A neurodevelopmental model of eating disorder risk in midlife to older adulthood

Abstract

Prior research suggests that rapid neurobiological changes and their interaction with psychosocial stressors serve as major precipitants of disordered eating in adolescence. However, there has been little discussion of how similar processes may facilitate disordered eating in later life, which is similarly associated with rapid biosocial change. We present a novel framework for the increased risk of eating pathology in midlife into older adulthood, suggesting testable mechanistic pathways to guide future research. First, we review critical neurobiological and cognitive alterations that occur during this unique developmental stage and link them to established risk and maintenance factors for disordered eating in younger populations. Second, we discuss how neurobiological aging processes interact with aging-related psychosocial and biological stressors to increase vulnerability for disordered eating. We then outline the following directions for future research in middle-aged and older adults: 1) directly investigate the influence of brain changes, and their interactions with age-related environmental and hormonal influences, on eating disorder behavior; 2) distinguish pathways from neurodevelopmental changes to eating pathology that differ as a function of aspects of identity, comorbidity, or specific developmental period; 3) establish the neurodevelopmental risk factors influencing risk for different eating disorder phenotypes; and 4) identify potential neural targets for prevention and treatment of eating disorders as vulnerable individuals age. Finally, we advocate for enhanced awareness of how neurobiological changes in midlife and older adulthood may increase the risk of eating pathology, and for a multidisciplinary approach that considers biological and environmental vulnerabilities in treating disordered eating in this population.

Eating disorders are often regarded as disorders of youth, arising from the convergence of rapid neurobiological changes and unique psychosocial influences occurring during adolescence. Yet, development continues beyond adolescence. In mid- to late-life, marked neurobiological, endocrinological, and metabolic shifts occur alongside major role transitions and sociocultural pressures, making this a neurodevelopmentally sensitive period for mental health concerns. Parallel to adolescence, the interaction of these disturbances may set the stage for the new onset, recurrence, or worsening of eating pathology (Matsumoto & Rodgers, 2020; Samuels et al., 2019). The epidemiological pattern of eating disorders in later life is still emerging. Several age-related factors, including medical and psychiatric comorbidities and cognitive decline, can obscure symptoms (Vaidyanathan et al., 2026) and gendered ageism among providers often precludes diagnosis (Kohestani et al., 2025), making true estimates of prevalence over the adult lifespan difficult. However, growing evidence points to a second heightened window of risk that begins later in life. For example, a significant majority of women with binge-eating disorder (BED) reported onset either in young adulthood (31.7%) or after age 55 (51.2%) (Kilpela et al., 2023). In a review of published cases of anorexia nervosa (AN), bulimia nervosa (BN), and eating disorder not otherwise specified in older individuals, nearly two-thirds described recurrence after a period of remission or new onset after age 50 (Lapid et al., 2010). Numerous additional eating disorder cases with late-life onset or recurrence have been described in the literature since (e.g., Cwikel, 2011; Malik et al., 2014; Rakusin et al., 2021; Schulte et al., 2026). This pattern is not unique to women; men also appear to have a later “peak” in eating disorder psychopathology between the ages of 55 and 64 (Hilbert et al., 2012).

Research suggests that the 12-month prevalence of eating disorders is around 4% among midlife women (Micali et al., 2017), and 0.6-1.2% among midlife and older men (Reas & Stedal, 2015), though the latter estimate is based on very few studies. Subthreshold disordered eating is also common in midlife, with 11-13% of midlife women (Fairweather-Schmidt et al., 2015; Gagne et al., 2012) and 7-25% of midlife men (Hilbert et al., 2012; Mangweth-Matzek et al., 2016; Rosenberger & Dorflinger, 2013) engaging in eating disorder behavior. Further, disordered eating among this age group has increased in prevalence over the last two decades (Ackard et al., 2013; Mitchison et al., 2014). Importantly, the physical consequences of eating pathology during midlife and beyond may be more severe than in adolescence or young adulthood (Samuels et al., 2019). For instance, a systematic review indicated that premature mortality secondary to eating pathology was 21% among older individuals with an eating disorder (Lapid et al., 2010). Nevertheless, many eating disorder studies, especially those focused on neural and cognitive mechanisms, exclude participants over 50. Thus, to date, very little is known about the unique neurobiological mechanisms that may influence the risk of eating pathology at this life stage.

Prior reviews have examined this period of elevated risk for eating disorders from a psychosocial perspective or have discussed the potentially important influence of changes in sex hormones that occur during mid- to late-life (e.g., Samuels et al., 2019; Matsumoto & Rogers, 2020; Anaya et al., 2023; Baker & Runfola, 2016). However, given that major neurobiological changes also occur during this period, we propose that these life stages represent a second biobehaviorally sensitive period that warrants significant further research. In this narrative, integrative review, we synthesize findings across psychology, neuroscience, endocrinology, and physiology to present a novel mechanistic framework for the promotion and maintenance of eating pathology in midlife and older adulthood. Though the age ranges used to define life stages vary across classification systems, based on prior neurobiological literature, we broadly define midlife as beginning in the early to mid-40s and older adulthood as beginning in the mid-60s (Dohm-Hansen et al., 2024; Ihle et al., 2016). Our review focuses on diagnosed eating disorders (e.g., AN, BN, BED) and subclinical disordered eating behaviors (e.g., restriction, loss-of-control eating, compensatory behaviors) and integrates evidence from both human and animal models, drawing on studies using a wide range of methods (e.g., structural and functional MRI, electrophysiological recordings, neuropsychological assessments).

Understanding the changes in brain structure and function that occur during this unique developmental stage may shed new light on potential new targets for psychopharmacology and psychotherapy in midlife and beyond. However, it should be noted that many of the pathways to risk discussed here are currently speculative and often based on research in non-clinical midlife and older adult populations. To date, very little research has directly examined associations between neurobiological functioning and disordered eating in adults over age 45 years. Therefore, we do not assume a single longitudinal trajectory of symptom engagement; rather, we use “risk” as an umbrella term to refer to either new onset, recurrence, or worsening of symptoms. This review and resulting proposed model propose transdiagnostic vulnerability for disordered eating and are intended to generate testable hypotheses that require additional empirical support. Based on this framework, we outline future research directions and potential implications for clinical practice.

Neurocognitive Changes

Aging is associated with significant changes in the brain that impact cognitive functioning. Beginning in midlife, demyelination of the nerve fibers that comprise white matter slows the transmission of nerve impulses, disrupting communication between brain regions implicated in coordinating multiple cognitive processes (Huang et al., 2025). Particularly affected are the major domains of reward processing, executive functioning, and interoception. As food is an innate, primary reward, age-related attenuation in neural activation across all phases of reward processing (e.g., anticipation, receipt, and learning) may set the stage for both under- and over-eating. Structural atrophy in the prefrontal cortex (PFC) disrupts executive functions crucial to guiding goal-directed behavior and effective decision-making (Ferguson et al., 2021; Zanto & Gazzaley, 2019), including decisions about eating and weight management. Downstream effects can be seen in cognitively demanding processes such as emotion regulation, where reduced flexibility may compromise adaptive responses to stressors (Pruessner et al., 2020). Finally, reduced functional connectivity in regions involved in self-referential processing may inhibit awareness of internal sensations (Dobrushina et al., 2024; Murphy et al., 2018), affecting eating behavior. Alterations in these domains have been implicated in eating disorders in younger populations (Steinglass et al., 2019), and the effects of aging on these processes may create an additional period of heightened vulnerability to eating disorders among adults in mid- to late-life.

