Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Dec 3.
Published in final edited form as: Death Stud. 2024 Oct 21;50(1):67–78. doi: 10.1080/07481187.2024.2414257

Remembering the Life and Dying Days of a Deceased Close Other: Memory Recall and Associations with Loss Context

Tabea Wolf 1,*, Emily L Mroz 2,*, Wendy G Lichtenthal 3
PMCID: PMC12010010  NIHMSID: NIHMS2028114  PMID: 39432434

Abstract

Grievers recall memories from both the life of the deceased and the dying days, but differences in recall across these memory types are not well-characterized. In this study, 100 bereaved German adults described up to ten important memories of a deceased close other (M = 7.86). Memories from the dying days were classified into: final memories, health transition events, last time events, and temporal markers. Among those who provided at least one dying days memory (73%), these memories were reported to be recalled privately and shared socially more often than memories from the deceased’s life. Memories from the dying days were rated as less emotionally positive than those from the life, and contextual factors from the loss shaped memory recall frequency and emotional valence. Results underscore the need for appropriate end-of-life care to lay a foundation for adaptive remembering and suggest the relevance of dying days memories in therapeutic settings.

One often calms one’s grief by recounting it.

— Pierre Corneille

In the wake of the death of a close other, grievers commonly attempt to forge a coherent story of how the deceased lived and died. This meaning-making endeavor involves constructing and reconstructing narratives based on personal memories of experiences shared with the deceased (Neimeyer & Levitt, 2021). Through this process, adaptive meanings made and strengthened bonds are associated with improved adjustment and psychological outcomes in bereavement (Lichtenthal et al., 2010; Neimeyer et al., 2006; Wolf et al., 2024). It is thus not surprising that supportive resources, from sympathy cards to grief therapies, center around grievers reflecting on and making meaning of memories from the life and loss of a deceased close other (Mroz & Bluck, 2018; Shear et al., 2022). Despite the centrality of memories in individuals’ bereavement journeys, little evidence exists to characterize memories that grievers retain and recall from the life and dying days of a deceased close other. The current study examines the relative extent, and emotional experience, of recalling memories from the life, compared to the dying days, of the deceased. We also identify contextual details from the end-of-life that may shape the retention, recall, and sharing of memories from the dying days.

Recalling the Deceased: Evolving Perspectives Towards Wellbeing in Bereavement

Early grief adjustment theory suggested that grievers would be best served by psychologically disengaging from a deceased close other after a period of initial mourning, to avoid remaining attached to the deceased (for reviews and commentaries on this topic, see Worden, 2015; Wortman & Boerner, 2011). Contemporary perspectives recognize remembering and continuing relationships with the deceased as critical components of grief adjustment (Field et al., 2003; Lichtenthal et al., 2019; Root & Exline, 2014; Stemen, 2020). Maintenance of these continuing bonds (Klass, 2006) is generally considered adaptive (Boerner & Heckhausen, 2003). This connection is thought to occur through mental representations, often in the form of reminiscing on personal memories from lived experiences with the deceased, that help grievers transform the relationship (Wolf et al., 2024). Indeed, therapeutic activities commonly elicit memories about the deceased (Coenen, 2020; Iglewicz et al., 2020) to help grievers to make adaptive meaning of their loss (Lichtenthal et al., 2019).

Researchers and practitioners thus now recognize grievers’ memory systems serve essential functions for processing the reality of their loss (Boelen et al., 2006) and maintaining continuing bonds. Some researchers, however, have proposed that fixation on these memories may be a indicate struggling to process the reality of the loss. To date, studies that explore memory use in grievers have mainly compared memories related to the griever’s ‘loss’ (i.e., including any memory featuring the deceased) with memories unrelated to their loss (excluding any memory featuring the deceased; (Boelen et al., 2010; Maccallum & Bryant, 2008; Robinaugh & McNally, 2013). Researchers have rarely explored distinctions between memories from the life of the deceased, and memories from the period in which the deceased was close to death, defined as the dying days (Mroz & Bluck, 2018).

Memories from the Life and the Dying Days

For this study, we were interested in examining differing or converging patterns of recall of these two types of memories. Researchers have described the essential role of dying days memories in grief adjustment (Baddeley & Singer, 2008). A major task for most grievers, for example, is to process the circumstances within which the deceased died (Tenzek & Depner, 2017). Memories from the dying days can also set the stage for reflection about the entire relationship the griever had with the deceased, influencing ongoing meaning making and post-loss wellbeing (Generous & Keeley, 2022; Sedney et al., 1994).

Few studies have explored different types of memories from the dying days. Existing work focuses primarily on one type of dying days memory, final memories or final conversations from the very last time the griever was with their deceased close other (Generous & Keeley, 2022; Keeley & Koenig Kellas, 2005; Weber et al., 2023). But other memories from the dying days, such as memories about the deceased’s illness or medical care, or memories about special activities enjoyed by the deceased near their death, may also often be retained and recalled by grievers. Characterization of the variety of memories from the dying days, can thus expand our understanding of grievers’ memory use. In addition, in this initial study comparing memories from the life and the dying days, we chose to examine memory features that are classic in the autobiographical memory literature: the frequency of thinking about memories, frequency of talking about memories, and the emotional valence of memories. These features are considered central components of event recall and autobiographical reasoning, and thus highly relevant to this investigation (Bluck & Alea, 2011; Holland & Kensinger, 2010).

