Reflective Functioning and Financial Exploitation Vulnerability in Older Adults: The Importance of Significant Others

ABSTRACT

Objectives

Financial exploitation of older adults results in devastating economic, social, and psychological losses to older adults, their families, and society at large. This study examined the relationship between reflective functioning and financial exploitation vulnerability (FEV) and whether relationship status moderated the association.

Methods

A community-based sample of 156 Israeli older adults age 60 and over responded to demographic questions and questionnaires assessing reflective functioning and FEV.

Results

A hierarchical linear regression analysis covarying for age, sex, education, income, and sum of illnesses, revealed that higher reflective functioning was associated with lower FEV (p = .011). A main effect of relationship status was not found, but a significant interaction of reflective functioning × relationship status was discovered (p = .008), adding 4.2% to the total variance of the model. Probing the interaction revealed that the reflective functioning-FEV association was significant only for older adults not in a relationship.

Conclusions

Findings suggest that low reflective functioning may be associated with increased risk of financial exploitation, specifically in certain populations of older adults.

Clinical Implications

Care providers of older adults may consider assessing for, and identifying older adults with low reflective functioning, in order to prevent or intervene in the event of a potentially exploitative situation.

KEYWORDS:

• Financial exploitation
• mentalizing
• older adults
• reflective functioning
• relationship status
• theory of mind

Introduction

Financial exploitation is one of the most common forms of elder mistreatment (Lachs & Berman, 2011; Peterson et al., 2014), yet it remains a relatively understudied phenomenon. Financial losses as a result of financial exploitation are significant. Older adults in the United States have been estimated by the AARP and the National Opinion Research Center (NORC) to lose 28.3 billion dollars annually to financial exploitation (including abuse, scams, and fraud (Gunther, 2023). Apart from its profound financial consequences, the experience of financial exploitation carries devastating psychological and social consequences for older adults, their families, and society at large (Lavery et al., 2020; Spreng et al., 2016; S. Wood & Lichtenberg, 2017). Risk and protective factors of financial exploitation are thus crucially important to identify so that targeted preventative measures can be developed and implemented.

The social cognitive neuroscience model of financial exploitation, proposed by Spreng et al. (2016), suggests two interacting pathways of change in older adulthood that can lead to increased financial exploitation vulnerability (FEV). The pathways include cognitive changes and social cognitive changes. Numerous studies have reported reduced cognitive functioning to be associated with poor financial decision making and increased FEV, and specifically fluid cognitive abilities such as poor attentional capacity, poor planning and goal direction, and other executive functioning skills (Spreng et al., 2016 for review; Boyle et al., 2012; James et al., 2014; Lichtenberg et al., 2020; S. A. Wood et al., 2016, 2014). Changes in such cognitive abilities can result in financial mismanagement and place a person at greater risk of experiencing financial exploitation. The social cognitive pathway reflects reduced social appraisal and perception abilities that occur with aging. The model posits that both the cognitive and social pathways interact to predict FEV.

While there is considerable research support for the cognitive pathway of the Spreng et al. (2016) model (see Fenton et al., 2022 for review), few studies have examined the direct relationship between social cognitive processes and FEV. Indirect support comes from studies demonstrating that older adults have reduced ability to appraise and detect deceptive information (Denburg et al., 2007; Ebner et al., 2023), increased difficulty in perceiving social cues, thereby reducing the quality of their trustworthiness judgments (Bailey et al., 2016; Webb et al., 2016), and increased attention to positively valanced information (i.e., positivity effect; Carstensen & DeLiema, 2018). Such social cognitive changes can increase susceptibility to undue social influence, such as that of a scammer (Spreng et al., 2016).

One social cognitive domain that has not received significant attention in the financial exploitation literature is mentalization, also known as reflective functioning (Fonagy et al., 2016). Reflective functioning is the human ability to reflect on the feelings, goals, attitudes, and wishes not only of oneself in a given relationship, but also on those of others (Fonagy et al., 2016). Such abilities have been reported by some to weaken with aging (Charlton et al., 2009; Henry et al., 2013; Moran et al., 2012; Pardini & Nichelli, 2009; but see; Castelli et al., 2010) and poor mentalization has also been associated with reduced functioning in other cognitive domains (e.g., executive functioning and working memory; Bottiroli et al., 2016; Charlton et al., 2009). Additionally, age-related changes in regions important for mentalization and other social cognitive processes, such as the default mode network, which includes brain regions such as the anterior and ventral medial prefrontal cortex, have been reported (Andrews-Hanna et al., 2014; Moran et al., 2012).

