The Role of Autobiographical Resilience Memories in Emotion Regulation: An Account of Age Differences in Mnemonic and Positive Reappraisal

Abstract

A lifetime of resilience through emotionally challenging experiences may benefit older adults, lending to emotion regulation mastery with time. Yet the influence of autobiographical experiences on momentary reappraisal, the reinterpretation of negative stimuli as more positive, has never been empirically tested. This online study examined the extent to which associating life memories of resilience with novel negative scenarios enhanced reappraisal efficacy and reduced difficulty to reappraise. Younger and older adults reappraised negative images by associating reappraisals to freely selected autobiographical resilience memories, cued autobiographical resilience memories, or by finding situational silver linings without mnemonic association (control). Change in image emotional intensity ratings revealed no difference across reappraisal conditions for younger adults, while older adults most effectively down-regulated emotional intensity using the control reappraisal strategy. Older adults found autobiographical memories more helpful for mood regulation and less difficult to implement, and identified greater similarity between novel negative scenarios and their memories than younger adults. Surprisingly, greater similarity between resilience memories and negative images was associated with lower reappraisal efficacy for both age groups. Findings demonstrate age-equivalent benefits of utilizing reappraisals associated with past narratives of resilience, and suggest a sacrifice of immediate hedonic benefit for disproportionately greater subjective benefits with age.

Accumulated life experience supports emotion regulation in late life, as motivation to make meaning of salient life events increases with age (Carstensen, Fung, & Charles, 2003). Indeed, age related improvements in emotional health are well established (Carstensen et al., 2011; Brady, Kneebone, Denson, & Bailey, 2018; Mikkelsen, Tramm, & O-Toole, 2021), and often accredited to greater use of emotion regulation strategies (Lawton, 2001; Gross, Richard, & John, 2006), and greater benefits from emotion regulation use (Nowlan, Wuthrich, and Rapee, 2015; Etezadi & Pushkar, 2013; Lohani & Isaacowitz, 2014). Relative to their younger counterparts, older adults place greater value on making meaning of past memories, and more readily interpret autobiographical memories in terms of personal growth and self-efficacy (Bauer, McAdams, & Sakaeda, 2005; Pasupathi & Mansour, 2006). Late-life strategies of meaning-making may achieve emotional well-being through learned resilience from challenging autobiographical events. However, the impact of autobiographical recall on momentary emotion regulation efficacy remains largely unexplored. Thus, the current study examined how associating resilience narratives with momentary cognitive reappraisals, one of the most well-studied and effective emotion regulation strategies (Aldao, Nolen-Hoeksema, & Schweizer, 2010; Webb, Miles, & Sheeran, 2012), supports late-life emotion regulation.

Resilience Narratives Uniquely Support Emotion Regulation in Later-Life

Identifying lessons learned from life turning points promotes greater well-being as individuals cultivate wisdom and shift focus to eudaimonic gains, a suite of goals that promote well-being through self-acceptance, autonomy, life purpose, and growth (Bauer, Graham, Lauber, & Lynch, 2019; Ong & Patterson, 2016; McLean & Pratt, 2006;). Finding the good in past hardship lowers subsequent rates of psychopathology (Dulin & Passmore, 2010). Narrating growth from times of crisis also predicts resilience to new stress with age (Park 2010; Park, Chmiewlewski & Blank, 2010; Park, Mills-Baxter & Fenster, 2005). Even in highly negative and violent scenarios, older age is associated with a tendency to enhance positive features of experienced negative events (Ford, DiBiase, & Kensinger, 2018). Recalling successful coping efforts from the past may promote resolution of momentary distress by analyzing past adaptive behaviors, and recycling successful coping narratives via reminiscence, in response to newly experienced stressors (Meléndez, Fortuna, Sales, & Mayordomo, 2015). Thus, interpreting a silver lining in past times of crisis may improve emotional outcomes and emotion regulation when navigating novel hardship. Only one study has examined autobiographical recall of critical life memories and its influence on reappraising, such that recalling a resilience memory, relative to any positive memory, was associated with enhanced reappraisal of past negative life experiences in young adults’ lives (Paersch, Schulz, Wilhelm, Brown & Kleim, 2021). Yet, older adults may identify threads of resilience across autobiographical memories more readily, in order to support emotional well-being in the moment.

Compared to younger people, older individuals demonstrate more stable sense of identity when reviewing accounts of hardship and resilience (Pasupathi & Mansour, 2006), and more often acknowledge positive meaning and self-growth in narratives of adverse autobiographical experiences (Pasupathi & Mansour, 2006; Singer, Rexhaj, & Baddeley, 2007; Ford et al., 2018). Older adults also show greater mental health benefits from identifying resilience and positive meaning in challenging life circumstances like terminal diagnoses and chronic illness (Nowlan er al., 2014). Importantly, older age, which allows for the development of a greater repertoire of memories, provides perspective and time to identify thematic similarity across challenging experiences. Indeed, older adults are more likely than younger adults to connect lessons learned across several emotionally salient and key life memories (Singer et al., 2007). For older individuals, abstract lessons learned from these hardships are more readily associated with imagined outcomes in the future (Addis, Wong, & Schacter, 2008), suggesting that older adults may be more adept at associating themes of resilience across unique experiences and applying them when needed (Martins-Klein, Orlovsky, & Heideman, 2022).

Cognitive Considerations in Autobiographical Recall Integrity

Older age is often associated with a time of cognitive loss yet utilizing autobiographical recall in emotion regulation may be particularly effective for older adults because it is less taxing on cognitive resources than other variants of cognitive reappraisal (Martins, Florjanczyk, Jackson, Gatz, & Mather, 2018). Situation focused cognitive reappraisals rely on coordinated ventrolateral prefrontal activation (Allard & Kensinger, 2014; Opitz, Rauch, Terry, & Urry, 2012; Opitz, Lee, Gross, & Urry, 2014) and cognitive control abilities, such as working memory and updating, that decline with age (Ocshner & Gross, 2005; Ochsner et al., 2004; Hedden & Gabrieli, 2004). In contrast, recall of autobiographical resilience narratives may rely less on cognitive control, and more on preserved midline regions of the prefrontal cortex supporting late-life self-reference and self-cognition (Martinelli, Anssens, Sperduti, & Piolino, 2013; Gutchess, Kensinger, & Schacter, 2007; Cabeza & St Jacques, 2007).