Similar to other types of memories about the deceased, grievers have the capacity to form subjective narratives from experiences during the dying days, and these memory narratives may serve foundational psychological functions (Marwit & Klass, 1995; Mroz & Bluck, 2018). Despite this, memories from the dying days continue to be poorly characterized in some research and grief therapeutic settings (Djelantik et al., 2017; Weber et al., 2023), possibly based on concerns that thinking of or speaking about the end-of-life experience may be upsetting, morbid, or indicative of pathological yearning for the deceased. We consider that these perceptions may stem from contexts where deaths are unexpected or disturbing, where losing a close other in a jarring way may contaminate memories from the dying days. To address this, when examining the recall of dying days memories, we also considered whether recall frequency would be associated with end-of-life contextual factors, such as the expectedness of the loss and the opportunity to say goodbye to the deceased. This investigation thus provides insights into whether, and in what contexts, recalling dying days memories is more common and may potentially be reflective of grievers’ attempts to process upsetting loss experiences.

Current Study

The current study compared recall of memories from the deceased’s life to memories from the deceased’s dying days in a non-clinical sample of adults residing in Germany, with the goal of advancing the literature on memory recall in bereaved individuals. Specifically, through prompting participants to share a discrete set of memories of the deceased, our aims were to: 1) identify the number of memories recalled about the deceased’s dying days, as compared to number of memories about the deceased’s life; 2) characterize the types of memories from the dying days by describing salient memory categories (2a) and comparing recall frequencies and emotional valence between types of dying days memories and memories from the life (2b); and 3) examine associations between contextual factors (i.e., expectedness of the loss, chance to say goodbye to the deceased) and the number, recall frequency, and valence of memories from the dying days, as compared to memories about the life of the deceased.

Method

Participants

Data for this study were part of a larger project on autobiographical memory reflection in the context of loss and bereavement (for other findings from this dataset, see Pociunaite and Wolf, 2023). Participants were recruited through advertisements in local newspapers in Southern and Western Germany, posted flyers in public spaces (i.e., local shops, district meeting places, or churches), as well as through a few local grief support groups. In this pragmatic study, we were interested in enrolling individuals across a broad adult age range and with no criteria for time since loss or close other’s cause of death. Participants were instructed to focus on the death-related loss of one close individual in their life; participants who had experienced multiple losses were asked to choose just one, often the closest death-related loss they had experienced to date. Additional eligibility criteria included: age of 18 years, sufficient knowledge of the German language, access to a computer or phone to participate in the online interview, and ability to share specific memories of the deceased. In total, 101 adults expressed their interest in study participation, and all of them met the above-mentioned inclusion criteria. One person did not provide memories and was therefore excluded from the present analyses. Study participants were compensated with 25 EUR. For data sharing inquiries, the first author can be contacted directly.

Measures and Materials

Eliciting Memories about the Deceased and Measuring Recall and Emotional Valence

Participants were asked to recall up to ten important memories from experiences with their deceased close other and briefly describe each memory. Instructions to participants were:

“Please try to remember up to ten memories including the deceased that you consider as personally important. It does not have to be anything extraordinary, and there are no right or wrong answers. It is only important that it is a specific event you have experienced personally and together with the deceased.”

Because the goal of the project was to understand the types and uses of important memories recalled by the griever, rather than to understand how these memories are narrated by the griever, it was considered less resource-efficient to record and transcribe full memory narratives. Instead, the interviewer took notes about the experience the participant described in each case, developing a summary of the participant’s memory (see Table 2). After all the memories were provided, the interviewer went through the summaries with the participant. The interviewer also guided the participant through follow-up questionnaires about each memory, including a single item for to how frequently the participant thinks about each memory and a single item for how frequently they talk about each memory (Bluck & Alea, 2011). Responses were made on a scale from almost never (1) to very often (7). Participants also rated the current emotional valence of each memory on a scale from very negative (1) to very positive (7).

Table 2:

Examples and Descriptive Characteristics of Memory Summaries

Types of Memories Number of Memories Provided (Percent of Participants) Memory Summary Example
From the life of the deceased N/A The participant describes an incident in which she was pleasantly surprised by her husband. He planted a gooseberry tree in their garden because he knew these were her favorite berries.
Final memories 0 (70) The participant describes visiting her mother at the nursing home with her brother. The participant had an hour alone with her mother and they could talk. Shortly after the participant’s brother returned, their mother died.
1 (30)
Health transition events 0 (55) The participant describes accompanying her son to a doctor’s appointment. During the appointment, her son was told that he would probably only live for another two or three months, due to his aggressive cancer.
1 (25)
2 (11)
3 (6)
4 (3)
Last time events 0 (78) The participant describes the last time that she and her now-deceased husband spent a weekend together with friends. She remembers that they had breakfast in the garden, before they returned home.
1 (14)
2 (6)
3 (2)
Temporal markers 0 (84) The participant describes getting a watch from her father at Christmas. Her father bought that watch on a business trip to Chicago one year before his death. This trip was a big dream of his.
1 (16)
Open in a new tab

Note. This Table presents examples of each of the types of dying days memory, and, for contrast, provides one example of a life memory. Memories are presented in third person as a reflection of the summaries written by the research assistant. The second column indicates the percentage of participants who provided different numbers of each type of memory, from providing no memories in a given category to providing one or more. For parsimony, the number of memories from the life of the deceased across participants is not included.

Loss Contextual Factors

Using single-item scales (Boelen, 2012; Reid et al., 2021), participants indicated the degree to which the loss felt expected and the degree to which they felt they had an opportunity to say goodbye. Responses were given on scales ranging from unexpected (1) to expected (7) and from not having had a chance to say goodbye (1) to having had a chance to say goodbye (7), with higher scores indicating more expectedness and greater opportunity to say goodbye, respectively. In addition, participants indicated how long ago the loss happened (in years). Based on time since loss and participants’ current age, we also calculated their age at the time of loss.

