Emerging Adults’ Perception of the Typical Dementia Patient Based on Gender and Race

Abstract

Introduction

Though Black Americans are twice as likely as White Americans to be diagnosed with Alzheimer’s Disease, they may be excluded from the mental representation of a person with dementia.

Methods

Participants (N = 143, median age = 19) created visualized mental representations of a person, a man, a woman, a Black man, or a Black woman diagnosed with dementia by repeatedly selecting which among 12 faces looked most like each target category.

Results

The visualized representation of a person with dementia was more similar to the representation of a man and a Black woman than to a Black man and a woman, respectively.

Discussion

These findings highlight how intersectionality shapes mental representations of dementia, revealing that certain combinations of marginalized identities are perceived differently rather than excluded entirely.

Over 7 million older adults were diagnosed with Alzheimer’s and Related Dementias (ADRD) in 2020.^1-3 By 2060, this number is expected to increase to 13.9 million. ⁴ However, members of some social groups are at a higher risk for the disease. Women account for about two-thirds of individuals diagnosed with dementia and Black Americans have higher incidence rates than White Americans.^4,5 In fact, dementia is the fifth leading cause of death among Black Americans. ⁶ However, compared to Non-Hispanic Whites, Black Americans are 35% less likely to be diagnosed when symptoms are present, ⁷ and therefore the actual health discrepancy for higher incidence rates for dementia among Black Americans is underestimated. This may be because, for many people, Black Americans might be missing from their mental representation of a typical person with dementia. To investigate this possibility, the current research examined the extent to which Black Americans are included in the visual representations of people with dementia.

Intersectionality Framework

Intersectionality is a framework for understanding how multiple forms of discrimination interact to create unique experiences of marginalization and oppression. ⁸ The framework argues that social identities such as race, gender, class, and health status do not operate independently, but intersect in ways that shape how individuals are perceived and treated within society. For example, a Black woman may experience discrimination differently than Black men or White women due to the ways in which racism and sexism intersect in their lived experience. ⁸ Furthermore, stereotypes about Black women are not simply the summation of stereotypes about Black people and women. ⁹ Instead, Black women are associated with many unique stereotypes not associated with either Black people or women. Thus, people with intersectional identities do not always experience a double-jeopardy ¹⁰ in which discrimination directed at their single-category identities (eg, race or gender) are summed or compounded. ¹¹ Rather, sometimes intersectional discrimination can operate by rendering people with multiple marginalized identities invisible. ¹² For example, research has shown that people are less likely to remember a Black women’s statements, ¹³ slower to categorize Black women as women ¹⁴ especially if they have darker skin, ¹⁵ and attitudes toward Black people are associated with attitudes toward Black men but not Black women. ¹⁶

While the intersectionality framework is often used to examine institutional and interpersonal discrimination, it can also be applied in health contexts where disparities in diagnosis and treatment often emerge due to the effects of multiple social identities. For example, women with coronary heart disease (CHD) are members of 2 subordinate groups (ie, women and having CHD) and, therefore, may experience invisibility that contributes to health disparities. Consistent with this reasoning, women are diagnosed with CHD later than men even though it is the leading cause of the death among both genders in the U.S.^17,18 In the current research, we included having an illness as membership in a subordinate group because this is often stigmatized. ¹⁹ These intersecting biases are observed in the context of dementia, a stigmatized health condition influenced by stereotypes related to race, gender, and age. Black Americans and women with dementia in particular may face misdiagnosis, delayed diagnosis, and inadequate medical care. ²⁰ For example, physicians have dismissed memory concerns in Black patients as signs of normal aging, while women’s memory complaints have been misattributed to other conditions such as depression.^21-24

Consequences of intersectionality may lead to missed or delayed diagnoses among Black people and women in that they may not look like the ‘typical’ dementia patient. Historically, White men have been oversampled in research, shaping not only diagnostic tools and clinical expectations, but also broader public perceptions of diseases. ²⁵ When asked to imagine a person with dementia, individuals may default to a mental image of an older White man. As a result, dementia symptoms in individuals with intersecting identities who do not match this prototype (ie, Black people and women) may be overlooked or dismissed. Therefore, using an intersectional lens is important when considering how people visually represent dementia, as certain combinations of identities (eg, Black women or Black men) may be excluded from mental representations.

