Facing Challenging Experiences in Life – Narrative Identity Development Processes and Associations with Wellbeing During the Transition to Midlife

ABSTRACT This study investigates identity development in narratives of difficult experiences and associations with wellbeing in the transition to midlife. The narrative processes of emotional sequencing (i.e. emotional tone of how people frame their experiences: redemptive, neutral/vague, negative, and a combination of positive and negative) and positive and negative impact on the identity (i.e. how people perceive the impact of the narrated event) were examined in narratives about difficult experiences in early adulthood (Mage = 33.28, SD=0.54) and midlife (Mage = 38.62, SD =0.61). Results concerning emotional sequencing showed two typical patterns one of stability in neutral/vague sequencing, and another pattern of change from a combination of positive and negative to redemptive sequencing. Furthermore, positive and negative impact in early adulthood was associated with wellbeing in midlife. Findings in this study show that people who are neutral in their way of framing challenges maintain this outlook on life’s difficulties in the transition to midlife. The results also indicate that perceived negative impact of narrated events, and not only positive impact, in early adulthood has bearing on wellbeing during midlife.


Narrative identity and difficult experiences
Narrative identity is the ongoing and internalized story of oneself (McAdams & McLean, 2013).It involves stories of experiences in life that situate the individual in the world and provide purpose and meaning, connecting the individual's past, present, and future.Investigations of people's narratives therefore allow a glimpse into their identity, shedding light on how they came to their present identity and how they view the impact of significant experiences on it.
How people narrate difficult and challenging experiences has been found to be associated with many different positive aspects.A well-known way to narrate difficult experiences is to tell these types of stories using redemptive story sequencing (McAdams et al., 2001).Redemptive story sequencing is a way of structuring a story that involves narratives in which the protagonist starts off in a good state of life, followed by something bad happening, which in the end leads to something good (or even better than before); that is, narrating challenges with a positive framing of the negative.This could involve, for example, gaining insight from the experience or recovering from illness (McAdams, 2006).Research has found that narrative processes involving redemptive sequencing can predict sobriety (Dunlop & Tracy, 2013) and are associated with positive socio-emotional personality traits (Guo et al., 2016), and that people who display more redemptive sequences display more prosocial behavior (McAdams & Guo, 2015;McAdams et al., 1997).Recently, it has also been suggested that there is a need to account for more than just the absence or presence of this emotional sequencing, and that there may be important qualitative differences in how people tell redemptive narratives (Booker & Perlin, 2022;Perlin & Fivush, 2021; for a discussion see Dunlop, 2021).Perlin and Fivush (2021) demonstrated two forms of redemption: The narrator either returns to their baseline after the challenging experience or emerges from their experience having gained something (e.g., new resources, perspective, self-understanding).Booker and Perlin (2022) also found that the emergent redemption form was connected to better wellbeing.This research suggests that a more nuanced view on redemption may facilitate further understanding of this form of structure in narratives and its relation to wellbeing.In line with this, research also points toward the value of examining other types of sequences besides redemption (Blackie et al., 2020;Eriksson et al., 2020).
The importance of redemption in narratives of challenges needs to be viewed within the context of culture and its meaning in the US, where this narrative process was first described (for a review see Dunlop, 2022), as culture has a great impact on how people tell their stories and how they interpret experiences in life (Fivush et al., 2011;Hammack, 2008;McAdams & Pals, 2006).Myths, metaphors, and different reoccurring plots within the cultural context can all help shape how an individual narrates stories as part of their own identity (Fivush et al., 2011;McAdams, 2006).Therefore, as culture has a bearing on the identity in many ways, redemptive sequencing may not hold the same value in cultural contexts outside the US.Indeed, research in a European context has found that other ways to narrate difficult experiences, besides redemption, are meaningful for understanding how people frame difficult events.In the context of the UK, Blackie et al. (2020) found that redemptive sequencing was not common in narratives of trauma survivors from a third-person perspective; instead, recuperation, which involved a lessening of symptoms over time and coping with emotions, was more common.In another European study of narratives, Turner (2022) found in their Danish study growth similar to redemption but with the difference that the growth involved communal growth, concerning help and helping others.Narratives also demonstrated balanced affect with a reasoning of both positive and negative outcomes from the difficulty as well as a neutrality toward difficulties as nothing unique for the individual.Research from Sweden, and within the same longitudinal study as the present study, found that four types of emotional sequencing -neutral/vague, redemptive, negative, and a combination of positive and negative -were equally common in narratives of difficult experiences (Eriksson et al., 2020).In relation to the cultural context, it is interesting to note that both the Scandinavian studies, Swedish and Danish, showed similarities in that narratives highlighted both positive and negative outcomes of difficult experiences (i.e., a combination of positive and negative sequencing and balanced affect).Taken together, these studies from outside the US show that redemption can be found in other cultural contexts but that it is not a dominant theme in other cultures, which highlights the importance of acknowledging other possible narrative processes, such as emotional sequencing, that may be meaningful in other cultural settings.However, research on emotional sequencing from a cultural perspective is scarce.

