Introduction
In its early and, perhaps, the broadest definition, resilience was equated with the capacity to cope with and overcome adversity. The origin of this now vast field of research and practice is typically attributed, at least partially (Masten, Reference Masten2007; Masten & Tellegen, Reference Masten and Tellegen2012), to the work of Norman Garmezy who was concerned with issues related to primary prevention of psychiatric disorders and proposed and propelled the investigation of children who are at risk for the manifestation of psychopathology later in life because they were born to parents (mothers, fathers, or both) with psychiatric disorders (Garmezy, Reference Garmezy, Cancro and Dean1985, Reference Garmezy1992). In particular, he focused on families of individuals with schizophrenia, stressing that both the very origin of these children and their family environment form a system of risk factors for the development of psychopathology. In his earlier writings, Garmezy (Reference Garmezy1971) juxtaposed vulnerable (whose unfavorable prognosis follows the expectations generated by their familial risk factors) and invulnerable (whose unfavorable prognosis upsets “our prediction tables” and bears “the visible indices that are hallmarks of competence,” p. 114) children underlying the importance of developing not only prevention models aimed at curtailing the incidence of susceptibility to psychopathology in vulnerable children, but also enhancement models leading invulnerable children to survival and adaptation.
Although this early work laid the foundation for the science of resilience and offered the first broad notion of the concept, the evolution and crystallization of both the definition and the field of resilience have emerged through the contributions of multiple researchers and multiple studies (Cicchetti & Garmezy, Reference Cicchetti and Garmezy1993; Masten & Tellegen, Reference Masten and Tellegen2012; Masten et al., Reference Masten, Tyrell and Cicchetti2023). As eloquently put, “Resilience research co-emerged with developmental psychopathology from the same nexus of influences (Cicchetti, Reference Cicchetti, Cicchetti and Cohen2006; Masten, Reference Masten2007)” (Masten & Tellegen, Reference Masten and Tellegen2012, p. 346). Cicchetti’s work was fundamental to this nexus as he has contributed to the understanding of the phenomenon by interrogating it at different levels of analyses (Cicchetti, Reference Cicchetti2010), approaching it from different domains and contexts (Cicchetti & Rogosch, Reference Cicchetti and Rogosch2007; Curtis & Cicchetti, Reference Curtis and Cicchetti2007; Denckla et al., Reference Denckla, Cicchetti, Kubzansky, Seedat, Teicher, Williams and Koenen2020), and developing and evaluating the much-needed interventions for children and adolescents who have lived through adverse experiences (Luthar & Cicchetti, Reference Luthar and Cicchetti2000). In this essay, I will briefly sample different instances of the utilization of the concept of resilience, attempting to complement a comprehensive representation of the field in the special issue of Development and Psychopathology inspired by the 42nd Minnesota Symposium on Child Psychology, hosted by the Institute of Child Development at the University of Minnesota and held in October of 2022 (Masten et al., Reference Masten, Tyrell and Cicchetti2023). Having established the general context of the field, I will zoom in on some of its features, which I consider particularly promising to focus on to advance the study of resilience in the context of the future of developmental psychopathology.
Definitions, definitions, definitions
Garmezy’s (Reference Garmezy, Anthony and Koupernik1974) initial broad interpretation of resilience as the capacity to maintain normative development and demonstrate adaptive outcomes in the presence of adversity has been reflected in the definition of resilience as presented by the American Psychological Association, APA: “Resilience is the process and outcome of successfully adapting to difficult or challenging life experiences, especially through mental, emotional, and behavioral flexibility and adjustment to external and internal demands” (2024). Alternatively, paraphrasing, resilience is the physical and mental (or behavioral and psychological) capacity to confront and manage adversity generating stress, adjust to change, recoup, regain composure, and rise from holdups and setbacks. So…. resilience is life itself!
Yet, although defined through broad strokes, such definitions were marked (Denckla et al., Reference Denckla, Cicchetti, Kubzansky, Seedat, Teicher, Williams and Koenen2020) as difficult to translate into research language constrained by verifiable theories, falsifiable hypotheses, and interpretable data. Correspondingly, there is an ongoing quest for the definition of resilience through developing theoretical interpretations that generate specific testable hypotheses addressable with collectible and explainable data. Given the range of existing definitions, can constant features of the concept of resilience be noted? It seems so!
First, resilience assumes the presence of exogenous risk in response to which it arises. Thus, it has been defined as having a lower vulnerability to the presentation of such risk and maintaining the overall normative developmental trajectory in the face of the need to overcome or to continue facing adversity (Rutter, Reference Rutter2006). This notion of resilience stresses its “relative” nature (Holz et al., Reference Holz, Tost and Meyer-Lindenberg2020), as it arises in response to outer factors of various types (from wars and natural catastrophes to school bullying). Yet, it is important to stress that resilience is a dynamic developmental process (Rutter, Reference Rutter2012), endogenous to the individual, and, therefore, characterized by person-specific characteristics at multiple levels of functioning (e.g., genome, brain, behavior). This facet of resilience stresses the consistency of positive outcomes for resilient individuals in the context of various adversities, alluding to “true (or overall)” resilience (Kaufman et al., Reference Kaufman, Cook, Arny, Jones and Pittinsky1994). To capture this juxtaposition, resilience has been discussed to have both state (in response to adverse exogenous impacts) and trait (an ever-present endogenous denominator) presentations and manifest as both domain-specific (e.g., in response to physical stress) and domain-general (e.g., in response to any stress) adaptation to the changed environment. However, Denckla et al. (Reference Denckla, Cicchetti, Kubzansky, Seedat, Teicher, Williams and Koenen2020, p. 7) objects to the view of resilience as a trait: “Resilience is multidimensional; it is not static or trait-like.”
Second, similarly to the need for the vision system to be exposed to light to trigger its maturation or for the ontogenetic emergence of language to be surrounded by language-producing talking heads, resilience emerges in response to stress; in essence, it is the other side of the story of stress reactivity. Therefore, resilience is often studied explicitly and implicitly in the literature on stress (McEwen, Reference McEwen1998). Yet, “While research is beginning to demonstrate the link from social adversity to negative outcomes via altered neural mechanisms, which is in line with the vulnerability perspective, the resilience perspective has only recently become a focus within neuroscience.” (Holz et al., Reference Holz, Tost and Meyer-Lindenberg2020, p. 380). Thus, the field knows much more about the neuroscience of adversity than about the neuroscience of resilience. Yet, these systems are interconnected, as adversity needs to be encountered, and stress needs to be experienced to develop resilience, which, in turn, modulates the response to stress and adversity. This interactive nature of resilience has been identified as the capability of a complex system to endure and/or rebound from disturbance; this notion of resilience has been used in ecology (Holling, Reference Holling1973), economics (Rose, Reference Rose2007), and sociology (McKeown et al., Reference McKeown, Hai Bui and Glenn2022). Correspondingly, resilience is one of the characteristics of a complex system capturing its dynamics sampled through the system’s adaptive cycles; specifically, resilience refers to the capacity of a system to successfully adapt to challenges that threaten the function, survival, or development of that system (Masten et al., Reference Masten, Lucke, Nelson and Stallworthy2021).