Activation of the medial prefrontal cortex is also implicated in the updating and integration of newly learned information with prior knowledge (Schlichting & Preston, 2015). Memory representations in the medial prefrontal cortex support retrieval of relevant life episodes during newly experienced events (St Jacques, Olm, &Schacter, 2013), drawing on past pertinent memories to inform behaviorally appropriate responses in the moment (Kroes & Fernandez, 2012). Preserved medial networks in later life may, therefore, support reactivation of past relevant memories for momentary use.

Autobiographical memory retrieval relies on hierarchical cognitive processes that may also present age-related cognitive costs (Conway & Pleydell-Pearce, 2000). Unsurprisingly, older adults demonstrate reduced retrieval of memory-specific autobiographical details (Peters, Fan & Sheldon, 2019; Conway & Pleydell-Pearce, 2000) and tend to retrieve more gist-based, abstract autobiographical memories (Wank, Andrews-Hanna, & Grilli, 2021). However, older adults may recall more detailed autobiographical memories when self-relevant cues are provided (Dijkstra & Kaup, 2005). Specificity of recalled memories is similar in older and younger persons when key life memories are paired with emotionally salient cues that carry personal relevance (e.g., “little brother’s 18^th birthday”) relative to generic cues (“birthday) (Cuervo-Lombard, Raucher-Chene, Linden, & Voltzenlogel, 2021; Martinelli et al., 2013; Uzer & Brown, 2017). Provision of cues may, therefore, help to level the playing field across younger and older adults when accessing autobiographical memories, thus reducing age-related cognitive burden and promoting emotional benefits of mnemonic recall.

The Current Study

In the present study, we examined how recall of resilience memories facilitates momentary use of reappraisal among younger and older adults. Participants were randomly assigned to reappraise negative meaning as more positive by: (1) associating the reappraisal with a specific cued autobiographical resilience memory, whereby participants initially chose critical life moments and narrated memory meaningfulness to later integrate during the task, (2) associating the reappraisal to any freely selected resilience memory in the moment, or (3) finding positive meaning without autobiographical memory recall (control reappraisal). To assess subjective benefits of autobiographical memory reappraisal, participants rated image intensity before and after reappraisal use, and the difficulty of employing the assigned strategy. In the autobiographical reappraisal conditions, participants also rated the similarity between their autobiographical memory and the negative image, and the subjective helpfulness of memory recall to their reappraisal attempts.

Across all conditions, we anticipated greater reappraisal efficacy among older adults, as compared to younger adults, with disproportionate advantages for the autobiographical reappraisal strategies. For instance, in the non-autobiographical control condition, we expected older adults to show greater emotion regulation efficacy because reappraisals were reframed by increasing positive meaning, thus maximizing late-life positive attentional bias. Relative to young adults, we predicted that older adults would also be better equipped to identify overlapping themes across past narratives of resilience and novel stressors (Singer et al., 2007; Addis et al., 2008; Martins-Klein et al., 2022), supporting their ability to benefit from autobiographical reappraisal. We attributed older adults’ superior meaning-making capacity, memory integration and initial narration of resiliency, a skill that is particularly developed with age, (Pasupathi & Mansour, 2006; Park et al., 2005; Singer et al., 2007) to greater emotion regulation efficacy when generating autobiographical reappraisals. Among just older adults, however, we anticipated potentially reduced efficacy of uncued autobiographical reappraisals, relative to the other two conditions, because no initial elaborative narration of meaning was provided, thus increasing cognitive load and reducing opportunity to identify thematic similarity between past resilience memories and novel stressful scenarios (Livingstone & Isaacowitz, 2018; Lohani & Isaacowitz, 2014; Shiota & Levenson, 2009). Because young adults are still developing skills of meaning-making and memory integration, we expected lower emotion regulation efficacy than older adults, and overall, equivalent efficacy across all reappraisal conditions.

We similarly expected that older adults would find more shared overlap in similarity among their autobiographical memories and novel stressors and find their autobiographical memories more helpful during reappraisal than younger adults because older adults more readily use resilience narratives to regulate emotions than younger adults (Pasupathi & Mansour, 2006; Singer et al., 2007). Among both age groups, we expected reports of greater similarity when using uncued memories, because memory search would be unconstrained to identify the most relevant autobiographical resilience event at time of reappraisal.

Finally, a step-wise effect of reappraisal difficulty was expected. Across age groups, we anticipated that non-autobiographical reappraisals would be easiest to employ because the strategy precludes memory search demands, thus minimizing cognitive burden. We predicted that cued autobiographical reappraisals would be easier to use than uncued reappraisals, as self-relevant and specific cues reduce cognitive load associated with autobiographical memory search (Uzer & Brown, 2017; Addis, Knapp, Roberts, & Schacter, 2012; Dijkstra & Kaup, 2005; Sheldon, Williams, Harrington, & Otto, 2020). In contrast, we expected that uncued autobiographical reappraisals would be rated by participants as most effortful to employ because of demanding task contexts and memory search (Conway & Pleydell-Pearce, 2000).

Materials and Methods

Participants

Two-hundred nineteen younger adults (age 18–29, M_age = 25.07, SD = 2.61, 56% Female) and 208 older adults (age 60–84, M_age = 65.63, SD = 4.45, 69% Female) were included in the final study sample (see Table 1 for demographic breakdown). Six hundred and two younger adults and 564 older adults were screened via CloudResearch for age (18–29 for younger adults, 60–85 for older adults), US residency, English fluency, greater than a 95% MTurk approval rating (Mitra, Hutto, & Gilbert, 2015), and/or congruent reports of age (e.g., “what year were you born?” versus “how old are you?”) (n=130 excluded). Additional participants were excluded for poor task engagement on effort tests sensitive to bot submissions and low task engagement (Allahbakhsh et al., 2013), evaluated via the Winograd Attentional Check questionnaire (Levesque, Davis & Morgenstern, 2012) and Instructional Manipulation Check (Oppenheimer, Meyvis, and Davidenko, 2009, n = 135 excluded).

Table 1.

Sample demographics for older and younger adults across cued, uncued and non-autobiographical reappraisal conditions.

