Explaining Variation in Individual Aging, Its Sources, and Consequences: A Comprehensive Conceptual Model of Human Aging

Research on aging is a multidisciplinary endeavor, with biologists, medics, nutrition and sports scientists, psychologists, and social scientists investigating the phenomenon from their specific perspectives. Due to its multifaceted and multidimensional nature, none of the different disciplinary approaches on their own is capable of providing a comprehensive account of individual human aging: Understanding human aging requires a thorough understanding of its underlying biological processes, resulting in objective changes in appearance and functioning. At the same time, it also has to take into account the subjective perception and interpretation of these changes, which characterize the individual aging experience, as well as the societal, cultural, and physical contexts that shape aging processes and their consequences.

Human aging is an immensely heterogeneous phenomenon. There is huge variability in longevity, with some people dying at the age of 60 or before, whereas others live up to an age of 120 years. Relatedly, symptoms of aging (e.g., appearance, physical fitness, cognitive functioning) vary greatly among same-aged people. The same individual differences are seen in the consequences of aging for well-being and adaptation. To fully account for this complexity of individual human aging, a truly interdisciplinary perspective is imperative, which not only captures the large variety of contributing and modulating factors but also investigates their interplay and interactions in a joint model-based approach.

In this paper, we will provide an outline of such a model that bridges the gaps between different disciplines of aging research. The model introduces a shared terminology, it specifies points of convergence and interaction, and it organizes and structures the research field by defining research foci for interdisciplinary research on aging, such as buffering mechanisms and their relation to aging, the relation between aging and the subjective experience of aging, and the consequences of aging. The model also allows for the derivation of specific hypotheses that can be investigated and tested empirically, such that changes in buffering mechanisms are a major source of aging or that consequences of aging are mediated by the subjective experience of aging. Since the model we propose is novel, it is premature to finally evaluate these hypotheses. Thus, rather than conducting a strict, meta-analytic review of the available empirical research that has tested specific hypotheses derived from the model, we provide examples of existing research in order to illustrate the basic claims and tenets of the model.

The basic structure of the model is captured by a conceptual distinction between sources of aging, objective aging, subjective experiences of aging, and consequences of aging (Fig. 1). We assume a directional causal chain which starts from sources of aging that determine actual aging, which in turn gives rise to subjective experiences of aging and ultimately affects a multitude of different consequences, including the motivation, well-being, and life expectancy of the organism. We acknowledge that many more of the possible paths in the model may indeed exist. For the sake of our argument though, we deliberately set these aside for the time being and will address such extensions once the basic model is established.

In addition to this structural outline, there are two further core elements of the model: (1) biological repair and personal adaptation mechanisms are placed at the center of the model, both of which counteract aging processes. These buffering mechanisms themselves can be subject to age-related changes, be this as age-related increases or decreases in the efficiency to counteract aging, and thus resulting in either increased resilience against or accelerated aging. (2) To account for the social and cultural construction of aging processes and their embeddedness in historical and environmental contexts, the variables of the model are framed by external factors that feed into the different components of the model.

Before we introduce the model in more detail, we start with a definition of the concept of aging because research on aging has been fraught with misunderstandings resulting from different (implicit) usages of basic concepts, which has been a major obstacle, especially for interdisciplinary collaborations due to different conceptions of the very phenomenon they investigate. Following the definition, we list several corollaries to highlight the implications of the definition (what is and what is not aging proper). Based on these conceptual clarifications, we then introduce a model of individual aging that distinguishes between aging proper, its sources, experience, and consequences. We also specify modulating factors that moderate associations between the components of the model, which allows us to explain variability in human aging.

Aging is an age-related deterioration in an observable attribute of an organism that typically occurs later in life.

It has become rather common to apply the concept of aging also to cells (e.g., cell age, aging cells, senescent cells) or organs (e.g., brain age, skin age). Such expressions are widespread in the literature – even journals are labeled accordingly (e.g., “Aging Cell”). According to our definition of aging, such expressions and labels should be understood as an abbreviation for a more complex causal statement that relates the functioning of cells or organs to the aging of the individual organism. Expressions like “cell aging” or “organ aging,” as they are commonly used in the literature, thus typically do not refer to the lifetime of cells or organs but are statements about causes and sources of the aging of the organism to which these cells or organs belong.