Third, as resilience is a systemic characteristic, it requires the amalgamation of numerous signals from peripheral and central sources, extending from short-range signaling of local circuits of the nervous to long-range signaling of humoral immunity factors of the immune system (Cathomas et al., Reference Cathomas, Murrough, Nestler, Han and Russo2019). The appreciation of and appeal to the multilevel investigation of resilience was made 30 years ago (Cicchetti & Garmezy, Reference Cicchetti and Garmezy1993). These multiple interdependent levels have been conceptualized as allostatic load, stress inoculation, developmental trajectories, epigenetic factors, and transgenerational effects, among others (Southwick & Charney, Reference Southwick and Charney2012). Yet, despite the conceptual and practical appearance of finding psychological, biological, and biopsychological protective markers characteristic of resilience, the field still lacks an understanding of what these markers are, especially when sampled at the individual levels. Although still not numerous and mostly correlational in nature, there are studies sampling resilience as a complex collection of systemic indicators at multiple levels of individual functioning (Holz et al., Reference Holz, Tost and Meyer-Lindenberg2020).
Fourth, although initially discussed as a rather rare positive outcome in the high-risk context of familial risk for schizophrenia, resilience has been reconceptualized as a common phenomenon (Masten, Reference Masten2001). A superficial analogy here is a differentiation of big C and little c creativity (Runco, Reference Runco2019), indicating both the commonality (or not) of the process and the magnitude of the product (small everyday creative moments or highly impactful creative products). As per this analogy, Resilience (big R) pertains to rare remarkable demonstrations of strength in the face of adversity (e.g., Nelson Mandela and Aleksei Navalny), whereas resilience (little r) is a quantitative trait measurable in the general population. With regard to the latter, research has identified a number of factors differently associated with resilience, such as positively correlated strong social support, both through family and the wider social network (Herrman et al., Reference Herrman, Stewart, Diaz-Granados, Berger, Jackson and Yuen2011) and negatively associated social isolation (Tost et al., Reference Tost, Champagne and Meyer-Lindenberg2015); and such as positively correlated active coping responses intended to gain actual or perceived control of a stressor by minimizing its physical, psychological, or social harm (Russo et al., Reference Russo, Murrough, Han, Charney and Nestler2012) and negatively correlated passive coping, such as avoidance and helplessness (Southwick et al., Reference Southwick, Vythilingam and Charney2004; Wood & Bhatnagar, Reference Wood and Bhatnagar2015). Relatedly, there is accumulating evidence to differentiate passive and active (or more active) resilience (Rakesh et al., Reference Rakesh, Morey, Zannas, Malik, Marx, Clausen, Kritzer and Szabo2019). There is an open task of cross-mapping these various concepts onto a single “quilt” of resilience.
This “quilt” of definitions makes studying resilience a challenge, which necessitates an introduction of a specific definition used prior to a study-based specification and operationalization of the concept. So, what is the one for this essay? Here, resilience is defined as an integral over an interval, where performance is sampled behaviorally and biologically repeatedly (continuously) as a process and cross-sectionally (categorically) as an outcome (see the APA definition above) in the face of adversity and under stress. This definition seems to reflect the four common features of the multitude of conceptualizations of resilience, namely: (1) the manifestation of resilience is substantiated by the presence of adversity and stress; (2) resilience is a dynamic systemic characteristic; (3) resilience is characterized by a complex collection of systemic indicators at multiple levels of individual functioning; and (4) resilience is a common feature of the general population (i.e., a source of individual differences).
This endorsement of the definition in this essay is three-fold. First, the current literature on the neuroscience of resilience has established numerous targets that serve both as the (neuro)physiological substrates and biomarkers of resilience. The analysis of this literature presents the “hot spots” of resilience as a source of individual differences in the general population. Second, this definition permits a description of resilience at multiple levels of human performance and through the characterization of a changing complex system as it interacts with adverse challenges. Third, the availability of portable wearable and carriable devices today permits the collection of the needed data to substantiate and interrogate the notion of resilience as dynamic readiness characterized as a continuum of (neuro)physiological states integrated over numerous contexts, situations, and tasks.
(Neuro)physiological substrates and markers of resilience
To restate, resilience has been referred to as the capacity to face adversity and manage stress within the normal range of psychological and physiological functioning (Wu et al., Reference Wu, Feder, Cohen, Kim, Calderon, Charney and Mathé2013). This definition assumes that there is a notion of what that normal baseline is and what deviations from the baseline are acceptable for readiness to perform optimally with and without different sources of stress. Importantly, there is evidence (Ledford et al., Reference Ledford, Dixon, Luning, Martin, Miles, Beckner, Bennett, Conley and Nindl2020) that self-reported appraisal of resilience (that calls for the awareness of both the amount of stress and the individual’s readiness) appears to be as accurate as the capture of resilience through various biomarkers (that do not call for any awareness and are easily recordable) in predicting outcomes (9.6% vs. 10.8%, respectively). Yet, the predictive power of their combination is almost additive (17.2%), indicating that only a partial and small amount of predictive information contributed through these different types of assessments overlap. Hence, although it is well-known that, among others, both psychological and (neuro)physiological factors contribute to resilience, it is important to focus on the latter, as many relevant indicators can be collected noninvasively and continuously.
It is accepted that the neurophysiological mechanisms of resilience are remarkably intricate (Murrough & Russo, Reference Murrough and Russo2019). Importantly, most of the data have been derived from studies of stress reactivity (see juxtaposition above), and the hypothesized apparatus has linked modifications in immune, hormonal, and microbiota-related pathways substantiated by specific neurocircuits and genetic and epigenetic events and processes. Notably, the general assumption is that salient features of resilience are likely to rely on the active engagement of key (neuro)physiological systems whose critical function is to maintain homeostasis rather than on the absence of maladaptive alterations that are expected to be generated by stress (Russo et al., Reference Russo, Murrough, Han, Charney and Nestler2012). To describe the systemic adaptive response of the brain and body to stress, Bruce McEwen coined the term “allostasis” (McEwen, Reference McEwen1998, Reference McEwen2012), which is a state of the multilevel activation of the hormonal and cytokine signaling, the sympathetic nervous system and hypothalamic-pituitary-adrenal (HPA) axis, which have the capacity to calm down and recover adequately (Osório et al., Reference Osório, Probert, Jones, Young and Robbins2017). Thus, resilience can be seen as derived from the specific (neuro)physiological allostatic reactions that are produced in response to stress (Murrough & Russo, Reference Murrough and Russo2019). These responses are “sufficient but not excessive” and indicative of “rapid and efficient psychobiological recovery” (Feder et al., Reference Feder, Fred-Torres, Southwick and Charney2019, p. 444). Importantly, for this discussion, these responses are assumed to be replicable, although modulated by the characteristics of stress.
