Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Oct 16.
Published in final edited form as: Clin Gerontol. 2024 Feb 19;47(5):832–845. doi: 10.1080/07317115.2024.2317972

Evaluation of a Photo Captioning Cognitive Empathy Intervention for Dementia Caregivers

James K Rilling a,b,c,d,e, Minwoo Lee f, Julie McIsaac g, Sophie Factor f, Paige Gallagher f, Joseph H Kim f, Jiajin Zhang f, Carolyn Zhou a, Thomas W McDade h, Kenneth Hepburn i, Molly M Perkins j
PMCID: PMC11331024  NIHMSID: NIHMS1988037  PMID: 38372125

Abstract

Objectives:

The goal of this study was to develop and evaluate an intervention aimed at increasing cognitive empathy, improving mental health, and reducing inflammation in dementia caregivers, and to examine the relevant neural and psychological mechanisms.

Methods:

Twenty dementia caregivers completed an intervention that involved taking 3–5 daily photographs of their person living with dementia (PLWD) over a period of 10 days and captioning those photos with descriptive text capturing the inner voice of the PLWD. Both before and after the intervention, participants completed questionnaires, provided a blood sample for measures of inflammation, and completed a neuroimaging session to measure their neural response to viewing photographs of their PLWD and others.

Results:

87% of enrolled caregivers completed the intervention. Caregivers experienced pre- to post-intervention increases in cognitive empathy (i.e. Perspective-Taking) and decreases in both burden and anxiety. These changes were paralleled by an increased neural response to photographs of their PLWD within brain regions implicated in cognitive empathy.

Conclusion:

These findings warrant a larger replication study that includes a control condition and follows participants to establish the duration of the intervention effects.

Clinical Implications:

Cognitive empathy interventions may improve caregiver mental health and are worthy of further investigation.

Keywords: Anxiety, burden, fMRI, inflammation

Introduction

Currently, approximately 6.7 million Americans are living with Alzheimer’s Disease, and this is projected to increase to almost 13 million by the year 2050. Many people living with dementia are cared for by unpaid family caregivers. Today, there are over 11 million unpaid dementia caregivers in the U.S., and this number is expected to increase to meet the demands of a growing population of people with dementia (“2023 Alzheimer’s disease facts and figures,” 2023). Dementia caregivers are confronted with a number of stressors that include managing the behavioral and psychological symptoms of dementia as well as assisting with activities of daily living. As a result, caregivers experience significant burden and high rates of depression and anxiety (Bin Sallim et al., 2015; Joling et al., 2015; Pinquart & Sorensen, 2003). A range of different interventions have been developed with the aim of decreasing caregiver burden and improving caregiver well-being, and many of these have shown efficacy. These include psychoeducational programs, psychotherapy, and mindfulness-based interventions (Cheng et al., 2019).

One factor that has been associated with lower levels of subjective stress and depression among dementia caregivers is high levels of cognitive empathy (Jutten et al., 2019; Lee et al., 2001). Cognitive empathy refers to the ability to adopt another individual’s mental perspective and to understand what they are thinking or feeling. A substantial literature shows that cognitive empathy, often referred to as reflective functioning, benefits other types of caregiving relationships such as the parent-child relationship. Reflective functioning in parents is associated with competent caregiving and strong parent-child relationships (Camoirano, 2017). Reflective functioning is believed to promote competent caregiving by placing thought, understanding and planning between impulse and action. Collectively, these studies raise the prospect that increasing cognitive empathy in dementia caregivers could improve both caregiver mental health and the quality of care they deliver.

In a recent study aimed at increasing cognitive empathy in fathers of 2–5 year old children, fathers were asked to caption photos in the voice of their child: to express with text what they felt their child was thinking, feeling or experiencing during that moment. Fathers collected 3–5 photos per day for a 10-day period. Data were gathered via observations, interviews pre- and post the intervention, and the photo captions themselves. Themes and categories emerged during the rigorous process of iteratively coding, categorizing, and analytically reflecting upon the data. Data analysis of the father’s captions and interviews revealed that the experience of the father during this photo captioning activity resulted in improved interactional patterns with their children, harnessing the power of reflection, empathy, and mindfulness (McIsaac, 2021).

In the current project, our goal was to implement this same photo captioning intervention in caregivers of people living with dementia (PLWD), and to collect pilot data regarding its feasibility, tolerability and effectiveness in improving caregiver emotional well-being. Given that caregiving has also been associated with increased levels of proinflammatory biomarkers that mediate many of the chronic diseases of aging (Kiecolt-Glaser et al., 2003; Roth et al., 2019), and that may even contribute to depressive symptomology (Raison et al., 2006), we also examined intervention effects on inflammation. Furthermore, we sought to investigate the psychological and neural mechanisms of any intervention effects. We did so by measuring changes in self-reported empathy and by imaging caregiver brain function with fMRI as they viewed photos of their PLWD and attempted to empathize with them. We focused our fMRI analysis on three a-priori brain regions of interest that have been consistently implicated in cognitive empathy: the dorsomedial prefrontal cortex (DMPFC), the precuneus, and the right temporo-parietal junction (rTPJ) (Molenberghs et al., 2016).

Methods

All procedures were approved by the Emory University Institutional Review Board (protocol # 00001824).

Participants

People living with dementia (PLWD) and their family caregivers were recruited for this study via word of mouth, physical flyers, Facebook advertisements and on-line recruitment (Schlesinger Group | Qualitative & Quantitative Research Services). To target the intervention at caregivers experiencing high levels of caregiving burden, prospective caregiver participants were screened with the Zarit Burden Scale, and only those scoring 24 or higher were eligible. A total of 23 pairs were enrolled. An additional 23 caregivers were screened but not enrolled because their Zarit Burden Scale score was below 24. Three enrolled caregivers did not complete the study, resulting in a total of twenty caregivers with complete data.

Caregivers who were interested in the study contacted researchers to receive additional information regarding the project. Researchers then provided complete study details, and participants that wished to continue with the study completed a consent form using REDCap (https://www.project-redcap.org).

