Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Dec 9.
Published in final edited form as: J Appl Gerontol. 2025 Oct 10;45(8):1464–1473. doi: 10.1177/07334648251378791

“One More Picture:” A Mixed-Methods Examination of Caption Content in a Photo Captioning Cognitive Empathy Intervention for Dementia-Caregivers

Emily L Mroz 1, Rarinsri P Assakul 2, Jenna L Wells 2, Molly M Perkins 3,4, Kenneth Hepburn 1, James K Rilling 5,6,7,8,9
PMCID: PMC12683352  NIHMSID: NIHMS2108524  PMID: 41070931

Abstract

Empathic perspective-taking may support mental health in caregivers of people living with dementia (PLWD). A photo captioning intervention can promote perspective-taking by guiding caregivers to capture the inner voice of the PLWD. Caregivers’ approaches to captioning may be an important mechanism of intervention efficacy. Using a mixed-methods design including subgroup parallel sampling and descriptive content-analysis of 310 captions from an intervention trial, we identified patterns in caregivers’ photo captioning. Captions from caregivers who experienced increases in perspective-taking (1) maintained first-person, PLWD focused language, (2) highlighted engagement and intention, and (3) illustrated resilient living. Captions from caregivers who experienced little-to-no increases in perspective-taking (1) offered inconsistent, incomplete phrasing, (2) provided generic, detached descriptions, and (3) adopted a negative, critical tone. Captions from both subgroups (1) connected PLWDs’ emotions to events and (2) recorded PLWDs’ wonders, wants, and wishes. Results support the development and refinement of perspective-taking interventions to maximize caregiver benefits.

Keywords: perspective taking, Alzheimer’s disease, family caregiving, behavioral intervention

Introduction

Caregivers to family members and close others living with dementia (i.e., PLWD) support PLWD to navigate their daily lives, cope with their symptoms, and make decisions for the future. Due to the unique strains of caregiving for PLWD, these caregivers experience significant and persisting health challenges, including elevated burden, depression, and anxiety (Sörensen & Conwell, 2011). Given the pressing need to support caregivers to maintain health and well-being through and after their caregiving roles, scholars have called for the development of accessible, evidence-based behavioral interventions for caregivers (Gaugler, 2022), as well as a determination of mechanisms of action that lead to intervention impacts on caregiver health outcomes (Gitlin et al., 2000).

One increasingly popular approach to improve caregiving outcomes is through bolstering caregivers’ cognitive empathy (Huo et al., 2025). Cognitive empathy, the ability to understand another person’s thoughts, emotions, and perspectives, is often operationalized in terms of caregivers’ perspective-taking abilities in scenarios with their care recipient (Lovell et al., 2023; Wijma et al., 2018). Higher cognitive empathy has been associated with improved mental health in caregivers (Jütten et al., 2019), likely because this form of empathy facilitates emotional regulation, effective engagement in problem solving, and responsiveness to the needs of PLWD (Kripal & Reiter-Palmon, 2024; Thompson et al., 2022). To promote the dissemination of promising cognitive empathy interventions, researchers have recently emphasized the need to understand components of these interventions that foster empathy, caregiver mental health, and care navigation (Han et al., 2021).

Our team recently completed pilot testing of a photo captioning intervention that demonstrated improvements to empathic perspective-taking, and decreases in felt burden and anxiety, in caregivers of PLWD who reported high baseline levels of burden (Rilling et al., 2024). The intervention involved coaching caregivers to—over the course of 10 days—take three to five photographs daily of the PLWD and caption each photo with descriptive text that captured the inner voice of the PLWD. Caregivers were instructed to write these captions from the point of view of their PLWD.

The study team hypothesized that improvement in perspective-taking may be a psychological driver of improved caregiver mental health. Empathic perspective-taking was measured prior to the photo captioning intervention and within 1 week of completing the intervention using the perspective-taking subscale of the Interpersonal Reactivity Index (Davis, 1983). Results indicated that perspective-taking increased significantly from pre- to post-intervention, but that changes varied across participants (change score SD = 2.76; Rilling et al., 2024). Mental health indicators also improved across intervention participation.

For the current study, we sought to examine a behavioral mechanism of intervention efficacy: a component that could be a critical driver of intervention success by promoting cognitive empathy gains. We considered that variation in improvements in perspective-taking from pre- to post-test may be influenced by caregivers’ approaches to generating photo captions, with participants who experienced high gains in perspective-taking applying approaches differently from those who experienced low gains, or no gains, in perspective-taking. We postulated that caregivers’ approaches to caption content generation may reflect their success with “practicing” perspective-taking (Gehlbach & Mu, 2023) by setting aside their own thoughts, feelings, and goals to recognize those of the PLWD.

Current Study

This primary objective of this study was to examine the content of the photo captions caregivers generated while participating in this intervention, identifying patterns in caregivers’ approaches for generating captions. By doing so, our goals were to determine whether and how caption content can serve as a behavioral mechanism of intervention efficacy and to specify guidelines for caption writing that may lead to optimal intervention engagement.