	Older Adults (n=208)			Younger Adults (n=219)
	Non-AM (n=83)	Uncued AM (n=76)	Cued AM (n=49)	Non-AM (n=73)	Uncued AM (n=77)	Cued AM (n=69)
Age a	65.88 (4.87)	65.43 (4.10)	65.49 (4.32)	25.25 (2.52)	25.03 (2.71)	24.94 (2.61)
Gender a	69% F	66% F	66% F	52% F	59% F	64% F
Race a	95%	92%	89%	75%	68%	68%
Asian	0 (0%)	0 (0%)	0 (0%)	6 (8%)	7 (9%)	6 (9%)
Black/AA	1 (1%)	1 (1%)	3 (6%)	6 (8%)	7 (9%)	9 (13%)
White	79 (95%)	70 (92%)	44 (90%)	56 (77%)	54 (70%)	47 (68%)
Bi-Racial	0 (0%)	4 (6%)	2 (4%)	4 (6%)	7 (9%)	6 (9%)
Other/Declined to respond	3 (4%)	1 (1%)	0 (0%)	1 (1%)	2 (3%)	1 (1%)
Ethnicity	99% NH	89% NH	94% NH	94% NH	89% NH	95% NH
Education a	15.48 (2.67)	15.14 (2.46)	15.51 (2.67)	14.85 (2.03)	14.92 (2.25)	14.45 (2.08)
Stress a	4.28 (2.19)	4.33 (2.09)	4.41 (1.93)	5.29 (2.25)	5.66 (2.19)	5.96 (1.82)
Health	6.61 (1.64)	6.67 (1.74)	6.22 (1.85)	6.63 (1.51)	6.23 (1.67)	6.33 (1.59)

Note: M(SD) unless otherwise reported, NH = Non-Hispanic. “Stress” and “health” were measured in a different session, before completing the study task. Bi-Racial categories included individuals representing Native Hawaiian/Pacific Island and American-Indian/Alaska Native races). Racial identification denotes % Caucasian.

^aSignificant effect of age group;

^bSignificant effect of condition;

^cSignificant interaction effect of age group and condition

A total of 473 younger adults and 428 older adults were screened into the study, and only participants who accepted the study invitation, consented to participate, and completed at least 60% of the task were included in the final sample (n=269 excluded; the high rate of exclusion was, in-part, due to invited participants forgoing their study invitation). Participants with study dependent variables 3 standard deviations outside the mean were also excluded (n=2 excluded). Analyses reported below were run on valid trials for participants who completed at least two-thirds (6 of 9) of trials accurately (n = 203 excluded). As noted in Tables 1 and 2, initial analyses examined age group and conditional differences in sociodemographic and functional/mental health variables using univariate ANOVA or χ² tests. All variables that demonstrated conditional or age differences were first modeled as covariates. However, only education influenced primary outcome variables, thus education was modeled as a covariate and reported in relevant analyses.

Table 2.

Descriptive Statistics for Older and Younger Adults: Functional Health Measures.

	Older Adults (n=207)			Younger Adults (n=217)
	Non-AM (n=82)	Uncued AM (n=76)	Cued AM (n=49)	Non-AM (n=73)	Uncued AM (n=76)	Cued AM (n=68)
Sleep Disturbancea	48.85 (5.88)	50.38 (5.84)	51.05 (5.86)	52.02 (6.01)	52.38 (7.15)	53.02 (5.93)
Fatiguea,b	47.82 (8.99)	49.05 (10.1)	51.89 (8.77)	51.98 (9.82)	56.95 (9.58)	57.10 (8.14)
Anxietya	50.66 (9.86)	51.44 (8.79)	52.06 (9.55)	56.01 (9.17)	58.76 (10.21)	59.45 (8.94)
Depressiona	48.15 (8.76)	48.36 (7.60)	49.49 (8.95)	54.04 (9.57)	56.63 (11.39)	56.76 (9.15)
Physical Functiona	47.41 (9.79)	48.41 (10.11)	47.82 (9.91)	53.01 (7.20)	51.45 (8.32)	52.21 (7.41)
Paina	49.89 (8.80)	51.91 (9.59)	50.68 (8.33)	47.47 (8.51)	47.86 (8.95)	49.02 (10.16)
Activitya	34.92 (7.72)	36.24 (8.02)	36.95 (8.26)	36.21 (7.36)	38.45 (8.99)	39.29 (7.85)

Note: PROMIS-29 outcomes, reported as M(SD).

^aSignificant effect of age group;

^bSignificant effect of condition.

Procedure

Participants were invited to complete a one-hour online session on Amazon’s Mechanical Turk (MTurk; Crowston, 2012), and randomly assigned to one of three between-subjects reappraisal conditions (non-autobiographical control, uncued autobiographical memory or cued autobiographical memory). All participants completed an autobiographical memory generation task, reappraisal training, emotion regulation task, and post-task measures and questionnaires. The session median completion time was 53.2 minutes across all conditions and no difference in completion time was observed across conditions (p > 0.05).

Autobiographical Memory Generation Task.

Participants recalled 3 unique autobiographical memories of adverse experiences in their personal life during which they overcame a challenge and/or learned a salient life lesson (modified version of the Self-Defining Memory Task, Singer & Blagov, 2000; Blagov & Singer, 2004). For each memory, participants provided typed elaborative narratives describing the relevance and impact of the memory on their life, focusing on the growth, meaning, or lesson learned from the experience. Participants generated a 1–3 word descriptive cues for each memory (i.e. “Marriage, Wife, Church”; “Imprisonment, DUI”).

Emotion Regulation Task.

Participants were instructed to reappraise a slideshow of negative images and asked to consider them as real and not think of images as fake (e.g., “it’s just a movie”).

In the non-autobiographical memory condition, participants reappraised images by considering how the situation in the image may turn out ok in the end, so that it feels more positive. Participants in the uncued autobiographical memory condition were prompted to choose any relevant autobiographical resilience memory that shared features with the situation in the image, and integrate lessons learned from that memory into their momentary reappraisal so that the situation depicted in the image felt more positive. In contrast, individuals in the cued autobiographical memory condition were prompted with one of the 1–3 word cues they had generated in the Autobiographical Memory Generation Task and asked to integrate lessons learned from their memory into their momentary reappraisal so that the situation depicted in the image felt more positive (see Appendix A for full condition instructions).

Each assigned strategy was trained via three demonstration images, which were not shown again during the task. Corrective feedback was provided during each demonstration, where an example of alternative acceptable responses was given. At the end of task training, all participants were provided with instructional reminders and asked to define task parameters in their own words, as a manipulation check of their understanding and compliance with instructions.