Resilience in the brain
As the capacity to regulate emotions is a recognized aspect of resilience, it has been hypothesized that the function of the (neuro)physiological systems substantiating the cognitive restraint of emotion (Kong et al., Reference Kong, Wang, Hu and Liu2015; Takeuchi et al., Reference Takeuchi, Taki, Nouchi, Hashizume, Sassa, Sekiguchi, Kotozaki, Nakagawa, Nagase, Miyauchi and Kawashima2014; Urry et al., Reference Urry, Nitschke, Dolski, Jackson, Dalton, Mueller, Rosenkranz, Ryff, Singer and Davidson2004) and reward circuitry might be relevant for the understanding of the (neuro)physiological bases of resilience. Behaviorally, resilience in such studies is typically defined through indicators such as inhibitory control (Maier & Watkins, Reference Maier and Watkins2010; Ochsner & Gross, Reference Ochsner and Gross2005) and emotional appraisal and regulation (Hänsel & von Känel, Reference Hänsel and von Känel2008; Maier & Watkins, Reference Maier and Watkins2010), and flexibility (Waugh et al., Reference Waugh, Wager, Fredrickson, Noll and Taylor2008). Specifically, it has been demonstrated that individual differences in these processes substantiated by the related capability to employ prefrontal cortical (PF, or PF cortex, PFC) control systems that control affective processing by the amygdala and related structures mediate a response to adversity (Rodman et al., Reference Rodman, Jenness, Weissman, Pine and McLaughlin2019). Additionally, it has been hypothesized that the parietal lobe (PL) might play a role in resilience due to its centrality in the evaluation of the emotional relevance of the external stimuli in particular and emotion regulation in general (Bisley & Goldberg, Reference Bisley and Goldberg2010; Bzdok et al., Reference Bzdok, Hartwigsen, Reid, Laird, Fox and Eickhoff2016). Particularly, the primary somatosensory cortex is responsible for initiating the processing of sensory inputs, and the secondary somatosensory cortex engages cognitive processes to complete their processing. Both areas are located in PL, and, therefore, the lobe orchestrates a complex interplay between incoming stimuli and pre-existing somatosensory representations, generating fresh memory traces (Wagner et al., Reference Wagner, Shannon, Kahn and Buckner2005). Hence, it is sensible to wonder whether children who experience early life stress (ELS) but exhibit enhanced parietal functioning (i.e., children with, presumably, stronger resilience) are less susceptible to the impact of the novel external stimuli when they are somehow related to, reminiscent of, or can trigger their past neglected experiences (Luo et al., Reference Luo, Zou, Nie, Wu, Du, Chen, Li and Peng2023).
As the number of empirical studies has grown (Dedovic et al., Reference Dedovic, D’Aguiar and Pruessner2009; Pitman et al., Reference Pitman, Rasmusson, Koenen, Shin, Orr, Gilbertson, Milad and Liberzon2012; Shin & Liberzon, Reference Shin and Liberzon2010; van der Werff et al., Reference van der Werff, van den Berg, Pannekoek, Elzinga and Van Der Wee2013), numerous brain structures have been implicated as substantiations of resilience, including the anterior cingulate cortex (ACC), hippocampus, insula, orbitofrontal cortex (OFC), and ventro-medial and dorsolateral prefrontal cortical (vmPFC). An early comprehensive review of the literature on the (neuro)physiological bases of resilience has demonstrated that it arises from a sophisticated orchestra of tuned-up structures connected by the interchange between distributed brain systems, including the amygdala, ACC, and PFC, which are neuromodulated (Feder et al., Reference Feder, Fred-Torres, Southwick and Charney2019). A later review (Holz et al., Reference Holz, Tost and Meyer-Lindenberg2020) fine-tuned the “convergent resilience circuit,” still stressing its reliance on frontal regions, particularly the perigenual anterior cingulate cortex (adjacent to OFC), numerous regions of the PFC, and key limbic structures such as the ventral striatum (VS). To illustrate, a concurrent stimulation of the dorsolateral prefrontal cortical and OFC leads to enhanced resilience (Salehinejad et al., Reference Salehinejad, Nejati and Derakhshan2017). In addition, there is evidence of a positive association between resilience and a larger hippocampal structure (Moreno-López et al., Reference Moreno-López, Ioannidis, Askelund, Smith, Schueler and van Harmelen2020).
Special attention has been given to findings of altered hippocampus dentate gyrus (DG) development, as it has been previously related to stress reactivity (Anacker et al., Reference Anacker, Luna, Stevens, Millette, Shores, Jimenez, Chen and Hen2018; Boldrini et al., Reference Boldrini, Galfalvy, Dwork, Rosoklija, Trencevska-Ivanovska, Pavlovski, Hen, Arango and Mann2019; Roddy et al., Reference Roddy, Farrell, Doolin, Roman, Tozzi, Frodl, O’Keane and O’Hanlon2019) and depression and suicidal behavior (Boldrini et al., Reference Boldrini, Galfalvy, Dwork, Rosoklija, Trencevska-Ivanovska, Pavlovski, Hen, Arango and Mann2019; Boldrini et al., Reference Boldrini, Santiago, Hen, Dwork, Rosoklija, Tamir, Arango and Mann2013; Huang et al., Reference Huang, Coupland, Lebel, Carter, Seres, Wilman and Malykhin2013; Roddy et al., Reference Roddy, Farrell, Doolin, Roman, Tozzi, Frodl, O’Keane and O’Hanlon2019; Wang et al., Reference Wang, Neylan, Mueller, Lenoci, Truran, Marmar, Weiner and Schuff2010), and contains structures involved in adult neurogenesis (Toni & Schinder, Reference Toni and Schinder2016). Experimental manipulations of the ventral DG in model organisms have been related to resilience to chronic stress and depressive-like behaviors, suggesting a substantive and even causal role for the DG in resilience (Anacker et al., Reference Anacker, Luna, Stevens, Millette, Shores, Jimenez, Chen and Hen2018; Boldrini et al., Reference Boldrini, Galfalvy, Dwork, Rosoklija, Trencevska-Ivanovska, Pavlovski, Hen, Arango and Mann2019; Hill et al., Reference Hill, Sahay and Hen2015; Tunc-Ozcan et al., Reference Tunc-Ozcan, Peng, Zhu, Dunlop, Contractor and Kessler2019; Veena et al., Reference Veena, Srikumar, Raju and Shankaranarayana Rao2009). Specifically, DG neurogenesis made adult mice in the experimental group more resilient to chronic stress compared to control rodents (Anacker et al., Reference Anacker, Luna, Stevens, Millette, Shores, Jimenez, Chen