These biases play an important role not only in clinical decision-making, but also the early stages of family members recognizing symptoms and seeking care. Family members are commonly the first to observe behavioral and cognitive changes in loved ones and play a vital role in encouraging them to seek a diagnosis. ²⁶ Emerging adults typically have 2 generations above them (ie, parents and grandparents) for which they may be able to identify atypical cognitive decline and engage in preventative intervention. Perceptions of who represents a typical dementia patient may determine how quickly and accurately symptoms are recognized. However, due to racial and cultural disparities, this process is also shaped by racial and cultural differences, as families from non-White backgrounds may be more likely to interpret early cognitive changes as normal aging or depression rather than symptoms of dementia. ²⁷ Furthermore, healthcare providers often rely on family members during the diagnostic process to understand symptom onset and behavioral changes. If early changes are misinterpreted or overlooked by family members, diagnoses may be delayed even further.

Emerging adulthood, a developmental period spanning ages 18-29, represents a period when individuals begin to take on significant social and professional responsibilities. As future healthcare providers or caretakers to family and friends, emerging adults may play a crucial role in supporting older adults.^28-30 However, to our knowledge, previous research has not explored their visual representations of people diagnosed with dementia. Studies have identified that emerging adults often hold unfavorable attitudes toward older adults and aging. For example, college students in Israel typically avoid physical contact with older adults, ³¹ and third-year undergraduate nursing students in the United States associate aging with frailty, physical decline, and mental decline. ²⁷ To extend this literature, the present study was designed to investigate the extent to which intersecting identities (ie, gender and race) shape emerging adults’ visual representations of people with dementia. Given the important role that family members play in recognizing symptoms and initiating care, understanding the biases that shape how emerging adults perceive dementia may help explain broader disparities in diagnosis and treatment. If they fail to associate dementia with individuals from marginalized groups, they may be less likely to recognize symptoms in family members who do not fit the prototypical image of someone with dementia and consequently, be less likely to advocate for timely medical care.

Using Reverse Correlation to Study Mental Representations of People with Dementia

The current study utilized the reverse correlation (RC) procedure to examine the mental representations of people with dementia. RC has been used to study biases based on facial appearances related to race and ethnicity, ³² gender, ³³ age, ³⁴ and stereotypes and prejudices. ³⁵ The benefits of this method include reduced social desirability since participants do not know how their responses will be used in the study, and it does not force participants to focus on a set of stereotypes we choose. ³⁶ RC captures a person’s expectation about a target (eg, what does a person diagnosed with dementia look like?) and, therefore, allows researchers to test intersectionality by comparing how similar or dissimilar participants imagine various groups to appear.

The current research used the RC technique to investigate the extent to which women and Black people are included in the visual mental representations of people with dementia. Based on the intersectionality framework, we explored 2 preregistered hypotheses (https://aspredicted.org/sr24w.pdf), and an exploratory hypothesis across the full sample and subgroups of non-Hispanic White women and emerging adults who identified as races or ethnicities other than non-Hispanic White. Although the hypotheses remained the same across groups, we sought to examine whether patterns of visual representations varied by participants’ own racial and gender identities, as these factors may influence who is visually imagined when thinking about dementia.

Hypotheses

Preregistered Hypothesis 1. Visual representations of people with dementia will be more similar to visual representations of men with dementia than women with dementia.

Preregistered Hypothesis 2. Visual representations of people with dementia will be more similar to visual representations of men with dementia than Black men with dementia.

Exploratory Hypothesis 3. Visual representations of people with dementia will be more similar to visual representations of women with dementia than Black women with dementia.

Method

Participants

One hundred and forty-three undergraduate students were recruited through the University of North Florida’s (UNF) SONA Research Participation System. Participants were compensated with partial course credit and self-reported their age, racial/ethnic identity, and gender (Female, Male, Nonbinary Trans, Other). If participants self-reported more than one race, they were coded as ‘Mixed’. The median age of participants was 19 (age range = 18-29). The racial/ethnic composition of participants was 69.9% White, 14.7% Black, 6.3% Asian, 4.9% Mixed, 2.8% Hispanic, 0.7% Filipino and 0.7% Other. The gender composition of participants was 85.3% Female, 9.8% Male, 3.5% Nonbinary Trans, and 1.4% Other.