Reward System Functioning

Circuits linking the ventral striatum (particularly, the nucleus accumbens [NAc]), medial PFC, midbrain/ventral tegmental area (VTA), and the orbitofrontal cortex comprise the brain’s reward system. Starting in midlife, the aging brain undergoes changes in these regions that affect reward processing, most prominently through declines in dopamine (DA) system function (Marschner et al., 2005). Here, we discuss how age-related reward-processing alterations may increase risk for maladaptive eating; however, conclusions about these relations are limited by a significant lack of neuroimaging studies in adults with disordered eating over 50 years. S significant efforts are needed to advance research that may elucidate these reward-related mechanisms in older adults with disordered eating.

Reward Anticipation

Dopaminergic neurons, originating in the VTA, play a crucial role in reward signaling in the ventral striatum (Arias-Carrión & Pŏppel, 2007; Schultz, 2016; Tzschentke & Schmidt, 2000). Several studies in mice and humans have documented aging-related DA receptor loss in the PFC and striatum over the adult lifespan, particularly in the NAc, a structure implicated in goal-seeking and motivated behavior (Karrer et al., 2017; Konar-Nié et al., 2023). Reduced DA concentrations may drive significant NAc volume loss and alterations in neuronal morphology during aging, leading to motivational and reward-processing disruptions often observed in older adults (Konar-Nié et al., 2023). For example, the adaptive “shift” of striatal activation from reward receipt to anticipation—a process that signals upcoming reward and aids in learning— seems to diminish during midlife, and this may reflect a decline in DA receptor density and concentration (Vink et al., 2015). Additionally, neuroimaging studies have shown reduced striatal activation in anticipation of monetary rewards among older (e.g., mid-60s) compared to younger (e.g., mid-20s) adults (Dhingra et al., 2020; Dreher et al., 2008). Blunted activation during reward anticipation could lead to abnormal motivation for and processing of food-related reward, which has been associated with the development and maintenance of dietary restriction and binge eating (Steinglass et al., 2019). Further, reduced reward-circuit engagement during reward anticipation could reduce the ability to motivate the pursuit of rewards that may buffer against eating pathology, such as social rewards (Haynos et al., 2021). Indeed, social-reward anticipation deficits have been linked to eating pathology, alongside other mental-health concerns (Aldridge-Waddon et al., 2020).

Reward Receipt

Weaker DA functioning during reward anticipation predicts similarly abnormal neural signaling once a reward is obtained. Studies on monetary-reward response show consistent age-related patterns. Early older adults (M = 65 years) have shown less activation of the bilateral fronto-parietal network than younger adults (M = 24.5 years) upon reward delivery, coupled with reduced deactivation of default-mode network regions (Dreher et al., 2008). Another study comparing older adults aged 65-75 years to younger adults aged 20-30 years found similar age-related attenuation in brain electrophysiological signals following feedback about negative and positive outcomes, and reduced signal strength differentiation between gains and losses (Hämmerer et al., 2011). Together, findings suggest that aging could be associated with a more diffuse and less efficient neural response to monetary reward due to reduced striatal dopaminergic signaling. A similar lack of differentiation between monetary rewards and losses has been observed in young adult women recovered from AN (Wagner et al., 2007) and BN (Wagner et al., 2010); given that these samples did not have an eating disorder at the time of measurement, this may reflect a premorbid risk factor. Blunted neural reward responses have also informed reward deficit theories across the eating disorders field. For instance, some theories suggest that individuals binge eat to compensate for a lower initial response to food (Vrieze & Leenaerts, 2023). Other theories suggest that reward deficits maintain AN symptoms because people affected by this illness are not motivated to pursue food rewards (Wierenga et al., 2015). Lending further support to the idea that a reduced reward response can promote risk for dysregulated eating in older adults, evidence suggests that the size and density of taste buds—the sensory organs responsible for first detecting information about food reward—decrease steadily with age beginning between ages 40-50 (Karikkineth et al., 2021; Shimizu, 1997). However, investigations comparing neural responses to food-related reward receipt between older and younger adults are too methodologically varied to support definitive conclusions about whether aging specifically impacts responses to food rewards (Bennett et al., 2021). More research is needed to determine whether age-related shifts in neural response to monetary rewards extend reliably to food.

Reward Prediction Errors

Learning from and about rewards is driven by prediction errors (PEs), or the discrepancies between anticipated and actual rewards, which are reflected in the phasic activity of midbrain DA neurons projecting to the ventral striatum (Schultz, 2016; Schultz et al., 1997). Positive PEs (when an outcome is better than expected) are associated with an increase in dopaminergic firing. In contrast, negative PEs (when an outcome is worse than expected) are associated with a depression of dopaminergic activity (Schultz et al., 1997); though it is now recognized that DA neuron populations are heterogenous, with some subpopulations responding to aversive outcomes with excitation rather than inhibition (Bromberg-Martin et al., 2010), and that other neuromodulators (e.g., serotonin) also contribute to negative PE signaling (Moran et al., 2018). Abnormally blunted or exaggerated PEs reflect disrupted reward learning processes that may result in maladaptive behavior (Hauser et al., 2019).

A decline in DA signaling may underlie impaired PE tracking in the ventral striatum in later life (Eppinger, Schuck, et al., 2013). During value-based decision-making, while younger adults predictably show PE-associated activation in the ventral striatum, adults between 60 and 80 years show greater PE-associated activation in regions associated with working memory (Don et al., 2022; Worthy, Davis, et al., 2015). Coupled with suboptimal task performance, this indicates that older adults rely more on a short-term strategy based on gains and losses in recent trials, rather than a strategy based on long-term averaging of outcomes across trials (Don et al., 2022). Further, negative computed PEs appear larger in late-middle-aged and older adults (e.g., ages 60-80) than in younger adults, suggesting an age-related heightened sensitivity to negative outcomes (Worthy, Otto, et al., 2015). If engaging in disordered eating becomes associated with avoidance of negative outcomes (e.g., weight gain) or affective states (e.g., guilt, anxiety), this age-related learning bias may impede the adoption of healthier and more rewarding behaviors. Though studies of PE in older adults have not incorporated food-related rewards, reward behaviors that pertain to eating may be particularly vulnerable to aging due to concurrent changes in taste processing.

Studies of PEs and related neural circuitry reveal distinct effects between ED diagnoses. In AN, food-related and monetary reward stimuli are associated with stronger PE responses—for positive and negative outcomes—compared to healthy controls (DeGuzman et al., 2017; Frank et al., 2012, 2018). This is presumed to be related to dopaminergic sensitivity, a compensatory response to sustained caloric restriction that would ordinarily stimulate food-seeking behavior (Frank, DeGuzman, et al., 2019). Despite elevated PE signals, individuals with AN appear to learn less effectively from PEs and show greater links between negative PEs and disordered eating (Uniacke et al., 2024; Wierenga et al., 2022). For example, restrictive eating may be maintained when the magnitude of “punishment” associated with weight gain exceeds that associated with an unmet drive for nourishment (Frank, DeGuzman, et al., 2019). This process could represent a particular vulnerability for older adults to become stuck in a pattern of restrictive eating, as individuals in this developmental period are less able to integrate longer-term outcomes, and they learn less quickly from reward and more quickly from punishment (Worthy, Otto, et al., 2015).