Persons living with dementia (PLWD) were consented via Zoom with their caregiver present. They were told that their caregiver would take 3–5 photographs of them each day for 10 days, and that the caregiver would caption each photograph with what the caregiver believed the PLWD was thinking and feeling in each photo. We used NIH guidelines to assess the ability of persons with dementia to provide informed consent (NIH, 2009). We approached potential participants with cognitive impairment about the study, explained the study, including our approach, risks, and benefits (NIH, 2009), and following Palmer et al. (Palmer et al., 2005), asked participants: “(1) What is the purpose of the study? (2) What are the risks? and (3) What are the benefits?.” For those who were unable to answer these questions and provide informed consent, we used proxy consent along with established assent procedures (Black et al., 2010; Slaughter et al., 2007). In proxy consent, consent is obtained from a legally authorized representative on behalf of a person who is incapable of providing informed consent. In our case, written proxy consent was provided by the caregiver, who was also a family member. After obtaining proxy consent, participant assent was also obtained by the researcher who identified herself/himself, explained the study, and asked the person with dementia for permission to ask some questions (see (Black et al., 2010; Slaughter et al., 2007). Data were only collected from individuals who assented.

An additional group of 37 non-caregiver control participants were also recruited for comparison with caregivers to determine if there were any baseline differences in mental health or inflammation.

Study design

Both before and after the photo captioning intervention, all caregivers completed questionnaires, provided a bloodspot sample for measures of inflammation, and completed a neuroimaging session to measure their neural response to viewing photographs of their PLWD and others. All caregivers completed their post-intervention data collection session within one week of the last day of the intervention.

Control participants completed questionnaires, provided blood spot samples and completed a neuroimaging session in which a structural MRI scan was acquired but these MRI data are not reported here. Control participants did not receive a functional MRI scan.

Qualitative data

After the intervention, all caregivers were verbally asked, “Did this intervention have any effect on you or your relationship with your PLWD?,” and responses were transcribed by research staff in real time by typing the responses into an electronic form.

Questionnaires

Both caregivers and controls provided information about age, race, gender, body mass index, years of education, annual household income, weekly alcohol consumption, weekly exercise duration, and the presence/absence of diabetes, hypertension or heart disease.

Caregivers and control participants also completed several questionnaires via REDCap, including: 1) The Perceived Stress Scale (Cohen et al., 1983), 2) The Center for Epidemiological Studies Depression Scale (Radloff, 1977), 3) The State/Trait Anxiety Index (limited to the 20 items assessing state anxiety) (Spielberger, 1983), 4) the Zarit Burden Scale (Zarit et al., 1980), a measure of caregiving burden, 5) The Interpersonal Reactivity Index (Davis, 1983), and 6) the Barrett-Lennard Empathy Scale (BLES) (Barrett-Lennard, 1978).

Capillary blood collection

Caregiver and control participants provided a finger stick capillary dried blood spot (DBS) sample for measurement of C-reactive protein and cytokines IL-6, IL-10 and TNF-α (Mcdade et al., 2007). Research staff used a sterile, single-use micro-lancet to prick the finger of the participant. Between 2 and 5 drops of blood were applied to filter paper. After collection, a bandage was placed on the participant’s finger. Filter paper was then air dried for 4 hours and placed in a ziplock bag with desiccant and then frozen at −30°C until assay.

Neuroimaging session

Prior to the neuroimaging session, PLWD provided photographs of themselves to be used as stimuli for the fMRI scan of the caregiver. PLWD provided one photograph with a happy expression and another with a neutral expression.

Caregivers were positioned in the Siemens Trio 3T MRI scanner. Subjects lay motionless in a supine position in the scanner with padded head restraint to minimize head movement during scanning. Each scanning session began with a 15 s localizer scan, followed by a 5 min T1-weighted MPRAGE anatomical scan (TR = 1900 ms, TE = 2.27 ms, matrix = 256 × 256, FOV = 250 mm, slice thickness = 1.00 mm, gap = 0 mm). After collecting the anatomical scan, functional scans without contrast were acquired. Scans used an EPI sequence with the following parameters: TR = 1200 ms, TE = 30.0 ms, matrix = 74 × 74, FOV = 220 mm, slice thickness = 3.00 mm, gap = 0, 54 axial slices.

During the functional scan, participants viewed images of their PLWD (OP), as well as unknown PLWD of the same race and gender, and of a similar age (UP). They also viewed photographs of a friend or family member whom they do not provide care for, of the same race and gender and of similar age as their PLWD (OF). Participants were instructed to, “try to share the emotions of the person in the picture.” To verify participant attentiveness during scanning, participants were asked to press a button each time they saw their PLWD’s photo. Participants viewed pictures of four different individuals with both happy and neutral expressions (8 pictures total). The four individuals included the own PLWD (OP), two different unknown PLWD (UP1 and UP2) and the friend or family member (OF). Each of the eight pictures were viewed five times. Participants also viewed pictures of eight different objects (kitchen utensils), and each object was viewed five times. Pictures of four different individuals were shown sequentially, followed by pictures of four different objects. This was followed by a rest period with visual fixation. Afterward, participants viewed four different pictures of the same four individuals (with the opposite expression as the first four pictures), and then pictures of four more objects that were not viewed previously, followed by another rest period with visual fixation. This sequence was repeated five times. Each stimulus was presented for 5 seconds, and stimuli were followed by a variable inter-trial interval of 2, 3 or 4 seconds. Total task duration was 11 minutes and 15 seconds.

Intervention

Using their smartphone, caregivers were tasked with taking 3–5 daily photos of their PLWD over a period of 10 days and captioning those photos with descriptive text capturing the inner voice of the PLWD. In other words, the caregivers were asked to put themselves in their person’s shoes and caption the photo with what they thought they were thinking, feeling or experiencing in that moment; to consider what could be driving their behavior or emotional experience. Caregivers were provided with sample captions to help them better understand the exercise. Photos and captions were shared via Microsoft One Drive. A sample photo caption is provided in Figure 1, and additional sample captions are provided in Supplementary Table S1. Each caregiver’s captions were rated for caption quality as either “very good” (3), “adequate” (2) or “lacking” (1), as assessed by two members of the research team who were blind to intervention outcomes.

Figure 1.

Figure 1.

Open in a new tab

Sample photograph and accompanying caption produced by one of the enrolled caregivers.

