New directions for studying the aging social-cognitive brain

Abstract

The study of social cognition has extended across the lifespan with a recent special focus on the impacts of aging on the social-cognitive brain. This review summarizes current knowledge on social perception, theory of mind, empathy, and social behavior from a social-cognitive neuroscience of aging perspective and identifies new directions for studying the aging social-cognitive brain. These new directions highlight the need for (i) standardized operationalization and analysis of social-cognitive constructs; (ii) use of naturalistic paradigms to enhance ecological validity of social-cognitive measures; (iii) application of repeated assessments via single-N designs for robust delineation of social-cognitive processes in the aging brain; (iv) increased representation of vulnerable aging populations in social-cognitive brain research to enhance diversity, promote generalizability, and allow for cross-population comparisons.

Introduction

Decades of cognitive neuroscience research on older adults has generated robust findings regarding stability and change in general cognitive functions [1]. The impacts of aging and age-related disease on social cognition, in contrast, are less well understood, though the costs of social-cognitive dysfunction are high [2]. Social cognition refers to abilities relating to attending, perceiving, and remembering social information and accurately identifying and understanding thoughts, feelings, and intentions of the self and others, which influence social behaviors [3].

The current social-cognitive aging literature primarily revolves around the concepts of social perception, theory of mind (ToM), empathy, and social behavior [3–5]. Specific “social-cognitive” brain regions and networks have been linked to these constructs (e.g., anterior insula, amygdala, temporoparietal junction, and medial prefrontal cortex) [6]. Table 1 provides definitions and an overview of research findings on social cognition in aging. In brief, social cognition in healthy aging is characterized by both stability (e.g., intact affective empathy [7]) and change (e.g., decreased emotion identification and deception detection [8,9], but enhanced prosociality [10,11]). Both structural and functional alterations of the brain [12] as well as motivational changes [13] have been discussed as underlying mechanisms.

Table 1.

Four Domains of Social Cognition Impacted by Aging based on Henry et al., 2023.

Domain	Functional definition	Brain regions/networks	Age-related findings	Caveats
Social Perception	Ability to recognize, interpret, and make judgments based on socioemotional characteristics of others (e.g., faces, non-verbal gestures, prosody, eye gaze, and other physical and behavioral cues) Involves integration of sensory (e.g., visual/auditory signals) and contextual (e.g., environment, social norms, past experiences) information to meaningfully interpret social cues	Temporal, occipital, and parietal cortices for the processing of sensory information; frontal cortex for integration of social meaning	↓ Facial emotion identification ↑ Trustworthiness attributions	Findings indicating decline in social perception in older adults are dependent on stimulus features, social contexts, and motivational biases that emerge with increased age
Theory of mind (ToM)	Ability to perceive and understand others’ mental (i.e., cognitive ToM) and emotional (i.e., affective ToM) states, and recognize that they differ from one’s own	Default mode network (e.g., medial prefrontal cortex, posterior temporoparietal junction, precuneus, and posterior superior temporal sulcus)	↓ Cognitive ToM ↓ Affective ToM	Decline in cognitive ToM may be more substantial than decline in affective ToM in aging
Empathy	Emotional responses toward others, including sharing emotional states or appropriateness of emotional response	Medial prefrontal cortex, temporal lobe, posterior superior temporal sulcus, hippocampus, anterior insula, and anterior/posterior cingulate cortex	↓ Empathic accuracy ↨/↑ Emotional congruence, sympathy	Methodological differences in the measurement of empathy may explain mixed findings
Social behavior	Actions relating to self and others including how individuals interact and bond with each other and themselves, including decision-making	Decision-making: Prefrontal cortex (PFC) (specifically ventromedial PFC and orbitofrontal cortex), insula, caudate, amygdala, and anterior paracingulate cortex (PCC)	↓ Trust-related decision-making ↑ Susceptibility to deception ↨/↑ Prosociality	Limited by issues relating to operationalization and measurement

Notes: ↑ increased, ↓ decreased, ↨ maintained.