Conversely, reduced PE responses to reward and punishment may constitute a pathway towards loss-of-control eating. Decreased striatal DA receptor density has been observed in those with BN, contributing to a hypodopaminergic state consistently seen in this population (Broft et al., 2011; Yu et al., 2022). In BN and those with obesity, unexpected gains and losses of food-related reward have been associated with a blunted PE response compared to healthy controls (Frank et al., 2011, 2012; Olsavsky et al., 2019). Thus, it is hypothesized that a diminished PE response predisposes individuals to increased reward-seeking behavior, such as excessive intake of palatable food (Frank et al., 2011). Age-related alterations in the neural tracking of PEs, beginning in the mid-40s, coupled with reduced neural processing and taste differentiation (Green et al., 2013; Shimizu, 1997), may result in a similar vulnerability to dysregulated food intake, leading to binge eating.

Reward Learning Flexibility

Beyond the integrity of PE signals, adaptive reward learning also requires behavior to remain responsive to changes in reward outcomes. With aging, behavior becomes increasingly insensitive to shifts in reward contingencies. The degeneration of striatal DA transporters that occurs across adulthood (Karrer et al., 2017) may result in impairments in goal-directed (i.e., “model-based”) decision-making in late middle age and early older adulthood (Eppinger, Walter, et al., 2013), and instead promote the formation of automatic and inflexible (i.e., “model-free”) behaviors (Eppinger et al., 2011). Across all ages, an imbalance favoring model-free over model-based actions has been hypothesized to be a primary mechanism for repetitive, habit-like decision-making that could drive maladaptive restricting in AN, and the failure to adapt behavior in light of changing reward contingencies is seen in a range of eating pathology (Berner, Fiore, et al., 2023; Davis et al., 2020; Favier et al., 2020; Haynos et al., 2022; Reiter et al., 2017; Ritschel et al., 2017; Wang et al., 2023). As this imbalance amplifies with aging, maladaptive eating behaviors could become more easily entrenched and harder to extinguish in later life.

Executive Functioning

Following a general pattern of “last-in, first-out,” prefrontal brain regions are particularly vulnerable to age-related white-matter degradation that accelerates around age 50 (Gunning-Dixon et al., 2009). Volume decreases in the PFC, and resulting functional changes that begin in midlife and extend into older adulthood can impact cognitive processes that may directly influence the regulation of eating behavior and body-related thoughts (Ferguson et al., 2021; Zanto & Gazzaley, 2019). Below, we review aging-related shifts in the relevant domains of cognitive and behavioral control, working memory, and cognitive flexibility. Impairments in these executive functions have been linked to various forms of eating pathology in adolescent and young adult populations (Hirst et al., 2017). While literature linking eating pathology and executive functioning in midlife and older adults is sparse, evidence from two new-onset cases in adults in their 70s suggests that early decline in executive functioning may coincide with emergence of maladaptive eating in late life (Schulte et al., 2026), highlighting the need for further research to examine this link.

Cognitive and Behavioral Control

Neural connections between the PFC and sensory cortex enable “top-down” attention regulation, ignoring distractions in favor of goal-directed behavior. This function, known as cognitive interference control, shows well-documented declines with age and is associated with decreased functional connectivity between the PFC and other brain regions (Gazzaley & D’Esposito, 2007). In particular, older adults are worse at ignoring goal-incongruent environmental stimuli and suppressing irrelevant information from memory than emerging adults (Pettigrew & Martin, 2014), leading to deficits in adjusting attention in response to changing demands. As a result, older adults may find it harder to shift attention away from internal anxious thoughts (Price et al., 2011) or environmental cues about food, shape, or weight. Similar interference control deficits are observed in populations with BN and BED. Women with BN show impaired performance — which scales with eating disorder symptom severity — and hypoactivation in frontostriatal circuitry during tasks that measure cognitive interference control and response inhibition (Lee et al., 2017; Marsh et al., 2009). While general interference-control deficits have not been observed in BED (Voon, 2015), deficits in this group have emerged in the presence of food-related cues, indicating an attentional bias that may underlie food preoccupation (Colton et al., 2023; Lee et al., 2017; Svaldi et al., 2014).

Results from studies measuring age-related changes in behavioral control, or the ability to suppress automatic behavioral impulses, are mixed. In general, older adults (M=60-80 years) demonstrate impaired performance on tasks that require reactive suppression of dominant responses (e.g., stop-signal) (Rey-Mermet & Gade, 2018). Several recent studies, however, suggest that this impairment may not extend to food-specific contexts. One study reported no behavioral or neural differences between younger (20-30 years) and older (65-91 years) adults during a food-specific reactive inhibition task (Aiello et al., 2025), but another found that while task performance was intact, healthy adults over 55 years showed attenuated recruitment of PFC regions when inhibiting responses to high-calorie foods (Allen et al., 2023). Task performance during proactive inhibitory control appears to remain intact with age. Depending on the task used, some studies have found that aging is linked to either increased (interpreted as compensatory) or decreased (interpreted as hyper-efficient) brain activation in key PFC regions during proactive inhibition (Hsieh & Lin, 2017; Sebastian et al., 2013). Although behavioral performance on food-specific tasks appears largely preserved in older adults, reduced PFC recruitment may represent a latent vulnerability that could manifest as impaired control in specific conditions (e.g., heightened emotional or appetitive demand).

Impairments in behavioral control are hypothesized to promote the loss of control over eating that characterizes BN (Steinglass et al., 2019) and BED (Lavagnino et al., 2016). Young adults with BN (relative to healthy controls) and young adults with BED (relative to non-BED individuals with obesity) show hypoactivation of PFC regions when attempting to inhibit responses to disorder-relevant stimuli (Balodis et al., 2013; Berner, Winter, et al., 2023; Hege et al., 2015). Therefore, age-related decreases in PFC activation during response inhibition to high-calorie foods could increase the risk of new-onset or re-emergence of binge eating and, potentially, subsequent compensatory behaviors in later life. Indeed, a majority of older adults experiencing binge eating report onset in mid- (defined as ages 40-55) to late-life (defined as 56 years and older) (Kilpela et al., 2023). However, given the mixed nature of the current literature, conclusions about the specific symptoms associated with these neural patterns are limited. More research is needed to elucidate potential age-related alterations reflecting insufficient, inefficient, or hyper-efficient engagement of inhibitory control circuitry, and examine their association with different eating disorder presentations.

Working Memory

Diminished ability to streamline attentional efforts with advancing age results in the overloading of working memory with irrelevant information, reducing one’s capacity to effectively store and manipulate goal-relevant information (Park & Reuter-Lorenz, 2009). Verbal and visuospatial working memory begin to slow after age 30, and are especially sensitive to decline after age 60 (D’Antuono et al., 2022; Ferguson et al., 2021). Functional neuroimaging studies reveal that, compared to their younger counterparts, older adults require greater neural engagement of prefrontal regions during working memory tasks to compensate for cognitive declines (Swirsky & Spaniol, 2019). Research in younger populations has found inconsistent direct associations between working memory deficits and eating pathology (Smith et al., 2018). There is also early evidence to suggest that, across the adult lifespan, working memory impairment strengthens the association between emotion regulation difficulties and eating disorder severity (Barnhart et al., 2024). Thus, older adults who experience the greatest decline in cognitive control and working memory may have more difficulty regulating emotions adaptively in response to various stressors (Liang et al., 2017; Pruessner et al., 2020), placing them at heightened risk for eating pathology, which has been consistently associated with emotion regulation concerns (Prefit et al., 2019; Wong et al., 2025).