As can be seen in Figure 1, participants viewed nine images (800 × 600 pixels) in randomized order, depicting negative interpersonal situations (i.e. medical procedures, car accident, pain, poverty) from the Nencki Affective Picture System (NAPS, Marchewka, Zurawski, Jednorog, & Grabowska, 2014). Images were selected for mild to moderate negative valence (M_val = 2.73, SD= 1.15) and moderate arousal (M_aro = 4.48, SD= 2.01) based on published ratings (Riegel et al., 2015). Images were organized into three blocks of three trials, which were self-paced and separated by a 3s intertrial fixation. Each block concluded with a self-paced break.

Figure 1.

Emotion Regulation Task Timeline

Note: Self-Paced = self-paced timing; S = seconds. Across three strategy conditions, participants completed an identical trial timeline. In the prepare slide, participants prepared their strategy in concordance with their trained strategy. In the situational condition, participants were to prepare and then use a reappraisal with no self-relevance. In contrast, AM conditions instructed preparation of any memory that may apply to the image (uncued), or preparation of a memory and self-generated cue from the AM generation task (cued).

At the start of each trial, participants were asked to “view” the negative image and rate baseline image intensity on a scale of 1 (not at all intense) to 9 (extremely intense). Participants were then shown a condition-specific cue to prepare reappraisal. For the cued autobiographical memory condition, participants were randomly presented with one of the 1–3 word generated cues, denoting the specific memory with which the image should be reappraised. Order of memory cue was randomized across 3 total blocks. Uncued autobiographical memory participants were presented with a “Prepare AM” cue to recall any relevant past memory of resilience that was helpful to their reappraisal of the image. Participants in the non-autobiographical memory condition were presented with a “Prepare” cue denoting that participants should prepare to reappraise in a moment, by attending to the situational features of the negative stimuli. After the preparation cue, participants reappraised the negative image and qualitatively described in writing how they carried out their reappraisal. At the end of each trial, all participants rated image intensity on the same scale of 1 (not at all intense) to 9 (extremely intense), and how difficult it was to reappraise the image with their instructed strategy on a scale 1 (not at all difficult) to 9 (extremely difficult). In both autobiographical memory conditions, participants also rated the similarity of their memory to the image on a scale of 1 (not at all similar) to 9 (very similar), and how helpful their memory was in reappraising the image on a scale from 1 (not at all helpful) to 9 (very helpful) at the end of the trial.

Qualitative Manipulation Checks

Transcripts of participant qualitative responses, which were collected for each trial, were coded for strategy adherence by a team of 3 undergraduate research assistants supervised by the first author. For the cued and uncued condition, responses were coded as valid when participants successfully associated their resilience narrative with a potential positive outcome of the negative scenario using either any life memory of resilience or their prompted cued memory. Non-autobiographical memory reappraisal responses were valid if they referenced plausible silver-linings or positive reframing of the stimulus meaning. Interrater reliability showed high concordance (90%) across trials. Consensus was made to break ties on discrepant trials via coding by the first author.

Self-report questionnaires

Affective and Functional Health.

A functional and emotional health measure (Patient-Reported Outcomes Measurement Information System 29, PROMIS-29- Version 2; Cella et al., 2010) was gathered to characterize the sample and identify age differences in factors of socio-emotional well-being that could influence reappraisal efficacy. Seven subscales of the PROMIS-29 capture domains of mood and daily function including anxiety, depression, fatigue, pain, sleep disturbance, physical functioning, and social role. Each domain subscore was summed across 4 items on a five-point likert scale, except for pain intensity which was assessed on one 0–10 rating scale. The total sum of item responses across all domains was converted to t-scores and compared to national norms (M = 50, SD = 10; Rothrock, Amtmann, & Cook, 2020).

Baseline and Post-trial Mood.

Baseline mood was assessed to determine age group differences that may have confounded observed age differences. Mood was also examined post-task, to determine if mood varied as a function of task completion, participant age, and experimental condition. Participants reported positive and negative affect via the 20-item Positive and Negative Affect Schedule (PANAS; Watson, Clark, & Tellegen, 1988) before and after completing the emotion regulation task. Participants were instructed to rate the extent to which they felt positive (Interested, Excited, Strong, Enthusiastic, Proud, Alert, Inspired, Determined, Attentive, Active) and negative (Distressed, Upset, Guilty, Scared, Hostile, Irritable, Ashamed, Nervous, Jittery, Afraid) emotions in the moment on a scale from 1 (very slightly or not at all) to 5 (very much).

Results

Participant Characteristics and Variability in Functional Health

A series of mixed ANOVAs were conducted in SPSS (Version 25.0, IBM SPSS Statistics, Switzerland) to determine the fixed effects of age (younger, older) and reappraisal condition (non-autobiographical memory, uncued autobiographical memory, and cued autobiographical memory) on strategy efficacy (Δ intensity) and difficulty to reappraise. Tukey HSD -adjusted post-hoc t-tests clarified interaction effects. Older adults (M = 6.78, SD = 1.12) rated images as more emotionally intense than younger adults (M = 6.05, SD = 1.37), F(1,411) = 32.52, p<0.001, η_p²= 0.073. Thus, all reported models included baseline image intensity, in addition to education, as covariates.

Non-Autobiographical Memory Reappraisals are Most Effective for Older Adults, but All Strategies are Equally Effective for Younger Adults

Compared to initial ratings of image emotional intensity, reappraisal successfully reduced image intensity for younger and older adults when using the cued, t(117) = 6.31, p <0.001, uncued t(152) = 6.27, p <0.001, and non-autobiographical reappraisal strategy, t(155) = 11.20, p <0.001 (See Appendix B). Results also revealed a significant main effect of condition, F(2,410) = 9.78, p<0.001, η_p²= 0.06, in which non-autobiographical reappraisal was more effective at reducing intensity, relative to the uncued, t(301.85) = 5.08, p<0.001 and cued strategy, t(272) = 3.24, p = 0.001 among both age groups. As can be seen in figure 2, a significant interaction of Age x Condition F(2,410) = 6.03, p = 0.003, η_p²= 0.03, revealed that older adults drove the stepwise effect, such that the non-autobiographical strategy was most effective (M_{Δ intensity} = 1.16, SD = 1.12) at reducing image intensity, followed by the cued strategy (M_{Δ intensity} = 0.57, SD = 0.88, t(154.02) = 5.33, p <0.001). The uncued strategy was least effective (M_{Δ intensity} = 0.31, SD = 0.89, t(119.64) = 3.39, p = 0.001) at mitigating image intensity. In contrast, strategy efficacy did not differ significantly across conditions for younger adults (p = 0.47).

Figure 2.