Social perception

One aspect of social perception that received particular attention in aging research is emotion identification [9], with evidence largely pointing to declines in older adults’ ability to accurately identify emotions in others [3,5]. Age-related differences in emotion identification are qualified by contextual specificities and characteristics of experimental tasks. This includes specific emotions (e.g., anger, sadness, fear) being particularly difficult to distinguish [14,15]. Age-related neural alterations such as decreased white matter integrity of the fusiform gyrus are associated with age-related decline in emotion identification [16]. Also, less functional connectivity in the amygdala, superior temporal sulcus, and medial prefrontal cortex has been observed for negative compared to positive faces [17].

Research, however, also supports maintained and even enhanced identification of certain emotions (e.g., disgust [14]). Relatively greater accuracy in identifying disgust is associated with greater insula activity in older adults [16] and may reflect an age-related shift away from the eyes and toward the mouth in face processing [18]. Comparatively less well understood are impacts of other stimulus features such as facial age and race/ethnicity [19] or non-facial social information [15].

Theory of mind

Older, compared to younger, adults experience declines in understanding and interpreting both mental (i.e., cognitive ToM) and emotional (i.e., affective ToM) states of others [20]. Cognitive ToM decline is associated with reduced medial prefrontal cortex activity (e.g., during false-belief stories [21], while thinking about another person’s mental state [22]). Affective ToM decline is associated with weakened default mode network connectivity, specifically intrinsic connectivity between the right temporoparietal junction and right temporal pole [23]. However, affective ToM appears relatively more preserved in aging than cognitive ToM [4]. Further, poorer performance in ToM tasks among older than younger adults may be due to differences in cognitive strategy use, especially when understanding emotions and detecting deception are involved [24].

Empathy

Perspective-taking, as one facet of empathy, declines with age, while other facets of empathy remain intact, reflected in age-group similarities in both self-report and behavioral performance in tasks that induce personal distress and/or empathic concern (e.g. pain mimicry videos) [7,25]. Maintenance of some aspects of empathy with age may be attributed to more widespread anterior cingulate functional connectivity [26]. Findings also suggest that older adults recruit the right anterior insula less than younger adults during empathy tasks involving pain and other unpleasant stimuli [7]. Perspective-taking, which relies on the ability to understand the thoughts of others, is negatively impacted by working load with aging. In fact, studies show that working load lowers insula and amygdala activation, in addition to levels of empathy while processing emotions, in older adults [7].

Social behavior

One type of social behavior particularly relevant for older adults’ well-being is social decision-making, which is subject to age-related affective, integrative, and motivational changes in both brain and behavior [27]. Older, compared to younger, adults tend to engage in behaviors that can incur greater susceptibility to deceit [8]. Deciding to trust others is a dynamic process that involves the interplay between accurately interpreting cues of trustworthiness in others and updating initial perceptions of trustworthiness as more reliable information becomes available. Brain regions, such as the orbitofrontal cortex and temporoparietal junction, are critically involved in trustworthiness impressions and posterior cingulate, frontoparietal cortex, and anterior insula in integrating information indicative of untrustworthiness [28]. Trust-related decision-making is susceptible to age-related biases such as older than younger adults’ greater tendency to perceive faces as more trustworthy [29], perhaps due to greater sensitivity to trustworthiness cues in reward and face-sensitive brain regions [30]. Older adults may also use visual cues to a greater extent to inform trust-related decision-making, reflected in over-recruitment of visual processing brain regions [31].

In parallel, prosocial and altruistic behaviors (e.g., cooperation, generosity) increase with age, perhaps motivationally driven by emotional gratification [10]. Greater prosociality not only fosters positive social relationships but can also be detrimental when older adults interact with untrustworthy partners, leading to greater susceptibility to fraud and exploitation [8]. Given the growing prevalence of fraud targeting older adults, a comprehensive understanding of cognitive and brain mechanisms underlying the ability to detect deception by others is critically needed.