Procedure

After providing electronic consent, participants completed a RC task to generate a visual representation of a person diagnosed with dementia, answered questions about their attitudes toward dementia from the Dementia Attitudes Scale (DAS), ³⁷ completed demographics questions, and read a debriefing screen. We do not report analyses related to the DAS as they were unrelated to our hypotheses. The study went through the process of approval with the UNF Institutional Review Board (IRB), ensuring that the rights of the participants were protected. All materials, data, and analysis code are available online (https://osf.io/g7pzb/).

Brief Reverse Correlation (RC) Task

Following the Brief RC procedures, ³⁶ a base image was created as a morph of 4 Black American men, 4 Black American women, 4 non-Hispanic White men, and 4 non-Hispanic White women ³⁸ with superimposed random noise (see Figure 1). 3000 stimuli were produced by adding and subtracting a sinusoidal noise pattern from the base image. The Brief RC was used because it can produce richer images using fewer trials. Participants were presented with 12 noisy faces at a time arranged in a 3 × 4 grid. Participants were randomly assigned to select the face that most looked like a person diagnosed with dementia (N = 26), a man diagnosed with dementia (N = 31), a woman diagnosed with dementia (N = 32), a Black man diagnosed with dementia (N = 23), or a Black woman diagnosed with dementia (N = 31). Although there are no specific guidelines on sample size determination for the RC task, 28 participants per condition have been used in previous designs.^32,36 Participants repeated this task for 250 trials which is far above the recommended number of trials (90) to produce an accurate visual representation. ³⁶ Following the completion of the RC task, visual mental representations of a man, woman, Black man, and Black woman with dementia, were generated using the rcicr R package. ³⁹ Visual representations were computed by averaging the noise extracted from their selections and superimposing it onto the original base image.

Figure 1.

The Base Image Used in the Reverse Correlation Task

Data Analysis

Because each image is represented as a matrix of pixel values, we calculated correlations between the matrices of pixel values representing each image then used the cocor package in R to test for differences in how the visual representations of men, women, Black men, and Black women with dementia are correlated with the visual representations of a person with ADRD. ⁴⁰ For more information on details related to Brief RC analyses, see Schmitz et al 2024. ³⁶

Results

Full Sample

Results of the full sample are shown in Table 1. As predicted in our second preregistered hypothesis, visual representations of people with dementia were more strongly correlated with visual representations of men with dementia than Black men with dementia (r = 0.88, z = −29.77, P < .001, 95% CI [-0.015, −0.013]). However, our first preregistered hypothesis and our third exploratory hypothesis were not supported. Visual representations of people with dementia were correlated more strongly with visual representations of women with dementia than men with dementia (r = 0.88, z = 12.95, P = < .001, 95% CI [0.005, 0.007]). Moreover, visual representations of people with dementia were more similar to visual representations of Black women with dementia than women with dementia (r = 0.83, z = 32.19, P < .001, 95% CI [0.015, 0.017]).

Table 1.

Differences and Similarities Between Visual Mental Representations by Sample

	r	z	P	95% CI
Full sample
H1: Man and woman	0.8725	12.947	<.001	0.0051, 0.0069
H2: Man and Black man	0.8832	−29.773	<.001	−0.0148, −0.0129
H3: Woman and Black woman	0.8326	32.1857	<.001	0.0147, 0.0167
Non-Hispanic White Women
H1: Man and woman	0.8751	−19.4136	<.001	−0.0089, −0.0073
H2: Man and Black man	0.884	−63.2404	<.001	−0.0276, −0.0259
H3: Woman and Black woman	0.8731	−39.8039	<.001	−0.0186, −0.0168
Non-white participants
H1: Man and woman	0.8531	30.1847	<.001	0.0142, 0.0162
H2: Man and Black man	0.8621	76.9146	<.001	0.0352, 0.0371
H3: Woman and Black woman	0.8192	−49.8428	<.001	−0.0256, −0.0237

Note. Full Sample = N = 143. Non-Hispanic White Women: n = 80. Non-white participants: n = 43. H1: man and woman = visual representations of people with dementia will be more similar to visual representations of men with dementia than women with dementia. H2: man and Black man = visual representations of people with dementia will be more similar to visual representations of men with dementia than Black men with dementia. H3: visual representations of people with dementia will be more similar to visual representations of women with dementia than Black women with dementia.