Narrative identity development across time
Narrative processes can shed light on how people understand and frame their personal experiences, and it is therefore important to examine them across time to capture identity stability and development.Examining several narrative processes over a three-year period among emerging adults (aged 18-25 years), McAdams et al. (2006) found the highest stability for the narrative process of overall positivity of emotional tone and conceptual complexity.A more recent study found a moderate degree of rank-order stability among emerging adults' narratives for narrative exploratory processing and self-event connections (McLean et al., 2022).Within the context of romance, this study also found development in the form of decreased positive connections within self-event connections, as well as decreased negative connections.Another study that examined mean-level change and rank-order stability in narratives from emerging adulthood to early adulthood showed stability for communion and coherent positive resolution, but not for agency or emotional tone (Sengsavang et al., 2018).This study also showed positive development in levels of communion and emotional tone, indicating development in narratives across this time period.Recent research has also noted that repeated narration -that is, whether or not the same event is repeated -may also have implications for development across time, and showed that growth declines over time in repeated narratives about low points (Booker & Perlin, 2022).
In research on the narrative process of emotional sequencing (i.e., the structure of how people frame their experiences), there has been little focus on identity stability and development.As mentioned earlier, redemption is one of the common sequences in the US.Another sequence that has been found in this context is contamination (McAdams et al., 2001).Contamination refers to the narrative sequence in which a story starts out in a positive scene but ends negatively.Research on redemption and contamination has found zero to moderate degrees of rank-order stability among early adults (r = 0.50 for redemption and r = 0.28 for contamination) and seniors/midlife adults (r = 0.31 for redemption and r = 0.07 for contamination) (Dunlop et al., 2016).
The previous research on stability and change within narrative processes presented above shows variation in rank-order stability for emerging adults and midlife adults.However, little is known about changes in narrative processes between early adulthood and early midlife.To our knowledge there is no research on the development of emotional sequencing apart from redemption and contamination, and if there is development, on how people view and narrate their experiences in life.