Cognitive Flexibility

While reward learning flexibility depends on the implicit updating of value associations, cognitive flexibility involves the explicit ability to shift between mental sets and behavioral strategies in response to new information. Cognitive flexibility can be measured through a variety of measures, but is most often assessed using set-shifting tasks that require participants to switch between strategies to achieve a goal. Cognitive flexibility requires attentional control and primarily involves recruitment of the frontoparietal network, including the dorsolateral PFC and parietal cortex regions (Uddin, 2021). It appears particularly sensitive to declines in white matter integrity, as it requires coordination of multiple brain networks (Jolly et al., 2017). However, the evidence for the effect of aging on set-shifting is not straightforward. Compared to younger adults, adults over 60 years may experience greater overall costs associated with maintaining multiple tasks simultaneously, but performance immediately following a task switch appears intact (E.-H. Chen & Hsieh, 2023). Some have suggested that an increased burden on working memory during switching tasks, rather than a true deficit in flexibility, may be driving these observations. Set-shifting impairments may promote compulsive eating behavior, causing individuals to remain “stuck” in maladaptive eating due to an inability to change behavior once it is no longer beneficial and potentially harmful. Inefficient set-shifting has been identified as an endophenotype across eating disorder presentations, except the AN binge-eating/purging subtype (Wu et al., 2014). Thus, age-related increases in these inefficiencies could be one contributor to disordered eating in midlife and older adulthood.

Emotion Regulation

In addition to neurocognitive changes affecting the brain’s response to rewards, aging also impacts the response to threats. Age-related alterations in systems that process emotional stimuli and support emotion regulation may amplify risk for maladaptive eating behavior.

Emotion regulation encompasses a range of processes by which an individual attempts to modulate their emotional responses to stressors in the service of their long-term goals. Emotion regulation strategies like cognitive reappraisal are supported by cognitive control regions within the fronto-limbic network, including the dorsomedial PFC, dorsolateral PFC, ventrolateral PFC, anterior cingulate cortex, and amygdala (Buhle et al., 2014; Kebets et al., 2021; Ochsner et al., 2012). However, specific strategies are not “one size fits all”; successful emotion regulation requires a balance between maintaining a particular strategy and adjusting to a different strategy in response to situational changes. This process relies on cognitive control to override initial emotional responses, working memory to store and update information, and cognitive flexibility to shift strategy use when contexts change (Pruessner et al., 2020). The ability to apply emotion regulation flexibly and adaptively requires engaging the anterior PFC to monitor and update the values of alternative regulation strategies (Adamczyk et al., 2025; Koch et al., 2018; Roelofs et al., 2023).

Flexible and adaptive emotion regulation is particularly important during middle and older adulthood. During these life stages, individuals encounter numerous stressors across multiple contexts and domains (Benson et al., 2019; Pruessner et al., 2020). Although emotion regulation does not seem to universally worsen with age (Isaacowitz, 2022), those who experience greater age-related declines in cognitive resources may struggle to cope effectively with stressors. Relative to emerging adults, early older adults show degradation in both gray- and white matter integrity in the fronto-limbic network, including the anterior PFC (Henry et al., 2025). These structural changes may have direct effects on emotion regulation ability. One study in adults aged 56-83 years found that decreased white-matter integrity in a fronto-limbic pathway was associated with an accelerated rate of decline in executive functioning (Lloyd et al., 2021). In turn, this decline was associated with poorer amygdala downregulation despite greater PFC activity during an emotion regulation task. Degeneration in networks involved in cognitive control could also lead to failures in monitoring and in shifting emotion-regulation strategies. As a result, ineffective strategies may be applied rigidly and inflexibly. While the literature supporting this framework in older adults is in its infancy, initial evidence points to a decline across the adult lifespan in emotion regulation flexibility in response to internal (i.e., affect) and external (i.e., context) changes (Benson et al., 2019; Eldesouky & English, 2018; Whitmoyer et al., 2023).

Alterations in the engagement and connectivity of the fronto-limbic network have been shown to underlie difficulties in emotion regulation across several mental-health disorders, including AN and BN (X. Chen et al., 2024; Kebets et al., 2021; Koch et al., 2018; Rangaprakash et al., 2018). Drawing from evidence of reduced cognitive flexibility and poor interoceptive awareness in those with eating disorders (Frank, Shott, et al., 2019; Wollast et al., 2022; Wu et al., 2014), regulatory flexibility has been hypothesized to play a role in the development and maintenance of maladaptive eating behaviors (Dougherty et al., 2023). Thus, reduced flexibility in emotion regulation may be a key contributor to the heightened risk of eating disorder onset in mid- and late-life, particularly following stressful life events (Kilpela et al., 2023; Ng et al., 2013).

Body Sensing and Perception

Concurrent changes in perceptual abilities likely lay the groundwork for the development of disordered eating. Accurate interoception, or the brain’s ability to perceive and interpret signals from the body, and proprioception, or the brain’s ability to perceive the relative position of the body and its parts in space without visual input, have both been shown to decline with advancing age across the adult lifespan (Khalsa et al., 2009; Murphy et al., 2018; Nusser et al., 2020; Proske & Gandevia, 2012). This is accompanied by reduced engagement of regions critical for processing and sensing the body (insular and frontal opercular cortex) during interoceptive attention with advancing age (Dobrushina et al., 2024). Considerable evidence has implicated alterations in interoception and its associated neural correlates in AN, BN, and BED (Jacquemot & Park, 2020; Klabunde et al., 2017; Sysko et al., 2007). For aging adults, these interoceptive changes may increase the risk for loss-of-control eating via reduced sensitivity to internal physiological changes. For example, weaker functional connectivity in the frontoparietal and default mode networks has been associated with self-reported disinhibited eating in adults aged 60 and above (Brennan et al., 2022). These networks interact with primary interoceptive regions (e.g., anterior insula, dorsal anterior cingulate) and are implicated in self-referential processing and the integration of internal and external information (Brennan et al., 2022); therefore, reduced connectivity in these networks may diminish the ability to anticipate and regulate the body’s caloric needs. Additional evidence is necessary to determine whether reduced sensitivity in more central interoceptive circuitry (e.g., insula, salience network) with age (Kuehn et al., 2018) is reliably associated with loss-of-control eating.

Age-related changes in body perception may also increase risk for dietary restriction, as has been observed in younger populations with AN. Declines in visuospatial abilities beginning in late middle age —similar to the impairments seen in younger populations with AN (Gaudio et al., 2016)— which are accompanied by actual weight gain and changes in body composition and fat distribution, may contribute to the observed increase in ratings of “feeling fat,” and body dissatisfaction from pre- to post-menopause (Mangweth-Matzek et al., 2013). Further, alterations in visual processing abilities could increase risk of dietary restriction or binge eating (Kronovsek et al., 2021; Li et al., 2015).