Change in Intensity Modeled Across Age and Reappraisal Strategy.

Note: Error bars represent standard error. Overall effects revealed an age by condition interaction that was driven by older adults. While younger adults demonstrated no difference of benefit across emotion regulation strategies, non-AM reappraisal was most effective for older adults. * p< 0.01

Reappraisal is More Challenging for Younger Adults and Autobiographical Memory Reappraisals Are Globally More Challenging to Use

Younger adult (M = 4.87, SD = 1.85) ratings denoted that reappraisals were more difficult to use compared to older adults (M = 4.26, p = 1.82) across all conditions, F(1, 410) = 8.49, p = 0.004, η_p²= 0.02, as shown in figure 3. Cued reappraisal use was rated as the most difficult strategy to employ by both groups M_Difficulty = 5.58, SD = 1.86; F(2,410) = 29.23, p < 0.001, η_p²= 0.12, while the non-autobiographical reappraisal was rated least difficult (M_Difficulty = 3.85, SD = 1.61), followed by the uncued reappraisal strategy (M_Difficulty = 4.53, SD = 1.80). An interaction effect of age and condition was not observed (p = 0.75)

Figure 3.

Average Difficulty to Reappraise Modeled Across Age and Reappraisal Strategy.

Note. Error bars represent standard error. A main effect of age revealed that younger adults found all reappraisals more challenging to employ compared to older adults. Across age, a main effect of condition demonstrated a step-wise effect such that Non-AM reappraisals were easiest to use, followed by uncued AM reappraisals, and cued AM reappraisals, respectively.

* p<0.01.

Autobiographical Memory Reappraisals are More Helpful to Older Adults

A main effect of age was detected, F(1, 257) = 6.62, p = 0.011, η_p² = 0.03, in which older adults found reappraisal strategies (cued and uncued) more helpful (M_Helpfulness = 6.23, SE = 0.1) than younger adults (M_Helpfulness = 5.48, SE = 0.12), t(268.98) = 4.49, p < 0.001. A main effect of condition was also observed F(1, 257) = 21.57, p < 0.001, η_p² = 0.08, in which uncued reappraisal was rated as more helpful (M_Helpfulness = 6.20, SE = 0.12) than cued memory reappraisal (M_Helpfulness = 5.50, SE = 0.14), t(220.37) = 4.03, p < 0.001. An interaction effect of age and condition on perceived helpfulness of strategy was not detected (p = 0.73).

Uncued Autobiographical Memories Used During Reappraisal are More Similar to Novel Stressors than Cued Memories

Older adults (M_Similar = 5.16, SD = 1.68) rated their memories as more similar to negative images reappraised across both autobiographical memory strategies (M_Similar = 4.25, SD = 1.92), revealing a main effect of age F(1,257) = 4.32, p = 0.038, η_p² = 0.02. A main effect of condition was also observed, F(1,257) = 120.25, p < 0.001, η_p² = 0.32. As expected, uncued memories chosen and recalled individually by the participant on each trial were rated as more similar to reappraised images (M_Similar = 5.55, SE = 0.12), than cued memories which were assigned randomly on each reappraisal trial (M_Similar = 3.60, SE = 0.13).

Additional independent samples t-tests examining the effect of age on memory and image similarity were conducted within each autobiographical memory condition. As can be seen in Figure 4, while the cued condition revealed no difference in similarity ratings across age groups (p = 0.126), the effect of similarity rating in the uncued condition significantly varied by age, t(151) = 3.97, p <0.001, as older adults rated their memories more similar to reappraised images (M_Similar = 5.97, SD = 1.23) than younger adults (M_Similar = 5.05, SD = 1.63)

Figure 4.

Average Similarity Rating Across Age and Condition.

Note. Error bars represent standard error. A significant effect of age revealed that older adults found greater shared similarity between their memories and stimuli compared to younger adults. Uncued memories were significantly more similar to the stimuli than cued memories, based on self-report.

** = p< 0.001, * = p<0.05

The Relationship between Autobiographical Memory and Image Similarity is Associated with Reduced Reappraisal Efficacy, but also Difficulty

Linear regressions examined the relationship between secondary variables of interest (similarity between autobiographical memory and stimulus) and dependent outcome variables. Surprisingly, within autobiographical memory conditions, similarity between the memory and stimulus was negatively associated with reappraisal efficacy, such that greater similarity between the negative stimulus and memory used during reappraisal was associated with poorer efficacy, regardless of cuing parameters (r =− 0.120; β = −0.055, p = 0.049, r² = 0.014).

In contrast, greater reported similarity (r = − 0.44, p < 0.001) and helpfulness of memory to reappraise (r = − 0.34, p < 0.001) were both associated with lower self-reported reappraisal difficulty. Memory/stimulus similarity and perceived helpfulness of memory to reappraise were highly collinear, r = 0.620, p <0.001, thus similarity and helpfulness were modeled independently as predictors in follow-up linear regressions. Both similarity, β= −0.394, p < 0.001, r² = 0.201, and helpfulness β_s= −0.412, p < 0.001, r² = 0.115, predicted the difficulty with which participants used autobiographical memory strategies.

Baseline Affect and Change in Affect Over Time.

Two univariate ANOVAs examined baseline positive affect and negative affect scores, respectively, with age and condition as fixed factors, covarying for education. Neither main effects of age, condition, nor interaction of age and condition were significant for positive and negative baseline mood scores (both p’s > 0.10) suggesting age-equivalent baseline mood. We also examined if participants’ positive and negative affect was affected by the experiment, by conducting age by condition by time (baseline to post-task) mixed ANOVAS for positive and negative affect. There were no main effects of age, condition, time, or interaction of these factors for positive affect. However, an interaction of time by age was observed for negative affect, F(1, 410) = 6.62, p =0.010, η_p² = 0.02. While older (M = 13.02, SE =0.32) and younger adults (M = 13.32, SE =0.31) reported the same degree of negative affect at baseline, t (416) = −0.99, p = 0.32, and both age groups reported greater negative affect after the task, the magnitude of post-task negative affect was more pronounced among younger (ΔNegativeAffect = 2.07), relative to older adults across all conditions (ΔNegativeAffect = 1.05, t(405.55) = −2.46, p = 0.014). Including baseline negative affect, negative affect after task completion, or change in negative affect as covariates did not alter primary findings.