Data Collection Procedures

The study was conducted in accordance with the Declaration of Helsinki and approved by the ethical committee of the Deutsche Gesellschaft für Psychologie [German Psychological Association]. Participants reviewed an informed consent form and provided written consent before beginning the study. In the first phase of data collection, a questionnaire set was sent to participants’ homes. Participants were asked to provide demographic information about themselves and their deceased close other. They also responded to the items about loss contextual factors. The second phase was conducted online in an interview setting (via Zoom). The meeting was scheduled about two weeks after participants had sent back the completed questionnaire set. In this phase, we collected participants’ ten most important memories of experiences with the deceased and asked participants to respond to the follow-up items (i.e., frequency of thinking and talking about each memory, emotional valence of each memory). Participants were interviewed individually. Interviews were conducted by research assistants who were trained to respond empathically by the study principal investigator (TW) and a psychotherapist.

Data Analysis

Memory Coding

We chose a closed, single-criterion card sorting procedure (Fincher & Tenenberg, 2005) to analyze the memories according to recall period (i.e., life, dying days). Using this procedure, categories for sorting responses were pre-established (i.e., closed), and each response was sorted into only one category (i.e., single-criterion). First, using a set of instructions developed for this study, two coders read each memory summary independently and sorted them into one of two categories: referring to the life of the deceased, or referring to their dying days. If a memory was sorted into the dying days category, coders provided a short note explaining this decision. The two independent coders showed strong agreement regarding the categorization memory summary at this level (kappa = .80). Inconsistencies were resolved by the principal investigator (TW).

Based on the preliminary card-sort coding notes, and using existing literature on reflection on experiences with deceased close others (Generous & Keeley, 2022; Mroz & Bluck, 2018), the principal investigator (TW) developed a codebook to categorize participants’ dying days memories as one of four types (see Table 2). Both coders from the initial coding step went through the memory summaries again and sorted each summary that was categorized as referring to a dying days memory into one of the four dying days categories. The inter-rater reliability was moderate (kappa = .75). Inconsistencies were again resolved by the principal investigator (TW).

Statistical Analysis

We considered three dependent variables: the number of memories, out of the maximum ten in each participants’ discrete set of memories, that referred to the coded categories (e.g., “memories about the life” and “memories about the dying days”), the self-reported frequency of recall of each category of each memory, and the emotional valence of each memory. Whereas the number of memories per category within each discrete set describes a between-participant variable, the self-reported frequency of recalling each memory and the emotional evaluation of the memory were analyzed on two different levels: the within-participant level and the between-participant level (i.e., responses aggregated within participants to form a mean score).

To address Aim 1, we used simple descriptive statistics (i.e., mean, standard deviation, range, percentage). We then used paired sample t-tests to examine how many memories about the life, in contrast to the dying days were recalled within each participant’s set. We then calculated mean scores of self-reported frequency of a) thinking about and b) talking about memories from the life and memories from the dying days. Again, we used paired sample t-tests to compare across memory types. The same procedure was used to compare the emotional valence of these memories. If a participant did not share any memories related to the deceased’s dying days, they were excluded from these sets of analyses.

To address Aim 2, we used linear mixed regression models. The frequency of thinking and talking about memories were the dependent variables and the dying days categories were entered as predictor variables. To do this, we created four dummy-coded variables, one for each dying days category, with “1” indicating that a memory would belong to the respective category. Memories that refer to the deceased’s life served as the reference category.

To examine the associations between contextual loss factors (e.g., perceived expectedness of the loss, opportunity to say goodbye to the deceased) and dependent memory variables (Aim 3), we used mixed regression models. With respect to continuous dependent variables (i.e., frequency of recall and emotional valence), we reran the linear mixed regression models described above and included contextual factors and their interaction with the dying days categories as predictor variables. To keep models concise, contextual factors were entered simultaneously and interaction terms were tested subsequently. To analyze the number of memories per dying days category, we used a multinomial mixed regression model. Memories about the deceased’s life served as references group, and contextual factors were included as predictor variables. Note that contextual factors represent between-person variables and were centered between persons (i.e., grand-mean centering).

Results

Sample Characteristics

The final sample comprised 100 adults between 18 and 82 years old (M = 51.43, SD = 14.79). The majority were female (79%). At the time of data collection, more than half of the participants reported being married or in a long-term partnership (53%). The remaining participants were either single (22%), divorced (15%), or widowed (10%). Sixty participants were living with their partner or a family member, one person reported living in a residential care home, and 39 reported that they lived alone or in a shared flat. The educational background was assessed based on the German school system: 33 participants graduated from university, 30 participants graduated from high school (i.e., approximately 13 years of formal education), 19 finished school after the lower secondary level, and one participant reported no education. The remaining 17 participants completed vocational training. At the time of data collection, two thirds of the sample were (self) employed (62%), 22 retired, and 10 university students.

Participants reported having lost their partner (14%), one of their parents (50%), or another relative (24%). Seven participants had lost a child (7%). The remaining five participants reported to have experienced another type of loss such as a good friend (5%). Time since loss varied from a few weeks to more than twenty years prior (M = 11.87 years; SD = 14.84; range: 2 months to 72 years). On average, participants reported that the death of their lost close other came somewhat unexpectedly (M = 3.41, SD = 2.41, range: 1 to 7). On average, participants felt they were somewhat able to say goodbye to their lost close other, though there was a wide distribution of ratings in response to this question (M = 3.76, SD = 2.79, range: 1 to 7).

Memories about the Life and the Dying Days

In total, participants shared 786 memories; the average number shared was 7.86 (SD = 2.01; range: 3 to 10). Of these 786 memories, 155 referred to the dying days. The remaining majority, 631, referred to the life of the deceased. The majority of participants provided at least one dying days memory (73%), and three participants shared six dying days memories. On average, participants provided more memories about the life of the deceased (M = 6.31, SD = 2.15) than about their dying days (M = 1.55, SD = 1.45), t(99) = 15.53, p <. 001, Cohen’s d = 1.55). However, among participants who provided at least one dying days memory, participants reported thinking more about memories from the dying days (M = 4.30, SD = 1.50) than memories from the deceased’s life (M = 3.11, SD = 1.12), t(72) = −6.89, p < .001, Cohen’s d = −.81) in everyday contexts. We also found that these participants reported talking more about memories of the dying days (M = 2.48, SD = 1.33) than memories from the deceased’s life (M = 1.99, SD = 0.77), t(72) = −3.85, p < .001, Cohen’s d = −.45). While both types of memories were perceived, on average, as somewhat positive (M = 5.12, SD = 1.94), memories about the life were perceived as more positive (M = 5.32, SD = 1.03) than memories about the dying days (M = 4.42, SD = 1.95), t(72) = 3.52, p < .001, Cohen’s d = .41).