Narrative identity and its connections to wellbeing
Difficult experiences can represent critical points for identity development in adulthood and depending on how they are integrated into the identity they may also have an impact on wellbeing (Mitchell et al., 2021).Identity integration, bringing together and integrating different parts of the identity into a wholeness, represents an important aspect of the identity development in adulthood.Mitchell et al. (2021) highlight the importance of integration concerning negative experiences, unexpected and major life transitions in adulthood as well as different cultural identities.They also stress the importance of identity integration (or rather lack thereof) within clinical diagnosis such as borderline personality disorder and trauma.In the present study we focus on difficult experiences and the ongoing process of how people handle and integrate these as part of the identity development from early adulthood to midlife.
The way people construct narratives about their experiences in life has been found to be related to psychological wellbeing and positive development, and can also predict factors associated with wellbeing, as well as wellbeing over time (for a review see Adler et al., 2016).For example, research has found that more elaborative narratives of the experience of divorce were associated with increased ego development over the next two years (King & Raspin, 2004).Furthermore, research has demonstrated that positive resolutions of difficult experiences at age 52 enhanced resiliency and were connected to higher life satisfaction ten years later (Pals, 2006).Research with emerging adults has also found that affective processing (i.e., a combination of a high degree of positive valence, low negative valence, and contamination) and exploratory processing (i.e., a combination of redemption sequencing, narrative coherence, and causal change) were connected to increased emotional health across the college years (Lodi-Smith et al., 2009).Another narrative process that has been linked to wellbeing in narratives of difficult experiences is positive growth (Pals, 2006), also referred to as positive and negative impact on the identity (Eriksson et al., 2020).Indeed, positive impact has been showed to be associated with higher wellbeing and negative impact with more psychological distress (Eriksson et al., 2020).As most of these studies show, valence of the narrative process may be important for connections to wellbeing (Lilgendahl, 2015).
There are also studies that show that the importance of valence may vary depending on what type of experience the narrative involves.In their study with emerging adults, Lilgendahl and McLean (2020) found that a higher degree of self-change connections in positive stories about romantic relationship was associated with increased life satisfaction across time.They also found that higher levels of depression and anxiety predicted more negative self-event connections in areas of academic and romantic low points.Furthermore, examining narratives that describe experiences of high and low points in a sample of midlife adults (from 55 to 58 years), researchers found that meaning making with a positive valence was associated with emotion regulation two years later (Cox & McAdams, 2014).The same study found that meaning making with a negative valence was associated with negative emotion regulation, but only for narratives of low points.This research demonstrates that identity development can serve as a solid foundation for future wellbeing, and that valence and type of experience matter.However, overall there is little research on how narrative processes in early adulthood, such as emotional sequencing and impact on identity, are associated with wellbeing later in adulthood.

The present study
To shed light on how people handle and integrate negative experiences in life and how this develops from early adulthood to midlife we examine people's narratives and connections to wellbeing.This part of life is important for identity and wellbeing as the transition to midlife requires an ongoing process of reflexivity and possibly even identity revision when facing challenging experiences.In this time of life challenges can involve health issues, divorce, and loss of family members which may destabilize roles and assumptions that previously served as identity foundations (Lilgendahl, 2015).How people integrate these experiences may thus have bearing on both aspects of the identity and wellbeing.The purpose of the present study was therefore to examine identity development in narratives of difficult experiences and their associations with wellbeing in the transition to midlife.To do this, we examine how adults in the context of Sweden narrate difficult experiences told five years apart, first in early adulthood (age 33) and then in midlife (age 38), through three narrative processes: emotional sequencing, positive and negative impact on the identity.The specific research questions are: 1) What are the typical and atypical patterns of development of emotional sequencing between ages 33 and 38? 2) How are narrative processes of positive and negative impact on the identity at age 33 associated with narrative processes at age 38? 3) How are narrative processes of positive and negative impact on the identity at ages 33 and 38 associated with wellbeing at age 38? 4) How are narrative processes of emotional sequencing and positive and negative impact on the identity at age 38 associated with wellbeing at age 38?

Participants
Participants were part of a longitudinal study, the Gothenburg Longitudinal study of Development (GoLD), that was initiated in Sweden in 1982 with a sample of 144 one-to twoyear-olds and their families (Lamb et al., 1988).At the start of the study, families from all social living strata were represented.The present study uses data from the 10 th and 11 th waves of the project.At the 10 th wave there were 124 participants (ages ranging from 32 to 34 years, M = 33.28,SD=.54), and at the 11 th wave there were 116 participants (ages ranging from 38 to 40 years, M = 38.62, SD =. 61).This study involves participants who shared a story about a challenge at the 10 th and 11 th waves of the project (N=95).Among these 95 participants there were 50 women and 45 men.Three of these participants did not answer questionnaires on wellbeing.

Procedure
All participants from the previous waves of GoLD, except those who had declined further participation in the project altogether, were sent written information about the upcoming 11 th wave of the project.Thereafter, they were contacted via telephone and/or e-mail and asked if they were willing to participate.Due to the ongoing COVID-19 pandemic, participants who agreed to take part in the interviews were given the choice to conduct them either via Zoom or face-to-face (at the Department of Psychology at the University of Gothenburg, or at a location of their choice).All interviews were audio-recorded and transcribed.