Inflammation assays

Dried blood spot samples were shipped to the Laboratory for Human Biology Research at Northwestern University (PI: Thomas McDade) where they were assayed for C-reactive protein, IL-6, IL-10 and TNF-α using immunoassay protocols previously validated for use with DBS samples (McDade et al., 2004, 2021). Briefly, for both assays a semi-automated hole punch was used to cut a fixed diameter of sample, which was eluted overnight in buffer and transferred to 96 well assay plates for incubation with capture and detection antibodies. CRP was quantified with a high sensitivity enzyme immunoassay protocol with a lower limit of detection of 0.03 mg/L. Cytokines were quantified on a multiplex electrochemiluminescent immunoassay platform with lower limits of detection of <0.5 pg/mL for IL6 and IL10, and <1.0 pg/mL for TNFα. Due to the clinical significance of CRP measured in plasma, DBS results for CRP were converted to plasma equivalent values using an established formula (McDade et al., 2004).

Statistical analyses

Questionnaire data and cytokine levels were compared between caregivers and controls using a two-sample t-test, with the type-I error rate set to α = 0.05. We report the effect sizes (Cohen’s d) for all significant findings. To control for possible confounds such as BMI and gender, we also conducted multiple linear regressions with caregiver status as the independent variable of interest, and BMI and gender as additional covariates. Given that CRP levels were non-normally distributed, CRP values were log transformed before comparing average levels between groups. We also compared the proportion of caregivers and non-caregivers with CRP levels exceeding 10 mg/L, using a Chi-square test.

Pre- to post-intervention changes in questionnaire and cytokine levels were evaluated using a paired t-test. Pre- to post-intervention changes in CRP levels were highly skewed and were therefore evaluated using the nonparametric Wilcoxon signed rank test.

fMRI analyses were conducted with the Oxford Center for Functional Magnetic Resonance Imaging of the Brain’s software library (FSL, http://www.fmrib.ox.ac.uk/fsl/). The preprocessing pipeline of the fMRI data involved (1) motion correction using the MCFLIRT (Jenkinson et al., 2002), (2) non-brain tissue removal using the Brain Extraction Tool (BET), (3) slice timing correction, (4) high-pass temporal filtering with a cut-off of 100 s, (5) spatially smoothing with a Gaussian kernel of full-width at half maximum (FWHM) of 5 mm, and (6) normalizing to MNI space via corresponding extracted T1 brain using Boundary-Based-Registration (Greve & Fischl, 2009).

Preprocessed fMRI data were analyzed using the general linear model (GLM) for univariate statistical analysis. Regressors were specified for 1) photos of the caregiver’s PLWD, 2) photos of unfamiliar PLWD of the same gender, race and approximate age, 3) photos of a friend or relative that the caregiver does not provide care for, also of the same sex, race and approximate age, and 4) photos of objects. For each individual GLM, we specified contrast between the BOLD response to 1) own PLWD and unknown PLWD presentation periods and 2) own PLWD and friend/relative presentation periods. The individual-level GLM was implemented using FILM (FMRIB’s Improved Linear Model).

At the group level, we performed both region of interest (ROI) and whole brain analyses. We included three a priori ROIs in brain areas involved with cognitive empathy (rTPJ, DMPFC, Precuneus), as well as a control ROI implicated in emotional empathy (anterior insula) for comparison. The DMPFC and anterior insula ROIs were defined based on the Harvard-Oxford cortical structural atlases implemented in FSL (i.e., paracingulate gyrus and insula). To define the anterior insula specifically, the insula mask was limited to slices greater than or equal to Y = 0. The rTPJ and precuneus ROIs were defined functionally by creating a 10 mm-spherical mask around the peak activation coordinates drawn from previous meta-analyses on Theory of Mind (i.e., TPJ (Krall et al., 2015);: Precuneus: (Schurz et al., 2014)). For ROI analyses, contrast values were averaged across all ROI voxels and paired samples t-tests were used to compare these values between pre- and post-intervention scans. For whole brain analyses, pre- and post-intervention scan contrasts were subtracted at the individual level, and voxel-wise one-sample t-tests were conducted to determine where the group average contrast value differed from zero. Given the exploratory nature of our study, whole brain analyses were thresholded liberally using clusters determined by Z > 2.3 (1-tailed p < 0.01), and a family-wise error (FWE)-corrected cluster significance threshold of p < 0.05.

Assessment of intervention feasibility and tolerability

Tolerability of the intervention was assessed as the number of participants who completed the intervention relative to the number who dropped out. Feasibility was assessed by our ability to properly implement the intervention and successfully collect the desired data from caregivers.

Results

Demographics and health status of caregivers and controls

Caregivers averaged 53.20 years of age (SD = 9.34) and were mostly women (18 women and 2 men). The racial distribution of caregivers was as follows: 9 White, 7 Black, 2 Hispanic, 1 Asian, 1 Mixed Race.

Control participants did not significantly differ from caregivers on age (caregiver M = 53.20, SD = 9.34, non-caregiver M = 54.95, SD = 8.96, t(55) =−0.69, p = 0.49), gender (90% vs. 89% female, X2 (1, 57) = 0.009, p=0.92, or racial distribution (27 White, 5 Black, 1 Hispanic, 3 Asian, 1 Mixed Race, X2(4, 57) = 6.14, p=0.19). Caregivers and controls also did not significantly differ with respect to years of education (t(53)=−0.45, p=.64), annual household income (t(51) = 0.87, p=.39), weekly alcohol consumption (t(53) = 0.66, p=.51), weekly exercise duration (t(53)= −0.85, p=.40), or the prevalence of diabetes (X2(1, 57) = 3.01, p=.08), hypertension (X2(1, 57) = 2.17, p=.14) or heart disease (X2(1, 57) = 1.88, p=.17). On the other hand, caregivers had a significantly higher BMI compared with controls (t(32.86) = 2.70, p=.011) (Table 1).

Table 1.

Comparison of demographic and health variables between caregivers and non-caregiver controls.