Psychosocial Stressors

The impact of aging-related psychosocial stressors on eating disorder symptoms in midlife and beyond has been the focus of several existing review papers (Luca et al., 2015; Mangweth-Matzek et al., 2013; Marshall et al., 2012; Matsumoto & Rodgers, 2020; Samuels et al., 2019). Here, we consider how these stressors interact with neurobiological risk factors that emerge during this period. In adolescent populations, eating pathology has been linked to the interaction of neurobiological shifts and developmental challenges, including body composition changes with puberty, identity and self-esteem development, and navigating peer relationships (Lena et al., 2004). We describe how the interaction of the unique neurobiological changes and development-specific stressors during midlife and early older adulthood may similarly increase vulnerability to eating pathology. Increasing stressors may exacerbate age-related declines in neurocognitive processes that govern reward processing, executive functions, and emotion regulation. Compounding this, neural changes that influence how individuals respond to environmental stimuli, especially threats, may be particularly problematic during later adulthood due to the inevitability of numerous and overlapping stressors. These stressors range from new and unfamiliar experiences to the cumulative effect of stress over the lifespan.

Acute Stress

The stressful life events experienced by middle and early older adults can degrade psychological well-being and cognitive functioning (Moore et al., 2020). In midlife, such life events include caring for elderly parents, children leaving the home, divorce, and perimenopause/menopause. Older adulthood can be marked by retirement, bereavement, and significant changes in health, functional abilities, and social support. Evidence from younger populations shows that heightened stress following negative events predicts subsequent engagement in behaviors such as restriction, binge eating, and purging (Donofry et al., 2016). This appears to extend to midlife and older adulthood, as disordered eating during this period often arises following stressful life experiences (Kilpela et al., 2023; Ng et al., 2013).

In addition to psychosocial stressors, aging also introduces stressors specific to the body and appearance. Due to impaired energy expenditure regulation, adults in their 60s and 70s appear to lose more weight after transient decreases in energy intake, and gain more weight after increases in intake, compared to younger adults (Das et al., 2001; Roberts et al., 1994). Further, physical signs of aging, such as wrinkles and gray hair, are often a focus of discrimination (Sabik, 2015) and media pressure (Kvaka & Bardone-Cone, 2026), making one’s changing physical appearance and body image increasingly salient and distressing for aging adults. This is highlighted by current trends in medical aesthetics, as middle-aged and older adults account for a disproportionately large share of glucagon-like peptide-1 (GLP-1) agonist use (Bozick et al., 2025) and cosmetic procedures (American Society of Plastic Surgery, 2024) compared with other age groups. In middle-aged women, the link between societal pressures and disordered eating appears to be moderated by stress (Kvaka & Bardone-Cone, 2026). The intersectional experience of both weight- and age-related beauty standards during this stage is thus a unique source of acute stress that may be especially linked to disordered eating risk.

Just as cognitive deficits during adolescence are thought to impair adaptation to pubertal physical changes, leading to maladaptive behaviors like restriction, binge eating, and purging (Lena et al., 2004), age-related neurobiological and cognitive changes may make it particularly difficult to adjust to physical changes and other acute sources of stress during this stage (King Johnson et al., 2023). Aging adults, who exhibit weaker top-down control due to PFC degradation, may be disproportionately vulnerable to the stress-induced shift away from engaging in adaptive, top-down cognitive control, and toward a maladaptive, emotion-driven driven response (Datta & Arnsten, 2019; Hermans et al., 2014).

Chronic Stress

Older adults, compared to their younger counterparts, naturally accumulate greater psychological and physiological burden from prolonged exposure to stressful daily experiences and adverse events throughout their lives. Chronic stress further accelerates cognitive aging processes (James et al., 2023; Kulshreshtha et al., 2023), leading to increased risk for disordered eating. The hypothalamic-pituitary-adrenal (HPA) axis mobilizes energy and regulates physiological responses to stress via the release of cortisol. However, this typically adaptive response to acute, discrete stressful experiences becomes maladaptive when stress is persistent. In aging mice, glucocorticoid receptor reactivity appears reduced, inhibiting negative feedback on the HPA axis, prolonging activation, and resulting in maintenance of elevated cortisol (Yamada, 2021). This hormone can cross the blood-brain barrier and disrupt cognitive processes (Gupta & Morley, 2014).

Higher allostatic load, or the accumulation of stress over time, has been shown to be related to lower gray and white matter volume and density throughout the aging brain (Lenart-Bugla et al., 2022). Specific brain regions involved in cognition and memory, such as the hippocampus, amygdala, and PFC, appear to be particularly vulnerable to the damaging effects of persistent cortisol hypersecretion. Structural atrophy in these areas has consistently been associated with elevated cortisol levels (Lenart-Bugla et al., 2022). The reward-related imbalance favoring habitual (i.e., model-free) over goal-directed (i.e., model-based) decision-making with increasing age can be amplified under conditions of prolonged stress (Palamarchuk & Vaillancourt, 2021), facilitating an inflexible and often maladaptive pattern of behavior. The neurocognitive vulnerabilities leading to increased eating disorder risk are thus exacerbated by chronic stress related to the accumulated experience of challenging life events in later life

Hormonal and Metabolic Changes

Several prior papers have highlighted changes in the endocrine system, particularly sex-hormone changes, that may increase risk for eating disorders in midlife (Anaya et al., 2023; Baker & Runfola, 2016; Khalil et al., 2022; Kummer et al., 2019; Mikhail et al., 2021). Though the empirical literature remains limited, the existing evidence suggests an association between the menopausal and andropausal transitions and disordered eating. Here, we highlight how disordered eating could arise from interactions among age-related changes in endocrine signals and other metabolic processes with neural circuits that govern executive function, reward processing, and appetite. Estrogen and testosterone, known to be neuroprotective, decline more rapidly starting in midlife, compounding the neurobiological and cognitive vulnerabilities described above (Russell et al., 2019; Saleki et al., 2023). Hormonal shifts and increased adipose tissue further contribute to disruptions in metabolic signaling (Chowen & Garcia-Segura, 2020), exacerbating impairments in satiety signaling and reward responsiveness (J.-J. Liu et al., 2015; Y. Liu et al., 2025), which may increase the risk of maladaptive eating behavior. Longitudinal, multimethod studies incorporating measures of hormone levels, neuroimaging, cognitive functioning, and eating pathology are necessary to elucidate whether the association between midlife hormonal shifts and disordered eating may be mediated or moderated by alterations in neurobiological functioning.

Sex Hormones

Rapid reproductive hormone changes in midlife may serve as early catalysts for many of the mechanisms proposed here, accelerating neurobiological changes that increase risk for disordered eating. In females, major hormonal changes occur during the transition into menopause, beginning in the mid-40s. Findings across preclinical and clinical research indicate that age-related decreases in estrogen interact with neurobiological structure and functioning, particularly within prefrontal, cholinergic, and dopaminergic systems (Russell et al., 2019). Such an effect is thought to be mediated by a reduction in synaptic plasticity, neurogenesis, and neurochemical communication (Henderson, 2008). As a result, menopause represents a critical period during which neurocognitive functioning is acutely vulnerable to decline (Russell et al., 2019). Many women experience transient or lasting cognitive changes associated with the menopausal period, including in working memory, attention, and executive functioning (Conde et al., 2021).