Narrative prompt
We used the same prompt for difficult and challenging experiences as in Eriksson et al. (2020), asking participants "When you look back over your entire life, what is the most difficult experience, or some of the most difficult events, you have faced?"We also asked participants when the event occurred and whether/how the experience had impacted their identity.The narrative prompt "Challenges" was an adaptation based on the Life Story Interview (McAdams, 2008).

Emotional sequences
To code for emotional sequences, we used a codebook that was developed within the same longitudinal study and therefore within the same cultural context (Eriksson et al., 2020).This codebook describes four emotional sequences.Neutral/vague sequencing involves narratives that use a neutral or vague emotional tone to frame the experiences; for example, stating that the difficult experiences no longer affect the respondent.A combination of positive and negative sequencing refers to narratives in which the protagonist frames their experiences as not only positive but also negative in some way.Redemptive sequencing involves only a positive way of framing the difficult challenge, whereas negative sequencing involves a negative or continued negative state in the respondent's evaluation of the experience.

Positive impact on the identity
We coded for perceived impact of the experience using a previously developed coding system (Pals, 2006), and coded for positive impact on the identity on a 4-point scale ranging from 1 (no evidence whatsoever for positive impact on the self) to 4 (positive impact is a strong theme of the narrative and is elaborated, important, and transformative).For example, positive impact involved gained insight, gained self-strength, appreciation of life, improvement, and better relationships with/support from others.

Negative impact on the identity
To code for perceived negative impact on the identity from the experience, a coding system for growth (Pals, 2006) was adapted for negative impact (Eriksson et al., 2020).This coding system ranged from 1 (no evidence whatsoever for negative impact on the identity) to 4 (negative impact is a strong theme of the narrative and is elaborated, important, and negatively transformative).Examples of negative impact include a loss of abilities, a negative or limiting self-perception, insecurity, burdening responsibilities or memories, and relational difficulties.Reliability was acceptable for positive impact (intraclass r = .92)and negative impact (intraclass r = .83).

Wellbeing
We measured wellbeing using the self-rating Satisfaction With Life Scale (SWLS) and the Brief Symptom Inventory 18 scale (BSI-18).

The Satisfaction With Life Scale (SWLS)
This scale contains five items to measure global satisfaction with life (Diener et al., 1985), and has been validated in Sweden (Hultell & Gustavsson, 2008).Participants were asked to rate statements such as "In most ways my life is close to my ideal" on a range from 1 (strongly disagree) to 7 (strongly agree) (α=.85).

The Brief Symptom 18 Inventory scale (BSI-18)
This scale of 18 items was used to measure psychological distress (somatization, depression, and anxiety; Derogatis, 2001; validated in Sweden by Fridell et al., 2002).Level of distress was rated by participants based on the last seven days from 0 (not at all) to 4 (very much) (α=.84).

Data analyses
First, a person-centered approach was used to investigate typical and atypical patterns of individual stability and change in emotional sequencing between ages 33 and 38, using the cross-tabulation procedure EXACON (Bergman et al., 2003).This person-centered approach, built on the Fisher fourfield hypergeometric distribution test, identifies which individual patterns of change and stability occur more frequently than expected by chance, and which occur less frequently than expected by chance.As these analyses were explorative and we did not want to omit valuable findings attributable to sample size, we used alpha level p < .05.
Second, rank-order stability was examined between narrative processes (positive and negative impact).Third, to examine predictive associations between narrative processes and wellbeing we examined associations between narrative processes (positive and negative impact on the identity) at age 33, and wellbeing (BSI-18, SWLS) five years later at age 38.Fourth, cross-sectional associations were examined between narrative processes (positive and negative impact on the identity, and emotional sequence) and wellbeing at age 38 (BSI-18, SWLS).

Results
We first present descriptive results of emotional sequences and patterns of development between early adulthood (age 33) and midlife (38).Second, we present results from analyses on rank-order stability for the two narrative processes of positive and negative impact on the identity.Third, we present analyses on associations between narrative processes at age 33 and wellbeing at age 38.Fourth, associations between narrative processes at age 38 and wellbeing at age 38 are presented.