Method

This study involved a secondary analysis of data generated from a pilot trial of a photo captioning intervention conducted with caregivers of PLWD (Rilling et al., 2024). The trial was Institutional Review Board approved (#00001824). We employed a mixed-methods intervention design (Creswell & Clark, 2017). Captions were first divided into two groups based on quantified changes in caregivers’ empathic perspective-taking across the intervention. Descriptive information about captions (i.e., length of captions and number of captions per participant) and the participants who wrote the captions (e.g., baseline psychological traits) were compared across groups. In the qualitative phase, captions were inductively explored, resulting in categories of approaches to caption content used by caregivers in each group. We followed the COnsolidated criteria for REporting Qualitative research (Tong et al., 2007).

Participants

For full details about the recruitment, enrollment, and retention of participants for the pilot trial, see Rilling et al. (2024). This study focused only on the family caregiver participants (herein: participants) who were enrolled in the experimental condition. The pilot trial targeted family caregivers who were experiencing high levels of caregiving burden at baseline. As such, prospective caregiver participants were screened by completing the Zarit Burden Interview (Zarit et al., 1980) and enrolled if they scored 24 or higher on this scale. Twenty participants met eligibility and completed the study. Participants completed an online-administered informed consent form prior to engaging in study procedures.

Study Design and Procedures

Data for this study was sourced from the original pilot trial of this photo captioning intervention. The intervention guided participants to use their smartphone to take photos of their PLWD over a 10-day period. They were prompted to “take 3–5 pictures a day, captioning each picture with 2–3 sentences with what you think [PLWD’s name] is thinking or feeling at that time.” The study team provided caregivers with examples of captions to help them understand the exercise, including, “I’m enjoying being outside and listening to my classical music. I’m feeling relaxed and glad for this down time. So much of my life feels complicated and confusing and this is simple and ‘me’.” and “I wish I could remember how to use my phone. I am not sure what I am trying to do with it. I guess I will keep pushing buttons until something looks familiar.” When caregivers submitted captions that did not align with study instructions, research staff would clarify instructions, providing specific feedback tailored to the caregivers’ initial captions. For example, if one caption that was submitted was more in-line with intervention instructions than others, the staff member would encourage the participant to write captions that were similar to that caption. The research staff noted that this clarification process was effective in improving the alignment of the captions with study instructions. Microsoft One Drive was used to collect photos and captions from participants.

To examine the photo captions, we focused on captions provided by participants who experienced the most and least change in perspective-taking. We disaggregated the caption data by applying a subgroup parallel extreme case sampling design (Onwuegbuzie & Leech, 2007) to create two subgroups. Extreme case sampling is recommended for studies examining program implementation, and as such, this method is appropriate for our goals (Suri, 2011).

To develop the two subgroups, we used pre- to post-test change scores in perspective-taking from the relevant subscale of the Interpersonal Reactivity Index (Davis, 1983), administered during the original pilot trial. This subscale included seven items, with participants responding to each item using response options from 1 to 4. As such, possible total scores ranged from 1 (i.e., the lowest possible report of perspective-taking) to 28 (i.e., the highest possible report of perspective-taking). The first subgroup included the captions of those participants who experienced high changes in perspective-taking from pre- to post-test, as operationalized by participants reporting changes in perspective-taking that were ≥½ standard deviation more than the average change score (i.e., ≥ a 3.23-point increase). The second subgroup included the captions of those participants who experienced low or no changes in perspective-taking from pre-to post-test, as operationalized by participants reporting changes in perspective-taking that were ≥½ standard deviation less than the average (i.e., ≤ a .47-point increase). In total, 5 participants with 120 captions were included in the subgroup experiencing high change in perspective-taking, and 8 participants with 190 captions were included in the subgroup experiencing low change. Thus, 310 captions from 13 of the 20 participants who completed the original intervention were included in this analysis. The sample sizes are considered an appropriate fit given our analytical approach (Onwuegbuzie & Leech, 2007).

Data Analysis

Features of photo captions (i.e., length of captions and number of captions per participant) from each subgroup were assessed using descriptive statistics and independent samples t-tests. We also compared baseline self-reported perspective-taking and caregiving burden1 scores using independent sample t-tests to examine potential variation in these factors between subgroups at the outset of the intervention.

Once the data were prepared for qualitative analysis, a team of researchers applied descriptive content-analysis to the study data (Sandelowski, 2000; Vaismoradi et al., 2013). We chose this method because it was consistent with our goal of providing a comprehensive description of a phenomenon across a larger amount of textual data (Vaismoradi et al., 2013). The analysis team comprised three female researchers with complementary backgrounds: one (EM) was a PhD-trained behavioral health researcher and gerontologist with expertise in qualitative methods, the second (RA) was a psychology undergraduate student with an interest in familial relationships and dynamics of caregiving, and the third (JW) was a PhD-trained psychologist with expertise in cognitive empathy in dementia caregiving contexts.

The team’s research questions and interpretations of findings were guided by the instructions for this intervention, which encouraged participants to fully engage by writing captions that featured first-person language from the point-of-view of the person living with dementia and write complete sentences with subjects and verbs. Interpretation of data was also guided by sensitizing frameworks. Most prominent was a dementia positivity lens, which suggests that individuals, communities, and societies can benefit from working towards discovering and supporting the strengths of PLWD (Lin & Lewis, 2015). This lens favors communication and behaviors that encourage PLWDs’ life engagement, future goal-setting, and dignity. Our work was also guided by Personhood Theory (Kitwood, 1997), which asserts that, despite cognitive decline, PLWD retain a sense of self, a range of emotional responses to events, and a desire for active interaction with their surroundings.