and Hen2018). Contrary to this, inhibiting immature neurons increased susceptibility to stress (Bagot et al., Reference Bagot, Parise, Peña, Zhang, Maze, Chaudhury, Persaud, Cachope, Bolaños-Guzmán, Cheer, Deisseroth, Han and Nestler2015; Jimenez et al., Reference Jimenez, Su, Goldberg, Luna, Biane, Ordek, Zhou, Ong, Wright, Zweifel, Paninski, Hen and Kheirbek2018; Padilla-Coreano et al., Reference Padilla-Coreano, Bolkan, Pierce, Blackman, Hardin, Garcia-Garcia, Spellman and Gordon2016). Relatedly, antidepressant treatments, exercise, and environmental enrichment have been reported to enhance DG structure and function (Boldrini et al., Reference Boldrini, Santiago, Hen, Dwork, Rosoklija, Tamir, Arango and Mann2013; Erickson et al., Reference Erickson, Voss, Prakash, Basak, Szabo, Chaddock, Kim, Heo, Alves, White, Wojcicki, Mailey, Vieira, Martin, Pence, Woods, McAuley and Kramer2011; Nuninga et al., Reference Nuninga, Mandl, Boks, Bakker, Somers, Heringa, Nieuwdorp, Hoogduin, Kahn, Luijten and Sommer2020; Van Praag et al., Reference Van Praag, Kempermann and Gage1999; Veena et al., Reference Veena, Srikumar, Raju and Shankaranarayana Rao2009). Antidepressants have been stated to cause neurogenesis increase in mice (Malberg et al., Reference Malberg, Eisch, Nestler and Duman2000; Wang et al., Reference Wang, David, Monckton, Battaglia and Hen2008), rats (Lyons et al., Reference Lyons, ElBeltagy, Umka, Markwick, Startin, Bennett and Wigmore2011), nonhuman primates (Perera et al., Reference Perera, Dwork, Keegan, Thirumangalakudi, Lipira, Joyce, Lange, Higley, Rosoklija, Hen, Sackeim and Coplan2011), and perhaps patients with depression (Boldrini et al., Reference Boldrini, Santiago, Hen, Dwork, Rosoklija, Tamir, Arango and Mann2013; Boldrini et al., Reference Boldrini, Underwood, Hen, Rosoklija, Dwork, John Mann and Arango2009). It is well established that successful recovery and adaptation after ELS (Charney, Reference Charney2004) is marked by active coping style, effective and efficient emotional regulation, and adequate cognitive functioning, which might either result from or lead to brain circuit remodeling, such as alterations in DG cellular plasticity, encoding of emotion-related memories, and strengthening the amygdala–PFC connectivity (Boldrini et al., Reference Boldrini, Galfalvy, Dwork, Rosoklija, Trencevska-Ivanovska, Pavlovski, Hen, Arango and Mann2019). Given this pattern of these findings, a possible translational outcome might be related to the potential causal function of the DG in the manifestation of depressive symptomatology after exposure to maternal stress, as it can open pathways to novel interventions.
In substantiating the involvement of different brain structures and networks, the literature cited illustrations of not only resilience itself but also related behaviors. Specifically, positive coping has been associated with an increased volume of perigenual anterior cingulate cortex (Holz, Boecker, Jennen-Steinmetz, et al., Reference Holz, Boecker, Jennen-Steinmetz, Buchmann, Blomeyer, Baumeister, Plichta, Esser, Schmidt, Meyer-Lindenberg, Banaschewski, Brandeis and Laucht2016), and the neighboring OFC has been labeled a neural marker of optimism (Nes & Segerstrom, Reference Nes and Segerstrom2006). The latter, in turn, has been linked to resilience (Feder et al., Reference Feder, Nestler and Charney2009; Ozbay et al., Reference Ozbay, Fitterling, Charney and Southwick2008), especially in adulthood and older age (Feder et al., Reference Feder, Fred-Torres, Southwick and Charney2019), and reported to mediate the connection between anxiety and lateral OFC (Dolcos et al., Reference Dolcos, Hu, Iordan, Moore and Dolcos2016). Relatedly, the enlarged vmPFC was observed to foster resilient functioning and safeguard against internalizing disorders (Morey et al., Reference Morey, Haswell, Hooper and De Bellis2016). There was also a report of the dynamic changes in the mobilization of vmPFC in response to stress, with prolonged stress exposure increasing engagement in this area (Sinha et al., Reference Sinha, Lacadie, Constable and Seo2016). Additionally, various elements of this circuitry have been implemented in research capturing the role of social support in pain-related adversity (Coan et al., Reference Coan, Schaefer and Davidson2006; Eisenberger et al., Reference Eisenberger, Master, Inagaki, Taylor, Shirinyan, Lieberman and Naliboff2011; Eisenberger et al., Reference Eisenberger, Taylor, Gable, Hilmert and Lieberman2007; Younger et al., Reference Younger, Aron, Parke, Chatterjee and Mackey2010).
The major chunk of this literature pertains to group studies utilizing structural and functional MRI (e.g., Amico et al., Reference Amico, Meisenzahl, Koutsouleris, Reiser, Möller and Frodl2011; Fischer et al., Reference Fischer, Ellwood-Lowe, Colich, Cichocki, Ho and Gotlib2019; Hopper et al., Reference Hopper, Frewen, van der Kolk and Lanius2007; Peres et al., Reference Peres, Foerster, Santana, Fereira, Nasello, Savoia, Moreira-Almeida and Lederman2011; Phan et al., Reference Phan, Fitzgerald, Nathan, Moore, Uhde and Tancer2005; Rauch et al., Reference Rauch, Shin, Segal, Pitman, Carson, McMullin, Whalen and Makris2003; Rauch et al., Reference Rauch, Whalen, Shin, McInerney, Macklin, Lasko, Orr and Pitman2000; Rodman et al., Reference Rodman, Jenness, Weissman, Pine and McLaughlin2019; Shin et al., Reference Shin, Bush, Milad, Lasko, Handwerger Brohawn, Hughes, Macklin, Gold, Karpf, Orr, Rauch and Pitman2011; Sun et al., Reference Sun, Haswell, Morey and De Bellis2019; van Dijk et al., Reference van Dijk, Talati, Kashyap, Desai, Kelsall, Gameroff, Aw, Abraham, Cullen, Cha, Anacker, Weissman and Posner2024), where individuals who experienced ELS are compared to typically developing individuals or to themselves when stratified by such variables as absence or presence of a particular disorder (e.g., PTSD, anxiety, or depression) or high-risk clinical groups are compared to low-risk controls. Thus, post-ELS, in non-PTSD, compared to PTSD, functional connectivity was decreased between the insula and the right amygdala (Etkin & Wager, Reference Etkin and Wager2007) but increased between the thalamus and the right medial frontal or the left rostral ACC, rACC (Yin et al., Reference Yin, Jin, Hu, Duan, Li, Song, Chen, Feng, Jiang, Jin, Wong, Gong and Li2011).