Inspection of bivariate correlations, reported in Table 1, guided our inclusion of covariates in subsequent analyses. The number of memories recalled of the dying days was positively associated with loss expectedness and having the opportunity to say goodbye. The frequency of recalling memories of the dying days was positively associated with age at loss and negatively associated with time since loss. The emotional valence of memories from the life was positively associated with age at loss and negatively associated with time since loss. The emotional valence of dying days memories was positively associated with having the opportunity to say goodbye. For each analysis, we only included covariates that were significantly correlated with the relevant main variables.

Table 1:

Bivariate Correlations between Study Variables

1 2 3 4 5 6 7 8 9 10 11
1 Age at loss 1
2 Time since loss −.603** 1
3 Expectedness of lossa .169 −.044 1
4 Opportunity to say goodbyea .226* −.164 .697** 1
5 Total number of memories reported .027 −.128 .145 .175 1
6 Number of memories about dying days .139 −.109 .328** .403** .289** 1
7 Frequency of thinking about the life −.002 −.083 .030 .014 .142 .105 1
8 Frequency of thinking about dying days .240* −.358** −.120 −.001 .240* .021 .397** 1
9 Frequency of talking about the life −.094 .007 .061 .015 .070 .034 .595** .176 1
10 Frequency of talking about dying days .104 −.171 .025 .009 .098 .039 .346** .437** .584** 1
11 Emotional valence: life memories .222* −.304** .025 .174 .203* .210* .233* .236* .193 .180 1
12 Emotional valence: dying days memories .071 −.011 .161 .242* −.177 .069 .030 .132 .038 .178 .014
Open in a new tab

Note.

*

p < .05,

**

p < .01.

a

Higher scores indicate more expectedness and a greater opportunity to say goodbye, respectively.

Characteristics of Types of Memories about the Deceased’s Dying Days

Our codebook development process resulted in four categories of dying days memories. The first category, final memories, included events that described the final transition from living to deceased, including memories in which the griever spent time with the deceased near death or witnessed the deceased’s final moments of life. The second category included memories from health transition events that led up to the death of a close other, including the diagnosis of a serious, terminal illness or times when irreversible deterioration of health status were apparent (e.g., falls, hospital visits). The third category included memories describing the last time the participant did a particular activity with or engaged with the deceased in a particular way, signifying closeness to death. Finally, the fourth category included memories that referred to the death of the deceased as a temporal marker. These denoted the evolution of the relationship across the life and death context, centering the death as a core aspect of the memory. These memories were often described important events related to the death (e.g., the funeral) or involved use of the death to anchor the memory in the participant’s timeline. Table 2 provides descriptions and brief examples of each category of dying days memories.

Memories from each of the dying days categories were compared to memories about the deceased’s life in terms of rehearsal frequency and emotional valence (see Appendix Table A1). Overall, participants’ increased thinking about memories from the dying days, compared to the life (i.e., as demonstrated in Aim 1), was strongest for final memories (estimate = 2.01, p < .001), followed by memories with death as a temporal marker (estimate = 1.67, p < .001), memories about last-time events (estimate = 1.49, p < .001), and memories about health transition events (estimate = 0.53, p = .005). Participants’ increased talking about memories from the dying days, as compared to memories from the life, was strongest for last-time events (estimate = 1.04, p < .001), followed by final memories (estimate = 0.78, p < .001), and memories with death as a temporal marker (estimate = 0.63, p = .039). Memories about health transition events were talked about similarly frequently as those about the deceased’s life. Finally, in terms of emotional valence, memories from the life were rated as more positive than memories for health transition events (estimate = −1.23, p < .001) and final memories (estimate = −0.70, p = .037), with no differences for memories of last-time events or those using death as temporal marker.

Associations between Contextual Loss Factors and Dying Days Memories

We found associations between loss expectedness, opportunity to say goodbye, and the number of memories describing health transition events: the more expected the loss was perceived to be, the more frequently a participant’s memories reflected health transition events (estimate = .176, p = .032). Likewise, having a greater opportunity to say goodbye was associated with providing marginally more memories pertaining to health transition events (estimate = .133, p = .066). Having an opportunity to say goodbye was also positively associated with providing a final memory: participants who felt able to say goodbye also more commonly included a final memory (estimate = .336, p = .001).

Results from the following analyses are presented in more detail in the Appendix (see Table A1). Examination of how frequently participants thought about each memory revealed interactions between age at loss and the frequency of thinking about both final memories and memories about health transition events. Specifically, both types of dying days memories were reported to be more frequently thought about than memories about the life of the deceased, and this effect was stronger the older the participant was at the time of loss (final memories: estimate = .032, p = .037; health events that led up to the death: estimate = .028, p = .009). We also found an interaction between thinking about final memories and time since loss: the greater frequency of thinking about these types of dying days memories as compared to memories of the life decreased the longer ago the loss happened (estimate = −.052, p = .007). Put differently, the more recently the loss occurred, the more frequently participants thought about final memories. A similar pattern was found regarding memories with death as a temporal marker, but the effect was only marginally significant (estimate = −.040, p = .051).