When changes in the functioning of cells are labeled as “cell aging,” this indicates that certain changes in cell functioning were identified as a source of aging of the individual organism. Such a usage is completely in line with our definition of aging and with our model. We just want to highlight that the label “cell age” is actually a stand-in for a statement that certain changes in cell functioning can cause individual aging.

If, on the other hand, “cell aging” is actually meant to indicate that cells themselves are aging rather than indicating a possible cause or source of aging of the organism, then this usage has to be clearly distinguished from what is typically labeled as “cell aging.” Changes in cell functioning that cause individual aging at the organismic level do not refer to the lifetime or life expectancy of cells, nor do they refer to changes typically occurring during the later life of a cell (if anything, there is a limit to the number of divisions in a given cell strain, the so-called “Hayflick limit” [typically ∼50 cycles] [1]). It is not that single cells typically change their functioning during their lifetime or that “senescent cells” have a shorter residual life expectancy or survival probability compared to non-senescent cells. Instead, with increasing age of the organism, cells that show certain deficiencies in functioning (which we then might call “senescent cells”) accumulate, which increases the risk that the organism to which these cells belong ages.

For organs, which may have a similar life expectancy as the organism to which they belong, the situation is similar. In many cases, the organism’s life expectancy does not directly depend on the organ (e.g., we can live a healthy life without showing any signs of aging even after our tonsils have been removed or even when one of our kidneys has died). It thus makes sense to speak of “organ age” – like of “cell age” – as an abbreviation, when we refer to changes in organ functioning as a source of aging of the individual organism. “Organ age” is then just a stand-in for a more complex causal statement and is fully compatible with our definition and model. Of course, one can speak of “organ age” to indicate the status of an organ’s functional capacity or that the organ has completely or even irreversibly lost its functional capacity. However, this is then a different concept of aging (here, at the level of organs) that should not be confused with aging of the individual.

Thus, although it is possible to describe the aging of cells or organs, this is not what is typically meant when researchers identify a phenomenon as cell aging or organ aging (e.g., brain age). Instead, this research typically identifies changes in cell or organ functioning or in the regenerative capacity of cells [2‒4] that are unrelated to the “life expectancy” of the cells or organs themselves but instead accelerate aging of the entire organism and thus should be labeled as sources of aging according to our model.

With this view, we move beyond other conceptualizations (e.g., “hallmarks of aging” [5]) and argue that age-related changes in the structure or functioning of the components of an organism (i.e., cells, organs) reflect either sources or modulating factors of aging. Our concept of aging comprises features that Ferrucci et al. [6] listed under the labels of functional and phenotypic aging, but our definition is different from theirs in that it comprises other features (e.g., appearance) that have no direct functional relevance, and we clearly distinguish between aging and its sources and consequences (such as Fried’s frailty phenotype [7]).

Aging always refers to changes in a specific observable attribute of the organism (e.g., decline in sensory, physical, or mental functioning; age-related changes in appearance; so-called “age cues” or “markers of aging” [8]). It is presumably a sign and a characteristic feature of normal and healthy aging for older adults who move through their 60s and 70s that aging does not occur in parallel for different attributes (e.g., wrinkles in the skin are typically unrelated to memory deficits or physical weakness). It is only in very old age when people are in their 80s and 90s and at the very end of their lives that risks increase for a scenario in which a large set of attributes are affected simultaneously, a phenomenon labeled as “de-differentiation” [9‒11].

The risk to develop certain diseases often increases with age. However, illness does not fulfill the criteria of our definition of aging (prototypical change that occurs during later life). Thus, even age-related illnesses or their consequences should not be considered a part of aging proper [12]. Importantly, however, certain illnesses can cause or accelerate aging, which is why certain illnesses are a potential source of aging (e.g., Werner syndrome [13]).

Similarly to illness, terminal decline – although occurring at the very end of life – is not part of aging proper because such processes are tied to approaching death rather than to age (e.g., changes relating to terminal decline could occur even at younger ages [14]). In addition, processes of terminal decline typically involve rates of change that are massively exacerbated as compared with those of age-related changes.

