average α = .805), so were averaged across scenarios to create a continuous score of for each strategy for each participant ranging from −1 to 1. MANOVA tested age differences in beliefs about the likelihood of use across 13 strategies, also controlling for gender. The multivariate test was significant for age, Wilk’s Λ = .804, F(26, 454) = 2.01, p = .003, ηp2 = .104. Younger adults were more likely to prefer avoidance, leaving, distraction, and rumination compared to middle-aged and older adults (see Table 1). They were also less likely to prefer attentional deployment—away from negative and toward positive aspects—and to prefer positive reappraisal. Latent Profile Analysis on Emotion Regulation Use We were also interested in identifying the extent to which younger, middle-aged, and older adults could be differentially classified by their emotion regulation beliefs. Because we found more significant age differences in the likelihood of using different strategies, we focused on preferences for use rather than on effectiveness beliefs. We used latent profile analysis (LPA), a person-centered method for classifying types of people into latent variables using continuous indicators. It is statistically similar to latent class analysis, using continuous rather than categorical indicators (Muthén, 2001), and conceptually similar to cluster analysis, using a probabilistic rather than a distance based model (Muthén & Muthén, 2000; Pastor, Barron, Miller, & Davis, 2007). LPA additionally allows for the estimation of a latent model structure, which enables the calculation of goodness-of-fit indices and comparisons among nested model structures, and the inclusion of covariates in predicting latent profile membership (Muthén, 2004). We applied a mixed-process approach to test a series of conceptually-driven, nested solutions that were examined based on statistical criteria and ease of interpretation (Marsh, Lüdtke, Trautwein, & Morin, 2009). Specifically, we tested whether the beliefs about emotion regulation strategy use of younger, middle-aged, and older adults were best classified into 2-, 3-, 4-, or 5-profile solutions. This decision was based on prominent emotion regulation theory (the five stages of the process model, Gross, 1998) and recent empirical tests of emotion regulation profiles in clinical samples (e.g., Chesney & Gordon, 2017). Data were analyzed using Mplus version 7.4 (Muthèn & Muthèn, 1998–2015). The 13 likelihood of use strategy scores were modeled as continuous observed indicators. We included age group and gender as covariates in the estimation of profile membership. At least 500 sets of random parameter start values were generated for each model to protect against a local maxima solution (Muthèn, 2001). Missing data (6%) were likely missing at random (Enders, 2011) and thus addressed using full information maximum likelihood estimation. Model selection was based on the examination of several criteria including the Bayesian Information Criterion (BIC), entropy value, Lo-Mendell-Rubin Test (LMRT), Parametric Bootstrapped Likelihood Ratio Test (BLRT), and ease of interpretation of the solution. Decreasing BIC values, entropy ≥ .90, and significant LMRT and BLRT indicate better-fitting solutions (Nylund, Asparouhov, & Muthén, 2007). After the profile groups were identified, multinomial logistic regressions were conducted to identify age differences in latent profile class membership, which allowed us to simultaneously examine the possibility that middle-aged and older adults were more likely to be in a profile group than were younger adults. Profile solution Fit statistics for the two-, three-, four-, and five-profile solutions are presented in Table 2. The BIC values decreased as the number of profiles increased. Entropy values were highest in the three- and four-profile solutions. Results from the LMRT indicated that the three-profile solution was significantly more favorable than the two-profile solution (p < .05); however, the four-profile solution was not significantly better than the three-profile solution (p > .05). Given that solutions with increased profiles are generally better fit to the data (Hagenaars & McCutcheon, 2009), BLRT results were less informative. In examining the specific profile solutions for the three- and four-profile structures, a clear theoretically-relevant pattern was observed for the three-profile solution. These profile patterns were replicated in the four-profile solution, with the addition of a fourth fragmented profile consisting of few people and no clear emotion regulation strategy likelihood-of-use pattern. Thus, we interpret the more parsimonious three-profile solution as the best-fitting structure. Table 2. Model Fit Statistics for Two-, Three-, Four-, and Five-Latent Profile Solutions Model  Log-likelihood  BIC  Entropy  LMRT  LRMTp  BLRT  BLRTp  Two-profile  −2,226.10  4,696.30  0.82  485.40  .001  485.40  .000  Three-profile  −2,078.54  4,497.69  .93  295.12  .016  292.10  .017  Four-profile  −2,001.74  4,440.58  .91  153.80  .075  153.61  .000  Five-profile  −1,959.90  4,453.41  .90  83.68  .240  83.68  .000  Model  Log-likelihood  BIC  Entropy  LMRT  LRMTp  BLRT  BLRTp  Two-profile  −2,226.10  4,696.30  0.82  485.40  .001  485.40  .000  Three-profile  −2,078.54  4,497.69  .93  295.12  .016  292.10  .017  Four-profile  −2,001.74  4,440.58  .91  153.80  .075  153.61  .000  Five-profile  −1,959.90  4,453.41  .90  83.68  .240  83.68  .000  Note: Bolded profile indicates best-fitting, parsimonious solution. BIC = Bayesian Information Criterion; BLRT = Parametric Bootstrapped Likelihood Ratio Test; LMRT = Lo-Mendell-Rubin Likelihood Ratio Test; Better fit is indicated by decreasing BIC, entropy ≥ .90, and significant LMRT and BLRT. View Large Table 2. Model Fit Statistics for Two-, Three-, Four-, and Five-Latent Profile Solutions Model  Log-likelihood  BIC  Entropy  LMRT  LRMTp  BLRT  BLRTp  Two-profile  −2,226.10  4,696.30  0.82  485.40  .001  485.40  .000  Three-profile  −2,078.54  4,497.69  .93  295.12  .016  292.10  .017  Four-profile  −2,001.74  4,440.58  .91  153.80  .075  153.61  .000  Five-profile  −1,959.90  4,453.41  .90  83.68  .240  83.68  .000  Model  Log-likelihood  BIC  Entropy  LMRT  LRMTp  BLRT  BLRTp  Two-profile  −2,226.10  4,696.30  0.82  485.40  .001  485.40  .000  Three-profile  −2,078.54  4,497.69  .93  295.12  .016  292.10  .017  Four-profile  −2,001.74  4,440.58  .91  153.80  .075  153.61  .000  Five-profile  −1,959.90  4,453.41  .90  83.68  .240  83.68  .000  Note: Bolded profile indicates best-fitting, parsimonious solution. BIC = Bayesian Information Criterion; BLRT = Parametric Bootstrapped Likelihood Ratio Test; LMRT = Lo-Mendell-Rubin Likelihood Ratio Test; Better fit is indicated by decreasing BIC, entropy ≥ .90, and significant LMRT and BLRT. View Large The three latent profiles of emotion regulation likelihood of use are illustrated in Figure 1 and descriptive results are presented in Table 3. The first profile, Situation Modifiers, included people (N = 137) who said they were most likely to use situation modification and least likely to use response modulation strategies. The second profile, Classically Adaptive Regulators, consisted of people (N = 137) who said they were most likely to use situation modification, attentional deployment, and positive reappraisal strategies—all nominally adaptive strategies—and least likely to use rumination and situation selection strategies. The final profile, Avoiders and Ruminators, consisted of people (N = 18) who said they most were most likely to use rumination and avoidant situation selection strategies and least likely to use situation modification and attentional deployment strategies. Additional profile descriptive statistics are presented in Supplementary Materials. Figure 1. View largeDownload slide Latent profiles of emotion regulation preferences. A three-profile solution provided the best fit to the data. The first profile (N = 137) included participants who preferred situation modification over other strategies. The second profile (N = 137) included participants who preferred to use classically adaptive strategies like situation modification, attentional deployment, and reappraisal. The third profile (N = 18) included participants who preferred to use avoidance and rumination, which are often considered less adaptive strategies. Figure 1. View largeDownload slide Latent profiles of emotion regulation preferences. A three-profile solution provided the best fit to the data. The first profile (N = 137) included participants who preferred situation modification over other strategies. The second profile (N = 137) included participants who preferred to use classically adaptive strategies like situation modification, attentional deployment, and reappraisal. The third profile (N = 18) included participants who preferred to use avoidance and rumination, which are often considered less adaptive strategies. Table 3. Standardized Means and Standard Errors for Emotion Regulation Strategy Likelihood of Use across the Three Latent Profiles   Profile 1 Situation Modifiers  Profile 2 Classically Adaptive Regulators  Profile 3 Avoiders and Ruminators  Strategy  M  SE  M  SE  M  SE  SS Avoidance  −0.21  0.10  −1.00  0.13  1.16  0.27  SS Leave  −0.38  0.10  −1.11  0.13  0.86  0.34  SM Modification  1.83  0.13  2.21  0.15  −1.99  0.29  AD Negative  0.43  0.09  1.18  0.13  −1.55  0.39  AD Positive  0.03  0.09  0.92  0.11  −1.12  0.30  CC Detachment  0.06  0.08  0.11  0.10  0.21  0.23  CC Distraction  −0.03  0.08  0.11  0.10  −0.32  0.20  CC Pos. Reappraisal  −0.42  0.09  0.52  0.12  −0.39  0.24  CC Rumination  0.34  0.15  −3.04  0.33  3.17  0.50  CC Acceptance  0.08  0.08  −0.11  0.10  −0.32  0.25  RM Suppression  −0.20  0.08  −0.32  0.10  0.22  0.36  RM Positive  −0.52  0.08  −0.06  0.11  −0.08  0.35  RM Physio  −0.33  0.09  0.31  0.10  −0.26  0.29    Profile 1 Situation Modifiers  Profile 2 Classically Adaptive Regulators  Profile 3 Avoiders and Ruminators  Strategy  M  SE  M  SE  M  SE  SS Avoidance  −0.21  0.10  −1.00  0.13  1.16  0.27  SS Leave  −0.38  0.10  −1.11  0.13  0.86  0.34  SM Modification  1.83  0.13  2.21  0.15  −1.99  0.29  AD Negative  0.43  0.09  1.18  0.13  −1.55  0.39  AD Positive  0.03  0.09  0.92  0.11  −1.12  0.30  CC Detachment  0.06  0.08  0.11  0.10  0.21  0.23  CC Distraction  −0.03  0.08  0.11  0.10  −0.32  0.20  CC Pos. Reappraisal  −0.42  0.09  0.52  0.12  −0.39  0.24  CC Rumination  0.34  0.15  −3.04  0.33  3.17  0.50  CC Acceptance  0.08  0.08  −0.11  0.10  −0.32  0.25  RM Suppression  −0.20  0.08  −0.32  0.10  0.22  0.36  RM Positive  −0.52  0.08  −0.06  0.11  −0.08  0.35  RM Physio  −0.33  0.09  0.31  0.10  −0.26  0.29  Note: Estimates are standardized on the observed indicators and covariates (i.e., STDYX in Mplus). AD = Attentional deployment; CC = Cognitive change; RM = Response modulation; SM = Situation modification; SS = Situation selection. View Large Table 3. Standardized Means and Standard Errors for Emotion Regulation Strategy Likelihood of Use across the Three Latent Profiles   Profile 1 Situation Modifiers  Profile 2 Classically Adaptive Regulators  Profile 3 Avoiders and Ruminators  Strategy  M  SE  M  SE  M  SE  SS Avoidance  −0.21  0.10  −1.00  0.13  1.16  0.27  SS Leave  −0.38  0.10  −1.11  0.13  0.86  0.34  SM Modification  1.83  0.13  2.21  0.15  −1.99  0.29  AD Negative  0.43  0.09  1.18  0.13  −1.55  0.39  AD Positive  0.03  0.09  0.92  0.11  −1.12  0.30  CC Detachment  0.06  0.08  0.11  0.10  0.21  0.23  CC Distraction  −0.03  0.08  0.11  0.10  −0.32  0.20  CC Pos. Reappraisal  −0.42  0.09  0.52  0.12  −0.39  0.24  CC Rumination  0.34  0.15  −3.04  0.33  3.17  0.50  CC Acceptance  0.08  0.08  −0.11  0.10  −0.32  0.25  RM Suppression  −0.20  0.08  −0.32  0.10  0.22  0.36  RM Positive  −0.52  0.08  −0.06  0.11  −0.08  0.35  RM Physio  −0.33  0.09  0.31  0.10  −0.26  0.29    Profile 1 Situation Modifiers  Profile 2 Classically Adaptive Regulators  Profile 3 Avoiders and Ruminators  Strategy  M  SE  M  SE  M  SE  SS Avoidance  −0.21  0.10  −1.00  0.13  1.16  0.27  SS Leave  −0.38  0.10  −1.11  0.13  0.86  0.34  SM Modification  1.83  0.13  2.21  0.15  −1.99  0.29  AD Negative  0.43  0.09  1.18  0.13  −1.55  0.39  AD Positive  0.03  