First, the analysis team developed a working set of research questions. Then, two coders (EM and RA) completed a process of data familiarization using the whole, original dataset (N = 20 participants; 503 captions). They recorded memos and used these observations to refine the research questions. This process also helped the study team to ensure saturation of ideas prior to formal analysis. At this point, these coders discussed their positionality, including how their personal backgrounds may guide study findings.

Next, these two coders iteratively analyzed the captions from the high perspective-taking change subgroup (n = 5; 120 captions), producing initial codes that were used to generate draft categories through discussion, case comparison, and observation of category saturation. Then, these coders analyzed the captions from the low perspective-taking change subgroup (n = 8; 190 captions) in the same way. The researchers were careful not to develop categories based on direct comparison of subgroups (i.e., “this subgroup is exhibiting this behavior more than the other subgroup…”); rather, categories were generated based on discrete observations within subgroups. Draft categories from each separate analytic process were then compared, and overlapping categories across subgroups were combined and refined. At this stage, any contrasting findings across subgroup categories helped to refine within-subgroup categories, and observations about these contrasting findings were documented as memos. Through this process, the two coders developed final category names and descriptions. The third team member (JW) acted as a validator, enhancing credibility by applying the drafted codebook back to the subgroups to confirm that the categories reflected the study data.

Results

Participant characteristics2 and baseline burden and perspective-taking scores for each subgroup are presented in Table 1. T-tests revealed no differences in the number of captions produced by participants in the high perspective-taking change subgroup (M = 23.75 captions; SD = 10.48) compared to the low perspective-taking change subgroup (M = 24.00 captions; SD = 5.70; t (11) = −0.05; p = .96), and no differences in the average number of words within captions produced by the high perspective-taking change participants (M= 13.39 words per caption; SD = 5.02) compared to the low perspective-taking change participants (M = 15.84 words per caption; SD = 9.10; t (11) = −0.63; p = .27). Similarly, t-tests revealed no differences in terms of baseline caregiving burden, t (11) = −0.39; p = .71, or caregiver perspective-taking, t (11) = 1.78; p = .10. Below, we present our qualitative findings paired with exemplar quotes, which are presented in quotations with participant ID numbers. See Figure 1 for a visual representation of findings.

Table 1.

Descriptive Characteristics of Caregivers in High- and Low Perspective-Taking Change Subgroups

Characteristic High change subgroup (n = 5) n (%)/M (SD) Low change subgroup (n = 8) n (%)/M (SD)
Gender
 Man 0 (0%) 1 (13%)
 Woman 5 (100%) 7 (88%)
Age 56.2 55.75
Race
 Black 3 (60%) 1 (13%)
 Hispanic 0 (0%) 1 (13%)
 White 1 (20%) 6 (75%)
 Mixed race 1 (20%) 0 (0%)
Care recipient’s relationship to caregiver
 Spouse 1 (20%) 4 (50%)
 Parent 4 (80%) 3 (38%)
 Grandparent 0 (0%) 1 (13%)
Caregiver’s educational background
 High school or equivalent 0 (0%) 2 (25%)
 Some college without degree 2 (40%) 0 (0%)
 Associate’s degree 1 (20%) 0 (0%)
 Bachelor’s degree 1 (20%) 1 (13%)
 Master’s degree 0 (0%) 5 (63%)
Baseline caregiving burden 47.80 (10.11) 44.63 (16.23)
Baseline caregiver perspective-taking 18.00 (3.08) 21.75 (3.99)
Open in a new tab

Note. One participant in the high change subgroup did not report their educational background, and as such, their information is not included. Race was recorded as an aggregate representation of race and Hispanic ethnicity.

Figure 1.

Figure 1.

Open in a new tab

Categories Generated From Analysis of Photo Captions of Caregivers who Experienced High Perspective-Taking Change and Low Perspective-Taking Change Across the Intervention.

Note. Captions are Written by the Participant and From the Perspective of the Person Living With Dementia (PLWD)

Photo Captioning by Participants who Experienced High Perspective-Taking Change

Analysis generated approaches to photo captioning that were used by caregivers who reported a high, positive change in perspective-taking. These categories described how participants (1) maintained first-person, care-recipient focused language, (2) highlighted care recipients’ engagement and intention, and (3) illustrated care recipients’ resilient living.

Maintenance of First-Person, Care-Recipient Focused Language.

Participants typically constructed full sentences that included subjects and verbs, for example, “I am very tired. I couldn’t sleep last night” (participant 21). They adopted a first-person perspective consistently in their captions, narrating experiences from the perspectives of the PLWD, as demonstrated by participant 23, “I want to go to bed, it’s that way, not in this chair!” Of note, rather than focusing on only positive moments, this person-focused approach included giving voice and insights to PLWDs’ more challenging or emotionally ambiguous experiences, as with participant 16’s caption, “Smile? Okay. This is the best I got right now. It’s early!” While the events described in captions varied across these participants, these structural elements were relatively consistent.