Even if certain changes are highly characteristic for the species and typically occur at a specific age, these changes are not labeled as aging if they occur early in the life cycle of the organism. These changes are typically characterized by gains in functioning and abilities, and we refer to these changes with other labels like maturation or growth. This is in opposition to the view that “humans age from the moment of conception” [8]. According to our model, processes that precede and cause aging in later life should be considered as part of the sources of aging. Even typical negative changes occurring earlier in life do not reflect aging, due to their early placement in the life cycle of an organism. For example, teen depression [15], midlife crisis [16], and menopause [17] do not fall under the definition of aging, whereas loss of sexual functioning in older men does [18].

Although growing older is also associated with specific benefits, we refrain from subsuming these changes under the label of aging because this does not conform to the general usage of this term in everyday language. That is, although characteristic for old age, these changes do not fall under the definition of aging and are referred to as “experience” or “learning.” The well-documented positive changes in old age are covered in our model as age-related changes in modulating mechanisms, or they may reflect positive consequences resulting from coping with old age and aging. In the following parts of this paper, we zoom into different parts of the model in order to provide more details and further elucidate the relations between the core components of our model.

Aging originates from and is intimately linked to changes that are located at the levels of molecules, cells, organelles, or organs (Fig. 2). These changes comprise damage to molecules like proteins or DNA [19, 20], epigenetic alterations [21], as well as changes in metabolism [22] or organ functioning [23] that ultimately lead to observable changes in functioning and appearance of the organism. These sources of aging correspond to what Ferrucci et al. [6] labeled as “root mechanisms of biological aging” in their model but also comprise aspects of what they list under “phenotypic aging” (e.g., immunosenescence, arterial stiffness). These sources of aging themselves are caused by different determinants. Proximal causes relate to changes in the expression of genes [24], and to accumulation of damage at cellular levels resulting from wear and tear [25], or from behavioral changes [26]. Distal causes and stressors stem from the environment of the organism (e.g., radiation, pollution, viruses).

Importantly, buffering mechanisms exist at different levels (biological, personal) and shield an organism against aging by counteracting the effects of its sources. These mechanisms endow organisms with an enormous resilience against aging and may even reverse the underlying processes of aging. Thus, although aging is generally considered to be a continuous and irreversible process, it has been shown that it can be reversed upon recovery from stress due to the effects of buffering and repair mechanisms that counteract the sources of aging [27]. These mechanisms are key for explaining resilience against aging and for understanding the dynamics in individual aging trajectories.

Evolution has endowed cells with a highly efficient set of mechanisms that shield the functioning of the cell against disturbances. These mechanisms, including chaperones [28], detoxifying enzymes [29], and various DNA repair systems [30], respond to damage and deviations from optimal functioning by reestablishing balance. At the organ level, stem cells fulfill this function [2]. Typically, the operation of these mechanisms can be described as a cybernetic control system: deviations from an ideal state trigger a response that aims to reestablish the desired state [31]. This bidirectional relation between biological sources of aging and corresponding repair mechanisms is highlighted by arrows pointing in both directions between sources of aging and corresponding repair mechanisms.

Sources of aging are also influenced by the lifestyle and behavior of the organism (e.g., preparation for old age, amount of physical activity, dietary preferences, engagement in vs. withdrawal from activities [32, 33]). These behaviors accelerate or reduce the speed of aging by creating conditions that are either harmful or beneficial for the functioning of cells and organs.

The effects of behavior and lifestyle elements on aging have recently become a major research topic in the aging literature. Systematic investigations of physical activity, dietary habits, and social engagement have documented strong effects on functioning and longevity in older adults [34‒37]. The exact causal pathways mediating these effects, however, are not yet well understood. Our proposed model assumes that effects of lifestyle on objective aging are mediated via the sources of aging. Some behaviors have a direct influence on the sources of aging by changing the availability and concentration of substances that influence metabolism (e.g., dietary habits regarding the uptake of substances like alcohol, nicotine, or trace elements [38‒40]). For many other behaviors, the influence is more indirect in that psychological modulators strengthen or weaken biological repair functions. One such modulator is chronic stress that affects the immune system and accelerates aging [41]. In turn, it also stands to reason that some of the effects of lifestyle may themselves be buffered or undone, at least within certain limits, by biological buffering mechanisms (e.g., red blood cells in endurance sport [42]). As of now, it is mostly unclear in which way the amount of physical activity, engaging in or disengaging from other activities (mental tasks, different forms of social contact, and activities), influences sources of aging or biological repair mechanisms at the organic, cellular, or molecular levels.