There also have been studies of healthy individuals investigating the neural correlates of resilience (Burt et al., Reference Burt, Whelan, Conrod, Banaschewski, Barker, Bokde, Bromberg, Büchel, Fauth-Bühler, Flor, Galinowski, Gallinat, Gowland, Heinz, Ittermann, Mann, Nees, Papadopoulos-Orfanos, Paus and Imagen Consortium2016; Gupta et al., Reference Gupta, Love, Kilpatrick, Labus, Bhatt, Chang, Tillisch, Naliboff and Mayer2017; Kong et al., Reference Kong, Wang, Hu and Liu2015; Reynaud et al., Reference Reynaud, Guedj, Souville, Trousselard, Zendjidjian, El Khoury-Malhame, Fakra, Nazarian, Blin, Canini and Khalfa2013; Salehinejad et al., Reference Salehinejad, Nejati and Derakhshan2017; Waugh et al., Reference Waugh, Wager, Fredrickson, Noll and Taylor2008). Similarly to the research in clinical samples, the generated findings are also quite mosaic, demonstrating both convergence and divergence with the findings from studies focusing on clinical diagnoses. Of interest is the differential engagement of the insula by low- (nonspecific activation to both the neutral and aversive stimuli) and high- (specific activation only to aversive stimuli) resilient people (Waugh et al., Reference Waugh, Wager, Fredrickson, Noll and Taylor2008), indicative of the capacity to use brain resources adequately under threat. Similarly, the salience network (i.e., the bilateral insula, dorsal ACC, dACC, and rACC) demonstrated less spontaneous activation in healthy young adults (Kong et al., Reference Kong, Wang, Hu and Liu2015), indicating, perhaps, higher capacity for emotional regulation (Etkin et al., Reference Etkin, Egner and Kalisch2011). There are also studies linking resilience to brain morphology (i.e., cortical thickness and surface area) of cortical-limbic regions engaged with the inhibition systems (Gupta et al., Reference Gupta, Love, Kilpatrick, Labus, Bhatt, Chang, Tillisch, Naliboff and Mayer2017). A large sample study has documented that those adolescents who experienced ELS and demonstrated positive life outcomes had larger volumes of gray matter in the right middle and superior frontal gyrus (Burt et al., Reference Burt, Whelan, Conrod, Banaschewski, Barker, Bokde, Bromberg, Büchel, Fauth-Bühler, Flor, Galinowski, Gallinat, Gowland, Heinz, Ittermann, Mann, Nees, Papadopoulos-Orfanos, Paus and Imagen Consortium2016). Later, the resilience-higher gray matter volumes were reported to be differentiated by sex so that the sex-by-resilience interaction differentiated the role of the enlarged gray matter in different areas of the brain for males and females (Cornwell et al., Reference Cornwell, Toschi, Hamilton-Giachritsis, Staginnus, Smaragdi, Gonzalez-Madruga, Rogers, Martinelli, Kohls, Raschle, Konrad, Stadler, Freitag, De Brito and Fairchild2023). Moreover, in a sample from a healthy population with ELS (Luo et al., Reference Luo, Zou, Nie, Wu, Du, Chen, Li and Peng2023), the engagement of PL was reported to be characteristic of resilience. Yet, there are some contradictory results on the involvement of PL (Barzilay et al., Reference Barzilay, Rosen, Moore, Roalf, Satterthwaite, Calkins, Ruparel, Patrick, Scott, Wolf, Gur and Gur2020; Grieder et al., Reference Grieder, Homan, Federspiel, Kiefer and Hasler2020), suggesting that it may have a complex and context-dependent role in resilience. Healthy samples have also been used to study functional connectivity, resulting in a complex map of positive and negative correlations with various brain regions and connectivity between them (Shi et al., Reference Shi, Sun, Wei and Qiu2019). There are studies where resilience is defined as a continuous indicator (e.g., adaptive psychosocial functioning adjusted for the severity of childhood adversity; González-García et al., Reference González-García, Buimer, Moreno-López, Sallie, Váša, Lim, Romero-Garcia, Scheuplein, Whitaker, Jones, Dolan, Fonagy, Goodyer, Bullmore and van Harmelen2023) that can be correlated with different characteristics of brain functioning, such as the nodal degree, which indexes the number of associations that various brain regions form in a given network (González-García et al., Reference González-García, Buimer, Moreno-López, Sallie, Váša, Lim, Romero-Garcia, Scheuplein, Whitaker, Jones, Dolan, Fonagy, Goodyer, Bullmore and van Harmelen2023).
Thus, the understanding of the brain foundation of resilience is only emerging, and as of today, the relevant literature lacks consistency (Eaton et al., Reference Eaton, Cornwell, Hamilton-Giachritsis and Fairchild2022; Méndez Leal & Silvers, Reference Méndez Leal and Silvers2021; Zhang et al., Reference Zhang, Rakesh, Cropley and Whittle2023). As the literature grows, there is hope for convergence on the definition of resilience, which, in turn, should aim to converge the findings on its brain bases. However, at this point, the bottom line is that the prefrontal and subcortical structure, function, and functional connectivity are engaged with and relevant to the manifestation of resilience (Zhang et al., Reference Zhang, Rakesh, Cropley and Whittle2023). Therefore, these systems, together or separately, depending on the instrumentation, should be consistently sampled while brain readiness to perform is recorded under different stressogenic situations.