0.09  0.92  0.11  −1.12  0.30  CC Detachment  0.06  0.08  0.11  0.10  0.21  0.23  CC Distraction  −0.03  0.08  0.11  0.10  −0.32  0.20  CC Pos. Reappraisal  −0.42  0.09  0.52  0.12  −0.39  0.24  CC Rumination  0.34  0.15  −3.04  0.33  3.17  0.50  CC Acceptance  0.08  0.08  −0.11  0.10  −0.32  0.25  RM Suppression  −0.20  0.08  −0.32  0.10  0.22  0.36  RM Positive  −0.52  0.08  −0.06  0.11  −0.08  0.35  RM Physio  −0.33  0.09  0.31  0.10  −0.26  0.29  Note: Estimates are standardized on the observed indicators and covariates (i.e., STDYX in Mplus). AD = Attentional deployment; CC = Cognitive change; RM = Response modulation; SM = Situation modification; SS = Situation selection. View Large Age differences in profile membership Consistent with the MANOVA analyses (see Table 1), multinomial logistic regression results revealed significant age differences in the odds of being in one profile versus another. Younger adults were chosen as the reference category a priori given our interests in comparing young to older adults; the inclusion of middle-aged adults was largely exploratory but could examine whether age differences emerged earlier (YA vs MA and OA) or later (YA vs OA only). Middle-aged and older adults had greater odds of being Classically Adaptive Regulators than Situation Modifiers (ORoa = 3.59, p < .001; ORma = 69.69, p < .001) or Avoiders and Ruminators (ORoa = 18.90, p = .006; ORma = 51.33, p = .008) relative to younger adults. Profile membership odds did not vary by gender (ps = .111–.610). Profile and age differences in emotion malleability beliefs We explored whether these profile categorizations predicted people’s beliefs about the malleability of emotion. Specifically, we ran a regression model predicting emotion malleability beliefs with profile membership. Age group was also included as a predictor in the regression models given the observed age differences in profile membership odds. Results indicated a significant effect of profile membership on malleability beliefs (β = 0.19, p ≤ .002). Controlling for age group, Situation Modifiers were more likely to hold fixed beliefs about emotion than Classically Adaptive Regulators (β = −0.17, p = .009) and Avoiders and Ruminators (β = −0.27, p = .039). However, there were no significant differences in emotion malleability beliefs between Classically Adaptive Regulators and Avoiders and Ruminators (β = −0.10, p = .457). Age group was not significantly related to emotion malleability beliefs (β = −0.06, p = .298). Profile membership and age group together accounted for 3.5% of the variance in emotion malleability beliefs. Discussion If we aim to understand how older adults can maintain high levels of emotional well-being, it is important to investigate emotion regulatory processes and the beliefs that guide those processes. The purpose of the present research was to characterize age differences in beliefs about emotion regulation strategies—how effective people perceive various tactics to be and how likely they believe they are to use them. The findings provide more nuanced insight into the factors that may influence age differences in the use and effectiveness of emotion regulation strategies. Age Differences (and Lack Thereof) in Emotion Regulation Beliefs Age differences do emerge in beliefs about emotion regulation strategies, though they are more likely to appear in beliefs about use than in beliefs about effectiveness. The findings suggest that younger adults generally agree with older adults on which strategies are effective, but diverge in which ones they use. This result is interesting because actual effectiveness of at least some regulation strategies may vary by age (e.g., Noh, Lohani, & Isaacowitz, 2011; Shiota & Levenson, 2009; Urry & Gross, 2010). It may therefore be important in future research to distinguish between beliefs about effectiveness in general and beliefs about personal effectiveness (see De Castella et al., 2013), as well as to assess whether some age groups are more likely than others to admit to utilizing strategies that they do not necessarily believe to be most effective. Of the age differences that did emerge, younger adults ranked cognitive and attentional distraction as more effective than middle-aged and older adults (though the latter effect was only marginally significant), but reported being less likely to use attentional deployment. These results generally align with past work showing consistent patterns of attention away from negative material in older adults (use) but mixed findings regarding the relation between attention and affect (effectiveness). For example, attentional deployment can effectively reduce negative affect for those adults who have stronger attentional control abilities, but not those without (Isaacowitz, 2012). In our data, older (and middle-aged adults) reported being more likely to use attentional deployment than younger adults, but rated it as less effective. It may be that individual differences in cognitive resources drive this discrepancy: Older age groups show a positivity effect, but it is only effective for some of them. There was some support for the idea that older (and middle-aged) adults turn to more putatively effective strategies than younger adults. Although participants generally agreed on the effectiveness of the strategies, younger adults were more likely to use avoidance and rumination (which have been linked to negative mental health outcomes; see Aldao et al., 2010) than middle-aged and older adults and less likely to use positive reappraisal (which has been linked to positive emotional outcomes in older adults (Nowlan et al., 2015); reappraisal in general has been linked with more positive outcomes (e.g., Aldao et al., 2010; Webb et al., 2012). The observed pattern of age differences did not support the idea that older adults more often try to intervene early in the emotion-generative process and avoid later strategies of cognitive change and response modulation, or that older adults prefer positive varieties of emotion regulation (Livingstone & Isaacowitz, 2016). Instead, a third possibility emerged, suggesting that younger adults may be more inclined to disengage from (rather than engage with) emotions compared to middle-aged and older adults. In particular, younger adults rated directing attention away from negative and cognitive distraction—two forms of mental disengagement—as more effective than older adults. They were also more likely to report using avoidance, cognitive distraction, and rumination than middle-aged and older adults. These results suggest a tendency for younger adults to try to disengage—cognitively and behaviorally—from emotional situations. This tendency may decline with age, as people learn to more effectively deal with emotions, and draw on knowledge and experience to engage with emotions effectively, though longitudinal research is needed to test this idea. Profiles of Emotion Regulation Using latent profile analysis, we identified three profiles of preferences for emotion regulation strategies. Slightly less than half of participants were classified as classically adaptive regulators, relying on a variety of situation-based and cognitive strategies that have been generally shown to be effective. A similar percentage of people were classified as situation modifiers, who showed a strong preference for changing situations. A small percentage of participants were classified as avoiders and ruminators, relying relatively more on these traditionally maladaptive strategies and less on situation modification and attentional deployment. Middle-aged and older adults were more likely to be classified as classically adaptive regulators than situation modifiers or avoiders and ruminators, compared to younger adults, which supports the idea that emotion regulation skills may improve with age (Charles, 2010). These people appear to demonstrate regulatory flexibility, reporting that they would rely on a greater variety of strategies than other profiles, suggesting a larger repertoire from which to choose, which is likely to benefit them across numerous contexts and domains (Bonanno & Burton, 2013). In contrast, avoidance and rumination are linked with poorer mental health (Aldao et al., 2010); it may be that younger adults have not yet learned that these strategies are ineffective. On the other hand, as an antecedent-focused strategy, situation modification is often considered an adaptive way of regulating one’s emotions (Gross, 1998). In contrast to classically adaptive regulators, situation modifiers were less likely to believe that emotions are malleable. For people with these beliefs, then, if strategies that target the emotion are unlikely to work, the most effective way to manage emotions may be to target the situation itself. The strategy is likely to be effective to the extent to which a situation can actually be modified. When a situation is uncontrollable, an emotion-focused approach may be more effective (Zakowski, Hall, Klein, & Baum, 2001). These profiles did not align with the families of strategies proposed by the process model, or with the idea that older adults may prefer more “positive” versions of strategies (e.g., pay more attention to positive than negative, use positive reappraisal). Instead, the pattern seems more nuanced. The distinction between putatively adaptive and maladaptive strategies is clear, with only a small number of people showing a distinctly maladaptive profile. In addition, there is a distinction between situation-focused and emotion-focused regulation, which parallels the distinction in the coping literature between problem-focused and emotion-focused coping (Lazarus & Folkman, 1984). Limitations and Future Directions A clear next step is to investigate the extent to which effectiveness and use beliefs are related to actual emotion regulation behavior in the lab and in real life (e.g., using experience sampling). Research in children suggests that beliefs about emotion regulation strategies predict in-lab behavior (Cole et al., 2009). Emotion regulation theory also identifies strategy selection as an important step in the emotion regulation process (Gross, 2015). Open questions include whether beliefs about effectiveness predict effective emotion regulation, beliefs about which strategies people would use predict what strategies they do use, and the extent to which these sets of beliefs (effectiveness and preferences) are related. Although we might predict a model in which age is related to differences in these beliefs, which in turn predict differences in likelihood of use, the current study examined beliefs in separate samples, to avoid beliefs in one category influencing the other. It would be interesting to examine within-person beliefs systems, and whether a feedback loop exists between the two over time. For example, if emotion regulation skills improve with age, we would expect a stronger correlation between effectiveness and likelihood of use beliefs to develop with age. Another limitation is that situation selection was assessed here as a disengagement from a negative situation that had already occurred. Theory and research suggest that when situation selection is implemented before the emotional event has occurred, it can be an effective strategy for managing short-term emotions (Livingstone & Isaacowitz, 2015). Moreover, situation selection is theorized to be especially effective for older adults because it can effectively intervene before emotional processes are underway (Urry & Gross, 2010). Due to the way the questions were phrased, this possibility was not fully explored in the current study. Similarly, all the strategies included in the current research were intrapersonal, though there are also likely opportunities for interpersonal emotion regulation (e.g., seeking social support) for many of the scenarios. Finally, this study examined the effectiveness and use of strategies for hedonic emotion regulation; however, findings may be different for contra-hedonic emotion regulation for instrumental purposes (see Tamir, 2009). Conclusion The findings reported here expand our knowledge of emotion regulation across adulthood, specifically the extent of age differences in beliefs about the