Patterns of emotional sequencing
Frequency of the narrative emotional sequences at age 38 are presented in Table 1.A chi-square goodness of fit showed equal expected values across the four emotional sequences: χ2 (1, N = 95) = 2.39, p = .496.Thus, no emotional sequence stood out as more or less frequent in the sample.
To examine development in the narrative process of emotional sequencing between ages 33 and 38, we examined patterns of individual stability.Emotional sequencing refers to the emotional framing of an event and is separate from positive and negative impact on the identity, as this process focuses more on structure and on how the narrator concludes the experience.These analyses that investigated typical and atypical individual patterns between ages 33 and 38 showed that it was typical for participants who told their stories of difficult experiences at age 33 in a neutral/vague way to do so at age 38 as well (see Figure 1).Furthermore, results showed that it was also typical for participants with a combination of positive and negative sequencing at age 33 to tell their narratives using redemptive sequencing and framing of the difficult event at age 38.
Two atypical patterns were also found.Results showed that it was not typical for participants who narrated their difficult experiences at age 33 with a neutral/vague sequencing to change at age 38 to narrating experiences using redemptive sequencing.Neither was it common for participants with   (Bergman et al., 2003).Numbers on the arrows indicate how many times more or less the pattern was observed than would be expected from chance alone.
redemptive sequencing at age 33 to change at age 38 to narrating their stories with neutral/vague sequencing (see patterns in Table 2).

Rank-order stability within positive and negative impact on the Identity
To examine stability and change within narrative processes across time, rank-order stability was examined between positive and negative impact on the identity at ages 33 and 38.Results showed significant rank-order stability for negative impact on the identity (r=.31, p < .01)and for positive impact on the identity (r = .39,p < .01),indicating rank-order stability within these two narrative processes across a five-year period.These results show that the degree of positive impact that the participants perceived that the difficult experience had had on their identity at age 33 (e.g., gained insight, gained self-strength, and improvement) showed moderate stability in relation to the same narrative process coded at age 38.Analyses also showed that the degree of negative impact (e.g. a loss of abilities, negative or limiting self-perception, and insecurity) that participants perceived from their experiences told at age 33 demonstrated stability with the degree of negative impact they perceived from those told at age 38.

Narrative processes and their associations with wellbeing
For analyses concerning wellbeing, we first examined longitudinal associations between narrative processes in stories about difficult experiences at age 33 and wellbeing at age 38.These analyses examined whether narrative processes at age 33 could predict wellbeing at age 38.The results showed a positive association between negative impact on the identity at age 33 and BSI-18 at age 38 (see Table 3).This means that participants whose narratives involved a higher degree of negative impact on their identity at age 33 rated higher psychological distress at age 38.There was also a small positive association between positive impact on the identity at age 33 and SWLS at age 38, indicating that participants who narrated more positive impact and growth at age 33 showed higher satisfaction with life at age 38.Together, these results indicate that how early adults narrate the impact of difficult experiences on their identity is important for their wellbeing at age 38.Bold text indicates the most common patterns between ages 33 and 38 among those who participated at both these ages (N=95).Next, we examined cross-sectional associations between positive and negative impact on the identity at age 38 and wellbeing at age 38.The correlations between positive and negative impact on the identity and wellbeing are presented in Table 3.These analyses showed associations between negative impact on the identity narrated at age 38 and BSI-18 (psychological distress) at age 38, indicating a small positive association between negative impact on the identity and negative aspects of wellbeing at age 38.That is, negative impact at age 38 was associated with more psychological distress at age 38.
Lastly, one-way ANOVAs were performed to examine associations between emotional sequences at age 38 and life satisfaction (SWLS) and psychological distress (BSI-18) at age 38.There were no differences between the emotional sequences with regard to BSI-18 -F(3, 91) = 2.78, p = .066,ω2= .05-and no differences between the emotional sequences with regard to SWLS: F(3, 91) = 0.741, p = .503,ω2=.008 (see Table 1 for means and standard deviations for each group).These analyses showed that there were no associations between emotional sequencing in the narratives of difficult events at age 38 and wellbeing at age 38.