Non-Hispanic White Women Subsample

Among the non-Hispanic white women sub sample, our 3 hypotheses were supported (Table 1). Visual representations of people diagnosed with dementia were correlated more strongly with visual representations of men with dementia than women with dementia (r = 0.88, z = −19.41, P < .001, 95% CI [-0.009, −0.007]). Furthermore, visual representations of people with dementia were more strongly correlated with visual representations of men with dementia than Black men with dementia (r = 0.88, z = −63.24, P < .001, 95% CI [-0.028, −0.026]). Finally, visual representations of people with dementia were more similar to CIs of women with dementia than Black women with dementia (r = 0.87, z = −39.80, P < .001, 95% CI [-0.020, −0.017]).

Non-white Subsample

Among the non-white subsample, our 2 pre-registered hypotheses were not supported (Table 1). However, our third exploratory hypothesis was supported. Visual representations of people with dementia were more similar to women with dementia than Black women with dementia (r = 0.82, z = −49.84, P < .001, 95% CI [-0.026, −0.024]). However, due to the small sample size of participants identifying as races or ethnicities other than non-Hispanic White (n = 43), we are uncertain whether this effect reflects a reliable pattern or is a result of sample variability.

Discussion

The current study examined whether intersecting identities shaped mental representations of people with dementia among emerging adults. Our findings provide support for predictions derived from the intersectionality framework which suggests that an individual’s experiences, opportunities, and disadvantages are shaped by the intersections of multiple overlapping social identities and systems of oppression or privilege. ⁸

In support of Hypothesis 2, participants’ visual representations of people with dementia were more similar to images of men with dementia than Black men with dementia. This finding is consistent with the intersectionality framework, which suggests that individuals with multiple marginalized identities (ie, gender, racial minority, and having a dementia diagnosis) may experience compounded stigma that influences how they are perceived. ⁴¹

Hypothesis 1 was not supported as visual representations of people with dementia were more similar to women than men which is inconsistent with previous work demonstrating that individuals imagine a typical person as a man as opposed to a woman.^42,43

Hypothesis 3 was also not supported as visual representations of people with dementia were more similar to Black women than race not specified women. This finding is inconsistent with the intersectional invisibility hypothesis which suggests that being a part of more than one minority group causes persons to be seen as ‘invisible’, ¹² as Black women were not ‘invisible’ compared to women.

Strengths

The present study focused on differences in perceptions of dementia based on race and gender. Previous research has focused efforts on identifying knowledge of and attitudes toward aging, however, there is a limited focus on dementia perception while considering the intersectionality of demographic characteristics (ie, gender and race) among a young sample.^44,45 It is important to examine the differences in perceptions of dementia, specifically when perceiving the intersection of race and gender in an emerging adult sample which represents future caregivers and healthcare providers of the aging American population. ⁴⁶ Findings can be used to inform dementia education and stigma intervention programs targeting what dementia looks like to combat underdiagnosis and differential treatment of Black American individuals with dementia.

Second, the current study is the first to study visual mental representations of people diagnosed with dementia. Previous studies have focused on data generated through questionnaires and survey data rather than pictures.^30,47 Therefore, the present study is important because it utilized the reverse correlation method which has several advantages in assessing implicit biases. Persons tend to base social judgments rapidly on faces, which influences social outcome responses.^48-51

Limitations

Future research should focus specifically on marginalized groups like Black Americans and Hispanics. Study participants were predominantly non-Hispanic White participants; (n = 100, 69.9%) and women (n = 122, 85.3%). Previous research suggests that predominantly White samples prevent insight into disparities among racial and ethnic minorities.^52,53 Regarding the present study, utilizing a predominantly White sample does not provide an accurate perception of dementia among other racial/ethnic groups. Black Americans and Hispanics are at a higher risk for cognitive decline but are less likely to be clinically diagnosed and included in clinical trials targeting dementia.^54-56 Barriers including lack of knowledge and dementia stigma are prominent in communities like Black Americans and Hispanics causing persons to not seek help when symptoms are present or for members of a community to not identify when someone is demonstrating signs of mild cognitive impairment.^57-59 Because emerging adults will be the future of caretaking in both a personal and professional capacity, focusing a study on a predominantly marginalized sample can provide insight into their perceptions of dementia and an avenue to educate communities about dementia to improve attitudes toward dementia.