effectiveness of and likelihood of using different emotion regulation strategies. Although there was relative consensus on strategy effectiveness and popularity of use, there was also significant variability across age groups. Beliefs about emotion regulation likely play an important role in the process of selecting a strategy to use (see Gross, 2015). Continuing to investigate such beliefs—where they come from, whether they change over time, and how they influence emotion regulation processes—will help us understand how people of different ages implement different emotion regulation strategies, and thus how to maintain and enhance well-being across the life span. Supplementary Material Supplementary data is available at The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences online. Funding This research was supported in part by the National Institute on Aging Grant R01 AG948731 to D. M. Isaacowitz. Conflict of Interest None reported. References Aldao, A. ( 2013). The future of emotion regulation research capturing context. Perspectives on Psychological Science , 8, 155– 172. doi: 10.1177/1745691612459518 Google Scholar CrossRef Search ADS PubMed  Aldao, A., Nolen-Hoeksema, S., & Schweizer, S. ( 2010). Emotion-regulation strategies across psychopathology: A meta-analytic review. Clinical Psychology Review , 30, 217– 237. doi: 10.1016/ j.cpr.2009.11.004 Google Scholar CrossRef Search ADS PubMed  Bonanno, G. A., & Burton, C. L. ( 2013). Regulatory flexibility: An individual differences perspective on coping and emotion regulation. Perspectives on Psychological Science , 8, 591– 612. doi: 10.1177/1745691613504116 Google Scholar CrossRef Search ADS PubMed  Bushman, B. J., Baumeister, R. F., & Phillips, C. M. ( 2001). Do people aggress to improve their mood? Catharsis beliefs, affect regulation opportunity, and aggressive responding. Journal of Personality and Social Psychology , 81, 17– 32. doi: 10.1037//0022-3514.81.1.17 Google Scholar CrossRef Search ADS PubMed  Cabello, R., & Fernández-Berrocal, P. ( 2015). Implicit theories and ability emotional intelligence. Frontiers in Psychology , 6, 700– 707. doi: 10.3389/fpsyg.2015.00700 Google Scholar CrossRef Search ADS PubMed  Carstensen, L. L., Isaacowitz, D. M., & Charles, S. T. ( 1999). Taking time seriously. A theory of socioemotional selectivity. The American Psychologist , 54, 165– 181. doi:10.1037/0003-066X.54.3.165 Google Scholar CrossRef Search ADS PubMed  Charles, S. T. ( 2010). Strength and vulnerability integration: A model of emotional well-being across adulthood. Psychological Bulletin , 136, 1068– 1091. doi: 10.1037/a0021232 Google Scholar CrossRef Search ADS PubMed  Chesney, S. A., & Gordon, N. S. ( 2017). Profiles of emotion regulation: Understanding regulatory patterns and the implications for posttraumatic stress. Cognition and Emotion , 31, 598– 606. doi: 10.1080/02699931.2015.1126555 Google Scholar CrossRef Search ADS PubMed  Cohen, P., Cohen, J., Aiken, L. S., & West, S. G. ( 1999). The problem of units and the circumstance for POMP. Multivariate Behavioral Research , 34, 315– 346. doi: 10.1207/S15327906MBR3403_2 Google Scholar CrossRef Search ADS   Cole, P. M., Dennis, T. A., Smith-Simon, K. E., & Cohen, L. H. ( 2009). Preschoolers’ emotion regulation strategy understanding: Relations with emotion socialization and child self-regulation. Social Development , 18, 324– 352. doi: 10.1111/j.1467-9507.2008.00503.x Google Scholar CrossRef Search ADS   De Castella, K., Goldin, P., Jazaieri, H., Ziv, M., Dweck, C. S., & Gross, J. J. ( 2013). Beliefs about emotion: Links to emotion regulation, well-being, and psychological distress. Basic and Applied Social Psychology , 35, 497– 505. doi: 10.1080/01973533.840632 Google Scholar CrossRef Search ADS   Dennis, T. A., & Kelemen, D. A. ( 2009). Preschool children’s views on emotion regulation: Functional associations and implications for social-emotional adjustment. International Journal of Behavior Development , 33, 243– 252. doi: 10.1177/0165025408098024 Google Scholar CrossRef Search ADS   Dweck, C. S. ( 1999). Self-theories: Their role in motivation, personality, and development . Philadelphia, PA: Psychology Press. Enders, C. K. ( 2011). Analyzing longitudinal data with missing values. Rehabilitation Psychology , 56, 267– 288. doi: 10.1037/a0025579 Google Scholar CrossRef Search ADS PubMed  Gross, J. J. ( 1998). The emerging field of emotion regulation: An integrative review. Review of General Psychology , 2, 271– 299. doi: 10.1037/1089-2680.2.3.271 Google Scholar CrossRef Search ADS   Gross, J. J. ( 2015). Emotion regulation: Current status and future prospects. Psychological Inquiry , 26, 1– 26. doi: 10.1080/1047840X.2014.940781 Google Scholar CrossRef Search ADS   Hagenaars, J. A., & McCutcheon, A. L. (Eds.). ( 2009). Applied latent class analysis . Cambridge, UK: Cambridge University Press. Isaacowitz, D. M. ( 2012). Mood regulation in real time: Age differences in the role of looking. Current Directions in Psychological Science , 21, 237– 242. doi: 10.1177/0963721412448651 Google Scholar CrossRef Search ADS PubMed  John, O. P., & Gross, J. J. ( 2004). Healthy and unhealthy emotion regulation: Personality processes, individual differences, and life span development. Journal of Personality , 72, 1301– 1334. doi: 10.1111/j.1467-6494.2004.00298.x Google Scholar CrossRef Search ADS PubMed  Lazarus, R. S. ( 1991). Emotion and adaptation . Cary, NC: Oxford University Press. Lazarus, R. S., & Folkman, S( 1984). Stress appraisal and coping . New York: Springer. Livingstone, K. M., & Isaacowitz, D. M. ( 2015). Situation selection and modification for emotion regulation in younger and older adults. Social Psychological and Personality Science , 6, 904– 910. doi: 10.1177/1948550615593148 Google Scholar CrossRef Search ADS PubMed  Livingstone, K. M., & Isaacowitz, D. M. ( 2016). Age differences in use and effectiveness of positivity in emotion regulation: The sample case of attention. In A. D. Ong & C. E. Löekenhoff (Eds.), Emotion, aging, and health  (pp. 31– 48). Washington, DC: American Psychological Association. Google Scholar CrossRef Search ADS   Marsh, H. W., Lüdtke, O., Trautwein, U., & Morin, A. J. ( 2009). Classical latent profile analysis of academic self-concept dimensions: Synergy of person-and variable-centered approaches to theoretical models of self-concept. Structural Equation Modeling , 16, 191– 225. doi: 10.1080/10705510902751010 Google Scholar CrossRef Search ADS   Mather, M., & Carstensen, L. L. ( 2005). Aging and motivated cognition: The positivity effect in attention and memory. Trends in Cognitive Sciences , 9, 496– 502. doi: 10.1016/j.tics. 2005.08.005 Google Scholar CrossRef Search ADS PubMed  Muthén, B. ( 2001). Latent variable mixture modeling. In G. A. Marcoulides & R. E. Schumacker (Eds.), New developments and techniques in structural equation modeling  (pp. 1– 33). Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Muthén, B. ( 2004). Latent variable analysis. The Sage handbook of quantitative methodology for the social sciences  (pp. 345– 368). Thousand Oaks, CA: Sage Publications. Muthén, L. K., & Muthén, B.O. ( 1998–2015). Mplus user’s guide . 7th ed.. Los Angeles, CA: Muthén & Muthén. Muthén, B., & Muthén, L. K. ( 2000). Integrating person-centered and variable-centered analyses: Growth mixture modeling with latent trajectory classes. Alcoholism: Clinical and Experimental Research , 24, 882– 891. doi: 10.1111/j.1530-0277.2000.tb02070.x Google Scholar CrossRef Search ADS   Noh, S. R., Lohani, M., & Isaacowitz, D. M. ( 2011). Deliberate real-time mood regulation in adulthood: The importance of age, fixation, and attentional functioning. Cognition and Emotion , 25, 998– 1013. doi: 10.1080/02699931.2010.541668 Google Scholar CrossRef Search ADS PubMed  Nolen-Hoeksema, S., & Aldao, A. ( 2011). Gender and age differences in emotion regulation strategies and their relationship to depressive symptoms. Personality and Individual Differences , 51, 704– 708. doi: 10.1016/j.paid.2011.06.012 Google Scholar CrossRef Search ADS   Nowlan, J. S., Wythrich, V. M., & Rapee, R. M. ( 2015). Positive reappraisal in older adults: A systematic literature review. Aging & Mental Health , 19, 475– 484. doi: 10.1080/13607863.2014.954528 Google Scholar CrossRef Search ADS PubMed  Nylund, K. L., Asparouhov, T., & Muthén, B. O. ( 2007). Deciding on the number of classes in latent class analysis and growth mixture modeling: A Monte Carlo simulation study. Structural Equation Modeling , 14, 535– 569. doi: 10.1080/10705510701575396 Google Scholar CrossRef Search ADS   Ortner, C. N. M., Briner, E. L., & Marjanovic, Z. ( 2017). Believing is doing: Emotion regulation beliefs are associated with emotion regulation behavioral choices and subjective well-being. Europe’s Journal of Psychology , 13, 60– 74. doi: 10.5964/ejop.v13i1.1248 Google Scholar CrossRef Search ADS PubMed  Papageorgiou, C., & Wells, A. ( 2001). Metacognitive beliefs about rumination in recurrent major depression. Cognitive and Behavioral Practice , 8, 160– 164. doi: 10.1016/S1077-7229(01)80021–3 Google Scholar CrossRef Search ADS   Pastor, D. A., Barron, K. E., Miller, B. J., & Davis, S. L. ( 2007). A latent profile analysis of college students’ achievement goal orientation. Contemporary Educational Psychology , 32, 8– 47. doi: 10.1016/j.cedpsych.2006.10.003 Google Scholar CrossRef Search ADS   Reed, A. E., Chan, L., & Mikels, J. A. ( 2014). Meta-analysis of the age-related positivity effect: Age differences in preferences for positive over negative information. Psychology and Aging , 29, 1– 15. doi: 10.1037/a0035194 Google Scholar CrossRef Search ADS PubMed  Sands, M., Livingstone, K. M., & Isaacowitz, D. M.(in press). Characterizing age-related positivity effects in situation selection. International Journal of Behavioral Development . doi: 10.1177/0165025417723086 Shiota, M. N., & Levenson, R. W. ( 2009). Effects of aging on experimentally instructed detached reappraisal, positive reappraisal, and emotional behavior suppression. Psychology and Aging , 24, 890– 900. doi: 10.1037/a0017896 Google Scholar CrossRef Search ADS PubMed  Tamir, M. ( 2009). What do people want to feel and why? Pleasure and utility in emotion regulation. Current Directions in Psychological Science , 18, 101– 105. doi: 10.1111/j.1467-8721.2009.01617.x Google Scholar CrossRef Search ADS   Tamir, M., John, O. P., Srivastava, S., & Gross, J. J. ( 2007). Implicit theories of emotion: Affective and social outcomes across a major life transition. Journal of Personality and Social Psychology , 92, 731– 744. doi: 10.1037/0022-3514.92.4.731 Google Scholar CrossRef Search ADS PubMed  Trincas, R., Bilotta, E., & Mancini, F. ( 2016). Specific beliefs about emotions are associated with different emotion- regulation strategies. Psychology , 7, 1682– 1699. doi: 10.4236/psych. 2016.713159 Google Scholar CrossRef Search ADS   Troy, A. S., Shallcross, A. J., & Mauss, I. B. ( 2013). A person-by-situation approach to emotion regulation: Cognitive reappraisal can either help or hurt, depending on the context. Psychological Science , 24, 2505– 2514. doi: 10.1177/0956797613496434 Google Scholar CrossRef Search ADS PubMed  Urry, H. L., & Gross, J. J. ( 2010). Emotion regulation in older age. Current Directions in Psychological Science , 19, 352– 357. doi: 10.1037/14857-004 Google Scholar CrossRef Search ADS   Webb, T. L., Miles, E., & Sheeran, P. ( 2012). Dealing with feeling: A meta-analysis of the effectiveness of strategies derived from the process model of emotion regulation. Psychological Bulletin , 138, 775– 808. doi: 10.1037/a0027600 Google Scholar CrossRef Search ADS PubMed  Zakowski, S. G., Hall, M. H., Klein, L. C., & Baum, A. ( 2001). Appraised control, coping, and stress in a community sample: A test of the goodness-of-fit hypothesis. Annals of Behavioral Medicine , 23, 158– 165. doi: 10.1207/S15324796ABM2303_3 Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The Journals of Gerontology Series B: Psychological Sciences and Social Sciences Oxford University Press

Loading next page...
 
/lp/ou_press/age-differences-in-beliefs-about-emotion-regulation-strategies-XYBMuKK6Mo
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
ISSN
1079-5014
eISSN
1758-5368
D.O.I.