Discussion
There are several important findings from this study examining identity development processes in narratives of difficult experiences and associations with wellbeing in the transition to midlife.First, results indicate that common developmental patterns of emotional sequencing involve stable patterns of a neutral framing of events.Furthermore, stable patterns of moving from combining negative and positive emotions to redemption were also found.Second, the development of narrative processes involving positive and negative impact on the identity showed moderate rank-order stability.Third, results concerning associations between narrative processes and wellbeing showed that narrative processes in early adulthood have a bearing on wellbeing during midlife, and few associations between narrative processes in midlife and wellbeing.Particularly negative impact on the identity stood out in the analyses of connections to wellbeing.The results will be further discussed below.

Narrative development in the transition to midlife
Results involving developmental patterns in the transition to midlife demonstrated that stability in neutral sequencing was common, and that it was atypical for individuals with neutral sequencing in early adulthood to narrate their stories with a redemptive conclusion in midlife, and the other way around, from redemptive sequencing to neutral.Findings of these common and also atypical patterns of neutral sequences are important, as a neutral emotional tone may often be viewed as simply an absence of emotional tone or a scale step in the middle between endpoints of positive and negative tone (McLean & Lilgendahl, 2019;Pals, 2006), and has been given less attention than other emotional sequences such as positive endings (Adler et al., 2016;Dunlop & Tracy, 2013;McAdams, 2006;McAdams et al., 2001;McLean & Lilgendahl, 2008;Pals, 2006).However, together the results involving developmental patterns of neutral/vague sequencing demonstrate that narrating difficult experiences with a neutral or vague tone appears to be a way of its own to view difficulties in life in the transition to midlife.That is, this developmental pattern shows that people who are neutral in their way of viewing challenges appear to maintain this outlook on life's difficulties.As this structure of framing a difficult experience appears to be a stable characteristic, it is therefore an emotional sequence that has its own importance alongside positive and negative tones (or variations of these).Future research could therefore benefit from examining neutral/vague sequencing separate from positive and negative emotional tone.
In previous research, Booker and Perlin (2022; see also Perlin & Fivush, 2021) have suggested that neutral/vague sequencing could be viewed as similar to the form of redemption that involves the return to a baseline, which is a narrative that starts out in a good place but in which something bad happens, and then the individual returns to the same good place.It is important to note that in the present study neutral sequences involve a neutral, or vague, tone in regard to the experience, and the emotional quality of the framing of the narrative does not necessarily signal that the individual has returned to either a good or bad state.Thus, it is solely the view of how the person narrates the difficult experience that has been examined.However, in line with Perlin and Fivush (2021), we also believe that qualitative differences in emotional sequences can shed light on further aspects of the meaning for identity development, and as Booker and Perlin (2022) have also demonstrated, the meaning for wellbeing and narrative processes involving growth.
While we have argued that the common patterns of how people narrate difficult experiences using a neutral/vague tone show importance as a characteristic way of framing challenges, it could be that these patterns indicate something in relation to the Swedish context.For example, in Sweden there is a social norm, "the Law of Jante," that states that you should not stand out in relation to others (Trost, 2012;law of Jante, jantelagen -;Sandemose, 1933).Narratives with a neutral or vague tone do not highlight any emotional tone in the framing of the event and thus do not stand out.In line with the law of jante, conformity and modesty are also appreciated traits in Sweden (Trost, 2012).This is interesting given that Sweden is also viewed as an individualistic country and people score high on selfexpression values (World Values Survey, 2015).As such, it could be that individualistic expressions such as those part of narrative identity in the Swedish context perceived as needing to be expressed in a way that still fit into the norms of conforming and acting in a modest way.It could therefore be that framing difficult experiences with a neutral tone is a valued characteristic from the view of others in Sweden, as this way of framing hardship in life complies with both conformity and modesty and is therefore also a common stable pattern of emotional sequencing in narratives.However, future research is needed to examine how others perceive narrators who structure stories of difficult experiences with a neutral or vague conclusion (see McLean et al., 2020).
Furthermore, findings involving developmental patterns indicated that it was common to narrate stories with a combination of positive and negative sequencing in early adulthood and with redemptive sequencing in midlife.These findings are interesting in the light of developmental research, which has found that people tend to view things in a more positive way as they grow older (Agostino & Sheldon, 2021;Bauer et al., 2005;Mather & Carstensen, 2005).Results from the present study may therefore suggest a development of emotional sequencing between early adulthood and the transition to midlife with less focus on negative aspects in the framing of difficult experiences.Nevertheless, given the findings involving patterns of stability in neutral sequencing, and the frequent attention to either positive or negative evaluations of experiences, it would be valuable for future research to focus on the role of neutral sequencing in emotion regulation in the face of difficult experiences, and how it develops later in life.
The results from this study regarding rank-order stability for positive and negative impact on the identity show that these processes are moderately stable within narratives of difficult experiences.These findings are thus in line with previous research across emerging adulthood and during midlife (Dunlop et al., 2016;McAdams et al., 2006;McLean et al., 2022).However, this study adds to the knowledge about autobiographical stability and change within narratives of difficult experiences in the transition to midlife, and how narratives continue to unfold from early adulthood.