Caregivers Controls p
Age 53.20 ± 9.34 54.95 ± 8.96 .49
% Female 90% 89% .92
Years of Education 15.89 ± 2.32 16.19 ± 2.32 .64
Annual Income ($) 110,882 ± 62,255 94,471 ± 64,643 .39
BMI 31.20 ± 7.66 26.39 ± 4.80 .01
Alcohol Consumption (drinks/week) 2.22 ± 2.65 1.78 ± 2.11 .51
Exercise (hours/week) 4.83 ± 3.45 5.81 ± 4.23 .4
% with Diabetes 15% 2.70% .08
% with Hypertension 22% 8.82% .14
% with Heart Disease 5% 0% .17
Open in a new tab

Tolerability and feasibility

Twenty of twenty-three enrolled caregivers (87%) completed the full intervention, and we were successful in collecting demographic, mental health, inflammation and fMRI data from caregivers both pre- and post-intervention. Occasionally, caregivers provided captions that did not involve taking the mental perspective of the PLWD. When this happened, research staff would notify caregivers and attempt to clarify with example captions. This procedure was effective in improving the quality of the photo captioning.

Qualitative data

The majority of caregivers responded positively to the question, “Did this intervention have any effect on you or your relationship with your PLWD?” (see Supplementary Table S2). Below are quotes from three different caregivers that emphasize the effectiveness of the intervention at getting caregivers to adopt the mental perspective of their PLWD and at improving the caregiver/care recipient relationship.

It did start me thinking more about what’s going on inside of him (her spouse), and how he is feeling on a day-to-day basis; about what he’s going through. It took the focus off of me and put it on him a few times a day. We had some funny moments together in taking photos, and I think that brought us closer together. This is often a disease of seriousness, and so to laugh together about something was valuable.

[The intervention] made me more attentive to my grandfather’s every-day experience and I feel like I understand his feelings better. I feel that I spend more time to converse with my grandfather. Overall, my relationship with my grandfather seems to be improving.

It did help me be more aware of my mom’s feelings and emotions, which I usually take for granted because I’m with her all the time. I do pay attention to her moods, but I was more aware of them. It did help me to realize that sometimes she is more sensitive, and I need to consider her emotions more than I do instead of getting upset or frustrated. It helped me learn to be understanding.

Questionnaire data

Compared with the control group, caregivers at baseline scored higher on the Perceived Stress Scale (t(59) = 4.50, p < 0.001; d = 1.19) and the state anxiety component of the State-Trait Anxiety Scale (t(59) = 4.15, p < 0.001; d = 1.10). They did not significantly differ from controls on the Center for Epidemiological Studies Depression Scale (t(59) = 1.38, p = 0.17) (Table 2).

Table 2.

Comparison of questionnaire, inflammation and neuroimaging data between caregivers and non-caregiver controls.

Measure Caregiver Baseline (M ± SD) Control Baseline (M ± SD) p Caregiver Post-Pre (M ± SD) p
Questionnaires Perceived Stress 19.65 ± 6.61 12.58 ± 5.51 <.001 −1.55 ± 6.06 .27
STAI 42.35 ± 12.85 30.39 ± 9.55 <.001 −6.35 ± 9.74 .009
Depression Scale 20.17 ± 11.47 16.68 ± 8.29 .17 −2.8 ± 7.93 .13
Zarit Burden Scale 47.55 ± 13.47 n/a n/a −7.15 ± 9.23 .003
IRI PT (Perspective Taking) 19.30 ± 4.26 n/a n/a 1.85 ± 2.76 .007
IRI EC (Empathetic Concern) 22.80 ± 4.01 n/a n/a −.30 ± 2.81 .64
IRI F (Fantasy Scale) 13.15 ± 5.90 n/a n/a .15 ± 3.80 .86
IRI PD (Personal Distress) 8.40 ± 4.78 n/a n/a −.20 ± 3.55 .8
Barrett-Lennard Empathy 15.35 ± 11.91 n/a n/a .40 ± 12.40 .89
Inflammation biomarkers IL-6 0.81 ± 0.71 0.67 ± 0.59 .41 −.03 ± .25 .62
IL-10 0.57 ± 0.61 0.52 ± 0.33 .69 −.03 ± .19 .48
TNF-a 2.84 ± 0.58 2.92 ± 0.85 .69 .11 ± .58 .43
log CRP 0.53 ± 0.67 0.12 ± 0.57 .01
plasma CRP 8.56 ± 12.70 3.05 ± 6.21
BMI BMI 31.20 ± 7.66 26.28 ± 4.82 .01
Neuroimaging ROI rTPJ own PLWD - unknown PLWD 0.07 ± 0.19 .10 ± .26 .047+
own PLWD - own friend −0.12 ± 0.28 .15 ± 3.67 .047+
DMPFC own PLWD - unknown PLWD 0.07 ± 0.24 .11 ± .25 .04+
own PLWD - own friend −0.04 ± 0.32 .14 ± .40 .07+
Precuneus own PLWD - unknown PLWD 0.22 ± 0.29 .12 ± .39 .10+
own PLWD - own friend −0.14 ± 0.50 .20 ± .70 .11+
rAI own PLWD - unknown PLWD 0.08 ± 0.25 .05 ± .28 .23+
own PLWD - own friend 0.02 ± 0.42 .03 ± .40 .36+
lAI own PLWD - unknown PLWD 0.09 ± 0.23 .03 ±.27 .29+
own PLWD - own friend) 0.03 ± 0.44 .02 ± .42 .42+
Open in a new tab
+

denotes one-tailed p value

Among caregivers, there were pre- to post-intervention decreases on the Zarit Burden Scale (t(19) =−3.46, p = 0.003; d = 0.78) and the state anxiety component of the State-Trait Anxiety Scale (t(19) = −2.92, p = 0.009; d = 0.65). There were also pre- to post-intervention increases in the Perspective Taking subscale of the Interpersonal Reactivity Index (IRI) (t(19) = 3.0, p = 0.007; d = 0.67). There were no pre- to post-intervention changes on the Perceived Stress Scale (t(19)= −1.15, p = 0.27), the Center for Epidemiological Studies Depression Scale (t(19)= −1.58, p = 0.13), the Barrett-Lennard Empathy Scale (t(19) = 0.14, p = 0.89), or on the other subscales of the IRI (Empathic Concern (t(19)= −0.48, p = 0.64), Fantasy (t(19) = 0.18, p = 0.86) and Personal Distress (t(19)= −0.25, p = 0.80) (Figure 2, Table 2). Caption quality, as assessed by two raters, was not related to the degree of improvement in mental health variables (all p > 0.05).

Figure 2.

Figure 2.