10.1093/geronb/gby022
Publisher site
See Article on Publisher Site

Abstract

Abstract Objectives: Age shifts in emotion regulation may be rooted in beliefs about different strategies. We test whether there are age differences in the beliefs people hold about specific emotion regulation strategies derived from the process model of emotion regulation and whether profiles of emotion beliefs vary by age. Method An adult life-span sample (N = 557) sorted 13 emotion regulation strategies either by (a) how effective the strategies would be or (b) how likely they would be to use them, in 15 negative emotion-eliciting situations. Results Younger adults ranked attentional and cognitive distraction more effective than older adults, and preferred avoidance, distraction, and rumination more (and attentional deployment less) than middle-aged and older adults. Latent profile analysis on preferences identified three distinct strategy profiles: Classically adaptive regulators preferred a variety of strategies; situation modifiers showed strong preferences for changing situations; a small percentage of people preferred avoidance and rumination. Middle-aged and older adults were more likely than younger adults to be classically adaptive regulators (as opposed to situation modifiers or avoiders/ruminators). Discussion These findings provide insight into the reasons people of different ages may select and implement different emotion regulation strategies, which may influence their emotional well-being. Beliefs, Emotion, Emotion regulation, Implicit theories Theories of life-span emotional development propose that because aging is associated with both gains and losses in resources and functioning, older adults must shift their emotion regulation efforts to maintain high levels of well-being (Charles, 2010; Urry & Gross, 2010). Recent decades have seen a dramatic increase in research on emotion regulation in general. Studies have shown that (a) there are stable individual differences, including age differences, in the use of emotion regulation strategies (e.g., John & Gross, 2004), (b) strategies vary in their effectiveness (e.g., Webb, Miles, & Sheeran, 2012), and (c) effectiveness varies by age (e.g., Shiota & Levenson, 2009). Though research has investigated why such individual differences exist, few studies have examined the beliefs people hold about the effectiveness of the strategies themselves. We propose that due to age-related changes in motivation and resources, there may be systematic age differences in beliefs about the effectiveness and use of specific emotion regulation strategies, and these beliefs guide moment-to-moment choices about which regulation strategy is best for them. Socioemotional selectivity theory (SST) suggests that older adults are especially motivated to pursue emotional well-being goals due to a more limited future time perspective (Carstensen, Isaacowitz, & Charles, 1999). Older adults may therefore be more likely than younger adults to prefer emotion regulation strategies that are most effective in minimizing negative and maximizing positive emotions. The strength and vulnerability integration (SAVI) model proposes that greater life experience provides knowledge about which emotion regulation strategies are likely to support these goals, even as declining cognitive and physical resources may diminish the effectiveness of some strategies (Charles, 2010; see also Urry & Gross, 2010). Given their experience, older adults may better recognize which strategies are likely to be effective compared to younger adults. Meanwhile, older adults may also shift their use of such strategies away from those that rely on cognitive or physical resources that diminish with age and toward those that avoid the emotional experience altogether (Urry & Gross, 2010). Emotion Regulation in the Process Model The process model of emotion regulation proposes five families of emotion regulation strategies people can use to influence emotions, based on which part of the emotion-generative process they target (Gross, 1998). People can select or modify a situation based on their expectations for emotional change (situation selection and modification); direct their attention toward or away from aspects of the situation (attentional deployment); change the way they are thinking (cognitive change), such as by changing their appraisal of the situation or their own feelings (reappraisal); or target the emotional response—expression, behavior, or physiology (response modulation). Within each strategy family are specific tactics (e.g., using detached vs positive reappraisal; Shiota & Levenson, 2009). An expanded version of the process model proposes that in each case of emotion regulation, a person must decide whether to regulate, which strategy to use (i.e., which aspect of the emotion-generative cycle to target), and which specific tactic within that strategy to implement (Gross, 2015). In each case, a person evaluates the costs and benefits of regulating and of using each tactic. Survey and laboratory studies have identified age differences in both the use and effectiveness of emotion regulation strategies. Although studies generally do not find age differences in use or effectiveness of situation selection (Sands, Livingstone, & Isaacowitz, in press), older adults may use situation modification more (Livingstone & Isaacowitz, 2015). Eye tracking studies of spontaneous attentional deployment show older adults shift attention away from negative and toward positive stimuli, though this is not always associated with decreases in negative affect (Isaacowitz, 2012). Some self-report studies suggest older adults use more reappraisal and less suppression (John & Gross, 2004), whereas others find the opposite (Nolen-Hoeksema & Aldao, 2011). Studies that instruct reappraisal show mixed effects on emotion, with older adults generally successful at using positive reappraisal (Nowlan, Wythrich, & Rapee, 2015) but not detached reappraisal (e.g., Shiota & Levenson, 2009). Several studies have found no age differences in use or effectiveness of expressive suppression (e.g., Shiota & Levenson, 2009). In general, these studies assess either one or a small number of regulation strategies or tactics; to date, there has been no systematic study of age differences in strategy use and effectiveness across the process model. As a starting point, we propose to investigate how beliefs about these strategies and tactics vary by age. Emotion Regulation Beliefs Research on beliefs in other domains shows they have strong effects on motivation, affect, and behavior (Dweck, 1999). In the domain of emotion, negative beliefs about emotion (e.g., that emotions are irrational or damaging) have been linked to emotion regulation difficulties, less use of adaptive regulation strategies (e.g., cognitive reappraisal, problem-solving), and more use of maladaptive strategies (e.g., rumination, avoidance; Trincas, Bilotta, & Mancini, 2016). Beliefs about the effectiveness of regulation strategies can be identified early in development (Dennis & Kelemen, 2009), and greater recognition of effective strategies in children is linked to greater use of constructive strategies in response to frustration (Cole, Dennis, Smith-Simon, & Cohen, 2009). Adults are likely to engage in specific strategies when they believe they will be effective. For example, people engage in aggressive behaviors primarily when they believe it will make them feel better (Bushman, Baumeister, & Phillips, 2001). Likewise, positive beliefs about rumination (e.g., “I need to ruminate about the bad things that have happened in the past to make sense of them”) are linked with increased use (Papageorgiou & Wells, 2001). However, there are also individual differences in which strategies people find effective (Ortner, Briner, & Marjanovic, 2017). Age Differences in Emotion Regulation Beliefs The first goal of this research was to characterize age differences in beliefs regarding the effectiveness and use of a variety of emotion regulation strategies. To investigate this question, we collected data from younger, middle-aged, and older adults. Increasing attention to the role of context has highlighted that the effectiveness of strategies varies by situation (e.g., Troy, Shallcross, & Mauss, 2013). We therefore developed a scenario-based measure of emotion regulation beliefs and generated situations eliciting a range of negative emotions. Additionally, whereas most studies on emotion regulation investigate a small number of strategies, we included 13 emotion regulation tactics across the five strategy families proposed by the process model (Gross, 1998). Although emotion regulation strategies vary in their effectiveness depending on context (e.g., Troy et al., 2013), some strategies have been repeatedly linked with better or poorer outcomes. In particular, meta-analysis has linked problem-solving (akin to situation modification), reappraisal, and acceptance with more positive mental health outcomes, and rumination, suppression, and avoidance with negative ones (Aldao, Nolen-Hoeksema, & Schweizer, 2010). One possibility is that older adults, having more experience with emotions (Charles, 2010) and more motivation to maximize well-being (Carstensen et al., 1999), have stronger beliefs about the effectiveness of emotion regulation than younger adults, and are better able to identify putatively effective strategies. On the other hand, age-related declines may make certain strategies more demanding and less efficient (Urry & Gross, 2010), especially those that occur late in the emotion-generative process, once physiological arousal has occurred (Charles, 2010). SAVI proposes that reduced physiological flexibility makes older adults less effective at down-regulating from highly arousing emotional experiences once they are underway. Other strategies, such as reappraisal, may require the flexible use of cognitive resources, which decline with age (Urry & Gross, 2010). Consequently, some strategies, such as cognitive change and response modulation, may be less effective and less often used in old age. If this is the case, older adults’ beliefs are more likely to favor strategies that intervene earlier in the emotion-generative process: situation selection, situation modification, and attentional deployment. When examining age differences in emotion regulation, individual differences in preferences for strategies must be disentangled from differences in the effectiveness of those strategies (Isaacowitz, 2012). These two aspects of emotion regulation have different conceptual and real-life implications. For example, though a person may prefer to use certain strategies, they may not be especially effective for them. Conversely, a person may be able to effectively utilize a strategy in the lab, but be unlikely to employ the strategy in real life. Our focus in the current research was to characterize these beliefs, which we examined in two separate samples to avoid beliefs in one domain influencing beliefs in the other. Though this means that we cannot test the relationship between the two sets of beliefs, we can compare patterns of age differences in beliefs about both effectiveness and use. The second goal of this research was to identify profiles of emotion regulation beliefs and examine whether they vary with age. Profiles may emerge based on the families of strategies derived from the process model. For example, some people may believe that focusing emotion regulation efforts on the situation is most effective, others focusing on cognition, and others focusing on the response. On the other hand, age-related belief patterns could reflect a generalized positivity preference in line with positivity effects identified by SST (Mather & Carstensen, 2005). Age-related positivity effects refer to findings that older adults favor processing of positive over negative information in both attention and memory (Reed, Chan, & Mikels, 2014). This preference may also manifest in choice of emotion regulation strategies (see Livingstone & Isaacowitz, 2016): Older adults may prefer specific tactics like attending toward positive aspects of the situation and away from negative (Isaacowitz, 2012) and positive reappraisal (Shiota & Levenson, 2009). An exploratory third goal was to examine individual differences in beliefs about emotion more broadly as a possible correlate of beliefs about specific emotion regulation strategies. We focus on the degree to which a person believes emotions are malleable (and can be controlled) or fixed (and must play out on their own; Tamir, John, Srivastava, & Gross, 2007). Stronger malleability beliefs may be associated with beliefs that favor more putatively effective strategies, especially antecedent-focused (e.g., situation modification, attentional deployment) and emotion-engaging strategies that target cognition (e.g., cognitive change). In contrast, weaker malleability beliefs may be associated with beliefs that favor more response-focused strategies (e.g., expressive suppression) and emotion-disengaging strategies (e.g., avoidance). One study has found that older adults report weaker malleability beliefs than younger adults (Cabello & Fernández-Berrocal, 2015), but we know very little about how age and emotion malleability beliefs together predict beliefs about emotion regulation. We therefore assessed trait-level emotion malleability beliefs along with beliefs about the effectiveness of and preferences for emotion regulation strategies and examined how these beliefs vary with age. Method Participants Power analyses using G*Power (version 3.1.9) for a MANOVA with 13 strategies and three age groups showed that a total sample size of 171 would have 80% power to detect a small effect (ηp2 = .02) at an α level of .05. Therefore, for each version of the study, we collected at least 60 participants in each age group for each version before stopping data collection. The sample consisted of 557 adults (57% female, 73% White, 3% Asian, 6% Black, 7% Hispanic) ages 18–87 (Mage = 42.57, SD = 17.85). Most participants (N = 402) were recruited from www.mturk.com, an online platform for recruiting workers to perform tasks, and were paid $1.05. Additional participants were recruited from a subject pool at a large private university (N = 106) and were given partial course credit, and from an existing database of adults over 60 (N = 49) who had previously been recruited from the Boston community through print and online advertisements; they were entered into a raffle to win a $100 gift card. On MTurk, we used the quota feature in Qualtrics to limit enrolment by age group to ensure adequate numbers of middle-aged (ages 40–59) and older adults (ages 60+), who may be underrepresented on the platform. Participants were randomly assigned to complete either the effectiveness (Ns = 123 YA, 79 MA, 77 OA) or the likelihood version (Ns = 129 YA, 84 MA, 65 OA). Materials Emotion scenarios Rather than ask participants to report on global beliefs about the effectiveness and likelihood of use of emotion regulation strategies, we specified a range of emotion-eliciting situations and asked participants to classify the strategies for each scenario. This procedure was designed to encourage participants to visualize emotion regulation in specific situations, and allowed participants to acknowledge variability in effectiveness and use across contexts (Aldao, 2013; Bonanno & Burton, 2013). Fifteen emotion-eliciting scenarios (see Supplementary Materials) were generated to address a range of negative emotions (e.g., sadness, anger, disgust, guilt) based on core relational themes outlined by Lazarus (1991). The scenarios were phrased broadly to be applicable to people of differing ages and life situations (e.g., “Someone has said something hurtful to you”). An independent sample (N = 66, ages 20–73) rated scenarios for negativity, familiarity, recency, incidence, and likelihood of future occurrence. Descriptive statistics for scenarios are available in Supplementary Materials. Emotion regulation items Thirteen emotion regulation strategies (see Supplementary Materials) were designed to represent specific tactics within the five stages of emotion regulation as outlined by the process model (Gross, 1998, 2015). Beliefs about emotion malleability Participants completed the four-item implicit theories of emotion questionnaire (Tamir et al., 2007). Internal consistency was good (α = .797). Scores were rescaled to a range from 0 to 100 (Cohen, Cohen, Aiken, & West, 2010), with higher scores indicating stronger belief that emotions are malleable (e.g., agrees more strongly with items such as “If they want to, people can change the emotions that they have.”) This scale was used to examine how beliefs about emotion relate to profiles of emotion regulation. Procedure Surveys were hosted on Qualtrics.com (Qualtrics, Provo, UT). MTurk participants first completed a demographic screening questionnaire, for which they were paid $0.05. Participants who qualified based on age were given the option to complete an additional 20-min survey for $1.00. After providing informed consent, participants read instructions and completed a practice sorting task followed by the main sorting task. Participants then completed a second demographic questionnaire and the implicit theories of emotion questionnaire. Final demographic statistics were based on the second demographic questionnaire, which was also used to verify age; those with mismatched age reports were excluded. Participants recruited from campus and the local community simply saw the consent form, tasks, and demographic survey. Effectiveness sorting task Instructions read, “We are interested in the many ways that people can manage their emotions or feelings. For the following situations, imagine yourself in that situation and think about what would be the most effective way for you to deal with your emotion. Assume that you want to feel less negative emotion.” Participants were presented with one scenario at a time, written in large font at the top of the web page, in a randomized order. Strategies were presented by description (not name) in a fixed order (see Supplemental Materials) on the left side of the screen. On the right side of the screen were three boxes, labeled Most effective (“These strategies would be most effective in helping you feel less negative emotion”), Somewhat effective (“These strategies may be effective in helping you feel less negative emotion”), and Least effective (“These strategies would not be effective in helping you feel less negative emotion, or may even make things worse”). Three categories were chosen rather than rating scales to reduce participant fatigue across the 15 scenarios and to allow participants to visually compare strategies using different boxes. Participants sorted the strategies by dragging each strategy into the box that best matched their rated effectiveness in that scenario. To ensure adequate variability, participants were asked to include at least three strategies in each box and to sort all strategies into a box. Likelihood of use sorting task Instructions were the same as for the effectiveness version except for the third sentence: “We are not necessarily interested in what you would do to deal with the situation, but rather what you would do to deal with emotions that arise in that situation.” Materials and procedures were the same as for the effectiveness task, except that participants were asked to sort the strategies into those that they would be Most likely to use (“These are strategies that you would most likely use if faced with the situation”), Somewhat likely to use (“These strategies are the ones you may use if you were faced with the situation”), or Least likely to use (“These strategies are the ones you would be least likely to use, or would definitely not use, if faced with the situation”). Results Effectiveness Beliefs Responses were scored 1 (Most effective), 0 (Somewhat effective) or −1 (Least effective). Emotion regulation strategies generally had acceptable internal consistency across scenarios (average Cronbach’s α = .819), so were averaged across scenarios to create a continuous score of for each strategy for each participant ranging from −1 to 1. Multivariate analysis of variance (MANOVA) tested age differences in effectiveness beliefs across 13 strategies. Because of known gender differences in emotion regulation (e.g., Nolen-Hoeksema & Aldao, 2011), gender was included as a covariate. The multivariate test was marginally significant for age, Wilks Λ = .848, F(26,458) = 1.51, p = .054, ηp2 = .079. There was a significant age difference in cognitive change—distraction and a marginal age difference for attentional deployment—away from negative (see Table 1). Younger adults rated attentional deployment and cognitive distraction as more effective than older adults; middle-aged adults did not differ from either group, ps > .05. There was also an age difference for cognitive change—rumination, F(2,240) = 3.34, p = .037, ηp2 = .027; middle-aged adults rated rumination as less effective than older adults; younger adults did not differ from either group. Table 1. Descriptive Statistics and Inferential Tests of Age Differences in Emotion Regulation Beliefs   Mean Belief (SD)        Younger  Middle-Aged  Older  F  p  ηp2  Effectiveness        F(2,240)       SS Avoid  −.31 (.43)  −.38 (.36)  −.44 (.45)  1.42  .243  .012   SS Leave  −.05 (.33)  −.08 (.32)  −.15 (.45)  1.45  .237  .012   SM Modify  .56 (.33)  .61 (.29)  .59 (.35)  0.60  .551  .005   AD Negative  .41 (.30)  .37 (.31)  .30 (.28)  3.01  .051  .024   AD Positive  .35 (.35)  .28 (.39)  .42 (.41)  2.12  .123  .017   CC Detached  .03 (.42)  .09 (.44)  .13 (.44)  1.18  .310  .010   CC Distract  .09 (.34)  −.02 (.43)  −.11(.40)  5.24**  .006  .042   CC Positive  .25 (.37)  .16 (.40)  .27 (.43)  1.44  .238  .012   CC Ruminate  −.78 (.28)  −.87 (.24)  −.74 (.37)  3.33*  .037  .027   CC Accept  .02 (.49)  .03 (.49)  .14 (.49)  1.05  .350  .009   RM Suppress  −.18 (.34)  −.23 (.42)  −.21 (.39)  0.45  .640  .004   RM Mask  −.11 (.47)  −.09 (.46)  −.12 (.50)  0.07  .928  .001   RM Physiology  .12 (.43)  .12 (.49)  .03 (.48)  1.17  .313  .010  Preferences        F(2,238)       SS Avoid  −.12 (.38)  −.27 (.42)  −.37 (.43)  7.77**  .001  .061   SS Leave  .04 (.33)  −.06 (.34)  −.08 (.38)  3.11*  .046  .025   SM Modify  .51 (.29)  .55 (.27)  .54 (.30)  0.67  .515  .006   AD Negative  .16 (.33)  .32 (.38)  .31 (.35)  6.05**  .003  .048   AD Positive  .04 (.41)  .24 (.45)  .23 (.48)  5.87  .003  .047   CC Detached  .01 (.37)  .07 (.40)  −.01 (.44)  0.78  .462  .006   CC Distract  .02 (.37)  −.13 (.42)  −.14 (.43)  4.76**  .009  .038   CC Positive  −.13 (.43)  .06 (.46)  .05 (.53)  4.84**  .009  .039   CC Ruminate  −.03 (.49)  −.30 (.48)  −.34 (.51)  10.29***  <.001  .080   CC Accept  −.01 (.41)  .09 (.49)  .05 (.48)  1.23  .295  .010   RM Suppress  .02 (.39)  −.04 (.47)  −.01 (.35)  0.65  .523  .005   RM Mask  −.12 (.43)  −.16 (.45)  −.09 (.46)  0.41  .664  .003   RM Physiology  −.08 (.45)  −.02 (.55)  .07 (.38)  1.87  .157  .015    Mean Belief (SD)        Younger  Middle-Aged  Older  F  p  ηp2  Effectiveness        F(2,240)       SS Avoid  −.31 (.43)  −.38 (.36)  −.44 (.45)  1.42  .243  .012   SS Leave  −.05 (.33)  −.08 (.32)  −.15 (.45)  1.45  .237  .012   SM Modify  .56 (.33)  .61 (.29)  .59 (.35)  0.60  .551  .005   AD Negative  .41 (.30)  .37 (.31)  .30 (.28)  3.01  .051  .024   AD Positive  .35 (.35)  .28 (.39)  .42 (.41)  2.12  .123  .017   CC Detached  .03 (.42)  .09 (.44)  .13 (.44)  1.18  .310  .010   CC Distract  .09 (.34)  −.02 (.43)  −.11(.40)  5.24**  .006  .042   CC Positive  .25 (.37)  .16 (.40)  .27 (.43)  1.44  .238  .012   CC Ruminate  −.78 (.28)  −.87 (.24)  −.74 (.37)  3.33*  .037  .027   CC Accept  .02 (.49)  .03 (.49)  .14 (.49)  1.05  .350  .009   RM Suppress  −.18 (.34)  −.23 (.42)  −.21 (.39)  0.45  .640  .004   RM Mask  −.11 (.47)  −.09 (.46)  −.12 (.50)  0.07  .928  .001   RM Physiology  .12 (.43)  .12 (.49)  .03 (.48)  1.17  .313  .010  Preferences        F(2,238)       SS Avoid  −.12 (.38)  −.27 (.42)  −.37 (.43)  7.77**  .001  .061   SS Leave  .04 (.33)  −.06 (.34)  −.08 (.38)  3.11*  .046  .025   SM Modify  .51 (.29)  .55 (.27)  .54 (.30)  0.67  .515  .006   AD Negative  .16 (.33)  .32 (.38)  .31 (.35)  6.05**  .003  .048   AD Positive  .04 (.41)  .24 (.45)  .23 (.48)  5.87  .003  .047   CC Detached  .01 (.37)  .07 (.40)  −.01 (.44)  0.78  .462  .006   CC Distract  .02 (.37)  −.13 (.42)  −.14 (.43)  4.76**  .009  .038   CC Positive  −.13 (.43)  .06 (.46)  .05 (.53)  4.84**  .009  .039   CC Ruminate  −.03 (.49)  −.30 (.48)  −.34 (.51)  10.29***  <.001  .080   CC Accept  −.01 (.41)  .09 (.49)  .05 (.48)  1.23  .295  .010   RM Suppress  .02 (.39)  −.04 (.47)  −.01 (.35)  0.65  .523  .005   RM Mask  −.12 (.43)  −.16 (.45)  −.09 (.46)  0.41  .664  .003   RM Physiology  −.08 (.45)  −.02 (.55)  .07 (.38)  1.87  .157  .015  Note: Possible scores ranged from −1 to 1. SS = situation selection. AD = Attentional deployment; CC = Cognitive change; RM = Response modulation; SM = Situation modification. *p < .05, **p < .01, ***p < .001. View Large Table 1. Descriptive Statistics and Inferential Tests of Age Differences in Emotion Regulation Beliefs   Mean Belief (SD)        Younger  Middle-Aged  Older  F  p  ηp2  Effectiveness        F(2,240)       SS Avoid  −.31 (.43)  −.38 (.36)  −.44 (.45)  1.42  .243  .012   SS Leave  −.05 (.33)  −.08 (.32)  −.15 (.45)  1.45  .237  .012   SM Modify  .56 (.33)  .61 (.29)  .59 (.35)  0.60  .551  .005   AD Negative  .41 (.30)  .37 (.31)  .30 (.28)  3.01  .051  .024   AD Positive  .35 (.35)  .28 (.39)  .42 (.41)  2.12  .123  .017   CC Detached  .03 (.42)  .09 (.44)  .13 (.44)  1.18  .310  .010   CC Distract  .09 (.34)  −.02 (.43)  −.11(.40)  5.24**  .006  .042   CC Positive  .25 (.37)  .16 (.40)  .27 (.43)  1.44  .238  .012   CC Ruminate  −.78 (.28)  −.87 (.24)  −.74 (.37)  3.33*  .037  .027   CC Accept  .02 (.49)  .03 (.49)  .14 (.49)  1.05  .350  .009   RM Suppress  −.18 (.34)  −.23 (.42)  −.21 (.39)  0.45  .640  .004   RM Mask  −.11 (.47)  −.09 (.46)  −.12 (.50)  0.07  .928  .001   RM Physiology  .12 (.43)  .12 (.49)  .03 (.48)  1.17  .313  .010  Preferences        F(2,238)       SS Avoid  −.12 (.38)  −.27 (.42)  −.37 (.43)  7.77**  .001  .061   SS Leave  .04 (.33)  −.06 (.34)  −.08 (.38)  3.11*  .046  .025   SM Modify  .51 (.29)  .55 (.27)  .54 (.30)  0.67  .515  .006   AD Negative  .16 (.33)  .32 (.38)  .31 (.35)  6.05**  .003  .048   AD Positive  .04 (.41)  .24 (.45)  .23 (.48)  5.87  .003  .047   CC Detached  .01 (.37)  .07 (.40)  −.01 (.44)  0.78  .462  .006   CC Distract  .02 (.37)  −.13 (.42)  −.14 (.43)  4.76**  .009  .038   CC Positive  −.13 (.43)  .06 (.46)  .05 (.53)  4.84**  .009  .039   CC Ruminate  −.03 (.49)  −.30 (.48)  −.34 (.51)  10.29***  <.001  .080   CC Accept  −.01 (.41)  .09 (.49)  .05 (.48)  1.23  .295  .010   RM Suppress  .02 (.39)  −.04 (.47)  −.01 (.35)  0.65  .523  .005   RM Mask  −.12 (.43)  −.16 (.45)  −.09 (.46)  0.41  .664  .003   RM Physiology  −.08 (.45)  −.02 (.55)  .07 (.38)  1.87  .157  .015    Mean Belief (SD)        Younger  Middle-Aged  Older  F  p  ηp2  Effectiveness        F(2,240)       SS Avoid  −.31 (.43)  −.38 (.36)  −.44 (.45)  1.42  .243  .012   SS Leave  −.05 (.33)  −.08 (.32)  −.15 (.45)  1.45  .237  .012   SM Modify  .56 (.33)  .61 (.29)  .59 (.35)  0.60  .551  .005   AD Negative  .41 (.30)  .37 (.31)  .30 (.28)  3.01  .051  .024   AD Positive  .35 (.35)  .28 (.39)  .42 (.41)  2.12  .123  .017   CC Detached  .03 (.42)  .09 (.44)  .13 (.44)  1.18  .310  .010   CC Distract  .09 (.34)  −.02 (.43)  −.11(.40)  5.24**  .006  .042   CC Positive  .25 (.37)  .16 (.40)  .27 (.43)  1.44  .238  .012   CC Ruminate  −.78 (.28)  −.87 (.24)  −.74 (.37)  3.33*  .037  .027   CC Accept  .02 (.49)  .03 (.49)  .14 (.49)  1.05  .350  .009   RM Suppress  −.18 (.34)  −.23 (.42)  −.21 (.39)  0.45  .640  .004   RM Mask  −.11 (.47)  −.09 (.46)  −.12 (.50)  0.07  .928  .001   RM Physiology  .12 (.43)  .12 (.49)  .03 (.48)  1.17  .313  .010  Preferences        F(2,238)       SS Avoid  −.12 (.38)  −.27 (.42)  −.37 (.43)  7.77**  .001  .061   SS Leave  .04 (.33)  −.06 (.34)  −.08 (.38)  3.11*  .046  .025   SM Modify  .51 (.29)  .55 (.27)  .54 (.30)  0.67  .515  .006   AD Negative  .16 (.33)  .32 (.38)  .31 (.35)  6.05**  .003  .048   AD Positive  .04 (.41)  .24 (.45)  .23 (.48)  5.87  .003  .047   CC Detached  .01 (.37)  .07 (.40)  −.01 (.44)  0.78  .462  .006   CC Distract  .02 (.37)  −.13 (.42)  −.14 (.43)  4.76**  .009  .038   CC Positive  −.13 (.43)  .06 (.46)  .05 (.53)  4.84**  .009  .039   CC Ruminate  −.03 (.49)  −.30 (.48)  −.34 (.51)  10.29***  <.001  .080   CC Accept  −.01 (.41)  .09 (.49)  .05 (.48)  1.23  .295  .010   RM Suppress  .02 (.39)  −.04 (.47)  −.01 (.35)  0.65  .523  .005   RM Mask  −.12 (.43)  −.16 (.45)  −.09 (.46)  0.41  .664  .003   RM Physiology  −.08 (.45)  −.02 (.55)  .07 (.38)  1.87  .157  .015  Note: Possible scores ranged from −1 to 1. SS = situation selection. AD = Attentional deployment; CC = Cognitive change; RM = Response modulation; SM = Situation modification. *p < .05, **p < .01, ***p < .001. View Large Likelihood of Use Beliefs Responses of Most likely were assigned a value of 1, Somewhat likely a value of 0, and Least likely a value of −1. Emotion regulation strategies generally had acceptable internal consistency across scenarios (average α = .805), so were averaged across scenarios to create a continuous score of for each strategy for each participant ranging from −1 to 1. MANOVA tested age differences in beliefs about the likelihood of use across 13 strategies, also controlling for gender. The multivariate test was significant for age, Wilk’s Λ = .804, F(26, 454) = 2.01, p = .003, ηp2 = .104. Younger adults were more likely to prefer avoidance, leaving, distraction, and rumination compared to middle-aged and older adults (see Table 1). They were also less likely to prefer attentional deployment—away from negative and toward positive aspects—and to prefer positive reappraisal. Latent Profile Analysis on Emotion Regulation Use We were also interested in identifying the extent to which younger, middle-aged, and older adults could be differentially classified by their emotion regulation beliefs. Because we found more significant age differences in the likelihood of using different strategies, we focused on preferences for use rather than on effectiveness beliefs. We used latent profile analysis (LPA), a person-centered method for classifying types of people into latent variables using continuous indicators. It is statistically similar to latent class analysis, using continuous rather than categorical indicators (Muthén, 2001), and conceptually similar to cluster analysis, using a probabilistic rather than a distance based model (Muthén & Muthén, 2000; Pastor, Barron, Miller, & Davis, 2007). LPA additionally allows for the estimation of a latent model structure, which enables the calculation of goodness-of-fit indices and comparisons among nested model structures, and the inclusion of covariates in predicting latent profile membership (Muthén, 2004). We applied a mixed-process approach to test a series of conceptually-driven, nested solutions that were examined based on statistical criteria and ease of interpretation (Marsh, Lüdtke, Trautwein, & Morin, 2009). Specifically, we tested whether the beliefs about emotion regulation strategy use of younger, middle-aged, and older adults were best classified into 2-, 3-, 4-, or 5-profile solutions. This decision was based on prominent emotion regulation theory (the five stages of the process model, Gross, 1998) and recent empirical tests of emotion regulation profiles in clinical samples (e.g., Chesney & Gordon, 2017). Data were analyzed using Mplus version 7.4 (Muthèn & Muthèn, 1998–2015). The 13 likelihood of use strategy scores were modeled as continuous observed indicators. We included age group and gender as covariates in the estimation of profile membership. At least 500 sets of random parameter start values were generated for each model to protect against a local maxima solution (Muthèn, 2001). Missing data (6%) were likely missing at random (Enders, 2011) and thus addressed using full information maximum likelihood estimation. Model selection was based on the examination of several criteria including the Bayesian Information Criterion (BIC), entropy value, Lo-Mendell-Rubin Test (LMRT), Parametric Bootstrapped Likelihood Ratio Test (BLRT), and ease of interpretation of the solution. Decreasing BIC values, entropy ≥ .90, and significant LMRT and BLRT indicate better-fitting solutions (Nylund, Asparouhov, & Muthén, 2007). After the profile groups were identified, multinomial logistic regressions were conducted to identify age differences in latent profile class membership, which allowed us to simultaneously examine the possibility that middle-aged and older adults were more likely to be in a profile group than were younger adults. Profile solution Fit statistics for the two-, three-, four-, and five-profile solutions are presented in Table 2. The BIC values decreased as the number of profiles increased. Entropy values were highest in the three- and four-profile solutions. Results from the LMRT indicated that the three-profile solution was significantly more favorable than the two-profile solution (p < .05); however, the four-profile solution was not significantly better than the three-profile solution (p > .05). Given that solutions with increased profiles are generally better fit to the data (Hagenaars & McCutcheon, 2009), BLRT results were less informative. In examining the specific profile solutions for the three- and four-profile structures, a clear theoretically-relevant pattern was observed for the three-profile solution. These profile patterns were replicated in the four-profile solution, with the addition of a fourth fragmented profile consisting of few people and no clear emotion regulation strategy likelihood-of-use pattern. Thus, we interpret the more parsimonious three-profile solution as the best-fitting structure. Table 2. Model Fit Statistics for Two-, Three-, Four-, and Five-Latent Profile Solutions Model  Log-likelihood  BIC  Entropy  LMRT  LRMTp  BLRT  BLRTp  Two-profile  −2,226.10  4,696.30  0.82  485.40  .001  485.40  .000  Three-profile  −2,078.54  4,497.69  .93  295.12  .016  292.10  .017  Four-profile  −2,001.74  4,440.58  .91  153.80  .075  153.61  .000  Five-profile  −1,959.90  4,453.41  .90  83.68  .240  83.68  .000  Model  Log-likelihood  BIC  Entropy  LMRT  LRMTp  BLRT  BLRTp  Two-profile  −2,226.10  4,696.30  0.82  485.40  .001  485.40  .000  Three-profile  −2,078.54  4,497.69  .93  295.12  .016  292.10  .017  Four-profile  −2,001.74  4,440.58  .91  153.80  .075  153.61  .000  Five-profile  −1,959.90  4,453.41  .90  83.68  .240  83.68  .000  Note: Bolded profile indicates best-fitting, parsimonious solution. BIC = Bayesian Information Criterion; BLRT = Parametric Bootstrapped Likelihood Ratio Test; LMRT = Lo-Mendell-Rubin Likelihood Ratio Test; Better fit is indicated by decreasing BIC, entropy ≥ .90, and significant LMRT and BLRT. View Large Table 2. Model Fit Statistics for Two-, Three-, Four-, and Five-Latent Profile Solutions Model  Log-likelihood  BIC  Entropy  LMRT  LRMTp  BLRT  BLRTp  Two-profile  −2,226.10  4,696.30  0.82  485.40  .001  485.40  .000  Three-profile  −2,078.54  4,497.69  .93  295.12  .016  292.10  .017  Four-profile  −2,001.74  4,440.58  .91  153.80  .075  153.61  .000  Five-profile  −1,959.90  4,453.41  .90  83.68  .240  83.68  .000  Model  Log-likelihood  BIC  Entropy  LMRT  LRMTp  BLRT  BLRTp  Two-profile  −2,226.10  4,696.30  0.82  485.40  .001  485.40  .000  Three-profile  −2,078.54  4,497.69  .93  295.12  .016  292.10  .017  Four-profile  −2,001.74  4,440.58  .91  153.80  .075  153.61  .000  Five-profile  −1,959.90  4,453.41  .90  83.68  .240  83.68  .000  Note: Bolded profile indicates best-fitting, parsimonious solution. BIC = Bayesian Information Criterion; BLRT = Parametric Bootstrapped Likelihood Ratio Test; LMRT = Lo-Mendell-Rubin Likelihood Ratio Test; Better fit is indicated by decreasing BIC, entropy ≥ .90, and significant LMRT and BLRT. View Large The three latent profiles of emotion regulation likelihood of use are illustrated in Figure 1 and descriptive results are presented in Table 3. The first profile, Situation Modifiers, included people (N = 137) who said they were most likely to use situation modification and least likely to use response modulation strategies. The second profile, Classically Adaptive Regulators, consisted of people (N = 137) who said they were most likely to use situation modification, attentional deployment, and positive reappraisal strategies—all nominally adaptive strategies—and least likely to use rumination and situation selection strategies. The final profile, Avoiders and Ruminators, consisted of people (N = 18) who said they most were most likely to use rumination and avoidant situation selection strategies and least likely to use situation modification and attentional deployment strategies. Additional profile descriptive statistics are presented in Supplementary Materials. Figure 1. View largeDownload slide Latent profiles of emotion regulation preferences. A three-profile solution provided the best fit to the data. The first profile (N = 137) included participants who preferred situation modification over other strategies. The second profile (N = 137) included participants who preferred to use classically adaptive strategies like situation modification, attentional deployment, and reappraisal. The third profile (N = 18) included participants who preferred to use avoidance and rumination, which are often considered less adaptive strategies. Figure 1. View largeDownload slide Latent profiles of emotion regulation preferences. A three-profile solution provided the best fit to the data. The first profile (N = 137) included participants who preferred situation modification over other strategies. The second profile (N = 137) included participants who preferred to use classically adaptive strategies like situation modification, attentional deployment, and reappraisal. The third profile (N = 18) included participants who preferred to use avoidance and rumination, which are often considered less adaptive strategies. Table 3. Standardized Means and Standard Errors for Emotion Regulation Strategy Likelihood of Use across the Three Latent Profiles   Profile 1 Situation Modifiers  Profile 2 Classically Adaptive Regulators  Profile 3 Avoiders and Ruminators  Strategy  M  SE  M  SE  M  SE  SS Avoidance  −0.21  0.10  −1.00  0.13  1.16  0.27  SS Leave  −0.38  0.10  −1.11  0.13  0.86  0.34  SM Modification  1.83  0.13  2.21  0.15  −1.99  0.29  AD Negative  0.43  0.09  1.18  0.13  −1.55  0.39  AD Positive  0.03  0.09  0.92  0.11  −1.12  0.30  CC Detachment  0.06  0.08  0.11  0.10  0.21  0.23  CC Distraction  −0.03  0.08  0.11  0.10  −0.32  0.20  CC Pos. Reappraisal  −0.42  0.09  0.52  0.12  −0.39  0.24  CC Rumination  0.34  0.15  −3.04  0.33  3.17  0.50  CC Acceptance  0.08  0.08  −0.11  0.10  −0.32  0.25  RM Suppression  −0.20  0.08  −0.32  0.10  0.22  0.36  RM Positive  −0.52  0.08  −0.06  0.11  −0.08  0.35  RM Physio  −0.33  0.09  0.31  0.10  −0.26  0.29    Profile 1 Situation Modifiers  Profile 2 Classically Adaptive Regulators  Profile 3 Avoiders and Ruminators  Strategy  M  SE  M  SE  M  SE  SS Avoidance  −0.21  0.10  −1.00  0.13  1.16  0.27  SS Leave  −0.38  0.10  −1.11  0.13  0.86  0.34  SM Modification  1.83  0.13  2.21  0.15  −1.99  0.29  AD Negative  0.43  0.09  1.18  