The role of narrative processes in wellbeing in midlife
Findings involving the cross-sectional results in midlife showed only low associations between negative impact on the identity and psychological distress, and no associations between positive impact on the identity and satisfaction with life.These results are surprising, given that there is a fair amount of research supporting that positive growth is connected to wellbeing (Adler et al., 2016;Bauer & McAdams, 2004;Greenhoot et al., 2013) and with thought to the previous findings from the same longitudinal study as the present study is a part of, which showed that positive impact on the identity was connected to wellbeing in early adulthood (Eriksson et al., 2020).However, there are many other things in midlife that can be challenging and even if an individuals' experience a perceived positive impact of the narrated event this may not have an overall bearing on the satisfaction in life.In the development from early adulthood to midlife there may be many other domains in life that account for an individual's overall satisfaction with life.For example, the transition to midlife often involves adjusting one's roles in the areas of work (e.g. more responsibility) and family (e.g.extended family) -and navigating between them.
As the results in the present study only showed that negative impact on the identity in midlife is associated with more psychological distress, this indicates that this narrative process may play some role in negative aspects of wellbeing in midlife, but not positive impact.In contrast, findings from longitudinal investigations in the present study showed associations between the amount of impact, both positive and negative, on the identity in early adulthood and wellbeing in midlife.These findings are in line with the narrative research that has demonstrated the ability of narrative processes to facilitate outcomes of wellbeing over time (Cox & McAdams, 2014;King & Raspin, 2004;Lilgendahl & McLean, 2020;Lodi-Smith et al., 2009;Pals, 2006).However, as discussed above, there are many other aspects of life that may influence the overall wellbeing in the transition to midlife.More research is therefore needed to further understand what aspects that influence wellbeing in midlife.The differences in connections between perceived impact of an event on the identity and wellbeing in early adulthood and midlife could be understood by what characterizes these different parts of adult life with for instance midlife for many being a time of increased responsibilities.Future research could benefit from examining what aspects that are driving these differences in associations in different parts of adult life.
With regard to the lack of connections between positive impact and negative impact on the identity cross-sectionally and over time this shows that there is no interplay between these two narrative processes.The absence of connections longitudinally between positive and negative impact on the identity might be explained by the design of the study and that the design did not necessarily prompt a repeated event.In this way the event that the participants talked about could be two different events at the different time points.However, it could also be that these narrative processes are separate processes and therefore do not interact over time.
What is particularly notable regarding both the longitudinal and cross-sectional findings of associations between narrative processes and wellbeing is that the autobiographical process of negative impact on the identity stands out.The longitudinal findings in the present study showed associations between negative impact on the identity in early adulthood and psychological distress in midlife, and that only negative impact in midlife was cross-sectionally connected to wellbeing.Together, the results from the present study therefore indicate that it is important to examine negative identity development processes in early adulthood, as this may impact wellbeing later in life.Similarly, researchers have highlighted this importance of considering the valence of meaning making processes (Lilgendahl, 2015) and demonstrated, for example, that negative meaning making predicts negative emotion regulation while positive meaning making predicts positive emotion regulation (Cox & McAdams, 2014).However, an extensive part of narrative research focuses more on solely the positive development and connections to wellbeing (see e.g., Adler et al., 2016).Given the present findings and previous research, negative autobiographical reasoning deserves more attention.For example, it could be the case that negative reasoning involves some aspect of rumination and more negative thought processes, which in turn lead to an experience of more psychological distress (see Banks & Salmon, 2013).It could also be that negative reasoning is part of an ongoing process of resolving discrepancy between a global meaning and the meaning connected to the experienced event (Park, 2010).If the discrepancy is not reduced, for instance through change of some kind, this can lead to distress.As meaning making is multifaceted and has been demonstrated to account for variance in psychological distress (Waters et al., 2013), it is therefore of great value to consider several aspects such as valence within autobiographical processes in examinations of identity development and connections to wellbeing.
In a broader context of identity and wellbeing in adulthood, findings from this study show that it is important to consider valence in examinations of how people integrate their negative experiences in adulthood, as the valence of the perceived impact of an event on the identity may have different bearings on wellbeing at different phases in life.The present study demonstrated that negative identity development, such as perceived negative impact of an event, can play an important role in connection to psychological distress in the transition to midlife.