Estrogen, through modulation of DA signaling in the striatum, interacts with neurobiological reward systems and influences the regulation of food seeking behavior (Cho et al., 2025; Richard et al., 2017; Yoest et al., 2014). Deficiencies during menopause could thus form a pathway toward either restrictive or loss-of-control eating in midlife and older women. As a result of the downstream effects of estrogen depletion on DA signaling, reward-seeking behavior may be reduced or inefficient, which can manifest as decreased motivation to seek food-related rewards and increased cognitive rigidity (Cho et al., 2025). Conversely, other evidence from animal and human studies suggests that cravings for highly palatable foods and loss-of-control eating increases when estrogen levels are low, thought to be due to inhibitory effects of estrogen on areas of the brain associated with food intake regulation (Ma et al., 2020; Richard et al., 2017). The relation between estrogen, alterations in cognition and reward processing, and specific eating disorder phenotypes should be confirmed in future longitudinal research.

In males, testosterone levels decline with age, with the proportion of hypogonadal men steadily increasing after age 50 (Harman et al., 2001). In younger cohorts and animal studies, low testosterone has been linked to loss-of-control eating (e.g., Culbert et al., 2020). Early evidence in middle- to older-aged men similarly supports a link between self-reported symptoms of low testosterone and both binge eating and compensatory behaviors (Kummer et al., 2019). Similar to estrogen in women, testosterone exerts neuroprotective effects and promotes neuroplasticity throughout the lifespan (Saleki et al., 2023). Across midlife and early older adulthood, men with lower endogenous testosterone consistently exhibit impairments and higher rates of decline in cognitive abilities, particularly in visuospatial, constructional, and visual memory domains, compared to those with higher testosterone levels (Moffat, 2005; Zitzmann, 2006). The link between testosterone and neurobiological functioning is supported by more recent neuroimaging data in middle-aged and older adults that found a positive association between testosterone levels and functional connectivity in regions implicated in spatial cognition and visual processing (Magalhães et al., 2025). Thus, the deleterious effects of sex-hormone depletion in males and females may further accelerate the neurocognitive changes associated with eating disorder risk during this period.

Metabolic Alterations

As individuals reach midlife, reproductive hormonal changes have been implicated in an overall decrease in muscle mass and increases in adipose tissue (Matsumoto & Rodgers, 2020; Samuels et al., 2019). From young adulthood to middle age, fat mass and overall weight increase, and fat-free mass decreases, particularly around the transition to menopause in females (Greendale et al., 2019; Sternfeld et al., 2004). Among females, this increased fat mass is centrally concentrated. Visceral and abdominal subcutaneous fat mass increase during the transition from perimenopause to menopause as estradiol concentrations decrease and relative concentrations of androgens increase (Ambikairajah et al., 2019; Greendale et al., 2019; Marlatt et al., 2020).

The contribution of these changes in body composition to dysregulated eating is partially mediated by alterations in neurobiological functioning (Chowen & Garcia-Segura, 2020). Visceral adipose tissue not only passively stores excess fat, but acts as an active endocrine organ by modulating hormones involved in numerous metabolic processes (Coelho et al., 2013). One of these adipokines, leptin, plays a crucial role in the homeostatic regulation of energy by signaling to the hypothalamus about the body’s energy stores, modulating appetite and energy expenditure (Kwon et al., 2016). Adiponectin, on the other hand, is a protective hormone that enhances insulin sensitivity, promotes metabolism, and exerts anti-inflammatory effects (Coelho et al., 2013). Age-related increases in fat mass result in a hormonal imbalance, stimulating leptin production but decreasing adiponectin secretion (Coelho et al., 2013). These metabolic alterations have peripheral effects, including decreased metabolic rate and insulin resistance (Balaskó et al., 2014; Coelho et al., 2013), which may push individuals towards extremes in weight gain or weight loss as they age, exacerbating disordered eating.

Other metabolic shifts in midlife may specifically increase risk for restrictive eating. For example, older individuals show increased postprandial insulin after standardized meals (Melanson et al., 1998), which could contribute to prolonged satiety and reduced intake in later adulthood. However, this postprandial effect is also apparent in younger, middle-aged females after menopause (Bermingham et al., 2022), suggesting that changing sex hormones play a pivotal role in changing metabolism and appetitive drives with age.

The aging brain is also particularly vulnerable to the effects of these metabolic and hormonal imbalances. Microglia are brain cells whose primary function is to sense homeostatic changes and mobilize immune responses (Kim et al., 2024). In the hypothalamus, microglia modulate the activity of appetite- and energy-regulating neurons in response to signals indicating metabolic stress (Kim et al., 2024). However, microglia deteriorate with age and become overreactive to systemic signals (Chowen & Garcia-Segura, 2020). Simultaneously, expanding adipose tissue triggers the production of pro-inflammatory cytokines such as tumor necrosis factor (TNF)-α and interleukin (IL)-1β and IL-6 (Coelho et al., 2013). Senescent microglia exhibit exaggerated responses to this cytokine spike (Chowen & Garcia-Segura, 2020), thereby promoting neuroinflammation, worsening leptin and insulin resistance, and disrupting important eating-related functions, like satiety and reward signaling in response to food (J.-J. Liu et al., 2015; Y. Liu et al., 2025). Therefore, age-related metabolic changes not only alter peripheral appetite signaling but, through neuroinflammatory pathways, may compromise the brain's capacity to regulate eating behavior.

Summary of Review

In summary, middle age and early older adulthood represent unique life stages characterized by significant neurobiological shifts that compromise the ability to respond effectively to rewards, threats, and internal signals, coupled with considerable age-related stressors and alterations in endocrine and metabolic functions which can interact with neurobiological changes to further heighten risk (Table 1, Figure 1). In response to this neurobiological dysregulation, many individuals may develop or experience a re-emergence of unhealthy eating patterns. This process largely parallels the convergence of neurobiological, environmental, and hormonal factors that confer increased eating disorder risk in adolescence. Efforts to control weight may be especially prominent during this time, as metabolic slowing and/or medical conditions often result in weight gain and body image dissatisfaction. Stress-induced appetite suppression may further promote irregular and insufficient food intake. Due to neurocognitive aging processes leading to increased rigidity and rule-based decision-making in later life, once disordered eating and/or compensatory behaviors are established, they may become more rapidly fixed and resistant to change.

Table 1.

Synthesis of reviewed age-related neurodevelopmental changes and hypothesized links to increased eating disorder risk.

Domain/Subdomain	Nature of Change	Hypothesized Link to Risk for Disordered Eating
Reward System Functioning	↓ dopamine receptors, transporters in midbrain (VTA) ↓ reward signaling in ventral striatum
• Reward Anticipation	Blunted striatal activation during anticipation	Abnormal motivation/processing of food-related reward ↓ pursuit of protective social reward
• Receipt of Reward	↓ activation of fronto-parietal network upon reward delivery ↓ neural differentiation between gains and losses	Greater risk of binge eating to compensate for ↓ reward response to food ↓ motivation to pursue food rewards
• Reward Prediction Errors	↑ reliance on working memory vs. ventral striatum during value-based decision-making ↑ sensitivity to negative outcomes	↓ ability to consider long-term outcomes Avoidance of weight gain “punishment” outweighs drive for nourishment
• Reward Learning Inflexibility	Disrupted equilibrium between goal-directed and habitual behaviors ↓ goal-directed decision-making	Rigid, repetitive, habit-like eating behaviors
Executive Functioning	White matter damage → disrupted communication between brain regions ↓ prefrontal cortex volume
• Cognitive & Behavioral Control	↓ ability to ignore goal-incongruent stimuli and suppress irrelevant information ↓ ability to suppress automatic behavioral impulses	Harder to shift attention away from internal/external cues about food, shape, weight, ↑ preoccupation with eating ↑ loss of control eating
• Working Memory	↓ capacity to store and manipulate goal-relevant information	Difficult to adaptively regulate emotions in response to stressors
• Cognitive Flexibility	↓ ability to maintain multiple tasks at once	↑ compulsive eating behavior, risk of becoming “stuck” in maladaptive eating behaviors despite ↓ benefits and/or ↑ harm
Emotion Regulation	↓ cognitive control → inflexibility in ER strategy implementation → ↓ ability to tailor or adjust ER strategies in response to stressful events	Use of ED behaviors to regulate emotions in the short term
Body Sensing and Perception	↓ communication between brain regions implicated in interpretation of bodily sensations Inaccurate perception and interpretation of hunger/fullness cues	↑ loss of control eating Disruptions in body image

Note. VTA = ventral tegmental area; ER = emotion regulation; ED = eating disorder.