Analyses of how frequently participants talked about each memory also showed interactions between age at loss and frequency of talking about final memories. Final memories were more frequently talked about than memories about the deceased’s life, and the older the person was at the time of loss, the stronger this effect was (estimate = .028, p = .027). In addition, the higher frequency of talking about memories of last time events as compared to memories about the life decreased the longer ago the loss happened (estimate = −.055, p = .0123). A similar pattern was found regarding final memories, but the effect was only marginally significant (estimate = −.029, p = .057).

Finally, in examining the emotional valence of the recalled memories, we found a number of different interaction effects for final memories and memories of health transition events. Final memories were generally rated as less positive compared to memories about the deceased’s life (see Aim 2); however, this difference becomes smaller the older a person was at time of loss (estimate = .057, p = .003) and the more a person felt having the opportunity to say goodbye (estimate = .353, p = .018). Similar results were found regarding memories about health transition events. In general, those memories were perceived as less positive compared to memories about life (see Aim 2); but this difference becomes smaller the longer ago the loss happened (estimate = .030, p = .021) and the more a person felt the opportunity to say good-bye (estimate = .199, p < .030). Age at the time of loss showed the opposite effect, the difference in valence between memories about health transition events and memories about the deceased’s life becomes stronger the older a person was at the time of loss (estimate = −.044, p < .001).

Discussion

In this study, grievers bereaved by a range of causes were prompted to share a maximum of ten specific, personally significant memories featuring the deceased. Participants shared nearly eight memories on average. While most of the memories shared pertained to the life of the deceased, nearly three-quarters of participants shared at least one memory from the dying days, and up to six. Participants reported thinking and talking more frequently about memories from the dying days than memories of the life of the deceased, particularly final memories and memories of last time events. While all memories were reported to be somewhat emotionally positive on average, dying days memories, especially memories about health transition events and final memories, were reported as less positive than memories from the life of the deceased. Contextual loss factors shaped participants’ thinking about, talking about, and attributing emotional tone to dying days memories in their everyday lives. Notably, having had the opportunity to say good-bye was associated with positive emotional tone of these memories.

Differences in Frequency of Recall Across Memory Types

Thinking more frequently about certain types of memories from the dying days may reflect efforts to integrate these memories into one’s own broader life story. This may be an essential part of the process of accepting the reality of the loss (Boelen et al., 2006) and making meaning of the loss (Neimeyer et al., 2006; Neimeyer & Levitt, 2021), which are widely considered tasks of adaptive grieving. Indeed, our prior research found that memories from the serious illness and death of a close other (i.e., events represented within the dying days) were more strongly connected to grievers’ reports of personal growth than memories about other life challenges (e.g., financial hardship, relationship deterioration; Mroz et al., 2020). While previous studies have demonstrated an associated between preferential, compulsive retrieval of memories about the deceased and prolonged grief symptoms (Boelen et al., 2010; Maccallum & Bryant, 2010), these studies have not distinguished between memories of the deceased’s life and dying days. Additional research is needed to determine whether, when, and how more frequent thinking about memories from the dying days may be associated with adaptive meaning-making processes, versus rumination to understand its role in protracted intense grief reactions.

Frequent discussion of dying day memories, as compared to memories from the life of the deceased, may also facilitate processing and may additionally elicit social support. We recognize that, in general, speaking about the dying days may deepen opportunities for meaning-making, allowing for additional perspectives from others on how end-of-life events unfolded, and potentially easing rumination or searching for meaning when end-of-life was distressing. In other cases, these memories may serve different uses for the griever when memories are less distressing; interestingly, we found that participants provided more dying days memories in their sets of memories when the death was expected and when they felt they had an opportunity to say goodbye, and that the chance to say goodbye improved the emotional tone of these memories.

When circumstances surrounding the death experience are better in these ways, the griever may be more inclined to access and share dying days memories because they are more well-integrated and tolerable or even positive and meaningful to re-experience. This underscores the value of bereavement-conscious palliative and hospice care (Roberts et al., 2021), communication training for healthcare providers, and pre-death psychosocial support services. Each of these resources may increase the likelihood that grievers have positive experiences leading up to the patient’s death, including during the dying days. In fact, a recent study demonstrates one way this association occurs: when the bereaved experiences the death as involving appropriate care for the deceased, including treatment in line with the prognosis, they narrate final memories with more positive affect (Mroz & Bluck, 2023). Similar examination of other memories from the dying days (e.g., health transition events,) may further explain how end-of-life circumstances and care can shape use and emotional tone of these memories.

Uses of Memories from the Life and Dying Days of the Deceased during Bereavement

Our results suggest distinct ways that memories from the life, and from the dying days, can be used by grievers. For example, nostalgic reflection on the past has been shown to buffer grief-related distress in bereaved adults (Reid et al., 2021). Nostalgic responses can be generated by recall of good times spent with the deceased. Memories that spur nostalgia may be cued by unexpected reminders as well as intentional memorializing practices. Having a variety of significant, emotionally positive memories from across shared lived experiences with the deceased is thus likely to afford optimal conditions for nostalgic reflection.

Most memories about the dying days, in contrast, may be less well suited to support nostalgic reflection, and may instead serve other purposes. For example, conscious recall of memories from the dying days to may help grievers appraise their ability to cope with future challenges situations (Mroz et al., 2023), or prepare for their own eventual deaths (Webster, 1993). This may explain why we observed that older study participants reported more frequently thinking about memories from health transition events and final memories, as compared to younger participants. Middle-aged and older adults are most likely to be in or to anticipate caregiving roles for ill or frail family members (AARP, 2020) and may also be beginning to prepare for their own deaths. Memories about end-of-life experiences, i.e., those stemming from the dying days, may serve as meaningful anchors as they prepare to shepherd others through this life phase or contemplate their own changing health and mortality.