Highlighting Care Recipients’ Engagement and Intention.

Participants used photo captions to capture the ways that PLWDs’ day-to-day activities reflected their broader sense of purpose, life engagement, or thoughtful intention, as with participant 13, who wrote of their grandmother: “I think I can fit my drink on my rollator and push it to my room without spilling. Will (participant’s name) yell at me if I do spill? I am going to try it!” Captions showcase participants’ ambitions and directions from past experiences to current actions and into future planning, as with participant 23’s caption, “Sharing my feelings with my family always helps make the day better. Feeling blessed.”

Through the captions they wrote, participants characterized their PLWD as planful and purposeful by, for example, suggesting that PLWDs use personal goals and values to direct their engagement with the world around them, as with participant 9’s caption, “Time to take my pills. I don’t like taking pills, but I know they’re for my good health.” While actions could sometimes depict inflexibility or rebellion, participants were careful to describe even these actions in ways that reflected PLWDs’ right to autonomy and reasonable pursuit of preferences. For example, participant 21 wrote from the perspective of their mother, “I got a new bed and I’m going to transfer my things how I want to, not how they want me to. I’m not being stubborn, I just want what I want.” Participants used word choice, punctuation, and sentence structure to convey PLWDs’ enthusiasm, life engagement, and planfulness across photo captions.

Illustration of Resilient Living.

Captions depicted PLWDs’ responses to dynamism in their everyday lives, in other words, their approaches to addressing challenges that occur during a range of life events. These captions appear to embody the perspectives of the PLWD, capturing difficult moments in tandem with PLWDs’ efforts to resolve or endure difficulties, as in participant 13’s caption, “I don’t want to fall! Why can’t I keep my balance! I am going to call the doctor and get out the physical therapist again. My leg is completely numb.” In this example, the participant takes the PLWD’s point-of-view to convey the distress the PLWD faces with mobility problems, while also ascribing a solution-focused coping response by describing their plans to integrate additional medical care and alleviate symptoms. In these captions, participants sometimes describe PLWDs being thwarted in-the-moment, but persistent in trying again, as with participant 16’s caption, “My legs feel like jelly. Ok. Let me hold onto to this.”

Participants sometimes chose to focus their captions on capturing distressing moments for the PLWD, even if the PLWD’s perspective in-the-moment depicted the participant themselves negatively. For example, participant 21 wrote, “I took my pill already for today, there’s no way I will be forced to take another dose, they want me to overdose,” where the participant acknowledges the PLWDs’ real fears, despite the participant’s presumed lack of harmful intentions. Participant 13 goes further, depicting their mother’s terror and anger, directed at the participant themself, when the PLWD was hospitalized against their will, “Where am I going? Who are these people? Why am I strapped in? HELP ME!” In these cases, participants appear able to reconcile two versions of reality: their own version, where the choices they make are intended to support the PLWD’s well-being, and their PLWD’s version, which validates the emotional complexity of navigating health challenges and dementia symptoms.

Photo Captioning by Participants who Experienced Low Perspective-Taking Change

Analysis also generated categories describing approaches to photo captioning that were used by caregivers who reported low or no gains in perspective-taking. These categories describe that participants (1) offered inconsistent, incomplete phrasing in captions, (2) provided generic, detached descriptions, and (3) adopted negative, critical tone. Each category is described below, paired with illustrative quotes from captions.

Offering Inconsistent, Incomplete Phrasing.

Participants shifted between a first-person point-of-view from the perspective of the PLWD (in alignment with study instructions, e.g., participant 5, “I enjoy doing my puzzles”), a first-person point-of-view from the perspective of the participant themself (e.g., participant 8, “Looking at me take her picture”), and no distinguishable point-of-view, where captions included action phrases with no subject (e.g., participant 6, “Ready for dinner”) or a vague subject (e.g., participant 15, “Can you back up some”). Captions were sometimes brief or truncated, without clear links between the PLWD’s viewpoint and an action or experience occurring at the time, as with this caption written by participant 15, “3 hours later she is in the same spot.” Some participants offered captions that were longer or more complete at the beginning of the intervention, but then reduced their captions down to fewer words, sometimes without a subject, over time.

Providing Generic, Detached Descriptions.

Participants used captions to describe in-the-moment events, activities, or emotions experienced by PLWD, often with no deeper reflection of the PLWD’s expectations or goals (e.g., participant 15, “I love to eat”). These captions appeared to simply state what the participant saw in the photo or through the camera lens (e.g., participant 6, “Just chillin’ after dinner watching some funny television”). In these cases, the caregiver seemed focused on providing an accurate description of the moment, rather than a speculation about any deeper connections between what might be occurring in the moment and the PLWD’s thoughts, feelings, motivations, or identity. This is illustrated in participant 2’s caption, “Lunch is good, I’m enjoying scrolling while I eat. I feel calm,” where the PLWD is depicted as content and in touch with her emotions, but only described as engaging with her day on a surface level. Sometimes, these captions reinforce stereotypes of what living with dementia can look like by emphasizing the mundane and confusing aspects of life, for example, as written by participant 12, “I don’t know what to do with all this. I am tired of puzzling out this puzzle.” In some cases, this emphasis of the mundane was observed because participants would submit similar or repeated captions more than once during the intervention, implying a flat repetitiveness in the PLWD’s life (e.g., 6 of the 11 captions submitted by participant 12 involved depictions of the PLWD engaging with a puzzle).