Figure 1. Integrated neurobiological model of eating disorder risk in middle aged and older adults.

Note. This figure illustrates how age-related changes in neurobiological and neurocognitive domains may interact with psychosocial stressors, shifts in endocrine signals, and changes in body composition to increase risk for disordered eating in midlife and older adulthood. See Table 1 for mechanism-level detail and corresponding hypothesized links to risk for disordered eating. * = Reviewed in Mangweth-Matzek et al. (2013), Samuels et al. (2019), Matsumoto & Rogers (2020). † = Reviewed in Baker & Runfola, (2016). Portions of this figure were created with BioRender.com.

Future Research Directions

The above literature highlights midlife and early older adulthood as life stages in which widespread alterations occur across a range of biological and psychosocial processes, creating a sensitive developmental period of heightened risk for eating pathology. However, the literature investigating eating pathology in middle and early older adulthood is extremely limited in general (Mangweth-Matzek et al., 2014), and almost no research has investigated the impact of the neurobiological and neurocognitive aging changes that may increase the risk of eating pathology. It has been common practice to exclude adults over the age of 50 from such research to reduce sample heterogeneity related to the cognitive effects of aging (Sterling et al., 2022). Little to no research has included adults in their 80s and 90s, and thus, it remains unclear how these processes may extend and evolve into later older adulthood. This practice has resulted in a dearth of research regarding the neurobiological factors that might enhance the risk of disordered eating during these life stages, severely limiting the ability to develop age-adapted neurodevelopmental theories and treatment targets. Most of the research reviewed has examined influences that may indirectly affect eating disorder symptoms (i.e., through effects on appetite or metabolism), rather than directly investigating the role of different neurobiological risk factors in promoting and maintaining eating disorder behaviors. Further, existing research on this topic has been largely cross-sectional, precluding an ability to examine causal relations between neurobiological factors and eating pathology and limiting knowledge about the specific timing of different neurobiological risks. Another risk of these designs is potentially conflating aging with cohort effects (Yang et al., 2024), which may shape how aging-related neurobiological changes translate into eating disordered behaviors. Thus, a key component of addressing this gap is to encourage further age inclusion across the lifespan in research investigating the biobehavioral bases of eating disorders. In addition, more targeted investigations into neural mechanisms underlying eating pathology in middle and older adulthood, including those using rigorous longitudinal designs, are needed. Large-scale, longitudinal studies (e.g., Midlife in the US, Health and Retirement Study) tracking neurobiological and cognitive variables across aging are encouraged to incorporate more targeted measures of disordered eating symptoms. As a first step, this review presents a hypothesized model to serve as a foundation for generating testable hypotheses to be confirmed via future empirical research.

As more research is conducted on the pathways to disordered eating in middle and early older adulthood, several unanswered questions warrant investigation. First, although many of the neurobiological influences reviewed above impact all or most individuals as they age, the majority of individuals do not develop eating disorders (Mangweth-Matzek et al., 2023), and eating disorder diagnoses decrease on average as people age (Brown et al., 2020). Thus, it will be vital to better understand the individual differences in brain structure and function that may increase risk of eating pathology for some in later adulthood. Additionally, the existing literature largely does not differentiate between new onset, recurrence, or exacerbation of eating pathology among affected individuals in middle and older adulthood, reflecting a broader issue within older adulthood mental health (Carpenter et al., 2022). However, literature in younger samples suggests that the degree to which eating pathology is continuous versus intermittent may impact the severity of eating pathology (Venables et al., 2026). This research highlights the importance of differentiating how brain structure and function may vary according to distinct trajectories of eating pathology across the lifespan, including first versus repeated occurrence of eating disorder symptoms in midlife and older adulthood.

Further, older adults are not a homogeneous group, and several important variables may moderate the associations between aging-related neurodevelopmental processes and eating pathology. Although it is common practice to group middle and older adulthood together, these are very distinct life stages (Dohm-Hansen et al., 2024), and the neurodevelopmental neurobiological influences impacting eating and weight in each are likely to differ. Additionally, eating pathology in midlife and older adulthood demonstrates comorbidities with a number of mental (e.g., depression, anxiety) and physical (e.g., gastrointestinal, reproductive, and nutritional concerns) health disorders that are common during this life stage (Carpenter et al., 2022; Hambleton et al., 2022) and also impact neurobiological function (Jang et al., 2026; Navakkode & Kennedy, 2024). However, the degree to which these concerns interface with one another and underlying neurobiological vulnerabilities remains unknown and warrants further investigation.

The vast majority of the research on aging-related eating pathology has focused on women (Mangweth-Matzek et al., 2016). However, there is evidence that a significant subset of men experience elevated eating pathology in middle and older adulthood (Mangweth-Matzek et al., 2023; Mangweth-Matzek et al., 2016). Given the pronounced differences in the hormonal shifts that may occur among men and women during these periods, and the differential impact of these hormonal changes on brain function (Chowen & Garcia-Segura, 2020), it will be important to determine if there are differing sex-linked biobehavioral pathways to disordered eating during this period (Kummer & Mangweth-Matzek, 2023). Where the literature has examined differences between men and women in midlife and beyond, it has typically either focused on gendered social roles and pressures (e.g., changes in domestic responsibilities, gendered ageism, retirement; Midlarsky et al., 2018) or sex-related biological factors (e.g., menopause; Baker & Runfola, 2016) without clearly distinguishing sex and gender related influences on eating disorder risk. Further, there is currently no information on the effects of aging-related neurobiological risk factors for eating disorders in transgender and gender diverse individuals, a group that exhibits elevated eating disorder risk in the general population (Rasmussen et al., 2023), but may present with distinct neurobiological phenotypes (Mueller et al., 2021). Thus, additional research is needed to determine the role of sex, gender, and their interactions in middle and older adulthood.

It has also been noted that there is a considerable lack of racial and ethnic diversity in the research related to the neurocognitive changes that occur with aging (Brett et al., 2021), despite evidence that systemic factors disproportionately affecting marginalized populations can accelerate neural processes related to cognitive decline (Turney et al., 2023). Indeed, compared to non-Hispanic White adults, Black/African American and Hispanic adults experience faster declines in cognitive abilities—a disparity that appears to be widening with more recent cohorts (Yang et al., 2024). It has also been noted that the effects of the stress caused by stigma and discrimination can yield adverse impacts on neurobiological function (Berger & Sarnyai, 2015; Grasser & Jovanovic, 2022). The eating disorders field suffers from a lack of diversity, with most research overwhelmingly conducted on cisgender white females (Burnette et al., 2022; Egbert et al., 2022), limiting the ability to examine how specific aspects of identity interact with neurobiological processes to influence eating disorder risk as individuals age. The impact of social factors unique to marginalized communities on trajectories of neurobiological aging should not be ignored. Future research should prioritize the recruitment of racially and ethnically diverse individuals who are currently underrepresented in the literature and examine how these outcomes may vary across marginalized identities.