Discussion

This study aimed to investigate the degree to which autobiographical memories of past resilience facilitate momentary emotion regulation in response to novel stress, which could suggest novel implications for mnemonic reappraisal. All reappraisal strategies, including reappraisals utilizing autobiographical resilience memories, reduced the intensity of negative emotion for younger and older adults. Thus, reappraisals associated with past experiences of resilience, or those of perseverance during pivotal life events, may work equally well for older and younger adults to regulate negative emotions as they emerge.

Older but Not Younger Adults Capitalize on Mnemonic Reappraisal

Older adults reported that autobiographical reappraisals were more helpful and less difficult to use than did their younger peers. This effect persisted across conditions, despite objective efficacy of autobiographical reappraisal strategies reducing negative affect ratings less than the control reappraisal strategy. While efficacy of reappraisals that reference autobiographical memory may be age-invariant, features of engaging in this novel strategy, such as perceived helpfulness and difficulty of use, may preferentially benefit older adult emotion regulation.

Our finding that older adults found mnemonic recall easier and more helpful than younger adults is consistent with evidence that autobiographical recall – while not more common in older than younger adults – maybe more automatic and habitual for older adults. Indeed, younger adults report more frequent use of autobiographical memory recall to direct momentary behavior than older adults (Bluck & Alea, 2009; Vranić, Jelić, & Tonković, 2018). This may be because young adults face several developmentally appropriate, yet challenging decisions for the first time with limited prior experience or “evidence” that they may successfully overcome their life event. Young adults, in turn, may resort more frequently to their past experiences for guidance in the moment (Bluck and Alea, 2009). While this is a new developmental skill in young adulthood, it is possible that older adults engage in autobiographical memory recall more habitually in their day to day life to support emotional well-being, because they have a rich lifetime of evidence overcoming hardship. Older adults are thought to have more accumulated life experiences that are stored as a central set of resilience memories (Bluck and Alea, 2008). Advancing age is also associated with integrating distinct life challenges into cohesive narratives of resilience that promote emotion regulation benefits (Martins-Klein et al., 2022). With time, these practiced resilience narratives may be easily accessed by older adults as new negative events unfold. Our findings further corroborate this notion, as older adults found more similarity than younger adults between their own life experiences and the scenes depicted in the study.

Taken together, these findings may explain why older and younger adults demonstrated objective age-equivalent benefits when utilizing past autobiographical memories of resilience during reappraisal, but older adults found the strategy subjectively easier. That is, older adults’ positive attentional focus (Shiota & Levenson, 2009; Lohani & Isaacowitz, 2014) and proclivity to identify and integrate positive lessons learned from critical life experiences with time (Carstensen et al., 2003; Pasupathi & Mansour, 2006; Singer et al., 2007) may confer subjective benefits to mnemonic reappraisal that are not yet developmentally available to younger adults.

Implications for our observed age difference in reappraisal benefits may be further explained by the Selection Optimization and Compensation in Emotion Regulation model (SOC-ER; Urry & Gross, 2010), which proposes that people match the selected emotion regulation strategy to the situational demands in contexts of cognitive resources and physiological flexibility (Opitz et al., 2012; Tucker, Feuerstein, Mende-Siedlecki, Ochsner, & Stern, 2012). In the current study, when presented with the option to capitalize on age-related positive attentional focus, older adults benefited most from positive, non-autobiographical reappraisals. However, when taught to utilize novel strategies that build on unique late-life strengths, such as reminiscence and meaning-making, older adults endorsed greater subjective benefits from their reappraisals than younger people. It is important to acknowledge that older adults’ ability to manage their emotions well does not equate to regulatory success nor does it clarify the motivation to regulate one’s emotions. Our observed discrepancy of objective outcomes and subjective ratings is in line with findings from past studies in which self-reported measures differ from emotional outcomes, highlighting individuals’ difficulty in accurately estimating not only their performance on emotion regulation tasks, but also their affective states in relation to performance (Shiota & Levenson, 2009; Todd, Tennen, Carney, Armeli, & Affleck 2004; Reisenzein, Bordgen, Holtbern, & Matz, 2006).

In regard to motivational factors, eudaimonic goals, which are prioritized by older adults (Ong & Patterson, 2016), may utilize life experience and self-efficacy to guide momentary problem-solving, thus taking precedent over immediate hedonic benefits. Our findings may, therefore, be attributed to variability in motivations that drive emotion regulation success. Efficacy is one dimension of “successful” emotion regulation. Other features of regulation such as expectations of successful regulation (Bigman, Mauss, Gross, & Tamir, 2016), goals to modify affect (Tamir, 2016; Tamir, Bigman, Rhodes, Salerno, & Schreier, 2015; Tamir & Millgram, 2017), and beliefs about one’s emotion regulation skills (Ford & Troy, 2019) influence strategy outcomes. For instance, higher quality reappraisals, as measured by richer detail, unique perspective, and reappraisal plausibility, predict affective improvement from negative mood induction (Southward, Sauer-Zavala, & Cheavens, 2021). That is, one’s sense of emotion regulation self-efficacy and motivation to regulate predicts regulatory success, even in contexts where objective or quantitative measures, such as change in emotional intensity, do not. For older adults who regulate more often (Lawton, 2011; John & Gross, 2006) and prioritize self-efficacy, meaning, and growth in negatives contexts (Pasupathi & Mansour, 2006; Singer et al., 2007), regulatory motivations may capture a more holistic picture of emotion regulation success. Future studies may test motivational factors more directly, by integrating measures of confidence and reappraisal quality to subjective ratings (i.e. “How confident are you that you reappraised correctly?” or “Rate the quality of your reappraisal”).

As predicted, younger adults showed no significant differences in mood improvement across autobiographical reappraisals and the control strategy, though autobiographical reappraisals were rated as even more challenging to utilize than for older adults. Thus, while recalling life memories of resilience and accomplishment may aid the process of reappraising negative situations for younger adults, autobiographical regulatory strategies appear to come at a significant motivational and cognitive cost. Recent research demonstrates that recalling autobiographical episodes of self-efficacy – events where a difficult situation was managed successfully – enhanced reappraisal of past negative emotional memories above and beyond recalling unrelated positive memories prior to reappraisal (Paersch et al., 2021). Our findings extend this work by suggesting that deliberately associating lessons learned from one’s past directly to momentary reappraisals of novel stressors also reduced emotion intensity. However, less demanding non-mnemonic reappraisals may produce similar effects with less cognitive costs to younger adults.