Open in a new tab

Pre- to post-intervention changes in questionnaire scores; error bars = ± 1 SE; ranges for the various scales differed: the Zarit Burden Scale (0–88), the Center for Epidemiological Studies Depression Scale (0–60), the perceived stress Scale (0–40), the state trait anxiety index (20–80), the interpersonal reactivity Index; empathic concern (0–28), perspective taking (0–28), fantasy Scale (0–28), personal distress (0–28), Barrett-Lennard empathy Scale (−48–48).

Inflammation biomarkers

There was no difference between the caregivers and controls in baseline IL-6 (t(57) = 0.83, p = 0.41), IL-10 (t(55) = 0.40, p = 0.69), or TNF-α (t(57)=−0.40, p = 0.69), even after controlling for potential confounds of BMI and gender (see supplementary materials). On the other hand, log CRP was higher in caregivers (t(57) = 2.53, p = 0.01; d = 0.68), but this difference was eliminated after controlling for BMI and gender in a multiple regression model (see supplementary materials). BMI was higher in caregivers compared with controls (t(32.86) = 2.70, p = 0.011; d = 0.80), and BMI was positively correlated with log CRP among both caregivers (r = 0.67, p = 0.001) and controls (r = 0.47, p = 0.004) A higher proportion of caregivers than controls had CRP >10 mg/L (27% vs. 3%, X2 (1, 59) = 7.97, p=0.005), a level that is highly elevated and found in less than 5% of most populations (Pearson et al., 2003) (Table 2).

There were no significant pre- to post-intervention changes in IL-6 (t(18)= −0.51, p = 0.62), IL-10 (t(16)=−0.73, p = 0.48), TNF-α (t(18) = 0.82, p = 0.43) or CRP (Z = −0.59, p = 0.56). However, caregivers with higher baseline CRP had larger decreases in CRP across the intervention (r= −0.84, p < 0.001; Figure 3). Notably, three caregivers with CRP >10 mg/L had large decreases in CRP over the course of the intervention, such that the proportion of caregivers with CRP >10 mg/L did not differ from controls post-intervention (6% vs. 3%, X2 (1, 55) = 0.28, p=0.60), in contrast to pre-intervention (Table 2).

Figure 3.

Figure 3.

Open in a new tab

Scatterplot of the relationship between baseline, pre-intervention CRP levels and pre- to post-intervention changes in CRP levels.

Neuroimaging

ROI analysis

Cognitive Empathy Regions of Interest (ROIs): To identify brain regions that were specifically activated in caregivers while they viewed their care recipient, we examined the contrast between viewing photographs of their PLWD and an unknown PLWD (own PLWD – unknown PLWD). Results show significant increase in activation from pre- to post-intervention within the right TPJ (t(19) = 1.76, one-tailed p = 0.047, d = 0.39) and the DMPFC (t(19) = 1.89, one-tailed p = 0.04, d = 0.42), but not the precuneus (t(19) = 1.35, one-tailed p = 0.10) (Figure 4) (Table 2).

Figure 4.

Figure 4.

Open in a new tab

Comparison of the average value for the contrast (own PLWD – unknown PLWD) between pre- and post-intervention scans within the right temporo-parietal junction (left), dorsomedial prefrontal cortex (middle) and precuneus (right); error bars = ± 1 SE. * denotes significance with a one-tailed test.

For the contrast between viewing photographs of their PLWD and a friend or family member (own PLWD – own friend), there was a significant increase in activation from pre- to post-intervention within the right TPJ (t(19) = 1.76, one tailed p = 0.047, d = 0.39), but not the DMPFC (t(19) = 1.56, one tailed p = 0.07), or the precuneus (t(19) = 1.30, one tailed p = 0.11) (supplementary Figure S1) (Table 2).

Emotional Empathy ROIs: For the contrast (own PLWD – unknown PLWD), there was no significant change in activation from pre- to post-intervention within the either the right (t(19) = 0.76, one-tailed p = 0.23) or left (t(19) = 0.55, one-tailed p = 0.29) anterior insula. There was also no significant change within the right (t(19) = 0.36, one-tailed p = 0.36) or left (t(19) = 0.21, one-tailed p = 0.42) anterior insula for the contrast (own PLWD – own friend) (Table 2).

Whole brain analysis

We also conducted an exploratory analysis in which we searched beyond our a-priori regions of interest (ROIs) for brain regions that showed an effect of the intervention. For the contrast (own PLWD – unknown PLWD), there was a pre- to post-intervention increase in activation within the right ventrolateral prefrontal cortex. This was driven by an increased response to the own PLWD post-intervention (Figure 5; Supplementary Table S3).

Figure 5.

Figure 5.

Open in a new tab

Results of whole brain analysis, illustrating regions where the contrast (own PLWD – unknown PLWD) increased from pre- to post-intervention. Results are thresholded using clusters determined by Z > 2.3 (1-tailed p < 0.01), and a family-wise error (FWE)-corrected cluster significance threshold of p < 0.05. Data are plotted from the peak voxel within the right ventrolateral prefrontal cortex; error bars = ± 1 SE. OP = own PLWD, UP = unknown PLWD.

For the contrast (own PLWD – own friend), there was a pre- to post-intervention increase in activation within the right lateral occipital cortex and the right inferior and middle frontal gyrus (supplementary Figure S2, Supplementary Table S3).

Discussion

Here, we demonstrate the feasibility of a novel photo-captioning intervention designed to target cognitive empathy in dementia caregivers. The intervention had high tolerability, with 87% of enrolled caregivers successfully completing it. We also conducted a preliminary assessment of the intervention’s efficacy in improving caregiver well-being, as well as potential psychological and neural mechanisms for these effects.

As expected, caregivers showed evidence of worse mental health compared with controls at baseline, and this could not be attributed to group differences in age, gender, income, education, exercise, alcohol consumption, or chronic disease prevalence. Specifically, caregivers scored higher on the Perceived Stress Scale and on the state anxiety component of the State-Trait Anxiety Scale. This was paralleled by higher BMI in caregivers, which in turn explained higher CRP levels among caregivers. In particular, caregivers were more likely to have CRP levels over 10 mg/L. On the other hand, there was no evidence for elevated cytokine levels (IL-6, IL-10, TNF-α) in caregivers compared with controls. Although previous studies have shown that caregiving may accelerate age-related increases in pro-inflammatory cytokines such as IL-6 and TNF-α (Kiecolt-Glaser et al., 2003; Roth et al., 2020), cross-sectional associations between caregiving and inflammation have been characterized as weak (Roth et al., 2019). Combined with our limited sample size and statistical power in this pilot study, this suggests that the lack of group differences in cytokine levels may not be unexpected.