Taken together, aging is associated with decline but also maintenance and even improvement in different social-cognitive functions. While important discoveries regarding the aging social-cognitive brain have been made over the last two decades, limited scope and methodology impact comprehensive, robust, and generalizable knowledge gain. In the following, we have identified promising directions for future investigation into the aging social-cognitive brain. In particular, we call for (i) standardized operationalization and analysis of social-cognitive concepts to increase comparability and convergence across research studies; (ii) use of naturalistic paradigms to enhance ecological validity of social-cognitive processes assessed; (iii) application of repeated assessment (single-N designs) for robust delineation of social-cognitive response in the aging brain within individuals; (iv) inclusion of vulnerable aging populations to enhance diversity, promote generalizability, and allow for cross-population comparisons in the study of social-cognitive brain aging (Figure 1).

Figure 1.

Promising new research directions for advancing the social-cognitive neuroscience of aging.

Promising future directions into the aging social-cognitive brain

Achieving consensus on operationalization and analysis of social-cognitive constructs

Over many years of cognitive aging research, central concepts have been well-defined and validated (e.g., crystalized and fluid intelligence) [32]. The use of standardized assessment batteries, including computerized measurements, has been particularly beneficial for characterizing age-related stability and change in cognitive aging and delineating relationships between cognitive components [33]. In contrast, operational definitions and measurements of central concepts in social-cognitive aging are more inconsistent across studies, which makes meta-analytic consensus challenging [15,34]. For example, social behavior has received very little empirical attention with relevance to aging (but see some research on social inappropriateness, which has been associated with increases in normal and neurodegenerative aging [4,35]). In fact, the study of social behavior in aging can be regarded as a highly inconsistent social-cognitive domain in terms of operationalization due to the diverse range of behaviors possible which may not be adequately captured by singular measures or tasks (but see the study by Grainger et al. [4]).

A greater focus on specific social behaviors of particular relevance to older adults will be beneficial for generating solid directional conclusions in future research and forming a basis from which greater well-being can be fostered among older adults and their social partners (e.g., social engagement and bonding can enhance quality of life of older adults; decision-making is important for life management and planning needs in aging). Further, clear definitions and agreed-upon operationalizations, supported by both strong theory and rigorous psychometrics, will also facilitate data harmonization across studies and brain-behavior comparisons for greater knowledge convergence in the emerging field of social-cognitive aging neuroscience.

Enhancing ecological validity of social-cognitive assessments

Replicating real-world social cognition in a laboratory setting is difficult given the complexity and diversity of social-cognitive phenomena; most current social-cognitive paradigms lack ecological validity. The use of technology (e.g., ambulatory assessment, augmented reality) to create naturalistic social contexts in which realistic behaviors are produced is an emerging interest in social-cognitive research [36]. One recent attempt at designing ecologically valid tasks is the Phishing Internet Task paradigm, which assesses vulnerabilities to deception in aging in a naturalistic computing environment [37]. Using this paradigm has shown a deviation between laboratory-based self-reports of phishing susceptibility and actual behavior. This empirical pattern exemplifies the need for more behavior-based, ecologically valid social-cognitive assessments to ensure that observed study responses reflect behavior in the real world.

Building on these representative findings, brain processes underlying social cognition in real life among older adults can be more closely and validly tracked. For example, future research designs could leverage ambulant methods (e.g., mobile electroencephalography, eye-tracking, functional near-infrared spectroscopy) and dynamic neuroimaging techniques (e.g., hyperscanning to simultaneously measure brains active during social interactions) to capture social-cognitive neural processes in aging more naturally.