Positive reappraisal directs focus to emotional aspects of a scenario while reinterpreting in a way that makes a scenario feel more positive. As a result, positive reappraisal may rely on executive functions that support emotion regulation to a lesser degree (Shiota & Levenson, 2009). Instead, positive reinterpretation may rely more on life experience, including evidence that difficult life experiences may lead to important gains of self-efficacy and resilience – an area of socioemotional cognition that is still developing in young adulthood. Younger adults are less adept at allocating a greater proportion of their attention to positive information, compared to older adults (Shiota & Levenson, 2009; Lohani & Isaacowitz, 2014; Isaacowitz, Toner, Goren, & Wilson, 2008; Murphy & Isaacowitz, 2008). Rather, young adults excel at utilizing reappraisal strategies that instruct detachment or objective/unemotional approaches of rethinking and show attentional preference for negative over positive emotional information (Mikels, Larkin, Reuter-Lorenz, & Carstensen, 2005). Taken together, younger adults may demonstrate less variability in affective benefits from utilizing positively focused and mnemonic reappraisals that utilize life experience to serve momentary coping due to different developmentally appropriate socioemotional goals.

Finally, consistent with our age-specific predictions, positive, non-autobiographical reappraisals most effectively reduced emotional intensity for older but not younger adults. This age difference replicates past studies, which found that efficacy of reappraisals that highlight positivity without autobiographical context was greater among older adults, compared to their younger counterparts (Lohani & Isaacowitz, 2014; Shiota & Levenson, 2009). Of note, these studies instructed participants to increase positivity of stimulus meaning, in contrast to reappraisal that decreased negativity in emotion eliciting stimuli, mirroring the approach used in the present study. This key distinction in instructions supports late-life attentional deployment to more positive than negative information (Mather & Carstensen, 2003, 2005; Carstensen & Mikels, 2005), thus boosting reappraisal effects for older adults. Perhaps unlike cognitive reappraisals that instruct to decrease negative meaning, increasing positive meaning in cognitive reappraisal may be more effective for older adults to implement.

Memory and Negative-Stimulus Similarity is Linked to Lower Reappraisal Hedonic Benefits

Contrary to a priori predictions, greater similarity between autobiographical memories and negative images did not mitigate emotional intensity. Instead, greater similarity between memory and image predicted lower image emotional intensity change scores for participants across age groups. One possibility is that the impact of recalling emotionally salient memories of challenging life experiences induced negative emotions, which impacted hedonic benefits of strategy use. It is unclear if participants may have induced negative mood by thinking of resilience memories via rumination, the repetitive, cyclical negative thinking over one’s experiences or feelings resulting in depressed mood (McLaughlin & Nolen-Hoeksema, 2011; Nolen-Hoeksema, Wisco, Lyubomirsky, 2008; Watkins & Roberts, 2020). Younger adults are especially prone to rumination compared to older adults as they reminisce over past experiences to guide future decisions (Ricarte Rose, Serrano, Martinez-Lorca, & Latorre, 2016; Ricarte, Ros, Latorre, & Barry, 2020), and more readily attune to negative emotional information for adaptative gains (Mikels et al., 2005). Rumination may have driven the age differences in decreased perceived helpfulness and greater difficulty reported by younger adults that we found during autobiographical reappraisal use, compared to older adult reports. Our finding that participants, particularly younger adults, reported more negative affect after the regulation task is consistent with such a possibility, although including these scores of negative affect in the models did not influence the pattern of results. Future research should further examine this possibility by supplementing pre and post task PANAS measures with a rumination scale, such as the Ruminative Response Scale (RSS; Treynor, Gonzalez and Nolen-Hoeksema, 2003), which may elucidate the degree to which younger adults’ worsening affect during autobiographical memory reminiscence may contribute to ruminative thinking.

Limitations and Future Directions

The current findings clarify the impact of autobiographical resilience memory recall on reappraisal use across age groups, suggesting an increased perceived helpfulness and efficacy of strategy use and decreased perceived difficulty of strategy use for older, but not younger adults. While impactful, these findings must be considered within the context of important key limitations to the current study that warrant future investigation. First, our study stimuli may have featured scenarios that some of our participants never encountered, thus making them more challenging to directly associate with cued and uncued life events. For instance, one image depicts a man reacting emotionally to his farmland flooding in a rural location; this scenario may mirror the experience of helplessness due to natural disaster, yet geographic, occupational, and other life factors may limit the likelihood that our participants experienced a similar event in their lifetime. Utilizing participant ratings of stimulus salience may address this limitation in the future.

Due to concerns of cross-over effects and strategy perseveration difficulties for older adults, strategy conditions were manipulated between-subjects. Thus, we did not test for emotion regulation outcomes across strategies within-person. In addition, perceived helpfulness was only measured for the autobiographical memory reappraisal conditions in order to reduce task completion time, and we failed to directly inquire about emotion regulation self-efficacy beliefs. This limitation prevented us from directly comparing the subjective helpfulness of the autobiographical memory reappraisal conditions to control reappraisals. Clarifying this effect may elucidate whether immediate hedonic gains, such as reduced emotionally intensity, are sacrificed in exchange for eudaimonic benefits such as self-efficacy, environmental mastery, and wisdom gained. Future studies may integrate a within-subject manipulation including both autobiographical memory and control reappraisal trials and track perceived-efficacy of emotion regulation ratings across all conditions to allow for direct comparison of strategies.

Cognitive resources were also not assessed because study piloting revealed elevated drop-out rates when participants were redirected to external cognitive testing platforms. Prior work suggests a moderating effect of fluid intelligence on reappraisal success among both young and older adults, as tracked by cognitive measures from the Wechsler Adult Intelligence Scale (e.g. Block Design, Coding, and Digit Span subtests; Opitz et al., 2014). Tracking these cognitive measures, along with other measures of working memory, may help explain the differential impact of cognitive control on mnemonic reappraisals. Although objective benefits of strategy use, such as efficacy, may be matched across autobiographical memory and control conditions, individuals with low cognitive reserve may have capacity to benefit equally from utilizing autobiographical reappraisals. If so, integrating autobiographical resilience memories into reappraisals may hold key clinical importance for not only healthy older adults, but also individuals with subjective or early stages of cognitive decline.