Both state anxiety and caregiver burden decreased across the intervention group with medium to large effect sizes, suggesting that the intervention may have improved caregiver mental health. While there was no clear evidence of a parallel decrease in biomarkers of inflammation for the overall sample, those with higher baseline CRP showed greater decreases in CRP over the course of the intervention. In particular, three of the four caregivers with baseline CRP >10 mg/L experienced pronounced decreases in CRP across the intervention, whereas none of the caregivers transitioned in the opposite direction (i.e., from below to above 10 post-intervention). This raises the prospect that the intervention may be particularly effective for those with very high CRP levels at baseline. It is also possible that this correlation reflects regression to the mean. CRP >10 mg/L is often indicative of acute infection (Pearson et al., 2003). Therefore, three of the four caregivers with CRP >10 mg/L at baseline may have had infections that resolved over the course of the 10-day intervention. However, none of the caregivers seem to have developed new infections over the course of the intervention based on CRP levels. These combined findings raise the intriguing possibility that caregivers may be more prone to suppressed immunity at baseline, and that this vulnerability is ameliorated by cognitive empathy training. This possibility could be tested more directly in future studies by including markers of immune function.

The hypothesized psychological mechanism by which the intervention improved caregiver mental health was an increase in cognitive empathy. Cognitive empathy requires that individuals try to adopt the mental perspective of others. Indeed, scores on the Perspective-Taking subscale of the IRI increased across the intervention, whereas other IRI subscales that capture different components of empathy showed no change. This suggests that the intervention did particularly impact cognitive empathy. The specificity of the effect is desirable since high levels of emotional empathy have actually been linked with worse caregiver mental health (Hua et al., 2021). Given these findings for Perspective-Taking, the lack of change in the Barrett-Lennard Empathy Scale, another measure of cognitive empathy, is surprising. However, the two empathy measures were not correlated either pre- (r= −0.09, p = 0.70) or post-intervention (r = 0.31, p = 0.19), suggesting that they do not measure the same aspect of cognitive empathy. The perspective-taking scale of the IRI measures a person’s self-reported tendency to spontaneously adopt the psychological point of view of others in everyday life. On the other hand, the Barret-Lennard Empathy Scale measures self-reported empathic understanding, or the ability to accurately perceive and understand the emotions of others. It was originally developed to assess therapists’ ability to empathize with clients, but has since been extended to other types of relationships such as marital partners. Our intervention encourages caregivers to take time to adopt the mental perspective of their care recipient more often than they might normally do. Therefore, a possible explanation for this discrepancy is that our intervention works by increasing caregiver’s tendency to empathize, more-so than by improving their accuracy when doing so.

To investigate potential neural mechanisms by which our intervention improved caregiver well-being, we also imaged caregiver brain function as they viewed photographs of their PLWD and attempted to empathize with them. As hypothesized, activation within brain regions implicated in cognitive empathy, including the DMPFC and rTPJ, increased across the intervention. These effects were only significant with a one-tailed t-test, and tests were conducted for three different ROIs for each of two contrasts. Thus, these findings may be vulnerable to type I errors. It will therefore be important to determine if these findings replicate in a future study with a larger sample. Of interest, pre- to post-intervention increases in neural activity were specific to these cognitive empathy ROIs, and were not observed in the anterior insula, a control region that is instead implicated in emotional empathy (Craig, 2004; Singer et al., 2009).

Our exploratory fMRI analysis also revealed a region within the ventrolateral prefrontal cortex where activation to the PLWD increased over the course of the intervention. The ventrolateral prefrontal cortex is critically involved in emotion regulation (Dixon et al., 2017; Golkar et al., 2012). Many caregivers have ambivalent feelings toward their PLWDs (Losada et al., 2017). Therefore, one speculative interpretation of this finding is that empathizing requires effectively regulating negative emotional feelings toward the PLWD. Indeed, pre- to post-intervention changes in cognitive empathy ROIs were strongly correlated with changes within the ventrolateral prefrontal cortex (for OP-UP, vlPFC vs. rTPJ r = 0.80, vlPFC vs. DMPFC r = 0.79, vlPFC vs. precuneus r = 0.77, all p < 0.001). Nevertheless, the main effect of the intervention within ventrolateral prefrontal cortex will also require replication since the statistical threshold employed is below that generally considered necessary to adequately protect against false positive results for whole brain analyses (Eklund et al., 2016; Woo et al., 2014).

Our study is not the first to target cognitive empathy in dementia caregivers. Another study used a virtual reality simulation to help caregivers understand the experience of dementia. When combined with a short, on-line e-learning course, this intervention was also associated with a significant increase in the Perspective-Taking subscale of the IRI, measured three weeks after the intervention (d = 0.42). Thus, the intervention may have increased cognitive empathy. Moreover, caregivers reported an increased number of positive interactions with their PLWD across the intervention (Wijma et al., 2018). The effect size of our photo captioning intervention on the Perspective-Taking subscale of the IRI (d = 0.67) is medium to large. However, Wijma et al used a different formula for their effect size calculation, and our data yield an effect size of d = 0.44 using that formula. Thus, our intervention yields an effect size that is comparable to that reported for this virtual reality intervention (d = 0.42).

The evidence for intervention-related improvements in mental health outcomes and increased cognitive empathy were based on self-report and are therefore vulnerable to demand characteristics. It was obvious to caregivers that the intervention was designed to increase perspective-taking, and that the goal was to improve the caregiver-PLWD relationship. Therefore, participants may have been biased to report changes consistent with our hypotheses. Future studies should compare the intervention against a control condition that elicits similar expectations for improvements in well-being. To the extent that inflammation biomarkers track caregiver well-being, they should be less vulnerable to these biases. Comparison with a control condition is also necessary to discern whether intervention effects are due to the cognitive empathy training per se, or to other nonspecific aspects of the intervention such as interactions with study staff or participating in a meaningful activity. Finally, the observed decreases in caregiver burden and anxiety could reflect regression to the mean rather than intervention effects, and this possibility could also be addressed by the addition of a control condition.