Adopting a Negative, Critical Tone.

While participants described PLWD engaging in both positive and negative events or activities, the captions often conveyed neutral or negative tones and emphasized PLWDs’ mistakes, grumpiness, or disengagement, as in this caption from participant 10, “Not sure why I’m up, it’s too early.” Participants appear to be using these captions in some cases to reinforce unappealing qualities exhibited by the PLWD, stymieing any exploration of PLWDs’ perspectives or deeper needs. For example, on two occasions participant 6 wrote that the PLWD was, “arguing about something,” and “perturbed because I told her not to turn off the lights,” portraying the PLWD as both disagreeable and agitational. Some captions conveyed the participants’ disapproval of PLWDs’ emotions or challenging behaviors, portraying the PLWD as someone choosing to be uncooperative. For example, participant 15 wrote the caption, “I can pretend to be nice SOMETIMES,” implying the PLWD’s strained efforts to avoid a default unpleasant attitude.

Photo Captioning Approaches Used by all Participants

Analyses revealed two approaches to photo captioning that were adopted by participants across both subgroups. Specifically, participants (1) connected emotions to events and (2) recorded wonders, wants, and wishes. While all caregivers adopted these approaches, we noted that these approaches were enriched for caregivers who had reported experiencing high perspective-taking change due to their engagement in other captioning approaches.

Connecting Emotions to Events.

Participants consistently described their PLWDs’ emotions as extensions of their descriptions of the activity or experience with which PLWD engaged. They often wrote about what occurred in the day, how the participant felt about it, and why they might feel this way (e.g., participant 5 wrote, “I really am looking forward to going to the store, but my knee hurts and I’m tired”). By connecting emotions with current or recent events, participants seemed to reflect their understanding that the PLWD made these connections themselves throughout their days. For participants who experienced high changes in perspective-taking, these connections between emotions and events sometimes overlapped with highlighting engagement and intention, resulting in captions that illustrated how smaller actions or experiences, even from prior days, could feed into bigger emotional states of mind and expectations for the future. For example, participant 13, who had reported high perspective-taking change, wrote, “I like coming to the chiropractor. I hope he can help me today. Maybe I’ll be able to go a day without falling down and can walk on my own!” Participant 10, who experienced low perspective-taking change, wrote, “Not feeling too good tonight. Think I’m going to go to bed early,” depicting links between feelings and actions without a broader connection to precursors for these feelings, or future consequences of these actions.

Recording Wonders, Wants, and Wishes.

Finally, captions from both subgroups showed evidence of participants viewing PLWDs as inquisitive or introspective through writing about their internal wonders, wants, and wishes. In some cases, this was illustrated by participants using the caption to suggest a question that could be running through the PLWD’s mind at the time the photo was taken, for example, as with participant 16, who wrote, “Why do I have on all these clothes?” For participants who reported a high change in perspective-taking, instances of wonders, wants, and wishes sometimes overlapped with participants’ efforts to illustrate resilient living, enriching the captions. For example, participant 13 wrote of their mother, “I am going to take this medicine but I’m not going to tell [participant’s name] that it is making me feel tired and weak. Did the Dr. say that I should take it before or after I eat? I am going to take this because I don’t want another blood clot in my bladder.” This participant’s caption illustrates their speculation about the PLWDs’ wish to avoid another blood clot, their want to avoid revealing side effects to the caregiver, and their wonder about what their doctor’s recommendation had been. In combination, this content depicts this PLWD as someone who will overcome discomfort and uncertainty to attain a health goal. Participants who reported low perspective-taking change offered wonders, wants, and wishes that mainly corresponded to upcoming tasks (e.g., participant 10, “I’m hungry! Wonder what’s for dinner. Also, I’m dizzy ” participant 6, “Wondering about doing the dishes”).

Discussion

In this study, we characterized the content of photo captions from caregivers of PLWD who experienced high perspective-taking gains and those who experienced low perspective-taking gains, across their engagement in a 10-day photo captioning intervention. No differences were found between these subgroups in terms of the number of captions participants provided or the number of words per caption. Baseline self-report surveys also indicated that the caregivers in the high perspective-taking gains group were not naturally better at perspective-taking compared to those in the low gains group. Yet, the content of captions from each subgroup suggests distinct patterns in the ways caregivers approached the task of “capturing the inner voice of the PLWD” and suggests that the approaches used by caregivers who experienced the most perspective-taking change may have contributed to their favorable outcomes. Our sensitizing frameworks, emphasizing dementia positivity (Lin & Lewis, 2015) and personhood (Kitwood, 1997), shaped our interpretation of our findings in this section.