Given that reflective functioning involves the ability to understand the intentions of others, it is likely to play an important role in FEV, as financial exploitation usually occurs via a social interaction between an individual and the exploiter. Arguably, older adults who are better at understanding the underlying intentions of others (i.e., increased reflective functioning) will also be better at detecting the malicious intentions of a potential exploiter, and consequently more likely to avoid exploitation. Consistent with this line of thought, mentalization abilities have been demonstrated to improve deception detection (Stewart et al., 2019; Wang et al., 2023). For example, in a study of university students and staff by Stewart et al. (2019), cognitive theory of mind (i.e., effortful reasoning; see Shamay-Tsoory & Aharon-Peretz, 2007) was associated with detecting deceptive content in real-life video clips of family members pleading for information regarding missing/murdered loved ones. Additionally, in a study of 35 university students by Wang et al. (2023), mentalizing was positively associated with deception detection (i.e., deciding whether individuals in a series of videos were lying regarding stealing money) and mediated the relationship between romantic relationships and detection ability. The authors explain the connection between mentalizing and deception detection through the prism of leakage theory (Ekman & Friesen, 1969), which posits that the emotional and cognitive load that lying entails can cause the deceiver to inadvertently “leak” subtle bodily expressions and non-verbal cues that signal lying. Detection of such cues may rely at least partly on mentalizing abilities (for full discussion, see Wang et al., 2023).

Reflective functioning may also be a protective factor against FEV via its effects on additional social factors known to reduce FEV. Studies have demonstrated that reflective functioning abilities relate to improved psychosocial functioning, including better self-reported social skills (Yeh, 2013), more social participation (Bailey et al., 2008), and better self-reported friendship indicators (i.e., size, closeness, and frequency of contact; Lecce et al., 2017). In this regard, numerous studies have reported associations between social network size, social support, and loneliness and FEV (Beach et al., 2018; Lichtenberg et al., 2013; Liu et al., 2017). One reason for these findings may be that strong supportive relationships can help protect a person from being exploited or scammed. For example, a significant other may detect the malicious intentions of a potential scammer when the targeted person is not aware of it him/herself (see DeLiema, 2018) for discussion of this notion through the lens of Routine Activity Theory). In support of this possibility, studies have found that being married and/or living with a spouse or partner is associated with lower risk of financial exploitation (Laumann et al., 2008; Peterson et al., 2014; Podnieks, 1992). Thus, it may also be the case that in the absence of strong reflective functioning skills, a close partner can serve as a safeguard for detecting deceitful players.

In light of the paucity of literature linking reflective functioning to FEV, the main aim of this study was to examine whether better reflective functioning is associated with reduced FEV. A secondary aim was to examine whether being in a relationship would moderate the association. With regard to our first aim, we hypothesized main effects of reflective functioning and relationship status on FEV, such that better reflective functioning and being in a relationship would be associated with lower FEV. Second, we hypothesized that the association between reflective functioning and FEV would be strongest for those who are not in a relationship.

Methods

Participants and procedure

Data were collected from a convenience sample of 156 community-dwelling Israeli older adults (age range = 60–92, M = 66.97, SD = 6.46), of which 60 (38.5%) were male. Most participants reported that they were in a relationship (n = 112, 71.8%), had at least a high-school education (n = 131, 83.9%), and that their monthly household income was at least 8,000 NIS, equivalent to about $2,000 (n = 111, 71.1%; see Table 1 for additional information and correlation matrix for study variables). Inclusion criteria were being over the age of 60 and proficient in Hebrew. Participants were excluded if they had known cognitive or neurological impairments and/or a diagnosis of a cognitive or neurological disorder. Research assistants approached eligible participants who were supplied with an anonymous Qualtrics link which contained the informed consent form and the study questionnaires. Only participants who provided informed consent continued on to complete the study questionnaires. No personal information was required or provided. Unless otherwise specified, scales were back-translated into Hebrew by two bilingual psychologists. The study was approved by the IRB of Bar-Ilan University.