0.13  −1.55  0.39  AD Positive  0.03  0.09  0.92  0.11  −1.12  0.30  CC Detachment  0.06  0.08  0.11  0.10  0.21  0.23  CC Distraction  −0.03  0.08  0.11  0.10  −0.32  0.20  CC Pos. Reappraisal  −0.42  0.09  0.52  0.12  −0.39  0.24  CC Rumination  0.34  0.15  −3.04  0.33  3.17  0.50  CC Acceptance  0.08  0.08  −0.11  0.10  −0.32  0.25  RM Suppression  −0.20  0.08  −0.32  0.10  0.22  0.36  RM Positive  −0.52  0.08  −0.06  0.11  −0.08  0.35  RM Physio  −0.33  0.09  0.31  0.10  −0.26  0.29  Note: Estimates are standardized on the observed indicators and covariates (i.e., STDYX in Mplus). AD = Attentional deployment; CC = Cognitive change; RM = Response modulation; SM = Situation modification; SS = Situation selection. View Large Table 3. Standardized Means and Standard Errors for Emotion Regulation Strategy Likelihood of Use across the Three Latent Profiles   Profile 1 Situation Modifiers  Profile 2 Classically Adaptive Regulators  Profile 3 Avoiders and Ruminators  Strategy  M  SE  M  SE  M  SE  SS Avoidance  −0.21  0.10  −1.00  0.13  1.16  0.27  SS Leave  −0.38  0.10  −1.11  0.13  0.86  0.34  SM Modification  1.83  0.13  2.21  0.15  −1.99  0.29  AD Negative  0.43  0.09  1.18  0.13  −1.55  0.39  AD Positive  0.03  0.09  0.92  0.11  −1.12  0.30  CC Detachment  0.06  0.08  0.11  0.10  0.21  0.23  CC Distraction  −0.03  0.08  0.11  0.10  −0.32  0.20  CC Pos. Reappraisal  −0.42  0.09  0.52  0.12  −0.39  0.24  CC Rumination  0.34  0.15  −3.04  0.33  3.17  0.50  CC Acceptance  0.08  0.08  −0.11  0.10  −0.32  0.25  RM Suppression  −0.20  0.08  −0.32  0.10  0.22  0.36  RM Positive  −0.52  0.08  −0.06  0.11  −0.08  0.35  RM Physio  −0.33  0.09  0.31  0.10  −0.26  0.29    Profile 1 Situation Modifiers  Profile 2 Classically Adaptive Regulators  Profile 3 Avoiders and Ruminators  Strategy  M  SE  M  SE  M  SE  SS Avoidance  −0.21  0.10  −1.00  0.13  1.16  0.27  SS Leave  −0.38  0.10  −1.11  0.13  0.86  0.34  SM Modification  1.83  0.13  2.21  0.15  −1.99  0.29  AD Negative  0.43  0.09  1.18  0.13  −1.55  0.39  AD Positive  0.03  0.09  0.92  0.11  −1.12  0.30  CC Detachment  0.06  0.08  0.11  0.10  0.21  0.23  CC Distraction  −0.03  0.08  0.11  0.10  −0.32  0.20  CC Pos. Reappraisal  −0.42  0.09  0.52  0.12  −0.39  0.24  CC Rumination  0.34  0.15  −3.04  0.33  3.17  0.50  CC Acceptance  0.08  0.08  −0.11  0.10  −0.32  0.25  RM Suppression  −0.20  0.08  −0.32  0.10  0.22  0.36  RM Positive  −0.52  0.08  −0.06  0.11  −0.08  0.35  RM Physio  −0.33  0.09  0.31  0.10  −0.26  0.29  Note: Estimates are standardized on the observed indicators and covariates (i.e., STDYX in Mplus). AD = Attentional deployment; CC = Cognitive change; RM = Response modulation; SM = Situation modification; SS = Situation selection. View Large Age differences in profile membership Consistent with the MANOVA analyses (see Table 1), multinomial logistic regression results revealed significant age differences in the odds of being in one profile versus another. Younger adults were chosen as the reference category a priori given our interests in comparing young to older adults; the inclusion of middle-aged adults was largely exploratory but could examine whether age differences emerged earlier (YA vs MA and OA) or later (YA vs OA only). Middle-aged and older adults had greater odds of being Classically Adaptive Regulators than Situation Modifiers (ORoa = 3.59, p < .001; ORma = 69.69, p < .001) or Avoiders and Ruminators (ORoa = 18.90, p = .006; ORma = 51.33, p = .008) relative to younger adults. Profile membership odds did not vary by gender (ps = .111–.610). Profile and age differences in emotion malleability beliefs We explored whether these profile categorizations predicted people’s beliefs about the malleability of emotion. Specifically, we ran a regression model predicting emotion malleability beliefs with profile membership. Age group was also included as a predictor in the regression models given the observed age differences in profile membership odds. Results indicated a significant effect of profile membership on malleability beliefs (β = 0.19, p ≤ .002). Controlling for age group, Situation Modifiers were more likely to hold fixed beliefs about emotion than Classically Adaptive Regulators (β = −0.17, p = .009) and Avoiders and Ruminators (β = −0.27, p = .039). However, there were no significant differences in emotion malleability beliefs between Classically Adaptive Regulators and Avoiders and Ruminators (β = −0.10, p = .457). Age group was not significantly related to emotion malleability beliefs (β = −0.06, p = .298). Profile membership and age group together accounted for 3.5% of the variance in emotion malleability beliefs. Discussion If we aim to understand how older adults can maintain high levels of emotional well-being, it is important to investigate emotion regulatory processes and the beliefs that guide those processes. The purpose of the present research was to characterize age differences in beliefs about emotion regulation strategies—how effective people perceive various tactics to be and how likely they believe they are to use them. The findings provide more nuanced insight into the factors that may influence age differences in the use and effectiveness of emotion regulation strategies. Age Differences (and Lack Thereof) in Emotion Regulation Beliefs Age differences do emerge in beliefs about emotion regulation strategies, though they are more likely to appear in beliefs about use than in beliefs about effectiveness. The findings suggest that younger adults generally agree with older adults on which strategies are effective, but diverge in which ones they use. This result is interesting because actual effectiveness of at least some regulation strategies may vary by age (e.g., Noh, Lohani, & Isaacowitz, 2011; Shiota & Levenson, 2009; Urry & Gross, 2010). It may therefore be important in future research to distinguish between beliefs about effectiveness in general and beliefs about personal effectiveness (see De Castella et al., 2013), as well as to assess whether some age groups are more likely than others to admit to utilizing strategies that they do not necessarily believe to be most effective. Of the age differences that did emerge, younger adults ranked cognitive and attentional distraction as more effective than middle-aged and older adults (though the latter effect was only marginally significant), but reported being less likely to use attentional deployment. These results generally align with past work showing consistent patterns of attention away from negative material in older adults (use) but mixed findings regarding the relation between attention and affect (effectiveness). For example, attentional deployment can effectively reduce negative affect for those adults who have stronger attentional control abilities, but not those without (Isaacowitz, 2012). In our data, older (and middle-aged adults) reported being more likely to use attentional deployment than younger adults, but rated it as less effective. It may be that individual differences in cognitive resources drive this discrepancy: Older age groups show a positivity effect, but it is only effective for some of them. There was some support for the idea that older (and middle-aged) adults turn to more putatively effective strategies than younger adults. Although participants generally agreed on the effectiveness of the strategies, younger adults were more likely to use avoidance and rumination (which have been linked to negative mental health outcomes; see Aldao et al., 2010) than middle-aged and older adults and less likely to use positive reappraisal (which has been linked to positive emotional outcomes in older adults (Nowlan et al., 2015); reappraisal in general has been linked with more positive outcomes (e.g., Aldao et al., 2010; Webb et al., 2012). The observed pattern of age differences did not support the idea that older adults more often try to intervene early in the emotion-generative process and avoid later strategies of cognitive change and response modulation, or that older adults prefer positive varieties of emotion regulation (Livingstone & Isaacowitz, 2016). Instead, a third possibility emerged, suggesting that younger adults may be more inclined to disengage from (rather than engage with) emotions compared to middle-aged and older adults. In particular, younger adults rated directing attention away from negative and cognitive distraction—two forms of mental disengagement—as more effective than older adults. They were also more likely to report using avoidance, cognitive distraction, and rumination than middle-aged and older adults. These results suggest a tendency for younger adults to try to disengage—cognitively and behaviorally—from emotional situations. This tendency may decline with age, as people learn to more effectively deal with emotions, and draw on knowledge and experience to engage with emotions effectively, though longitudinal research is needed to test this idea. Profiles of Emotion Regulation Using latent profile analysis, we identified three profiles of preferences for emotion regulation strategies. Slightly less than half of participants were classified as classically adaptive regulators, relying on a variety of situation-based and cognitive strategies that have been generally shown to be effective. A similar percentage of people were classified as situation modifiers, who showed a strong preference for changing situations. A small percentage of participants were classified as avoiders and ruminators, relying relatively more on these traditionally maladaptive strategies and less on situation modification and attentional deployment. Middle-aged and older adults were more likely to be classified as classically adaptive regulators than situation modifiers or avoiders and ruminators, compared to younger adults, which supports the idea that emotion regulation skills may improve with age (Charles, 2010). These people appear to demonstrate regulatory flexibility, reporting that they would rely on a greater variety of strategies than other profiles, suggesting a larger repertoire from which to choose, which is likely to benefit them across numerous contexts and domains (Bonanno & Burton, 2013). In contrast, avoidance and rumination are linked with poorer mental health (Aldao et al., 2010); it may be that younger adults have not yet learned that these strategies are ineffective. On the other hand, as an antecedent-focused strategy, situation modification is often considered an adaptive way of regulating one’s emotions (Gross, 1998). In contrast to classically adaptive regulators, situation modifiers were less likely to believe that emotions are malleable. For people with these beliefs, then, if strategies that target the emotion are unlikely to work, the most effective way to manage emotions may be to target the situation itself. The strategy is likely to be effective to the extent to which a situation can actually be modified. When a situation is uncontrollable, an emotion-focused approach may be more effective (Zakowski, Hall, Klein, & Baum, 2001). These profiles did not align with the families of strategies proposed by the process model, or with the idea that older adults may prefer more “positive” versions of strategies (e.g., pay more attention to positive than negative, use positive reappraisal). Instead, the pattern seems more nuanced. The distinction between putatively adaptive and maladaptive strategies is clear, with only a small number of people showing a distinctly maladaptive profile. In addition, there is a distinction between situation-focused and emotion-focused regulation, which parallels the distinction in the coping literature between problem-focused and emotion-focused coping (Lazarus & Folkman, 1984). Limitations and Future Directions A clear next step is to investigate the extent to which effectiveness and use beliefs are related to actual emotion regulation behavior in the lab and in real life (e.g., using experience sampling). Research in children suggests that beliefs about emotion regulation strategies predict in-lab behavior (Cole et al., 2009). Emotion regulation theory also identifies strategy selection as an important step in the emotion regulation process (Gross, 2015). Open questions include whether beliefs about effectiveness predict effective emotion regulation, beliefs about which strategies people would use predict what strategies they do use, and the extent to which these sets of beliefs (effectiveness and preferences) are related. Although we might predict a model in which age is related to differences in these beliefs, which in turn predict differences in likelihood of use, the current study examined beliefs in separate samples, to avoid beliefs in one category influencing the other. It would be interesting to examine within-person beliefs systems, and whether a feedback loop exists between the two over time. For example, if emotion regulation skills improve with age, we would expect a stronger correlation between effectiveness and likelihood of use beliefs to develop with age. Another limitation is that situation selection was assessed here as a disengagement from a negative situation that had already occurred. Theory and research suggest