Participants who experienced high perspective-taking change provided captions that maintained a first-person, PLWD point-of-view and included full sentences with subjects and verbs. This stood in contrast to the captions from low perspective-taking change participants. These elements, the point-of-view and the sentence structure, were addressed in the instructions that all participants receive (Rilling et al., 2024). By captioning photos from the wrong point of view, participants may miss opportunities to consider what the PLWD may be thinking or feeling. By captioning photos with sentence fragments, participants sometimes leave the point-of-view ambiguous, acknowledging general events or sentiments without differentiating their own thoughts and feelings from those of the PLWD. These approaches can undermine the potency of interventions; indeed, when people reflect on what they might do or feel, versus imagining how another person feels, they can become distracted by their own emotional needs (Batson et al., 1997; Lamm et al., 2007).

Captions written by participants who experienced high perspective-taking change highlighted PLWDs’ engagement with life, intentionality, and resilience in the face of challenges. These captions were not uniformly positive in tone; many depicted PLWDs’ negative emotions and acknowledged functional limitations, emergency situations, and constraints on PLWDs’ autonomy. Indeed, participants illustrated these challenges in their captions by constructing small stories that linked PLWDs’ present circumstances to their broader ambitions, reflected important aspects of their identity. This approach suggests that participants are avoiding or overriding stereotypes about what living with dementia may entail, a hallmark of a dementia-positive worldview (Lin & Lewis, 2015). This approach has been demonstrated experimentally to lead to increased empathy and recognition that behaviors which are typically unacceptable may be understandable responses to challenging external circumstances (Bientzle et al., 2021).

Perhaps our most striking finding was that some caregivers captioned photos in ways that depicted the caregiver themselves in a negative light. These captions represent perspective-taking enhanced by constructive self-awareness (Decety & Jackson, 2006): by writing captions that are sometimes self-critical, participants monitor their own caregiving thought processes and limits, distinguish their own feelings and intentions from those of the PLWD, and deeply acknowledge the challenges of living with dementia.

Ultimately, successful engagement in perspective-taking relies on both motivation and skill (Gehlbach & Mu, 2023). This photo captioning intervention offered an opportunity for caregivers to find the motivation, and practice the skill, of taking the perspective of their PLWD, thereby improving their perspective-taking abilities. Those who exhibited low change in perspective-taking may not have fully found motivation, or may not have appropriately practiced the skill, and therefore did not experience benefits to the same extent as other caregiver participants. Specifically, some captions were structured around criticizing the PLWD, rather than truly exploring their inner voice. Caregivers who leaned into this approach to captioning may be grappling with challenging feelings about their caregiving roles or harboring harsh perceptions of the PLWD (Bloom et al., 2023; Shim et al., 2012). They may need additional support to address those feelings before being fully motivated to engage in perspective-taking activities. In addition, some captions provided generic, detached descriptions of the PLWD’s thoughts, feelings, and actions, adopting an approach to captioning that could be similarly adopted by a person with no prior relationship with the PLWD. These captions did not convey a sense of knowing or understanding the PLWD, thus potentially representing missed opportunities to practice perspective-taking skills.

That said, we recognize that some caregivers may have been unable to fully engage with the intervention for other reasons. Caregivers’ self-perceived burden is one common barrier to engagement in behavioral interventions (Waligora et al., 2019). Baseline reports of caregiving burden in this study indicate that caregivers across sub-groups were similarly burdened by their role, so it is unlikely that one subgroup of caregivers was more overwhelmed by caregiving than another. However, some caregivers may not have had the attention capacity, or the available time, to thoughtfully engage with the intervention, despite motivation to do so. Future trials of this intervention could measure caregivers’ time spent, and engagement, during the intervention, as well as offering resources for caregivers to build time into their schedules for this and other helpful interventions.

Limitation and Future Directions

For this study, we did not analyze captioned photos themselves but recognize that caregivers’ approaches to photo taking (e.g., time of day and aptitude with smartphone cameras) may also influence their experiences with this intervention. Also, the direct associations between photo captioning approaches and intervention outcomes should be analyzed in a larger sample of caregivers to determine which photo captioning approaches have the most influence on intervention outcomes.

Given that some captions refuted stereotypes about living with dementia, it is possible that participating in this intervention can shift caregivers’ perceptions of their own potential future cognitive changes. Experiencing high rates of burden while caregiving for a PLWD is a risk factor for developing negative views on aging (Sabatini et al., 2025), with some caregivers developing maladaptive personal health outlooks and behaviors as a consequence of caregiving experiences (Mroz et al., 2024). By guiding caregivers to caption photos with positive, yet still realistic, depictions of daily living with dementia, this intervention may have a bonus effect of reducing stereotype-focused thinking (Galinsky & Moskowitz, 2000), curbing exacerbation of negative aging views, and promoting positive health outcomes for caregivers.

Conclusions

This study characterizes caregivers’ approaches to captioning photos during a cognitive empathy intervention and suggests that these approaches may be important behavioral mechanisms of intervention impact. While most caregivers experienced positive outcomes from engaging in this intervention, this work provides context for variation in this and other perspective-taking interventions’ impacts. Given our findings, it may be critical to coach caregivers who engage in perspective-taking interventions to consistently adopt the PLWD’s point-of-view, create small narratives that link daily living experiences to the PLWD’s identity, goals, and values, and infuse self-awareness to recognize when well-intended caregiving efforts may fall short.

What this paper adds

  • This study characterizes behavioral mechanisms that may support effective participation in a perspective-taking intervention for caregivers of PLWD.