Although we did not examine associations between memory types and psychological or behavioral outcomes in this study, we found that participants who had experienced their loss more recently also reported more commonly thinking about memories of events where death was a temporal marker, that is, that the death provided an anchor for thinking about when the main narrated event occurred. Here again, memories may be serving specific functions for the griever. For example, according to the Transition Theory (Brown et al., 2009), when major life events change how we operate in our daily lives, they can also temporarily or permanently change our mental construction of our autobiography, influencing the ways we organize memories. While this theory is most often applied to historical, collective challenges (e.g., terrorism, war), similar effects may be seen when individuals experience the death of a close other. Orienting lived experiences around the death may help the grievers early in their grief adjustment to integrate the loss into their larger life story, completing a process of memory re-organization that may be vital during this time of transition (Bohn & Habermas, 2016)). Over time, grievers may be less inclined to anchor their memories around the death because the death has been adequately integrated into their personal life story. In connection with related work (Maccallum & Bryant, 2010), it will be important to examine in future studies whether continuing to orient one’s memories of the deceased around the death in the longer term, thus favoring temporal marker memories over long periods, signifies difficulty integrating these types of memories.

Clinical Implications of Recalling Memories from the Life and Dying Days

This study suggests that memories about both the life of the deceased and their dying days are commonly recalled in a non-clinical sample, and given this, grievers thinking and speaking about the deceased’s dying days, in and of itself, may not be a marker of maladaptive grieving. While memories from the dying days were described as less emotionally positive than memories from the life of the deceased, they were still described as somewhat positive on average. Those that were the least positive included occasions when the deceased’s illness or functional limitations were pronounced (e.g., health transition events, final memories), and thus were understandably likely to be emotionally complex. The process of intrapsychic reflection about memories of time spent with the deceased during the dying days may be to be part of natural meaning-making processes that often contribute to loss adaptation. That said, given the range of responses about memory emotional tone, private recall of memories of the deceased may still be accompanied by distress. These memories may be intrusive, and reflection may sometimes be ruminative (Boelen et al., 2006), particularly when grievers search for meaning or replay scenes with embedded conflict, guilt, or other complex emotions.

Sharing of memories from the dying days can be experienced as isolating if not received by a supportive other. Given a current lack of social guidance about how to respond to grievers who share memories from the dying days, well-meaning close others may unintentionally create distance or discomfort when grievers share these memories, even if the sharer is not exhibiting distress or disclosing painful experiences. Thus, we encourage broader efforts to increase grief literacy and (Breen et al., 2022) identify helpful ways of responding to grievers sharing memories like these. That said, grief counseling and psychotherapy can provide protected spaces for grievers to deepen their reflection on the dying days as well as to receive feedback from a support provider. Counseling may facilitate the development of adaptive meaning-making and continuing bonds with the deceased, even when social narrative exchange falls short. Interventions that incorporate meaning reconstruction (Neimeyer et al., 2006), such as Meaning-Centered Grief Therapy (Lichtenthal et al., 2019), may offer an opportunity for the griever to develop and revise narratives based on these memories. Intentional attention to memories about the deceased’s dying days in such interventions may thus have value. For example, Prolonged Grief Therapy (Shear et al., 2022), incorporates retelling of the events surrounding the death to promote acceptance of and integration of the loss and to habituate to related feelings.

Limitations and Future Directions

This study was limited by use of a non-clinical sample of grievers with a wide range of time since loss; implications for specialized support to address disordered grieving or memory use in the early aftermath of a loss cannot be determined from this work. Similarly, we did not directly examine the intersection of healthcare services, or communication with providers, and study variables. Examining links between markers of quality end-of-life care, features of memories of the deceased (including of their dying days), and grievers’ psychological outcomes is an important avenue of future investigation. In addition, our sample reflected German adult populations, and our findings may not represent those from other cultural or ethnic backgrounds, or populations that are less homogenous in general.

Other limitations include use of a cross-sectional, retrospective design with no memory aides. This may have prompted research artifacts, such as the griever feeling pressure to share more or different memories than they might have outside of research setting. Associations between number of dying days memories and contextual loss factors may have been bidirectional, and directionality should be established in future work. Finally, because we did not evaluate associations with psychosocial or behavioral outcomes, we recognize that determining specific functions of memories was outside of this study’s scope.

The approach we used to elicit memories is just one way to assess how bereaved individuals recall experiences with a deceased close other over time. This design allowed us to directly compare the frequencies of memories about life with the deceased to dying days memories. Other studies (diary entry, naturalistic observation, narrative analysis) may provide information to support or refine results of this study. Finally, building on this study’s assertion that recalling memories from the dying days is generally common and can be innocuous, future research should examine grievers’ experiences sharing such memories with others to inform the development of resources to improve positive outcomes from such sharing.

Conclusions

Memories of a deceased close other play a central role in sustaining a connection to them. Recognition of the utility of memories for this purpose has ushered in useful investigation of continuing bonds, but has been hampered by conflicting perspectives, including unsubstantiated myths pertaining to how, when, and what grievers “should” recall (Wortman & Boerner, 2011). Our study addresses research gaps around memory recall in bereavement using a sample of individuals bereaved for between two weeks and more than two decades. Participants provided an array of memories from the life and dying days, demonstrating the significance of dying days memories and indicating how loss context can shape memory recall. This work can support shifts in research distinguishing functions of memories from the deceased’s life and death, shifts in medical practice to support narration of constructive dying days memories, and refinement of interventions to support grief adjustment. This work can also be used to support community education around remembering the deceased, helping the griever, and those who support them, to feel comfortable hearing and sharing a range of recalled memories.

Acknowledgements:

We thank Justina Pociunaite for her support during data collection, Annika Münzer for supporting the interviewer training, all students who supported the recruitment of participants and conducted the interviews, and all of the participants who shared their memories with us.

Funding:

Tabea Wolf received funding from the Graduate & Professional Training Center Ulm University (ProTrainU) within the framework of the start-up research financing program for young scientists (Anschubsfinanzierung B). The funding source provided resources for hiring student assistants, recruiting participants, and participant compensation, but had no other role. Emily Mroz is supported by the National Institute on Aging (NIA) through an Institutional Training Grant (T32AG019134).