Resilience in the genome
Whereas there has been much research and progress in mapping out the genetic bases of stress reactivity, far less is understood regarding the genetic endowment of resilience. It has been argued (Elbau et al., Reference Elbau, Cruceanu and Binder2019; Murrough & Russo, Reference Murrough and Russo2019) that identifying genetic variation that differentiated disease risk in the face of adversity should enhance the understanding of mechanisms that trigger and advance resilience, thus detecting new pharmacological targets. Genetic variation, which has already been associated with signaling systems that modulate the structure and function of the relevant neural substrates in response to stress, has been deemed to be a good starting point as candidate genes for resilience (Elbau et al., Reference Elbau, Cruceanu and Binder2019; Holz et al., Reference Holz, Tost and Meyer-Lindenberg2020; Niitsu et al., Reference Niitsu, Rice, Houfek, Stoltenberg, Kupzyk and Barron2018). Of no surprise is that most of these genes have a role in the central nervous system functioning. Thus, the serotonergic pathway (Kiser et al., Reference Kiser, Steemers, Branchi and Homberg2012) is known to substantiate emotional processing (Cao et al., Reference Cao, Harneit, Walter, Erk, Braun, Moessnang, Geiger, Zang, Mohnke, Heinz, Romanczuk-Seiferth, Mühleisen, Mattheisen, Witt, Cichon, Nöthen, Rietschel, Meyer-Lindenberg and Tost2018). Increased serotonin turnover in the amygdala, hypothalamus, PFC, and VS has been reported under stress (Feder et al., Reference Feder, Nestler and Charney2009). Similarly, the dopaminergic pathway, substantiating motivation (Dreher et al., Reference Dreher, Meyer-Lindenberg, Kohn and Berman2008), operates in the PFC and gets inhibited in the VS (Charney, Reference Charney2004) following stress. It has been observed that levels of dopamine are amended in depression and PTSD (Charney, Reference Charney2004; Dunlop & Nemeroff, Reference Dunlop and Nemeroff2007); heightened dopamine turnover has been stated to substantiate exaggerated fear response to stress (Hoexter et al., Reference Hoexter, Fadel, Felício, Calzavara, Batista, Reis, Shih, Pitman, Andreoli, Mello, Mari and Bressan2012). Given these associations, it is plausible that genetic variation in serotonergic and dopaminergic signaling might also be relevant to substantiating individual differences in resilience. Similarly, a genetic variation known to alter stress responsivity (Matosin et al., Reference Matosin, Halldorsdottir and Binder2018) might play an important part in resilience. In addition to the involvement of the neurotransmitter systems, the literature on stress reactivity emphasizes the importance of hormonal signaling, engaging systems such as corticotropin (corticotropin-releasing hormone, CRH, a central regulator of the HPA axis) and oxytocin (a natural hormone managing reproductive system and engaged with many aspects of social behavior). There are now “classic” genetic variants that are both functional and common, which have been analyzed with regard to their association with resilience (Niitsu et al., Reference Niitsu, Rice, Houfek, Stoltenberg, Kupzyk and Barron2018): the serotonin-transporter-linked polymorphic region (5-HTTLPR) in the serotonin transporter gene (SLC6A4), and repeats and single-nucleotide polymorphisms, SNP, in dopamine receptor D4 (DRD4), corticotropin-releasing hormone receptor 1 (CRHR1), and oxytocin receptor (OXTR) genes.
A variable-number tandem repeat polymorphism in the promoter region of the SLC6A4 gene is a well-studied variant. The region 5-HTTLPR consists of a 14-repeat short variant, S-allele, and a 16-repeat long variant, L-allele (Heils et al., Reference Heils, Teufel, Petri, Stöber, Riederer, Bengel and Lesch1996; Lesch et al., Reference Lesch, Bengel, Heils, Sabol, Greenberg, Petri, Benjamin, Müller, Hamer and Murphy1996). Of note also is a single-base substitution (A > G) in the L type of 5-HTTLPR known as rs25531 (Hu et al., Reference Hu, Lipsky, Zhu, Akhtar, Taubman, Greenberg, Xu, Arnold, Richter, Kennedy, Murphy and Goldman2006). A large corpus of research has registered an association between the 5-HTTLPR L-allele and decreased activation of the amygdala (Munafò et al., Reference Munafò, Brown and Hariri2008). Moreover, carriers of the L-allele demonstrated heightened functional coupling between the PFC and the amygdala (Pezawas et al., Reference Pezawas, Meyer-Lindenberg, Drabant, Verchinski, Munoz, Kolachana, Egan, Mattay, Hariri and Weinberger2005). It has also been reported (Holz, Zohsel, et al., Reference Holz, Zohsel, Laucht, Banaschewski, Hohmann and Brandeis2018) that environmental adversity moderated the impact of 5-HTTLPR on amygdala activation and connectivity in a number of studies (Alexander et al., Reference Alexander, Klucken, Koppe, Osinsky, Walter, Vaitl, Sammer, Stark and Hennig2012; Canli et al., Reference Canli, Qiu, Omura, Congdon, Haas, Amin, Herrmann, Constable and Lesch2006). Additionally, certain genotypes of 5-HTTLPR/rs25531 were associated with resilience in children/adolescents, while others were connected to resilience in adults (Niitsu et al., Reference Niitsu, Rice, Houfek, Stoltenberg, Kupzyk and Barron2018). The variation in the promoter region of SLC6A4 has been featured in genetic association studies of a number of psychiatric conditions, stressing a lack of specificity of its action. Moreover, it was featured in several interaction studies (known as G × E or gene by environment), where it was treated as a genetic liability to a particular (similarly nonspecific) adverse environmental impact. Importantly, the field does not converge on the specific role of this variation but acknowledges its relevance for understanding the genetic underpinning of both susceptibility and resistance to the manifestation of negative outcomes in the face of adversity.
Another well-studied source of genetic variation is a variable number of tandem repeats in a 30-base repeat sequence (VNTR) polymorphism in the monoamine oxidase A (MAO-A) gene. MAO-A is an enzyme that is central to the catabolism of a number of neurotransmitters, including serotonin. The activity of MAO-A influences serotonin levels: high levels of MAO-A activity can lead to decreased serotonin availability, and low MAO-A activity can result in increased serotonin levels. High levels of MAO-A are controlled by the MAO-A-H genotype, which has been stated to partially differentiate resilience in men (Holz et al., Reference Holz, Tost and Meyer-Lindenberg2020). Specifically, a lower emotional sensitivity in the amygdala (Alia-Klein et al., Reference Alia-Klein, Goldstein, Tomasi, Woicik, Moeller, Williams, Craig, Telang, Biegon, Wang, Fowler and Volkow2009; Lee & Ham, Reference Lee and Ham2008; Meyer-Lindenberg et al., Reference Meyer-Lindenberg, Buckholtz, Kolachana, Hariri, Pezawas, Blasi, Wabnitz, Honea, Verchinski, Callicott, Egan, Mattay and Weinberger2006), along with increased recruitment of PFC-based (ACC, vmPFC) cognitive control networks, was reported in individuals with the MAO-A-H genotype (Fan et al., Reference Fan, Fossella, Sommer, Wu and Posner2003; Meyer-Lindenberg et al., Reference Meyer-Lindenberg, Buckholtz, Kolachana, Hariri, Pezawas, Blasi, Wabnitz, Honea, Verchinski, Callicott, Egan, Mattay and Weinberger2006; Passamonti et al., Reference Passamonti, Cerasa, Gioia, Magariello, Muglia, Quattrone and Fera2008; Passamonti et al., Reference Passamonti, Fera, Magariello, Cerasa, Gioia, Muglia, Nicoletti, Gallo, Provinciali and Quattrone2006). Furthermore, it has been observed that the unfavorable genotype (i.e., 3 versus 4 repeats, with 3R variant resulting in lower MAO-A activity) and environmental adversity can co-act, substantiating the manifestation of negative outcomes such as reactive aggression; importantly, these effects appear to be sex-specific (Byrd & Manuck, Reference Byrd and Manuck2014; Caspi et al., Reference Caspi, McClay, Moffitt, Mill, Martin, Craig, Taylor and Poulton2002). Thus, genetic variants associated with stress reactivity and resilience might exert sex-specific effects (Holz, Boecker, Buchmann, et al., Reference Holz, Boecker, Buchmann, Blomeyer, Baumeister, Hohmann, Jennen-Steinmetz, Wolf, Rietschel, Witt, Plichta, Meyer-Lindenberg, Schmidt, Esser, Banaschewski, Brandeis and Laucht2016), substantiating well-known sex differences in response to adversity.