  • Our study demonstrates that perspective-taking gains during a photo captioning intervention may be fostered by participants’ caption content, rather than the length or amount of captions they generate.

  • Our findings suggest that constructive photo captioning can promote self-reflection and refute negative stereotypes about dementia in addition to promoting perspective-taking.

Applications of study findings

  • Our findings contribute to a growing body of literature on the mechanisms that promote the impact of gerontological interventions.

  • Our findings can be used to refine perspective-taking interventions to optimize effects on caregiver cognitive empathy and mental health.

  • These findings can also promote examination of how perspective-taking interventions can bolster caregivers’ views on dementia and their own aging.

Funding

This work was supported by the Emory Roybal Center for Dementia Caregiving Mastery (National Institutes of Health P30AG064200), the Emory University Alzheimer’s Disease Research Center (National Institutes of Health P50AG025688), and the Emory Center for Health in Aging. Dr Mroz is supported by the National Institute on Aging, Grant No. K22 AG088144–01.

Footnotes

1.

Caregiver burden was assessed using aggregate scores from The Zarit Burden Interview. Perspective-taking was assessed using a subscale from the Interpersonal Reactivity Index, as described in the introduction. Full description of these questionnaires can be found in this existing publication (Rilling et al., 2024).

2.

Participant demographic characteristics are presented for description only. Demographically, the two groups appeared different on several dimensions, but due to the sample size, no conclusions can be drawn through quantitative comparison.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

IRB Approval

Institutional Review Board Approval Number: 00001824.