Appendix

Table A1:

Characteristics of Memories from the each of Dying Days Category compared to Memories about the Life

Frequency of Thinking Frequency of Talking Emotional Evaluation
Model 1 Model 2 Model 3 Model 1 Model 2 Model 3 Model 1 Model 2 Model 3 Model 4
Intercept 3.096* 3.095* 3.096* 1.988* 1.986* 1.987* 5.276* 5.281* 5.279* 5.277*
Final memory 2.007* 1.901* 1.886* 0.781* 0.691* 0.719* −0.702* −1.010* −0.775* −1.422*
Health events 0.528* 0.491* 0.535* 0.143 0.130 0.149 −1.227* −1.221* −1.292* −1.577*
Last-time event 1.489* 1.453* 1.369* 1.040* 0.962* 0.766* −0.544 −.620† −0.589 −0.519
Death as temporal marker 1.666* 1.712* 1.745* 0.632* 0.640* 0.626* 0.200 0.216 0.219 0.278
Age at loss −0.007 −0.003 −0.008 −0.006 0.002 0.001 0.002
Time since loss −0.013 −0.008 −0.007 −0.004 −0.014† −0.018* −0.012
Saying good-bye 0.075* 0.079* 0.050
Age at loss*final memory 0.032* 0.028* 0.057*
Age at loss*health events 0.028* 0.014 −0.044*
Age at loss*last-time event 0.012 0.023 0.012
Age at loss* death as temporal marker 0.020 0.000 0.005
Time since loss*final memory −0.052* −0.029† 0.016
Time since loss*health events −0.017 −0.012 0.030*
Time since loss*last-time event −0.025 −0.056* −0.003
Time since loss*death as temporal marker −0.040† 0.005 0.002
Saying good-bye *final memory 0.353*
Saying good-bye *health events 0.199*
Saying good-bye *last-time event 0.052
Saying good-bye *death as temporal marker 0.319†
Open in a new tab

Note:

*

= p < .05,

†

= p < .10. Model 1 compares memories from each of the dying days categories to memories about the deceased’s life in terms of the frequency of thinking and talking as well as their emotionality. In Model 2 to Model 4, relevant contextual factors and their interaction with the dying days categories were entered. Contextual factors were entered simultaneously, but interaction terms were tested separately.