Variations in several genes participating in the turnover of dopamine have also been investigated. The catechol-o-methyltransferase Val158Met polymorphism (rs4680) regulates the extra-synaptic dopamine degradation due to its impact on the catechol-o-methyltransferase enzyme (Holz et al., Reference Holz, Tost and Meyer-Lindenberg2020). There are numerous studies of the association between this polymorphism and various relevant brain structures, although the results are somewhat difficult to interpret. The literature reports the polymorphism’s associations, specifically with the Val allele, with lower punishment-related VS activity (Schmack et al., Reference Schmack, Schlagenhauf, Sterzer, Wrase, Beck, Dembler, Kalus, Puls, Sander, Heinz and Gallinat2008); alleviated activation during reward anticipation (Dreher et al., Reference Dreher, Kohn, Kolachana, Weinberger and Berman2009; Yacubian et al., Reference Yacubian, Sommer, Schroeder, Gläscher, Kalisch, Leuenberger, Braus and Büchel2007); potentiated activity of the nucleus accumbens, the ACC and the right inferior PL during reward receipt (Camara et al., Reference Camara, Krämer, Cunillera, Marco-Pallarés, Cucurell, Nager, Mestres-Missé, Bauer, Schüle, Schöls, Tempelmann, Rodriguez-Fornells and Münte2010); as well as and null findings (Forbes et al., Reference Forbes, Brown, Kimak, Ferrell, Manuck and Hariri2009). The polymorphism has also been reported to differentiate the impact of ELS in the reward circuit, with lower activity in the VS and ACC with increasing levels of childhood adversity in Val carriers and the opposite effect for Met homozygotes (Boecker-Schlier et al., Reference Boecker-Schlier, Holz, Buchmann, Blomeyer, Plichta, Jennen-Steinmetz, Wolf, Baumeister, Treutlein, Rietschel, Meyer-Lindenberg, Banaschewski, Brandeis and Laucht2016). Variations in the dopamine receptor DRD4, specifically, its D4 version (Van Tol et al., Reference Van Tol, Wu, Guan, Ohara, Bunzow, Civelli, Kennedy, Seeman, Niznik and Jovanovic1992) and a VNTR containing 3 to 11 repeats in the dopamine transporter DAT (also known as SLC6A3) gene has been associated with individual differences in reward processing, ostensibly by acting on VS ventral striatal activity (Aarts et al., Reference Aarts, Roelofs, Franke, Rijpkema, Fernández, Helmich and Cools2010; Dreher et al., Reference Dreher, Kohn, Kolachana, Weinberger and Berman2009; Filbey et al., Reference Filbey, Ray, Smolen, Claus, Audette and Hutchison2008; Forbes et al., Reference Forbes, Brown, Kimak, Ferrell, Manuck and Hariri2009; Hahn et al., Reference Hahn, Heinzel, Dresler, Plichta, Renner, Markulin, Jakob, Lesch and Fallgatter2011; McClernon et al., Reference McClernon, Hutchison, Rose and Kozink2007; Nikolova et al., Reference Nikolova, Ferrell, Manuck and Hariri2011; Paloyelis et al., Reference Paloyelis, Mehta, Faraone, Asherson and Kuntsi2012; Wittmann et al., Reference Wittmann, Tan, Lisman, Dolan and Düzel2013), although there are some contradictory findings (Hoogman et al., Reference Hoogman, Onnink, Cools, Aarts, Kan, Arias Vasquez, Buitelaar and Franke2013). Moreover, in DRD4 resilience scores were associated with the CC and CT genotypes of rs1800955 (Cicchetti & Rogosch, Reference Cicchetti and Rogosch2012) and the 7r7r and 4r7r genotypes of the VNTR (Das et al., Reference Das, Cherbuin, Tan, Anstey and Easteal2011).
Variants in CRHR1 and OXTR genes have received less attention, although both have been associated with resilience scores. The relevant variation in CRHR1 was captured not with a single polymorphism but with a combination of them, suggesting that the TAT haplotype might contribute to the biological foundation of resilience (Cicchetti & Rogosch, Reference Cicchetti and Rogosch2012). The OXTR polymorphism rs53576 has also been associated with resilience, although there is a disagreement on what particular genotype, GG (Cicchetti & Rogosch, Reference Cicchetti and Rogosch2012) or AA (Bradley et al., Reference Bradley, Davis, Wingo, Mercer and Ressler2013), carried the signal.
Other candidate genes, selected due to their specific biological function, have been considered. Among them are genes coding for brain-derived neurotrophic factor (BDNF), FK506 binding protein 5 (FKBP5), and regulator of G-protein signaling 2. BDNF confirms the survival of existing neurons and supports the growth and differentiation of new neurons and synapses, being especially active in the areas of the brain, substantiating learning, memory, and higher thinking. It has been reported that the GG genotype of its rs6265 polymorphism contributed to resilience (Nederhof et al., Reference Nederhof, Bouma, Riese, Laceulle, Ormel and Oldehinkel2010; van Winkel et al., Reference van Winkel, Peeters, van Winkel, Kenis, Collip, Geschwind, Jacobs, Derom, Thiery, van Os, Myin-Germeys and Wichers2014), but perhaps only in Caucasians (Niitsu et al., Reference Niitsu, Rice, Houfek, Stoltenberg, Kupzyk and Barron2018). The FKBP5 gene encodes a protein FKBP51, a member of the immunophilin protein family and known for its role in immunoregulation and basic cellular processes involving protein folding and trafficking. The protein binds to the immunosuppressants FK506 and rapamycin, mediates calcineurin inhibition, and regulates the affinity of the glucocorticoid receptor (GR) for cortisol (Binder, Reference Binder2009; Denny et al., Reference Denny, Valentine, Reynolds, Smith and Scammell2000; Wochnik et al., Reference Wochnik, Rüegg, Abel, Schmidt, Holsboer and Rein2005). In fact, FKBP51 co-chaperones to GR directly affect its sensitivity to circulating glucocorticoids; thus, an important role of the FKBP51 protein is the regulation of stress responsivity. Loss of FKBP51 in gamma-aminobutyric-acid or glutamate neurons leads to negative outcomes, especially under high-risk environments (van Doeselaar et al., Reference van Doeselaar, Stark, Mitra, Yang, Bordes, Stolwijk, Engelhardt, Kovarova, Narayan, Brix, Springer, Deussing, Lopez, Czisch and Schmidt2023). The rs1360780 polymorphism in this gene appears to differentiate glucocorticoid resistance of the GR in the CC genotype (Binder et al., Reference Binder, Bradley, Liu, Epstein, Deveau, Mercer, Tang, Gillespie, Heim, Nemeroff, Schwartz, Cubells and Ressler2008), generating a more effective negative feedback loop thought to be stimulated by glucocorticoid, coupled with a more rapid stress adaptation in the face of adversity (Matosin et al., Reference Matosin, Halldorsdottir and Binder2018; Zannas et al., Reference Zannas, Wiechmann, Gassen and Binder2016) and a blunted threat-induced amygdala activity in the context ELS (Holz et al., Reference Holz, Buchmann, Boecker, Blomeyer, Baumeister, Wolf, Rietschel, Witt, Plichta, Meyer-Lindenberg, Banaschewski, Brandeis and Laucht2015; VanZomeren-Dohm et al., Reference VanZomeren-Dohm, Pitula, Koss, Thomas and Gunnar2015; White et al., Reference White, Bogdan, Fisher, Muñoz, Williamson and Hariri2012). Finally, the regulator of G-protein signaling 2 gene encodes the Regulator of G-protein Signaling 2 protein, which modulates the activity of G proteins, where the GG genotype of the rs4606 polymorphism has been reported to contribute to resilience in Black parents (Dunn et al., Reference Dunn, Solovieff, Lowe, Gallagher, Chaponis, Rosand, Koenen, Waters, Rhodes and Smoller2014).