Table 1. Descriptive statistics and correlations among study variables.

	Variable	M/%	SD	1	2	3	4	5	6	7
1.	Age	66.97	6.46	–
2.	Sex^a (male)	38.5%	–	−.08	–
3.	Education^b	4.32	1.93	−.20*	−.16*	–
4.	Income	7.14	2.41	−.39***	−.28***	.60***	–
5.	Sum of illnesses	1.84	1.47	.41***	−.05	−.15	−.21*	–
6.	Reflective functioning	4.11	.84	−.03	.06	.08	.11	−.20*	–
7.	Relationship status^b (not in a relationship)	24.4%	–	−.19*	−.17*	.18*	.41***	−.07	−.13	–
8.	FEVS	6.37	4.34	−.02	.18*	−.23**	−.43***	.16	.24**	.16

n = 156.Correlation values represent Pearson coefficients except for coefficients for sex, and relationship status, which represent point-biserial coefficients, and for education, which represents Spearman’s Rho.

^a− 0=male; 1=female. ^b- 1= elementary; 2= partial high-school; 3= high-school; 4= partial undergraduate; 5= undergraduate degree/diploma; 6= partial graduate; 7= M.A or equivalent; 8= Ph.D or equivalent. ^c- 0=not in a relationship; 1=in a relationship

* = p < .05; ** = p < .01; *** = p < .001.

Measures

Reflective functioning was assessed by the Hebrew version of the Reflective Functioning Questionnaire (Fonagy et al., 2016), available on the authors’ website. Eight items concerning individuals’ ability to mentalize others’ feelings, wishes, and desires (e.g., “People’s thoughts are a mystery to me” [reversed item]) are rated on a scale ranging from 1 (“strongly disagree”) to 6 (“strongly agree”). Reverse items were recoded, and a mean score was calculated, with higher scores indicate higher reflective functioning abilities (Cronbach’s alpha = .76).

FEV was assessed by the 17-item Financial Exploitation Vulnerability Scale (Lichtenberg et al., 2020) which measures contextual factors associated with financial decision making (e.g., financial awareness; psychological vulnerability; see Lichtenberg et al., 2020). The measure is available online where it was renamed as the Financial Vulnerability Survey (https://olderadultnestegg.com). Items (e.g., “Has a relationship with a family member or friend become strained due to finances as you have gotten older?,” “Are your memory, thinking skills, or ability to reason with regard to financial decisions or financial transactions worse than a year ago?”) are rated on a varying response scale, with higher scores indicating greater FEV. This scale has been used to examine FEV in older Israeli adults (e.g., Weissberger & Bergman, 2022; Weissberger et al., 2022) and Cronbach’s alpha in this study was .80.

Subjects also provided information regarding age, sex (male/female), relationship status (not in a relationship/in a relationship), education level (rated from 1, “elementary education” to 8, “Ph.D or equivalent”), monthly household income (ranging from 0, “no income,” to 10, “over 30,000 NIS [about $7,700]”), and sum of medical illnesses (see Shrira et al., 2011), calculated as the sum of positive responses to 15 medical conditions diagnosed by a physician (e.g., respiratory illnesses; cancer).

Data analysis

Study variables of interest were examined for normality prior to conducting analyses

Initial correlations between the study variables were calculated, and the study hypotheses were examined by a hierarchical regression, with FEV as the dependent variable. The first step included the following potential confounding socio-demographic variables: age, sex, education level, income, and sum of illnesses. The second step included the hypothesized main effects of reflective functioning and relationship status (not in a relationship/in a relationship), and the third and final step included the reflective functioning × relationship status interaction, which was probed using Model 1 of the PROCESS 3.4 macro for SPSS (Hayes, 2018). A power analysis for detecting a medium to strong effect size (.20) with 8 predictors required a sample size of 122, indicating that the current sample was sufficient for the study model. Potential multicollinearity between the predicting variables was rejected, as the values of both tolerance and variance inflation factor (VIF) ranges were .59–.98 and 1.11–1.69, respectively.