The extant literature also includes conflicting accounts of the biological effects of aging on eating and weight patterns, with some studies highlighting ways in which appetite and weight might be suppressed (Melanson et al., 1998) and others demonstrating how appetite and weight gain may be stimulated (Duval et al., 2014; Roberts et al., 1994; Rolls et al., 1995) by the neurobiological and hormonal shifts with age. To date, there is so little research on the neural mechanisms supporting eating pathology in older adults that it has not been possible to define the precise factors that make some individuals vulnerable to binge eating, others to restriction, and others to both. Further, new onset, continuing, recurring, and worsening eating disorders have all been noted during middle adulthood (Samuels et al., 2019). However, it is currently unknown if all such cases of eating pathology result from the same or different collections of neurobiological risk factors.

Thus, based on these identified gaps in the literature, future research is needed to: 1) directly investigate the influence of neurobiological and neurocognitive aging changes, as well as their interactions with environmental and hormonal influences during middle and older adulthood, on eating disorder behavior, including through longitudinal designs; 2) distinguish pathways from neurodevelopmental processes to eating pathology that differ as a function of aspects of identity (i.e., sex, gender, race, ethnicity), comorbidity, or specific developmental period; 3) establish the neurodevelopmental risk factors that represent shared and distinct mechanisms influencing the emergence or worsening of different eating disorder diagnoses and presentations; and 4) identify potential neural targets for prevention and treatment of eating disorders as vulnerable individuals age.

Treatment Considerations

Ultimately, a better understanding of the unique biobehavioral factors increasing risk in midlife and early older adulthood is needed to inform prevention and treatment interventions during these life stages. Like neuroimaging studies, clinical trials for various physical and psychological concerns have routinely excluded older adults despite heightened health risks in this group (Helfand et al., 2020; Nanna et al., 2020). There is similarly little research to inform how treatment should be adapted to serve individuals with eating disorders that occur beyond young adulthood (Mulchandani et al., 2021). This can result in many individuals at these life stages feeling that their experiences are not effectively addressed by existing treatments. Given the unique developmental shifts and the complexity of the neurobiological, hormonal, metabolic, and psychological influences on eating pathology in midlife and early older adulthood, a novel, multidisciplinary treatment approach in this population is likely necessary. In addition to the providers commonly involved in eating disorder care (e.g., mental-health provider, primary care physician, psychiatrist, and/or dietician; Halmi, 2005), involvement of other providers, such as neurology, endocrinology, and gynecology, may be warranted to manage underlying physical factors that may heighten the risk of eating pathology in this group.

Further, enhanced psychoeducation is needed for aging adults and their providers on the distinct neurobiological alterations associated with aging that increase the risk of eating pathology. This can serve two important functions. First, such information can help increase awareness and screening for eating disorders in this population among healthcare providers. Despite the heightened risk factors for disordered eating during this time period, individuals in midlife and older are infrequently screened for eating pathology (Samuels et al., 2019), which may lead to reduced treatment access among this population (Ackard et al., 2013). The results of this review provide further evidence dispelling the “SWAG” (skinny, white, affluent, girl) stereotype of eating disorders. Recent research has demonstrated that eating disorders influence a wider range of body sizes (Duncan et al., 2017), racial and ethnic identities (Cheng et al., 2019; Simone et al., 2022), gender identities (Gorrell & Murray, 2019; Keski-Rahkonen, 2023), and socioeconomic statuses (Mitchison et al., 2014) than previously assumed. Similarly, the above literature suggests that biological sensitivity to eating pathology is not only possible, but may be heightened during midlife and early older adulthood. Thus, further education is needed to dispel the persistent myths about the likelihood of experiencing eating disorders during different life stages. Second, such information may assist in destigmatizing the experience of eating pathology among older individuals. Prior research has demonstrated that individuals with eating disorders often encounter stigma from others, including providers (Brelet et al., 2021), as well as internalized self-stigma (O’Connor et al., 2021). This may be especially true for adults over 40, for whom symptoms may be dismissed or trivialized (Kohestani et al., 2025; McCallum & Alaggia, 2021). Further, older individuals may feel ashamed of eating and weight changes that differ from their baseline at younger ages, or may feel that they should “know better” than to engage in eating disorder symptoms (Schulte et al., 2026). There is some suggestion that neurobiological explanations for mental health concerns can be destigmatizing (Kvaale, Gottdiener, et al., 2013; Kvaale, Haslam, et al., 2013), highlighting that the affected individual is not to blame for their psychiatric symptoms. However, it is also important to recognize that biological explanations of illness can induce more hopelessness about the potential for symptom improvement (Loughman & Haslam, 2018); therefore, it will also be important to stress the continued neuroplasticity characterizing the brain, even as an individual ages (Park & Bischof, 2013).

Indeed, despite the numerous vulnerabilities outlined above, later life is not solely a time of decline and loss. In many ways, older adults are a markedly resilient group, many maintaining or even enhancing their quality of life in the face of ostensibly negative circumstances. Neurobiologically, this is evidenced by “scaffolding,” or the diffusion of neural activity and/or overactivation of brain regions during tasks, thought to reflect a compensatory process in response to age-related decline (Park & Reuter-Lorenz, 2009; Reuter-Lorenz & Cappell, 2008). For example, several studies have demonstrated increased neural activation in task-related brain regions in older compared to younger adults; this activation has been positively correlated with task performance (Cappell et al., 2010; Vallesi et al., 2011). Behavioral and cognitive strategies to counteract age-related declines, including cultivating positive experiences, physical exercise, sleep hygiene, and engaging in cognitively-stimulating activities, can enrich existing neural resources, buffer against the depleting effects of stress and other detriments to brain health, and aid in this compensatory scaffolding process (MacLeod et al., 2016; Park & Bischof, 2013; Reuter-Lorenz & Park, 2024). While this evidence lends support to the potential efficacy of interventions for eating disorders in later life, additional research is critically necessary. Future applications of mechanistically-targeted interventions in this population will require empirical validation of the biobehavioral pathways outlined in our proposed framework.

Conclusion

The dearth of literature regarding eating disorders in older adults, including the neurobiological and neurocognitive changes that promote such behavior as individuals age, has impeded progress in advancing research, improving clinical practice, and informing policies that affect this critical yet neglected area of mental health. Growing evidence for the brain’s capacity to adapt amid these aging processes supports the potential for successful prevention and intervention efforts if disordered eating can be adequately detected in this age group. Thus, future research on the aging-related neurobiological influences contributing to eating pathology opens doors not only to enhanced understanding of risk factors in this under-researched group, but also to more precisely designed interventions for individuals who are especially vulnerable to the impairing and life-threatening effects of eating disorder symptoms.

Article Highlights.

• Evidence suggests a second window of risk for eating disorders (EDs) in later life

• Interaction of brain changes and psychosocial stressors mirrors ED risk in adolescence

• Neurocognitive alterations disrupt adaptive responses to body changes and life events

• Biobehavioral research on EDs in aging adults will improve prevention and intervention