Aging is embedded into environmental and societal/cultural contexts that shape aging processes [43]. Our model highlights that sources of aging are also influenced by environmental factors (e.g., germs, pollution, radiation, sunlight intensity, as well as the availability and quality of nutrition and medication). Cultural and societal arrangements relating to working conditions or medical care should thus be considered as more distal factors that either put stress on an organism and its parts, support the functioning of its components and repair mechanisms, or even compensate for biological repair mechanisms. It is also well established that the nature of environmental and societal conditions and the way these operate on individual functioning and aging are subject to historical change (for an overview, see [44]).

Most of the changes contributing to the sources of aging are cumulative in nature, and they represent the sediment of a lifetime history of being exposed to different stressors and strains. Under normal conditions, negative effects of these factors are neutralized by biological repair and personal adaptation mechanisms, resulting in a remarkable resilience of the organism and their components against signs of aging. In older age, however, these buffering mechanisms themselves may reach their limits, and their ability to compensate or buffer effects of stress and disturbances becomes compromised [45, 46]. Either the sheer load of accumulated age-related losses overwhelms people’s buffering mechanisms or the robustness of how such buffering processes operate becomes itself fragile, or both influences operate simultaneously [47]. Functional deficits then tend to become chronic, and levels of functioning of cells and organs fall below the threshold that is needed to guarantee an optimal functioning of the organism, resulting in visible effects of aging. Most of the underlying sources of this age-related decline in efficient repair functions have not yet been identified. According to our model, potential sources are located at all levels: biological, personal, societal, and environmental.

As already indicated in the above definition, aging consists of a characteristic late-life decline in observable attributes of an organism that are related to its appearance or to its levels of sensory, physical, or mental functioning [48, 49]. As postulated in the model, many of these changes can be traced to the sources of aging which compromise the functioning of the constituent parts of an organism (cell and organ functioning [3, 4, 50]). These objective indicators of aging, however, do not yet capture the personal experiences of aging that form an essential part of human aging (Fig. 3).

Adult human beings typically develop a differentiated set of beliefs about their own aging, comprising how old they feel (subjective age [51]), which changes they have experienced across their life, and whether and how they relate these changes to their chronological age (awareness of age-related change [52]; attributions of experienced changes to age [53]), which changes they expect with advancing age (future self-views [54]), and how they evaluate the process of aging in general and with respect to their own aging (attitudes toward one’s own aging [55]). Furthermore, each of these different types of subjective beliefs regarding one’s own aging has been shown to be highly contextualized and domain-specific [56], which already points to the strong influence of cultural and societal factors in shaping these beliefs.

Subjective views of one’s own aging are not simple copies of the objective aging status of an individual. Instead, they reflect the result of complex cognitive processes involving comparisons of oneself with learned and internalized conceptions of how old people are (descriptive age stereotypes [57]) and how they should be (prescriptive age stereotypes [58, 59]), as well as processes of age group identification or distancing from one’s age group [60].

Our model assumes that these latter processes – comparisons with internalized norms and stereotypes, age group (de-)identification – reflect the operation of personal, mostly psychological modulators. Depending on which standards of reference are chosen, the same objective aging status can result in positive or negative evaluations of one’s own aging. Variability in these processes can (partly) be explained by differences in age stereotypes that individuals have acquired earlier in life and that later become internalized and shape an individual’ views of their own aging [54, 61, 62]. These stereotypes also influence a person’s behavior (preparation for old age [63, 64], physical activity [65]) and health [66], which feeds back into the sources of aging (see above). Another source of variance for personalized views on aging are cultural factors (age limits, age norms, culturally shared age stereotypes) that set boundary conditions for individual beliefs.