References

  1. AARP. (2020). Caregiving in the U.S. https://www.aarp.org/content/dam/aarp/ppi/2020/05/full-report-caregiving-in-the-united-states.doi.10.26419-2Fppi.00103.001.pdf
  2. Baddeley JL, & Singer JA (2008). Telling losses: Personality correlates and functions of bereavement narratives. Journal of Research in Personality, 42(2), 421–438. [Google Scholar]
  3. Bluck S, & Alea N (2011). Crafting the TALE: Construction of a measure to assess the functions of autobiographical remembering. Memory, 19(5), 470–486. [DOI] [PubMed] [Google Scholar]
  4. Boelen PA (2012). A prospective examination of the association between the centrality of a loss and post-loss psychopathology. Journal of affective disorders, 137(1–3), 117–124. [DOI] [PubMed] [Google Scholar]
  5. Boelen PA, Huntjens RJ, van Deursen DS, & van den Hout MA (2010). Autobiographical memory specificity and symptoms of complicated grief, depression, and posttraumatic stress disorder following loss. Journal of Behavior Therapy and Experimental Psychiatry, 41(4), 331–337. [DOI] [PubMed] [Google Scholar]
  6. Boelen PA, Van Den Hout MA, & Van Den Bout J (2006). A cognitive-behavioral conceptualization of complicated grief. Clinical Psychology: Science and Practice, 13(2), 109. [Google Scholar]
  7. Boerner K, & Heckhausen J (2003). To have and have not: Adaptive bereavement by transforming mental ties to the deceased. Death studies, 27(3), 199–226. [DOI] [PubMed] [Google Scholar]
  8. Bohn A, & Habermas T (2016). Living in history and living by the cultural life script: How older Germans date their autobiographical memories. Memory, 24(4), 482–495. [DOI] [PubMed] [Google Scholar]
  9. Brown NR, Lee PJ, Krslak M, Conrad FG, GB Hansen T, Havelka J, & Reddon JR (2009). Living in history: How war, terrorism, and natural disaster affect the organization of autobiographical memory. Psychological Science, 20(4), 399–405. [DOI] [PubMed] [Google Scholar]
  10. Coenen C (2020). The Creative Toolkit for Working with Grief and Bereavement: A Practitioner’s Guide with Activities and Worksheets. Jessica Kingsley Publishers. [Google Scholar]
  11. Djelantik AMJ, Smid GE, Kleber RJ, & Boelen PA (2017). Early indicators of problematic grief trajectories following bereavement. European Journal of Psychotraumatology, 8(sup6), 1423825. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Field NP, Gal-Oz E, & Bonanno GA (2003). Continuing bonds and adjustment at 5 years after the death of a spouse. Journal of consulting and clinical psychology, 71(1), 110. [DOI] [PubMed] [Google Scholar]
  13. Fincher S, & Tenenberg J (2005). Making sense of card sorting data. Expert Systems, 22(3), 89–93. [Google Scholar]
  14. Generous MA, & Keeley MP (2022). Exploring the connection between end-of-life relational communication and personal growth after the death of a loved one. OMEGA-Journal of Death and Dying, 84(3), 792–810. [DOI] [PubMed] [Google Scholar]
  15. Holland AC, & Kensinger EA (2010). Emotion and autobiographical memory. Physics of life reviews, 7(1), 88–131. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Iglewicz A, Shear MK, Reynolds III CF., Simon N., Lebowitz B., & Zisook S. (2020). Complicated grief therapy for clinicians: An evidence-based protocol for mental health practice. Depression and anxiety, 37(1), 90–98. [DOI] [PubMed] [Google Scholar]
  17. Keeley MP, & Koenig Kellas J (2005). Constructing life and death through final conversation narratives. Narratives, health, and healing: Communication theory, research, and practice, 365–390. [Google Scholar]
  18. Klass D (2006). Continuing conversation about continuing bonds. Death studies, 30(9), 843–858. [DOI] [PubMed] [Google Scholar]
  19. Lichtenthal WG, Catarozoli C, Masterson M, Slivjak E, Schofield E, Roberts KE, Neimeyer RA, Wiener L, Prigerson HG, & Kissane DW (2019). An open trial of meaning-centered grief therapy: Rationale and preliminary evaluation. Palliative & supportive care, 17(1), 2–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Lichtenthal WG, Currier JM, Neimeyer RA, & Keesee NJ (2010). Sense and significance: A mixed methods examination of meaning making after the loss of one’s child. Journal of clinical psychology, 66(7), 791–812. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Maccallum F, & Bryant RA (2008). Self-defining memories in complicated grief. Behaviour research and therapy, 46(12), 1311–1315. [DOI] [PubMed] [Google Scholar]
  22. Maccallum F, & Bryant RA (2010). Impaired autobiographical memory in complicated grief. Behaviour research and therapy, 48(4), 328–334. [DOI] [PubMed] [Google Scholar]
  23. Marwit SJ, & Klass D (1995). Grief and the role of the inner representation of the deceased. OMEGA-Journal of Death and Dying, 30(4), 283–298. [Google Scholar]
  24. Mroz E, & Bluck S (2018). The loss of a loved one: Remembering the life and the dying days. Reminiscence, life review, and life story work: International perspectives, 162–179. [Google Scholar]
  25. Mroz EL, & Bluck S (2023). Narrating Final Memories From Spousal Loss: The Role of Place and Quality of Death. American Journal of Hospice and Palliative Medicine®, 10499091231204965. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Mroz EL, Bluck S, Sharma S, & Liao H. w. (2020). Loss in the life story: Remembering death and illness across adulthood. Psychological Reports, 123(1), 97–123. [DOI] [PubMed] [Google Scholar]
  27. Mroz EL, Monin JK, Gaugler JE, Matta-Singh TD, & Fried T (2023). Re-Writing the Story of Mid-and Late Life Family Caregiving: Applying a Narrative Identity Framework. The Gerontologist, gnad040. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Neimeyer RA, Baldwin SA, & Gillies J (2006). Continuing bonds and reconstructing meaning: Mitigating complications in bereavement. Death studies, 30(8), 715–738. [DOI] [PubMed] [Google Scholar]
  29. Neimeyer RA, & Levitt HM (2021). What′ s Narrative Got to Do With it? Construction and Coherence in Accounts of Loss. In Loss and trauma (pp. 401–412). Routledge. [Google Scholar]
  30. Reid CA, Green JD, Short SD, Willis KD, Moloney JM, Collison EA, Wildschut T, Sedikides C, & Gramling S (2021). The past as a resource for the bereaved: Nostalgia predicts declines in distress. Cognition and Emotion, 35(2), 256–268. [DOI] [PubMed] [Google Scholar]
  31. Roberts KE, Lichtenthal WG, & Ferrell BR (2021). Being a bereavement-conscious hospice and palliative care clinician. Journal of Hospice & Palliative Nursing, 23(4), 293–295. [DOI] [PubMed] [Google Scholar]
  32. Robinaugh DJ, & McNally RJ (2013). Remembering the past and envisioning the future in bereaved adults with and without complicated grief. Clinical Psychological Science, 1(3), 290–300. [Google Scholar]
  33. Root BL, & Exline JJ (2014). The role of continuing bonds in coping with grief: Overview and future directions. Death studies, 38(1), 1–8. [DOI] [PubMed] [Google Scholar]
  34. Sedney MA, Baker JE, & Gross E (1994). “The story” of a death: Therapeutic considerations with bereaved families. Journal of Marital and Family Therapy, 20(3), 287–296. [Google Scholar]
  35. Shear MK, Skritskaya N, & Bloom C (2022). Prolonged Grief Disorder Therapy (PGDT). In Evidence Based Treatments for Trauma-Related Psychological Disorders: A Practical Guide for Clinicians (pp. 329–344). Springer. [Google Scholar]
  36. Stemen SE (2020). “They’re in my heart, mind, and cells”: What continuing bonds reveal about social networks. The Gerontologist, 60(6), 1085–1093. [DOI] [PubMed] [Google Scholar]
  37. Tenzek KE, & Depner R (2017). Still searching: a meta-synthesis of a good death from the bereaved family member perspective. Behavioral sciences, 7(2), 25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Weber KM, Lemanski M, Lorenz M, & Bingle C (2023). What We said and what I wish We said: Regret during final conversations. Death studies, 47(9), 983–993. [DOI] [PubMed] [Google Scholar]
  39. Webster JD (1993). Construction and validation of the Reminiscence Functions Scale. Journal of gerontology, 48(5), P256–P262. [DOI] [PubMed] [Google Scholar]
  40. Wolf T, Nusser L, & Cappeliez P (2024). How reminiscing about deceased close others together with continuing bonds relates to grief severity and personal growth: a cross-sectional study with bereaved adults. Aging & mental health, 1–9. [DOI] [PubMed] [Google Scholar]
  41. Worden JW (2015). Theoretical perspectives on loss and grief. Death, dying, and bereavement: Contemporary perspectives, institutions, and practices, 91–103. [Google Scholar]
  42. Wortman CB, & Boerner K (2011). Beyond the myths of coping with loss: Prevailing assumptions versus scientific evidence. [Google Scholar]

RESOURCES