Another important limitation of the study is that outcomes were based on self-administered training and self-reported measures of emotional intensity and difficulty, which are subject to demand characteristics (McRae, Ciesielski, & Gross, 2012). Indeed, older adults can be influenced more by demand characteristics than younger adults (Allard & Kensinger, 2014; Martins et al., 2018; Tucker et al., 2012), so the degree to which the accuracy of these reports may have influenced our results is uncertain. Ratings of image intensity do not capture emotion regulation strategy efficacy alone, which may involve more than momentary hedonic improvement (Southward et al., 2021; Tamir, 2009, 2016). Additional objective and subjective measures of efficacy and emotion regulation motivations could complement image ratings. For instance, amygdala activation, amygdala-ventromedial prefrontal cortex functional connectivity or heart rate variability model physiological metrics that capture “successful” reappraisal (Ochsner et al., 2004). Similarly, pupil size variation is an involuntary metric of autonomic nervous system activity that provides an unbiased index of cognitive effort and arousal (Granholm & Steinhauer, 2004). Thus, task-evoked changes in pupil diameter may also serve as a robust psychophysiological index of both cognitive (Ayres, Lee, Pas & Merrienboer, 2021) and emotional load (Bradley, Miccoli, Escrig, & Lang 2008; Granholm & Steinhauer, 2004).

One of the biggest limitations of the study was that participants were recruited from Amazon Mechanical Turk during the COVID-19 pandemic, and thus the study was carried out online. Online data collection limits verbal feedback and observation of outcome-modifying behaviors such as cellphone use, distraction, or multi-tasking. Training time was also minimized in order to standardize procedures and ensure that participants would complete and be reimbursed for the time commitment stipulated in the consent form. The degree to which a lab-instructed format could influence outcomes remains unclear. Moreover, the COVID-19 pandemic was a major threshold event with potential socioemotional implications for social isolation, anxiety, fear of contagion, chronic stress, and uncertainty across the lifespan (Sher, 2020; Birditt, Turkelson, Fingermand, Polenick, & Oya, 2020). Despite these considerations, both our young and older adult samples did not demonstrate clinically sensitive psychopathology, as measured by functional and mood measures. Both quantitative and qualitative measures of instructional adherence also suggested competency in task expectations (e.g. >90% accuracy on self-defined task cues for “Prepare” and/or “Reappraise”). None the less, increasing structured time around task training and reproducing study methods in an in-person format may reduce instructional ambiguity, subsequent errors in task completion, and detect subtle differences in outcomes that we were not found in an online sample.

Finally, we did not explicitly track participant emotion regulation goals, and the degree to which the desire to improve affect, instrumental goals to do well on the task, and/or other regulatory motives may be driving study outcomes. Motivation to change specific emotions may influence not only the regulatory strategy used, but also the success of the strategy to modify emotional states (Tamir, Vishkin & Gutentag, 2020; Mauss & Tamir, 2014; Tamir, 2016). For instance, individuals who believe in their capacity to regulate negative emotions and expect emotion regulation success are indeed more successful in regulating their emotional responses relative to individuals who hold no expectations for their regulatory capacity (Bigman et al., 2015). Motivational factors in emotion regulation may also compound benefits of recalling resilience memories, as narrating positive life experiences and those of self-efficacy promotes mood repair and wellbeing (Pascuzzi & Smorti, 2017; Speer & Delgado, 2017; Hallford, Farrell, & Lynch, 2022). Future studies of mnemonic reappraisal should clarify underlying motivations and evaluate the degree to which motives influence benefits of autobiographical resilience memory recall in reappraisal strategies.

Conclusion

Although prior research suggests a mechanistic role of autobiographical memory in regulating emotions towards past challenging experiences, our study is the first to assess how autobiographical memory recall impacts momentary reappraisal across age by associating the meaning of one’s past resilience to novel stress. Compared to younger adults, older adults found autobiographical memory reappraisals to be less difficult, and memories to be more overlapping with negative images. Autobiographical memory approaches are costly to employ, but our findings provide initial evidence that resilience memories are perceived as helpful and effective to the regulation of momentary negative emotion by older individuals. These findings provide a foundational step for exploring the role of resilience narratives in intervention in older adult populations. Future research may provide more context and clarification to our initial findings in this novel paradigm.

Biography

Irina Orlovsky, M.A., M.S. is a Ph.D. student in the Department of Brain and Psychological Sciences at UMass Amherst. Rebecca E. Ready, Ph.D is a professor of psychology and director of the Aging, Cognition, and Emotion Laboratory at UMass Amherst. Angela Gutchess, Ph.D is a professor of psychology and director of Aging, Culture, and Cognition at Brandeis University. Kristin M. Heideman B.S. is a postgraduate associate at the Yale Child Study Center and former lab manager at UMass Amherst. Bruna Martins-Klein, Ph.D is assistant professor of psychology and director of the Neural Vitality Laboratory at University of Southern California, previously assistant professor of psychology at UMass Amherst.

Appendix A. Task Instructions by Condition

Participants were instructed accordingly, based on the condition into which they were randomized:

[Control] Non- Autobiographical Memory:	“When you rethink, focus on what you see and try to reinterpret the image in a way that makes it feel more positive. Consider how the situation will be okay in the end, or rethink by considering that the situation is only temporary. Remember, as you rethink, find a silver-lining in the image, so that the image feels more positive.”
Uncued Autobiographical Memory:	“As you rethink, find the silver-lining in the image by relating it to any meaningful memory that relates to the image. Consider how the situation will be okay in the end, similar to your own meaning memory. Rethink by relating the image to what your meaningful memory taught you, how it turned out better than you expected, or that you preserved through a challenging event, in the end. You will be asked to use one of your meaningful memories to rethink the image. Consider how that memory could apply to the situation in the image, so the image feels more positive.
Cued Autobiographical Memory:	“As you rethink, find the silver-lining in the image by relating it to the meaningful memory you remembered today. Consider how the situation will be okay in the end, similar to your own meaning memory. Rethink by relating the image to what your meaningful memory taught you, how it turned out better than you expected, or that you preserved through a challenging event, in the end. You will be asked to use one of your meaningful memories to rethink the image. Consider how that memory could apply to the situation in the image, so the image feels more positive.

Note. Participants in the Cued AM condition were cued to recall the same memory for each block of three image trials, with memory order randomized across blocks. That is, the same memory was to be utilized for trials 1–3 of each block, randomizing the generated memory across blocks.

Appendix B. Mean Rating of Image Intensity Before and After Regulating, Across Age and Emotion Regulation Strategy

Note. Error bars represent standard error. Both older and younger adults benefited from all 3 conditions, in reducing image intensity, however, pre and post intensity ratings did not differ across conditions.

* p< 0.001.