Results

All continuous variables included in the study model were examined for normality prior to conducting analyses, and separately by relationship status for the RFQ and FEVS. Regarding covariates, skewness, and kurtosis values for age (1.32, 1.54), education (.235, −.890), income (−.270, −.452), sum of medical illnesses (.599, −.195) fell within acceptable limits according to (Tabachnick et al., 2013; absolute values of < 3 for skewness and < 10 for kurtosis). Skewness and kurtosis values for the FEVS in the entire sample were .874 and .950, respectively. When splitting the sample by relationship status, skewness and kurtosis of the FEVS for those in a relationship were .909 and 1.595 respectively, and .645 and .216 for those not in a relationship. Regarding the RFQ, skewness and kurtosis were .365 and .111, respectively, for the entire sample. In examining these metrics separately by relationship status, skewness and kurtosis for the RFQ were −.400 and −.056 for those in a relationship and −.084 and −.837 for those not in a relationship. Thus, values of skewness and kurtosis for the RFQ and FEVS, and within each subgroup, also fell within acceptable limits (Tabachnick et al., 2013). Average responses by item on the FEVS and RFQ are presented in Supplementary Tables S1 and S2, respectively.

Initial correlations reveal a weak negative correlation between reflective functioning and FEV, such that high levels of reflective functioning were associated with reduced FEV (r= −.25, p < .01). Relationship status was not correlated with reflective functioning (r= −.13, p > .05) and FEV (r= −.16, p > .05; see Table 1). In partial corroboration of the first hypothesis, a main effect was found for reflective functioning predicting FEV (B= −1.11, SE= .43, β= −.21, p < .05); relationship status did not demonstrate a significant main effect (B=-.36, SE= .87, β= −.04, p > .05; see Table 2 for regression coefficients). In line with the second hypothesis, a significant interaction of reflective functioning × relationship status was discovered (B=-.21, SE= .08, β= −.09, p < .01), adding 4.2% to the total variance (Total R²= .305; see Table 2). Probing this interaction using PROCESS (Hayes, 2018) revealed that the negative association between reflective functioning and FEV was significant only for older adults who were not in a relationship (B= −3.51, SE= .98, β= −.68, p< .001), and was nullified for individuals who were in a relationship (B= −.62, SE= .46, β= −.12, p> .05; see Figure 1). The results remained unchanged when covariates were excluded.

Figure 1. The interaction between reflective functioning and relationship status in predicting FEV.

Table 2. Hierarchical regression coefficients for FEVS (n = 156).

			95% CI for B
	Variable	B	LL	UL	SE B	β	t	p	ΔR^2
Step 1									.223***
	Age	−.13	−.25	−.01	.06	−.19	−2.14	.035
	Sex^a	.42	−1.04	1.87	.74	.05	.57	.571
	Education^b	.07	−.38	.51	.23	.03	.29	.771
	Income	−.84	−1.20	−.48	.18	−.47	−4.58	<.001
	Sum of illnesses	.44	−.06	.94	.25	.15	1.75	.083
Step 2									.040*
	Reflective functioning	−1.11	−1.96	−.26	.43	−.21	−2.58	.011
	Relationship status^c	−.36	−2.08	1.35	.87	−.04	−.42	.675
Step 3									.042**
	Reflective functioning × Relationship status	2.41	.65	4.17	.89	.49	2.72	.008
Total R^2									.305

^a- 0=male; 1=female. ^b- 1= elementary; 2= partial high-school; 3= high-school; 4= partial undergraduate; 5= undergraduate degree/diploma; 6= partial graduate; 7= M.A or equivalent; 8= Ph.D or equivalent. ^c- 0=not in a relationship; 1=in a relationship.

CI = confidence interval; LL = lower limit; UL = upper limit; *= p < .05; **= p < .01; ***= p < .001.

Discussion

This study examined the relationship between reflective functioning and financial exploitation vulnerability (FEV). Consistent with the first hypothesis, increased reflective functioning abilities were associated with lower FEV after controlling for socio-demographic variables of age, sex, education level, income, and sum of illnesses. A main effect of relationship status on FEV was not found. However, consistent with the second hypothesis, the association between reflective functioning and FEV remained significant only among older adults who reported not being in a relationship; the reflective functioning-FEV link was nullified for those in a relationship.