Applying single-N designs to delineate social cognition in the aging brain

Individual brains differ in shape and size, but group-based analysis of brain imaging data is based on the notion of an average brain [38]. Social-cognitive aging research to date has largely relied on such comparisons, where brain and behavioral responses of older and younger adults are averaged between individuals with conclusions drawn based on these group averages. Although group-based inferences have revealed important age-related differences in social cognition, in-depth analysis of individuals via repeated scans is warranted to adequately describe social-cognitive processes in the aging brain (i.e., single-N design or deep imaging [38,39]).

The single-N design is an individual-subjects neuroimaging and analytics approach that allows for extensive measurements of responses within single brains from one (or several) individuals. This design circumvents limitations associated with group-level inferences and promotes greater precision in neuroscience by magnifying brain dynamics during social cognition within one individual. Group-averaged brains are fit into templates that represent a singular brain but are not necessarily representative of any actual individual brain tested. The morphological impacts of aging on the brain amplify this concern as there are further variabilities in the representation of brain structure, function, and physiology to consider when studying older adults. Other advantages of the single-N design include increased signal, reduced inter-subject/between-study variability, and reduced subjectivity of localization when investigating neural patterns. This is critical given that repeated sampling within individuals can highlight brain regions of interest active during social cognition that may have otherwise washed out through group-based averaging [40]. Cognitive neuroscience is already successfully leveraging single-N designs [41], but social-cognitive aging research has yet to incorporate this methodology to form reliable conclusions [38].

Including vulnerable populations in the study of social-cognitive brain aging

Some older adults experience greater social-cognitive changes than others [4]. Divergent trajectories may be due to alterations in brain morphology and function, which are especially compromised among older adults managing chronic health conditions, such as neurodegenerative disease (e.g., Alzheimer’s Disease and Related Dementias (ADRD), Parkinson’s disease (PD)) and chronic pain [42,43].¹ In ADRD, social-cognitive deficits are linked to the deterioration of limbic, frontal, and parietal areas [44] and in PD with degeneration of nigrostriatal and mesolimbic pathways [35]. These brain changes result in reduced inhibition, increased socially inappropriate behaviors, and greater caregiver burden [35]. However, the unique impacts of age-related neurodegenerative disease on specific components of social cognitive function are not well understood (but see the study by Setien-Suero et al. [44]). Affective ToM/cognitive empathy (i.e., understanding others’ mental states) and emotion identification are impaired in ADRD [35] and PD [45]. However, investigations directly linking social-cognitive abilities with brain regions impacted by neurodegenerative disease are still needed [46].

Chronic pain is another health condition associated with altered cognitive and emotional function that places strain on relationships with others [47]. Findings are limited but social-cognitive impairments have been observed in chronic pain conditions such as fibromyalgia (e.g., declines in ToM and emotional function [48]) and complex regional pain syndrome (e.g., impaired social emotion recognition [49]). Still warranted is research that specifically addresses social-cognitive change within and across chronic pain conditions disproportionately affecting older adults, with direct links to impacts on the brain as well as treatment outcomes.

Finally, promoting inclusivity of diverse brain health trajectories in aging requires increased representation of racial/ethnic, cultural, and socioeconomic backgrounds in research [50]. Much work on normal cognitive aging to date skews toward white, highly educated, and economically stable individuals, who do not necessarily represent older adults most vulnerable to disparities in health and well-being. In fact, an emerging literature suggests that factors such as race/ethnicity, socioeconomic status, and sex contribute to differences in brain structure, function, and related behavior [50]. Enhanced diversity is crucial to delineate generalizability and specificity of aging effects on social cognition (e.g., cross-cultural variability [51]) as well as to promote greater rigor and reproducibility in social-cognitive brain aging research.

Conclusions

In summary, knowledge is growing and converging on age-related stability and change in specific social-cognitive functions and their associated brain mechanisms. Promising new directions that emphasize increased operationalization of social-cognitive constructs, ecological validity of social-cognitive measures, use of single-N designs, and representation of vulnerable aging populations will further enhance understanding of the aging social-cognitive brain.

Data availability

No data was used for the research described in the article.