Secondly, the current sample is predominantly composed of non-Hispanic White and women individuals therefore, it is difficult to draw conclusions about differences in visual representations of people diagnosed with dementia based on participants’ race and/or gender. However, results from our subsamples of non-Hispanic White women were partially consistent with results of the full sample as men were the default visual mental representation for a person diagnosed with dementia compared to Black men. Interestingly, this pattern was not observed among the non-White subsample where women and Black men were the default. These findings partially support previous research suggesting that individuals’ perceptions of a person with dementia might be more reflective of their own identity rather than the target. ⁶⁰ Future research should explore further whether similarities in race and/or gender between the assigned target category and participants’ identity influence these perceptions.

Furthermore, present findings may not be representative of the general public, as the sample was recruited through a university research participation system. Utilizing student samples can lead to sampling bias and impact generalizability as students are reported to be more educated, of higher socioeconomic status, and perform higher in cognitive skill domains than non-university adults.^61,62 Future researchers should collaborate with other degree programs and community colleges which tend to have a more diverse pool of individuals and establish community partnerships to target individuals outside a school setting. Moreover, incorporating samples representing different periods of the lifespan (ie, middle-adulthood, late-adulthood) might provide an understanding of cohort differences in the perception of a person diagnosed with dementia. This can also help to provide insight into how other individual factors (ie, differences in education attainment, SES, and gender) and contextual factors (ie, culture) may affect visual perceptions that may not have been captured.

Finally, among the full sample of emerging adults, Black women were the default compared to race not specified women when asked to think about a person diagnosed with dementia which aligns closely with the intersectional framework. The full sample was composed of mostly White women; therefore, our findings may suggest that participants associated dementia with a person who holds multiple marginalized identities reflecting intersectional stigma. ⁶³ Knowledge of dementia can possibly explain this finding. Prevalence rates of dementia are higher among Black women compared to other racial/ethnic groups of women.^6,64 Persons who better understand illnesses tend to display less stigmatizing and discriminatory attitudes. ⁶⁵ If participants were previously educated about ADRD disparities, this can explain why Black women were rendered the default. Future researchers should control for knowledge of dementia prevalence among intersectional groups.

Theoretical Implications

The present study examined how intersecting identities shape mental representations of dementia. Results demonstrated that race not specified men were the default when making depictions of people diagnosed with dementia, compared to Black men, across the full sample. These findings extend research on the intersectional framework by identifying attributions toward other minoritized groups (ie, persons diagnosed with dementia). They also have broad implications for the social perception of intersecting identities. Previous research has speculated that there are differences in perceptions of medical conditions.^65,66 For example, gender inequities in cardiovascular research as studies have previously relied on research conducted solely on men as guidelines for diagnosing and treating coronary heart disease. ⁶⁵ This is the opposite to breast cancer research examining experiences of male breast cancer patients diagnosed with a ‘woman’s disease’. ⁶⁶ Certain combinations of identities when perceiving health conditions (ie, Black men diagnosed with dementia) may be uncommon. They may require more effortful processing compared to other combinations like persons diagnosed with dementia or men diagnosed with dementia. Because of the increased effort, persons tend not to engage in the process that would result in identifying characteristics consistent with cognitive impairment, leading to groups being seen as invisible (ie, Black men).

Furthermore, these biases in social perception may influence how emerging adults who may play a critical role as future healthcare professionals and caregivers to friends and family, visualize dementia. If their representations of dementia fail to include Black individuals they may be less likely to recognize symptoms in Black family members or patients—particularly Black men, reinforcing disparities in early detection and care. Educating emerging adults on the diverse representations of dementia across different racial and gender groups is important in reducing stigma and promoting more equitable healthcare outcomes. Anti-stigma efforts that challenge stereotypes of dementia can help future caregivers develop a more inclusive understanding of the condition, ultimately improving diagnosis and treatment for all individuals.

Conclusion

Overall, men were found to be the default compared to Black men when perceiving people diagnosed with dementia across our full sample of emerging adults. Among non-Hispanic White women, men were the default compared to both Black men and women, and women were the default compared to Black women. Similarly, among non-White individuals, women were the default compared to Black women when considering people with dementia. Our findings illustrate the compounded effects of race and gender in shaping perceptions of people with dementia. As the United States population becomes increasingly diverse it is important to understand how these intersecting biases operate, as women and Black Americans are at a higher risk for dementia compared to other groups.