Limitations and future directions
We note that the sample size in this study is small (N = 95), and may have influenced the findings regarding narrative processes and their associations with wellbeing.However, many studies of narrative processes and connections to wellbeing have employed similar sample sizes (N=70, Adler et al., 2006;N=87;Bauer & McAdams, 2010), which suggests that meaningful associations can be found within smaller sample sizes.
Another limitation in the present study concerns the use of only one type of narrative, as research has shown that there is variability in narrative processes depending on event type (McLean et al., 2017).Narrative processes' development patterns and stability may therefore vary depending on the type of event.However, difficult experiences may be particularly important for narrative identity development in the transition to midlife (Lilgendahl, 2015), and may therefore be particularly valuable to examine in this phase in life, in relation to both developmental patterns and wellbeing.
An avenue for future research that this work contributes to is the methodology of going from qualitative information about people's lives to quantitative ratings such as identity development and perceived impact of an event.This type of research is valuable as qualitative information may provide new important information that is not possible to gain using only quantitative measures.Although qualitative work is time consuming, qualitative methods starting with an inductive data driven bottom-up process can shed light on complex cultural information that adds to the knowledge of the field (see e.g., Eriksson et al., 2020).Adding quantitative ratings then enables the possibilities of statistical analyses.Therefore, we see great potential for applying this methodological approach within other fields examining human development.

Conclusion
Findings in this study highlight the role of narrative processes in the transition to midlife.The patterns of emotional sequencing show that identity development from early adulthood to midlife involves stability in neutral/vague patterns, which suggests that this is a characteristic way of framing difficult experiences and thus that it is valuable to regard it as an emotional tone that is separate from positive and negative affect.Furthermore, findings involving associations with wellbeing demonstrate the importance of considering valence in autobiographical processes.In the transition from early adulthood to midlife, particularly negative development processes correspond with negative aspects of wellbeing in midlife.Therefore, it is important to examine the valence of autobiographical processes in development across the lifespan, as this can shed light on various nuances of relations between narrative processes and wellbeing.

Figure 1 .
Figure 1.Typical (solid arrows) and atypical (dotted arrows) patterns of individual stability and change of emotional sequencing between adjacent time points, identified through the cross-tabulation procedure EXACON(Bergman et al., 2003).Numbers on the arrows indicate how many times more or less the pattern was observed than would be expected from chance alone.

Table 1 .
Frequencies of emotional sequences at age 38 and mean values and standard deviations (in parentheses) for wellbeing (SWLS, BSI-18), impact on identity for each emotional sequence.

Table 2 .
Patterns of stability and change in emotional sequencing between ages 33 and 38.

Table 3 .
Correlations between positive impact, negative impact, and wellbeing at age 38, and between at age 33 and age 38.