To our knowledge, this is the first study to report a relationship between reflective functioning and FEV. The findings are consistent with the social cognitive neuroscience model of financial exploitation proposed by Spreng et al. (2016). According to this model, changes to cognitive functioning and social cognitive processes associated with aging interact to predict FEV. Reflective functioning is a social cognitive process that involves the ability to reflect on one’s own and others’ feelings, goals, attitudes, and wishes (Fonagy et al., 2016). This ability may be particularly important when a person is confronted with individuals who have nefarious intentions such as scammers. According to leakage theory (Ekman & Friesen, 1969), a deceitful individual may leak subtle bodily expressions and non-verbal cues that may signal to a person that the deceiver is lying. Mentalization (i.e., reflective functioning) may help a person detect such cues and avoid a potentially exploitative financial situation. This notion is consistent with previous studies that have examined the relationship between mentalizing and deception detection in younger adults (e.g., Stewart et al., 2019; Wang et al., 2023).

While there was no main effect of relationship status on FEV, we found that it had a moderating effect on the association of reflective functioning and FEV. Specifically, the association between low reflective functioning and high FEV was significant only in those who were not in a relationship. This may suggest that a partner in some way compensates for low reflective functioning to ultimately mitigate FEV. For example, it may be the case that a spouse is able to detect deceitful players when the partner fails to do so him or herself. This notion is supported by DeLiema’s (2018) application of Routine Activity Theory to FEV. Routine Activity Theory (Cohen & Felson, 1979) posits that criminal acts occur when three factors converge: a) a motivated offender, b) a suitable target, and c) the absence of capable guardians. In the context of financial exploitation, DeLiema (2018) discusses that risk of fraud (i.e., financial exploitation by a stranger) is particularly high when there is an absence of capable guardians to safeguard the person’s assets. She explains that this may occur in situations in which a person is experiencing subtle cognitive changes associated with a potential underlying dementia process. In early stages of dementia, these changes go undetected and do not elicit intervention by family or friends in the management of finances, but may still render a person more vulnerable to bad actors such as fraudsters. Reflective functioning is one such social cognitive process that is known to decline with age (Charlton et al., 2009; Henry et al., 2013; Moran et al., 2012; Pardini & Nichelli, 2009; but see; Castelli et al., 2010) and is affected by early dementia processes such as Alzheimer’s disease (Bora et al., 2015; Castelli et al., 2011; Lucena et al., 2020 for reviews). In cases in which reflective functioning is low, having a spouse or partner can serve as a critical safeguard against FEV.

One notable methodological consideration of this study is that both reflective functioning and FEV were measured using self-report questionnaires. The FEVS measures FEV by assessing the context of a person’s financial reality, including financial awareness, psychological vulnerability, and relationship strain (Lichtenberg et al., 2020, 2021). The FEVS has been shown to be sensitive to actual financial exploitation experiences (Lichtenberg et al., 2020), though questions themselves relate to contextual factors that may increase risk of financial exploitation. For example, questions assess whether the individual has a person with whom they can talk to or receive assistance regarding financial matters, as well as whether they feel concerned about their own financial arrangements and behaviors. Thus, self-reported reflective functioning may relate indirectly to FEV by its association with these contextual financial factors. Relatedly, interpretation of the moderating role of relationship status should also be considered in the context of the FEVS. While it is possible that having a partner may directly reduce FEV during actual financial exploitation attempts in which the partner is present and can help identify malicious intentions, this was not measured in the present study and the self-report items on the FEV do not capture this type of real-world scenario. The interaction effect of relationship status may instead reflect an individual’s knowledge of the role his or her partner plays or could play in financial interactions. While this is the first study to our knowledge to examine the relationship between reflective functioning and FEV and thus holds merit on its own, similar investigations using more objective metrics of FEV and FEV-related constructs (e.g., confirmed history of financial exploitation, financial decision-making assessments) and reflective functioning are also warranted.