Objectively observable and subjectively experienced aging not only affects the performance of individuals in various activities and tasks (e.g., at work, in sports, during daily activities), aging also affects the lives of individuals in general (see Fig. 4). These global consequences of aging comprise psychological aspects like affective well-being ([reduced] life satisfaction [67], depression [68]), goals and aspirations (what is considered to be feasible and age-appropriate [58]), and social functioning (withdrawal from social relations and activities, lack of initiative to establish new relations [69]), but also more objective indicators like physical activity levels, health, and life expectancy [65, 66, 70‒72].

According to our model, most of these consequences of aging are mediated by the subjective experience of aging. That is, they depend more on individuals’ constructions and interpretations of their aging experiences in light of their age-related beliefs and expectations regarding what is typical and age-appropriate than on their actual levels of functioning and aging-related losses therein. It is thus no surprise that psychological factors also play a major role in modulating the translation of aging experiences into global consequences. Previous research has highlighted the importance of accommodative processes (flexible goal adjustment, acceptance, and reinterpretation of losses) in coping with age-related change [73, 74]. With advancing age, there is also a shift in how people cope with age-related change. Active-assimilative coping becomes less frequent and also less efficient because people’s resources are increasingly compromised. As a consequence, assimilative coping gets more and more replaced by accommodative forms of coping [75]. Although most previous research has investigated buffering effects of accommodative coping with specific problems and critical life events (e.g., illness, chronic pain, disability, critical life events [76‒81]), our prediction is that similar shifts in the quality of coping and its efficiency are also obtained when it comes to how people respond to their aging in general.

Our analysis of age-related changes in coping with age shows that buffering mechanisms can also be characterized by age-related gains. The ability to distinguish between situations that can be personally controlled and those that have to be accommodated, and the capacity to reappraise the latter situations in a way that fosters acceptance and meaning is a developmental achievement that reflects experience, wisdom, and realism [46, 82].

Aging is a universal phenomenon that is characterized by typical trajectories later in life for different attributes. Our model identifies direct sources of aging, mostly residing at the biological level (e.g., cumulative damage at cellular levels, age-related decline in the efficiency of repair functions) as well as more distal factors at the behavioral (lifestyle) and environmental levels that also affect aging. We assume that these influences are mostly mediated via biological pathways, but the exact transmission of these influences is a most important target for future research.

Despite its ubiquity and inevitability, aging is by no means homogeneous. Human aging is characterized by enormous individual differences in the onset and course of decrements, and also with regard to its subjective experience and consequences. Our model identifies possible explanations for these individual differences, relating to qualitative and quantitative differences in buffering mechanisms that mediate and/or moderate the relations between the core factors and phenomena of aging. Resilience against aging and its negative consequences is influenced by lifestyle factors but also by cognitive interpretations that are shaped by individually held beliefs about aging (internalized age norms and stereotypes). Differences in coping with age are assumed to change qualitatively and quantitatively as people move across the second half of life. An adaptive switch from active-assimilative toward more accommodative forms of coping is assumed to occur during the transition from the third to the fourth age, due to age-related changes in time perspective and action resources [73‒75, 82].

The model sets an agenda for interdisciplinary research that addresses the relations between behavioral lifestyle factors, cognitive appraisal processes, and their interactions with biological sources of aging and corresponding repair mechanisms. Aspiring for a full picture of human aging, we also take into account the environmental and societal conditions in which these processes are embedded [43]. Any endeavor that successfully bridges the gaps between the mostly isolated research areas of different disciplines will not only improve our theoretical and conceptual understanding of human aging processes but will also provide novel ideas for interventions aiming at preventing, buffering, and postponing aging or alleviating its negative consequences.

The authors have no conflicts of interest to declare.

This work was supported by grants of the Carl-Zeiss-Stiftung (IMPULS project) and the VolkswagenStiftung (Aging-as-Future project, Az. 93 272).

The authors have developed the ideas and model described in this article in joint discussions. KR has written a first draft of the manuscript that has then been revised (amended, commented, expanded) by C.E. and D.G. in two extensive rounds of revision.