that when situation selection is implemented before the emotional event has occurred, it can be an effective strategy for managing short-term emotions (Livingstone & Isaacowitz, 2015). Moreover, situation selection is theorized to be especially effective for older adults because it can effectively intervene before emotional processes are underway (Urry & Gross, 2010). Due to the way the questions were phrased, this possibility was not fully explored in the current study. Similarly, all the strategies included in the current research were intrapersonal, though there are also likely opportunities for interpersonal emotion regulation (e.g., seeking social support) for many of the scenarios. Finally, this study examined the effectiveness and use of strategies for hedonic emotion regulation; however, findings may be different for contra-hedonic emotion regulation for instrumental purposes (see Tamir, 2009). Conclusion The findings reported here expand our knowledge of emotion regulation across adulthood, specifically the extent of age differences in beliefs about the effectiveness of and likelihood of using different emotion regulation strategies. Although there was relative consensus on strategy effectiveness and popularity of use, there was also significant variability across age groups. Beliefs about emotion regulation likely play an important role in the process of selecting a strategy to use (see Gross, 2015). Continuing to investigate such beliefs—where they come from, whether they change over time, and how they influence emotion regulation processes—will help us understand how people of different ages implement different emotion regulation strategies, and thus how to maintain and enhance well-being across the life span. Supplementary Material Supplementary data is available at The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences online. Funding This research was supported in part by the National Institute on Aging Grant R01 AG948731 to D. M. Isaacowitz. Conflict of Interest None reported. References Aldao, A. ( 2013). The future of emotion regulation research capturing context. Perspectives on Psychological Science , 8, 155– 172. doi: 10.1177/1745691612459518 Google Scholar CrossRef Search ADS PubMed  Aldao, A., Nolen-Hoeksema, S., & Schweizer, S. ( 2010). Emotion-regulation strategies across psychopathology: A meta-analytic review. Clinical Psychology Review , 30, 217– 237. doi: 10.1016/ j.cpr.2009.11.004 Google Scholar CrossRef Search ADS PubMed  Bonanno, G. A., & Burton, C. L. ( 2013). Regulatory flexibility: An individual differences perspective on coping and emotion regulation. Perspectives on Psychological Science , 8, 591– 612. doi: 10.1177/1745691613504116 Google Scholar CrossRef Search ADS PubMed  Bushman, B. J., Baumeister, R. F., & Phillips, C. M. ( 2001). Do people aggress to improve their mood? Catharsis beliefs, affect regulation opportunity, and aggressive responding. Journal of Personality and Social Psychology , 81, 17– 32. doi: 10.1037//0022-3514.81.1.17 Google Scholar CrossRef Search ADS PubMed  Cabello, R., & Fernández-Berrocal, P. ( 2015). Implicit theories and ability emotional intelligence. Frontiers in Psychology , 6, 700– 707. doi: 10.3389/fpsyg.2015.00700 Google Scholar CrossRef Search ADS PubMed  Carstensen, L. L., Isaacowitz, D. M., & Charles, S. T. ( 1999). Taking time seriously. A theory of socioemotional selectivity. The American Psychologist , 54, 165– 181. doi:10.1037/0003-066X.54.3.165 Google Scholar CrossRef Search ADS PubMed  Charles, S. T. ( 2010). Strength and vulnerability integration: A model of emotional well-being across adulthood. Psychological Bulletin , 136, 1068– 1091. doi: 10.1037/a0021232 Google Scholar CrossRef Search ADS PubMed  Chesney, S. A., & Gordon, N. S. ( 2017). Profiles of emotion regulation: Understanding regulatory patterns and the implications for posttraumatic stress. Cognition and Emotion , 31, 598– 606. doi: 10.1080/02699931.2015.1126555 Google Scholar CrossRef Search ADS PubMed  Cohen, P., Cohen, J., Aiken, L. S., & West, S. G. ( 1999). The problem of units and the circumstance for POMP. Multivariate Behavioral Research , 34, 315– 346. doi: 10.1207/S15327906MBR3403_2 Google Scholar CrossRef Search ADS   Cole, P. M., Dennis, T. A., Smith-Simon, K. E., & Cohen, L. H. ( 2009). Preschoolers’ emotion regulation strategy understanding: Relations with emotion socialization and child self-regulation. Social Development , 18, 324– 352. doi: 10.1111/j.1467-9507.2008.00503.x Google Scholar CrossRef Search ADS   De Castella, K., Goldin, P., Jazaieri, H., Ziv, M., Dweck, C. S., & Gross, J. J. ( 2013). Beliefs about emotion: Links to emotion regulation, well-being, and psychological distress. Basic and Applied Social Psychology , 35, 497– 505. doi: 10.1080/01973533.840632 Google Scholar CrossRef Search ADS   Dennis, T. A., & Kelemen, D. A. ( 2009). Preschool children’s views on emotion regulation: Functional associations and implications for social-emotional adjustment. International Journal of Behavior Development , 33, 243– 252. doi: 10.1177/0165025408098024 Google Scholar CrossRef Search ADS   Dweck, C. S. ( 1999). Self-theories: Their role in motivation, personality, and development . Philadelphia, PA: Psychology Press. Enders, C. K. ( 2011). Analyzing longitudinal data with missing values. Rehabilitation Psychology , 56, 267– 288. doi: 10.1037/a0025579 Google Scholar CrossRef Search ADS PubMed  Gross, J. J. ( 1998). The emerging field of emotion regulation: An integrative review. Review of General Psychology , 2, 271– 299. doi: 10.1037/1089-2680.2.3.271 Google Scholar CrossRef Search ADS   Gross, J. J. ( 2015). Emotion regulation: Current status and future prospects. Psychological Inquiry , 26, 1– 26. doi: 10.1080/1047840X.2014.940781 Google Scholar CrossRef Search ADS   Hagenaars, J. A., & McCutcheon, A. L. (Eds.). ( 2009). Applied latent class analysis . Cambridge, UK: Cambridge University Press. Isaacowitz, D. M. ( 2012). Mood regulation in real time: Age differences in the role of looking. Current Directions in Psychological Science , 21, 237– 242. doi: 10.1177/0963721412448651 Google Scholar CrossRef Search ADS PubMed  John, O. P., & Gross, J. J. ( 2004). Healthy and unhealthy emotion regulation: Personality processes, individual differences, and life span development. Journal of Personality , 72, 1301– 1334. doi: 10.1111/j.1467-6494.2004.00298.x Google Scholar CrossRef Search ADS PubMed  Lazarus, R. S. ( 1991). Emotion and adaptation . Cary, NC: Oxford University Press. Lazarus, R. S., & Folkman, S( 1984). Stress appraisal and coping . New York: Springer. Livingstone, K. M., & Isaacowitz, D. M. ( 2015). Situation selection and modification for emotion regulation in younger and older adults. Social Psychological and Personality Science , 6, 904– 910. doi: 10.1177/1948550615593148 Google Scholar CrossRef Search ADS PubMed  Livingstone, K. M., & Isaacowitz, D. M. ( 2016). Age differences in use and effectiveness of positivity in emotion regulation: The sample case of attention. In A. D. Ong & C. E. Löekenhoff (Eds.), Emotion, aging, and health  (pp. 31– 48). Washington, DC: American Psychological Association. Google Scholar CrossRef Search ADS   Marsh, H. W., Lüdtke, O., Trautwein, U., & Morin, A. J. ( 2009). Classical latent profile analysis of academic self-concept dimensions: Synergy of person-and variable-centered approaches to theoretical models of self-concept. Structural Equation Modeling , 16, 191– 225. doi: 10.1080/10705510902751010 Google Scholar CrossRef Search ADS   Mather, M., & Carstensen, L. L. ( 2005). Aging and motivated cognition: The positivity effect in attention and memory. Trends in Cognitive Sciences , 9, 496– 502. doi: 10.1016/j.tics. 2005.08.005 Google Scholar CrossRef Search ADS PubMed  Muthén, B. ( 2001). Latent variable mixture modeling. In G. A. Marcoulides & R. E. Schumacker (Eds.), New developments and techniques in structural equation modeling  (pp. 1– 33). Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Muthén, B. ( 2004). Latent variable analysis. The Sage handbook of quantitative methodology for the social sciences  (pp. 345– 368). Thousand Oaks, CA: Sage Publications. Muthén, L. K., & Muthén, B.O. ( 1998–2015). Mplus user’s guide . 7th ed.. Los Angeles, CA: Muthén & Muthén. Muthén, B., & Muthén, L. K. ( 2000). Integrating person-centered and variable-centered analyses: Growth mixture modeling with latent trajectory classes. Alcoholism: Clinical and Experimental Research , 24, 882– 891. doi: 10.1111/j.1530-0277.2000.tb02070.x Google Scholar CrossRef Search ADS   Noh, S. R., Lohani, M., & Isaacowitz, D. M. ( 2011). Deliberate real-time mood regulation in adulthood: The importance of age, fixation, and attentional functioning. Cognition and Emotion , 25, 998– 1013. doi: 10.1080/02699931.2010.541668 Google Scholar CrossRef Search ADS PubMed  Nolen-Hoeksema, S., & Aldao, A. ( 2011). Gender and age differences in emotion regulation strategies and their relationship to depressive symptoms. Personality and Individual Differences , 51, 704– 708. doi: 10.1016/j.paid.2011.06.012 Google Scholar CrossRef Search ADS   Nowlan, J. S., Wythrich, V. M., & Rapee, R. M. ( 2015). Positive reappraisal in older adults: A systematic literature review. Aging & Mental Health , 19, 475– 484. doi: 10.1080/13607863.2014.954528 Google Scholar CrossRef Search ADS PubMed  Nylund, K. L., Asparouhov, T., & Muthén, B. O. ( 2007). Deciding on the number of classes in latent class analysis and growth mixture modeling: A Monte Carlo simulation study. Structural Equation Modeling , 14, 535– 569. doi: 10.1080/10705510701575396 Google Scholar CrossRef Search ADS   Ortner, C. N. M., Briner, E. L., & Marjanovic, Z. ( 2017). Believing is doing: Emotion regulation beliefs are associated with emotion regulation behavioral choices and subjective well-being. Europe’s Journal of Psychology , 13, 60– 74. doi: 10.5964/ejop.v13i1.1248 Google Scholar CrossRef Search ADS PubMed  Papageorgiou, C., & Wells, A. ( 2001). Metacognitive beliefs about rumination in recurrent major depression. Cognitive and Behavioral Practice , 8, 160– 164. doi: 10.1016/S1077-7229(01)80021–3 Google Scholar CrossRef Search ADS   Pastor, D. A., Barron, K. E., Miller, B. J., & Davis, S. L. ( 2007). A latent profile analysis of college students’ achievement goal orientation. Contemporary Educational Psychology , 32, 8– 47. doi: 10.1016/j.cedpsych.2006.10.003 Google Scholar CrossRef Search ADS   Reed, A. E., Chan, L., & Mikels, J. A. ( 2014). Meta-analysis of the age-related positivity effect: Age differences in preferences for positive over negative information. Psychology and Aging , 29, 1– 15. doi: 10.1037/a0035194 Google Scholar CrossRef Search ADS PubMed  Sands, M., Livingstone, K. M., & Isaacowitz, D. M.(in press). Characterizing age-related positivity effects in situation selection. International Journal of Behavioral Development . doi: 10.1177/0165025417723086 Shiota, M. N., & Levenson, R. W. ( 2009). Effects of aging on experimentally instructed detached reappraisal, positive reappraisal, and emotional behavior suppression. Psychology and Aging , 24, 890– 900. doi: 10.1037/a0017896 Google Scholar CrossRef Search ADS PubMed  Tamir, M. ( 2009). What do people want to feel and why? Pleasure and utility in emotion regulation. Current Directions in Psychological Science , 18, 101– 105. doi: 10.1111/j.1467-8721.2009.01617.x Google Scholar CrossRef Search ADS   Tamir, M., John, O. P., Srivastava, S., & Gross, J. J. ( 2007). Implicit theories of emotion: Affective and social outcomes across a major life transition. Journal of Personality and Social Psychology , 92, 731– 744. doi: 10.1037/0022-3514.92.4.731 Google Scholar CrossRef Search ADS PubMed  Trincas, R., Bilotta, E., & Mancini, F. ( 2016). Specific beliefs about emotions are associated with different emotion- regulation strategies. Psychology , 7, 1682– 1699. doi: 10.4236/psych. 2016.713159 Google Scholar CrossRef Search ADS   Troy, A. S., Shallcross, A. J., & Mauss, I. B. ( 2013). A person-by-situation approach to emotion regulation: Cognitive reappraisal can either help or hurt, depending on the context. Psychological Science , 24, 2505– 2514. doi: 10.1177/0956797613496434 Google Scholar CrossRef Search ADS PubMed  Urry, H. L., & Gross, J. J. ( 2010). Emotion regulation in older age. Current Directions in Psychological Science , 19, 352– 357. doi: 10.1037/14857-004 Google Scholar CrossRef Search ADS   Webb, T. L., Miles, E., & Sheeran, P. ( 2012). Dealing with feeling: A meta-analysis of the effectiveness of strategies derived from the process model of emotion regulation. Psychological Bulletin , 138, 775– 808. doi: 10.1037/a0027600 Google Scholar CrossRef Search ADS PubMed  Zakowski, S. G., Hall, M. H., Klein, L. C., & Baum, A. ( 2001). Appraised control, coping, and stress in a community sample: A test of the goodness-of-fit hypothesis. Annals of Behavioral Medicine , 23, 158– 165. doi: 10.1207/S15324796ABM2303_3 Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

Journal

The Journals of Gerontology Series B: Psychological Sciences and Social SciencesOxford University Press

Published: Feb 21, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off