References

  1. Batson CD, Early S, & Salvarani G (1997). Perspective taking: Imagining how another feels versus imaging how you would feel. Personality and Social Psychology Bulletin, 23(7), 751–758. 10.1177/0146167297237008 [DOI] [Google Scholar]
  2. Bientzle M, Eggeling M, Kanzleiter M, Thieme K, & Kimmerle J (2021). The impact of narrative writing on empathy, perspective-taking, and attitude: Two randomized controlled experiments on violations of Covid-19 protection regulations. PloS One, 16(7), Article e0254501. 10.1371/journal.pone.0254501 [DOI] [Google Scholar]
  3. Bloom R, Osso F, McCarthy M, & Falzarano F (2023). Motivations and role captivity among informal caregivers to community-dwelling persons with dementia. Innovation in Aging, 7(Suppl 1), 745. 10.1093/geroni/igad104.2411 [DOI] [Google Scholar]
  4. Creswell JW, & Clark VLP (2017). Designing and conducting mixed methods research. Sage publications. [Google Scholar]
  5. Davis MH (1983). Measuring individual differences in empathy: Evidence for a multidimensional approach. Journal of Personality and Social Psychology, 44(1), 113–126. 10.1037/0022-3514.44.1.113 [DOI] [Google Scholar]
  6. Decety J, & Jackson PL (2006). A social-neuroscience perspective on empathy. Current Directions in Psychological Science, 15(2), 54–58. 10.1111/j.0963-7214.2006.00406.x [DOI] [Google Scholar]
  7. Galinsky AD, & Moskowitz GB (2000). Perspective-taking: Decreasing stereotype expression, stereotype accessibility, and ingroup favoritism. Journal of Personality and Social Psychology, 78(4), 708–724. 10.1037//0022-3514.78.4.708 [DOI] [PubMed] [Google Scholar]
  8. Gaugler JE (2022). Unpaid dementia caregiving: A policy and public health imperative. Public Policy & Aging Report, 32(2), 51–57. 10.1093/ppar/prac002 [DOI] [Google Scholar]
  9. Gehlbach H, & Mu N (2023). How we understand others: A theory of how social perspective taking unfolds. Review of General Psychology, 27(3), 282–302. 10.1177/10892680231152595 [DOI] [Google Scholar]
  10. Gitlin LN, Corcoran M, Martindale-Adams J, Malone C, Stevens A, & Winter L (2000). Identifying mechanisms of action: Why and how does intervention work. In: Handbook on dementia caregiving: Evidence-based interventions for family caregivers (pp. 225–248). Springer Publishing Company. [Google Scholar]
  11. Han A, Kim TH, & Hong H (2021). A factorial randomized controlled trial to examine separate and combined effects of a simulation-based empathy enhancement program and a lecture-based education program on family caregivers of people with dementia. Aging & Mental Health, 25(10), 1930–1940. 10.1080/13607863.2020.1768214 [DOI] [PubMed] [Google Scholar]
  12. Huo M, Vu T, Yoneda T, Wei J, Abbott B, & Monin JK (2025). Empathy and psychological outcomes in informal caregivers of community-dwelling older adults: A systematic review and meta-analysis. The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences, 80(6), gbaf018. 10.1093/geronb/gbaf018 [DOI] [Google Scholar]
  13. Jütten LH, Mark RE, & Sitskoorn MM (2019). Empathy in informal dementia caregivers and its relationship with depression, anxiety, and burden. International Journal of Clinical and Health Psychology: IJCHP, 19(1), 12–21. 10.1016/j.ijchp.2018.07.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Kitwood T (1997). Dementia Reconsidered the person comes. first Open University Press. Maidenhead. [Google Scholar]
  15. Kripal SJ, & Reiter-Palmon R (2024). The role of empathy in problem construction and creative problem solving. Learning and Individual Differences, 114(4), Article 102501. 10.1016/j.lindif.2024.102501 [DOI] [Google Scholar]
  16. Lamm C, Batson CD, & Decety J (2007). The neural substrate of human empathy: Effects of perspective-taking and cognitive appraisal. Journal of Cognitive Neuroscience, 19(1), 42–58. 10.1162/jocn.2007.19.1.42 [DOI] [PubMed] [Google Scholar]
  17. Lin S-Y, & Lewis FM (2015). Dementia friendly, dementia capable, and dementia positive: Concepts to prepare for the future. The Gerontologist, 55(2), 237–244. 10.1093/geront/gnu122 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Lovell B, Newman A, & Wetherell MA (2023). Seeing it my way: A perspective taking intervention alleviates psychological distress in caregivers of autistic children. Research in Developmental Disabilities, 133(5), Article 104396. 10.1016/j.ridd.2022.104396 [DOI] [Google Scholar]
  19. Mroz EL, Ali T, Piechota A, Matta-Singh TD, Abboud A, Sharma S, Monin JK, & Fried TR (2024). Personal health planning in adult-child former caregivers of parents living with dementia. American Journal of Health Promotion: AJHP, 38(3), 402–411. 10.1177/08901171231204670 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Onwuegbuzie AJ, & Leech NL (2007). Sampling designs in qualitative research: Making the sampling process more public. Qualitative Report, 12(2), 238–254. [Google Scholar]
  21. Rilling JK, Lee M, McIsaac J, Factor S, Gallagher P, Kim JH, Zhang J, Zhou C, McDade TW, Hepburn K, & Perkins MM (2024). Evaluation of a photo captioning cognitive empathy intervention for dementia caregivers. Clinical Gerontologist, 47(5), 1–14. 10.1080/07317115.2024.2317972 [DOI] [PubMed] [Google Scholar]
  22. Sabatini S, Turner SG, & Stephan BC (2025). Correlates of self-perceptions of aging in dementia caregivers: Findings from the German aging survey. Aging & Mental Health, 29(7), 1–13. 10.1080/13607863.2025.2464705 [DOI] [PubMed] [Google Scholar]
  23. Sandelowski M (2000). Whatever happened to qualitative description? Research in Nursing & Health, 23(4), 334–340. 10.1002/1098-240x(200008)23:4<334::aid-nur9>3.0.co;2-g [DOI] [PubMed] [Google Scholar]
  24. Shim B, Barroso J, & Davis LL (2012). A comparative qualitative analysis of stories of spousal caregivers of people with dementia: Negative, ambivalent, and positive experiences. International Journal of Nursing Studies, 49(2), 220–229. 10.1016/j.ijnurstu.2011.09.003 [DOI] [PubMed] [Google Scholar]
  25. Sörensen S, & Conwell Y (2011). Issues in dementia caregiving: Effects on mental and physical health, intervention strategies, and research needs. The American Journal of Geriatric Psychiatry: Official Journal of the American Association for Geriatric Psychiatry, 19(6), 491–496. 10.1097/JGP.0b013e31821c0e6e [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Suri H (2011). Purposeful sampling in qualitative research synthesis. Qualitative Research Journal, 11(2), 63–75. 10.3316/qrj1102063 [DOI] [Google Scholar]
  27. Thompson NM, Van Reekum CM, & Chakrabarti B (2022). Cognitive and affective empathy relate differentially to emotion regulation. Affective Science, 3(1), 118–134. 10.1007/s42761-021-00062-w [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Tong A, Sainsbury P, & Craig J (2007). Consolidated criteria for reporting qualitative research (COREQ): A 32-item checklist for interviews and focus groups. International Journal for Quality in Health Care: Journal of the International Society for Quality in Health Care, 19(6), 349–357. 10.1093/intqhc/mzm042 [DOI] [PubMed] [Google Scholar]
  29. Vaismoradi M, Turunen H, & Bondas T (2013). Content analysis and thematic analysis: Implications for conducting a qualitative descriptive study. Nursing & Health Sciences, 15(3), 398–405. 10.1111/nhs.12048 [DOI] [PubMed] [Google Scholar]
  30. Waligora KJ, Bahouth MN, & Han H-R (2019). The self-care needs and behaviors of dementia informal caregivers: A systematic review. The Gerontologist, 59(5), e565–e583. 10.1093/geront/gny076 [DOI] [PubMed] [Google Scholar]
  31. Wijma EM, Veerbeek MA, Prins M, Pot AM, & Willemse BM (2018). A virtual reality intervention to improve the understanding and empathy for people with dementia in informal caregivers: Results of a pilot study. Aging & Mental Health, 22(9), 1121–1129. 10.1080/13607863.2017.1348470 [DOI] [Google Scholar]
  32. Zarit SH, Reever KE, & Bach-Peterson J (1980). Relatives of the impaired elderly: Correlates of feelings of burden. The Gerontologist, 20(6), 649–655. 10.1093/geront/20.6.649 [DOI] [PubMed] [Google Scholar]

RESOURCES