In this study, relationship status was not related to FEVS. This is in contrast to some studies that have reported an association between being married and lower FEV (Laumann et al., 2008; Peterson et al., 2014; Podnieks, 1992) and to a study by Lichtenberg et al. (2021; authors of the FEVS) which found that living alone was associated with increased scores on the FEVS. It is unclear why, regardless of the significant interaction, we did not find a direct association between relationship status and FEVS. Most studies that have reported associations between being married and FEV have been epidemiological in nature with very large samples. It is possible that the effect is small in nature and that our study is underpowered to detect an association between relationship status and FEV. Consistent with this, the r-value of the bivariate association between relationship status and FEV was .16, indicating a relatively small effect.

This study is not without limitations. Participants were a convenience sample of older adults recruited using snowball sampling techniques. Such sampling methods are inherently non-representative of the population. Relatedly, as questionnaires were answered via an online platform, older adults without sufficient computer literacy to independently complete the survey were excluded. A second limitation relates to the self-report nature of the study. Findings reflect both self-reported reflective functioning and self-reported FEV, which are prone to response bias. Objective measures of these constructs will help draw firmer conclusions regarding their relationships. Additionally, the study is cross-sectional in nature and thus directionality of the relationships cannot be determined. While our interpretation of the findings assume that reduced reflective functioning increases FEVS, it is possible that the opposite connection is true, that increased FEVS reduces reflective functioning. A longitudinal design can resolve such unknowns. This study also did not examine mechanisms that may explain the relationships observed. For example, poor reflective functioning has been shown to be associated with reduced prosocial behaviors (e.g., conduct problems in adolescents, Morosan et al., 2020), which can result in increased social isolation and loneliness. Social factors such as loneliness and low social support are known risk factors of FEV (Beach et al., 2018; Liu et al., 2017), and may therefore explain, at least partially, the connection between reflective functioning and FEV. Future studies are necessary to investigate the mechanisms behind the relationship between reflective functioning and FEV.

Despite these limitations, the present study is the first to our knowledge to directly examine the relationship between reflective functioning and FEV. We found lower reflective functioning to be associated with greater FEV, a connection that appears to be especially relevant for those older adults not in a relationship, as evidenced by a moderating effect of relationship status. Additionally, our findings may assist healthcare professionals in identifying individuals who are at high risk of experiencing financial exploitation. For example, a single patient who shows weak reflective functioning skills may be at greater risk of experiencing financial exploitation. Accordingly, healthcare providers may choose to implement preventative measures targeted at such particularly vulnerable individuals. In this regard, clinicians may choose to increase the individual’s awareness to specific situations that may become potentially exploitative, or more generally to educate the individual about popular fraud and scam methods. While the literature on financial exploitation risk factors is growing, more work is needed to develop targeted preventative techniques that help mitigate risk of financial exploitation. Such efforts can also focus on higher risk populations of older adults in order to further enhance our understanding of the most effective interventions.

The findings of this study also highlight newly identified risk factors of FEV that can themselves serve as targets for intervention. For example, it is possible that improving an older adult’s reflective functioning abilities may reduce FEV. While interventions aimed at improving reflective functioning have mostly been conducted amongst parents of young children (e.g., see meta-analysis by Lo & Wong, 2022) or populations with psychiatric disorders (Katznelson, 2014 for review), such interventions may also prove effective in improving reflective functioning in older adults, thereby mitigating FEV.

Findings also suggest that certain populations of older adults, such as those with illnesses and psychological disorders in which reflective functioning is particularly affected (Katznelson, 2014 for review), may carry an increased risk of FEV. Few studies have investigated differences in FEV across specific at-risk populations of older adults, such as those with mental illnesses. Thus, additional research is warranted to replicate the study findings in other older adult populations. Future research may also focus on utilizing objective measures of reflective functioning and FEV in order to confirm the associations reported in the present study.

Clinical implications

• The association of poorer reflective functioning with greater financial exploitation vulnerability (FEV) suggests that poorer reflective functioning may be a risk factor of financial exploitation among older adults, and that the risk may be particularly high for those who are not in a relationship.

• Healthcare providers and care workers who are in regular contact with older adults may consider assessing reflective functioning and/or identifying older adults who may have low reflective functioning due to specific psychological disorders that affect reflective functioning abilities.

• Remaining alert to any potential exploitative situations within these groups of older adults may help prevent financial exploitation from occurring or allow for quick intervention in the event of an ongoing exploitative situation.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Data can be made available upon request to the corresponding author.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/07317115.2024.2320921

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.