Another limitation of this study is that post-intervention measures were only collected immediately after the intervention such that it was not possible to assess the duration of effects. Future studies should evaluate outcomes at additional, later post-intervention time points, perhaps accompanied by periodic booster trainings to help sustain benefits. Future studies should also keep in mind evidence for a curvilinear relationship between cognitive empathy and caregiver mental health suggesting that cognitive empathy may need to be quite high in order to yield mental health benefits (Jutten et al., 2019).

In conclusion, we have established the feasibility and tolerability of a photo-captioning intervention for dementia caregivers that appears to have successfully targeted cognitive empathy and increased activation in related brain structures, while also decreasing caregiver burden and anxiety. There are also preliminary indications that the intervention may be able to decrease inflammation, as measured by C-reactive protein among a subset of caregivers with elevated levels at baseline. These findings warrant a large replication study in which participants are randomized to the cognitive empathy intervention or a control condition and followed to establish the duration of the intervention effects.

Supplementary Material

Supplemental Material

Clinical implications.

  • The photo captioning intervention was associated with increased self-reported cognitive empathy and decreased burden and anxiety in dementia caregivers

  • Enhancing cognitive empathy may be an effective strategy for improving caregiver mental health

Acknowledgments

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.

Funding

The work was supported by the, Emory Roybal Center for Dementia Caregiving Mastery, Emory Center for Health in Aging, Emory University Alzheimer’s Disease Research Center.

Footnotes

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplemental data for this article can be accessed online at https://doi.org/10.1080/07317115.2024.2317972.

Data availability statement

This study was not preregistered. Data are available at: https://osf.io/sweq2/?view_only=414d4010fa8d4f80b8610d32282f097e

References

  1. 2023 Alzheimer’s disease facts and figures. (2023). 2023 Alzheimer’s disease facts and figures. Alzheimer’s & Dementia, 19(4), 1598–1695. 10.1002/alz.13016 [DOI] [PubMed] [Google Scholar]
  2. Barrett-Lennard GT (1978). The relationship inventory: Development and adaptations. JSAS Catalog of Selected Documents in Psychology, 8, 68. 10.2307/584212 [DOI] [Google Scholar]
  3. Bin Sallim A, Sayampanathan AA, Cuttilan A, & Ho RCM (2015). Prevalence of mental health disorders among caregivers of patients with Alzheimer disease. Journal of the American Medical Directors Association, 16 (12), 1034–1041. 10.1016/j.jamda.2015.09.007 [DOI] [PubMed] [Google Scholar]
  4. Black BS, Rabins PV, Sugarman J, & Karlawish JH (2010). Seeking assent and respecting dissent in dementia research. The American Journal of Geriatric Psychiatry, 18 (1), 77–85. 10.1097/JGP.0b013e3181bd1de2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Camoirano A (2017). Mentalizing makes parenting work: A review about parental reflective functioning and clinical interventions to improve it. Frontiers in Psychology, 8, 14. 10.3389/fpsyg.2017.00014 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Cheng ST, Au A, Losada A, Thompson LW, & Gallagher-Thompson D (2019). Psychological interventions for dementia caregivers: What we have achieved, what we have learned. Current Psychiatry Reports, 21(7). 10.1007/s11920-019-1045-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Cohen S, Kamarck T, & Mermelstein R (1983). A global measure of perceived stress. Journal of Health & Social Behavior, 24(4), 385–396. 10.2307/2136404 [DOI] [PubMed] [Google Scholar]
  8. Craig AD (2004). Human feelings: Why are some more aware than others? Trends in Cognitive Sciences, 8(6), 239–241. S1364661304001032 [pii]. 10.1016/j.tics.2004.04.004 [DOI] [PubMed] [Google Scholar]
  9. Davis MH (1983). Measuring individual-differences in empathy - evidence for a multidimensional approach. Journal of Personality & Social Psychology, 44(1), 113–126. 10.1037/0022-3514.44.1.113 [DOI] [Google Scholar]
  10. Dixon ML, Thiruchselvam R, Todd R, & Christoff K (2017). Emotion and the prefrontal cortex: An integrative review. Psychological Bulletin, 143(10), 1033–1081. 10.1037/bul0000096 [DOI] [PubMed] [Google Scholar]
  11. Eklund A, Nichols TE, & Knutsson H (2016). Cluster failure: Why fMRI inferences for spatial extent have inflated false-positive rates. Proceedings of the National Academy of Sciences, 113(28), 7900–7905. 10.1073/pnas.1602413113 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Golkar A, Lonsdorf TB, Olsson A, Lindstrom KM, Berrebi J, Fransson P, . . . Ohman A. (2012). Distinct contributions of the dorsolateral prefrontal and orbitofrontal cortex during emotion regulation. Public Library of Science ONE, 7(11), e48107. 10.1371/journal.pone.0048107 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Greve DN, & Fischl B (2009). Accurate and robust brain image alignment using boundary-based registration. Neuroimage: Reports, 48(1), 63–72. 10.1016/j.neuroimage.2009.06.060 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Hua AY, Wells JL, Brown CL, & Levenson RW (2021). Emotional and cognitive empathy in caregivers of people with neurodegenerative disease: Relationships with caregiver mental health. Clinical Psychological Science, 9(3), 449–466. 10.1177/2167702620974368 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Jenkinson M, Bannister P, Brady M, & Smith S (2002). Improved optimization for the robust and accurate linear registration and motion correction of brain images. Neuroimage: Reports, 17(2), 825–841. 10.1006/nimg.2002.1132 [DOI] [PubMed] [Google Scholar]
  16. Joling KJ, van Marwijk HWJ, Veldhuijzen AE, van der Horst HE, Scheltens P, Smit F, & van Hout HPJ (2015). The two-year incidence of depression and anxiety disorders in spousal caregivers of persons with dementia: Who is at the greatest risk? The American Journal of Geriatric Psychiatry, 23(3), 293–303. 10.1016/j.jagp.2014.05.005 [DOI] [PubMed] [Google Scholar]
  17. Jutten 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]
  18. Kiecolt-Glaser JK, Preacher KJ, MacCallum RC, Atkinson C, Malarkey WB, & Glaser R (2003). Chronic stress and age-related increases in the proinflammatory cytokine IL-6. Proceedings of the National Academy of Sciences, 100(15), 9090–9095. 10.1073/pnas.1531903100 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Krall SC, Rottschy C, Oberwelland E, Bzdok D, Fox PT, Eickhoff SB . . . Konrad K. (2015). The role of the right temporoparietal junction in attention and social interaction as revealed by ALE meta-analysis. Brain Structure and Function, 220(2), 587–604. 10.1007/s00429-014-0803-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Lee HS, Brennan PF, & Daly BJ (2001). Relationship of empathy to appraisal, depression, life satisfaction, and physical health in informal caregivers of older adults. Research in Nursing & Health, 24(1), 44–56. 10.1002/1098-240x(200102)24:1&lt;44:aid-nur1006&gt;3.0.co;2-s [DOI] [PubMed] [Google Scholar]
  21. Losada A, Pillemer K, Marquez-Gonzalez M, Romero-Moreno R, & Gallego-Alberto L (2017). Measuring ambivalent feelings in dementia family caregivers: The caregiving ambivalence scale. Gerontologist, 57(3), E37–E46. 10.1093/geront/gnw144 [DOI] [PubMed] [Google Scholar]
  22. McDade TW, Burhop J, & Dohnal J (2004). High-sensitivity enzyme immunoassay for C-reactive protein in dried blood spots. Clinical Chemistry, 50(3), 652–654. 10.1373/clinchem.2003.029488 [DOI] [PubMed] [Google Scholar]
  23. McDade TW, Miller A, Tran TT, Borders AEB, & Miller G (2021). A highly sensitive multiplex immunoassay for inflammatory cytokines in dried blood spots. American Journal of Human Biology, 33(6). 10.1002/ajhb.23558 [DOI] [PubMed] [Google Scholar]
  24. Mcdade TW, Williams S, & Snodgrass JJ (2007). What a drop can do: Dried blood spots as a minimally invasive method for integrating biomarkers into population-based research. Demography, 44(4), 899–925. 10.1353/dem.2007.0038 [DOI] [PubMed] [Google Scholar]
  25. McIsaac JH (2021). “What Were You Thinking?” Using Technology to Enhance Mentalization Capacity Among Fathers: A Phenomenological Study. [Ph.D.]. Fielding Graduate University. [Google Scholar]
  26. Molenberghs P, Johnson H, Henry JD, & Mattingley JB (2016). Understanding the minds of others: A neuroimaging meta-analysis. Neuroscience & Biobehavioral Reviews, 65, 276–291. 10.1016/j.neubiorev.2016.03.020 [DOI] [PubMed] [Google Scholar]
  27. NIH. (2009). Research involving individuals with questionable capacity to consent: Points to consider. http://grants.nih.gov/grants/policy/questionablecapacity.htm
  28. Palmer BW, Dunn LB, Appelbaum PS, Mudaliar S, Thal L, Henry R, . . . Jeste DV. (2005). Assessment of capacity to consent to research among older persons with schizophrenia, Alzheimer disease, or diabetes mellitus - comparison of a 3-item questionnaire with a comprehensive standardized capacity instrument. Archives of General Psychiatry, 62(7), 726–733. 10.1001/archpsyc.62.7.726 [DOI] [PubMed] [Google Scholar]
  29. Pearson TA, Mensah GA, Alexander RW, Anderson JL, Cannon RO 3rd, Criqui M, . . . American Heart A. (2003). Markers of inflammation and cardiovascular disease: Application to clinical and public health practice: A statement for healthcare professionals from the Centers for disease Control and Prevention and the American Heart Association. Circulation, 107(3), 499–511. 10.1161/01.cir.0000052939.59093.45 [DOI] [PubMed] [Google Scholar]
  30. Pinquart M, & Sorensen S (2003). Differences between caregivers and noncaregivers in psychological health and physical health: A meta-analysis. Psychology and Aging, 18 (2), 250–267. 10.1037/0882-7974.18.2.250 [DOI] [PubMed] [Google Scholar]
  31. Radloff LS (1977). The CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurement, 1(3), 385–401. 10.1177/014662167700100306 [DOI] [Google Scholar]
  32. Raison CL, Capuron L, & Miller AH (2006). Cytokines sing the blues: Inflammation and the pathogenesis of depression. Trends in Immunology, 27(1), 24–31. 10.1016/j.it.2005.11.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Roth DL, Haley WE, Sheehan OC, Huang J, Rhodes JD, Durda P, . . . Cushman M. (2020). The transition to family caregiving and its effect on biomarkers of inflammation. Proceedings of the National Academy of Sciences, 117(28), 16258–16263. 10.1073/pnas.2000792117 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Roth DL, Sheehan OC, Haley WE, Jenny NS, Cushman M, & Walston JD (2019). Is family caregiving associated with inflammation or compromised immunity? A meta-analysis. The Gerontologist, 59(5), E521–E534. 10.1093/geront/gnz015 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Schurz M, Radua J, Aichhorn M, Richlan F, & Perner J (2014). Fractionating theory of mind: A meta-analysis of functional brain imaging studies. Neuroscience & Biobehavioral Reviews, 42, 9–34. 10.1016/j.neubiorev.2014.01.009 [DOI] [PubMed] [Google Scholar]
  36. Singer T, Critchley HD, & Preuschoff K (2009). A common role of insula in feelings, empathy and uncertainty. Trends in Cognitive Sciences, 13(8), 334–340. 10.1016/j.tics.2009.05.001 [DOI] [PubMed] [Google Scholar]
  37. Slaughter S, Cole D, Jennings E, & Reimer MA (2007). Consent and assent to participate in research from people with dementia. Nursing Ethics, 14(1), 27–40. 10.1177/0969733007071355 [DOI] [PubMed] [Google Scholar]
  38. Spielberger CD (1983). State-trait anxiety inventory for adults. Sampler set. Mind Garden, [1983] ©1983. [Google Scholar]
  39. 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] [PubMed] [Google Scholar]
  40. Woo CW, Krishnan A, & Wager TD (2014). Cluster-extent based thresholding in fMRI analyses: Pitfalls and recommendations. Neuroimage: Reports, 91, 412–419. 10.1016/j.neuroimage.2013.12.058 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. 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]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Material
NIHMS1988037-supplement-Supplemental_Material.docx (731.9KB, docx)

Data Availability Statement

This study was not preregistered. Data are available at: https://osf.io/sweq2/?view_only=414d4010fa8d4f80b8610d32282f097e

RESOURCES