Of note is that, to date, seemingly only two genome-wide association studies (GWAS) have been conducted. The first one featured resilience conceptualized through a self-report of perceived resilience collected on almost 15,000 US Army soldiers of European descent (Stein et al., Reference Stein, Choi, Jain, Campbell-Sills, Chen, Gelernter, He, Heeringa, Maihofer, Nievergelt, Nock, Ripke, Sun, Kessler, Smoller and Ursano2019). There were three results at the genome-wide level of significance. The first signal was within a locus on an intergenic region on chromosome 4 upstream from the DCLK2 (Doublecortin-Like Kinase 2) gene (4 SNPs in linkage disequilibrium; top SNP: rs4260523 [p = 5.65 × 10−9] is an eQTL in frontal cortex), which is a member of the doublecortin family of kinases that promote survival and regeneration of injured neurons. The second signal was in the gene KLHL36 (Kelch-Like Family Member 36) at p = 1.89 × 10−6. A polygenic risk score (PGS, a weighted additive score of all alleles that demonstrated associations with a trait in the framework of GWAS, reflecting a substantial amount of the trait-associated variance with a single measure) derived from the self-assessed resilience GWAS was not significantly associated with outcome-based resilience. In addition, when a subsample of soldiers (N = 581) exposed to the highest level of deployment stress was extracted, genome-wide significant association with outcome-based resilience was registered for one locus (top SNP: rs12580015 [p = 2.37 × 10−8]) on chromosome 12 downstream from SLC15A5 (solute carrier family 15 member 5). Notably, the estimate for the heritability of resilience was 16%. The second study (Cusack, Aliev, et al., Reference Cusack, Aliev, Bustamante, Dick and Amstadter2023) used a previously utilized (Amstadter et al., Reference Amstadter, Maes, Sheerin, Myers and Kendler2016; Cusack, Bountress, et al., Reference Cusack, Bountress, Sheerin, Dick and Amstadter2023) discrepancy-based indicator of resilience, calculated based on the information on trauma exposure and a checklist of psychiatric symptoms (depression and anxiety). Heritability estimates for resilience did not differ from zero. Zero variants met genome-wide level of significance, but nine passed the suggestive association threshold and mapped onto three genes: SEZ6L (a protein-coding gene that contributes to specialized endoplasmic reticulum functions in neurons), LINC02112 (Long Intergenic Non-Protein Coding RNA 2112), and FRK (fyn related Src family tyrosine kinase), and one cluster of genes (NKAIN3, GGH, TTPA, YTHDF3-AS1) on chromosome 8 related to metabolization and transport of various vitamins and minerals. Whereas for SEZ6L and the chromosome 8 cluster, the associations presented meaningful interpretations, but these were not obvious for the two other genes. Importantly, none of these candidates have been implicated in resilience earlier. Finally, researchers utilized PGS previously established for alcohol dependency, alcohol consumption, and PTSD. They demonstrated genetic overlap between resilience and AD, as well as resilience and PTSD.
The interactive notion of resilience is well suited for genome-wide G × E studies (Genome Environment Wide Interactions Studies [GEWIS]) to identify gene variants that are able to differentiate individual responses to adverse environmental stimuli. Although there are some GEWIS primarily with negative outcomes (e.g., depression) and adverse life events, their results are difficult to interpret (Arnau-Soler et al., Reference Arnau-Soler, Macdonald-Dunlop, Adams, Clarke, MacIntyre, Milburn, Navrady, Hayward, McIntosh and Thomson2019; Coleman et al., Reference Coleman, Peyrot, Purves, Davis, Rayner, Choi, Hübel, Gaspar, Kan and Van der Auwera2020; Dunn et al., Reference Dunn, Wiste, Radmanesh, Almli, Gogarten, Sofer, Faul, Kardia, Smith, Weir, Zhao, Soare, Mirza, Hek, Tiemeier, Goveas, Sarto, Snively, Cornelis and Smoller2016; Ikeda et al., Reference Ikeda, Shimasaki, Takahashi, Kondo, Saito, Kawase, Esaki, Otsuka, Mano, Kubo and Iwata2016; Otowa et al., Reference Otowa, Kawamura, Tsutsumi, Kawakami, Kan, Shimada, Umekage, Kasai, Tokunaga and Sasaki2016; Suppli et al., Reference Suppli, Andersen, Agerbo, Rajagopal, Appadurai, Coleman, Breen, Bybjerg-Grauholm, Bækvad-Hansen, Pedersen, Pedersen, Thompson, Munk-Olsen, Benros, Als, Grove, Werge, Børglum, Hougaard and Musliner2022). By virtue of their design, GEWIS requires large sample sizes and documented stressful life events. To thwart power issue, the field put forward the usage of PGS, which are constructed separately for different outcomes, for example, depression (Halldorsdottir et al., Reference Halldorsdottir, Piechaczek, Soares de Matos, Czamara, Pehl, Wagenbuechler, Feldmann, Quickenstedt-Reinhardt, Allgaier, Freisleder, Greimel, Kvist, Lahti, Räikkönen, Rex-Haffner, Arnarson, Craighead, Schulte-Körne and Binder2019; Mullins et al., Reference Mullins, Power, Fisher, Hanscombe, Euesden, Iniesta, Levinson, Weissman, Potash, Shi, Uher, Cohen-Woods, Rivera, Jones, Jones, Craddock, Owen, Korszun, Craig and Lewis2016; Peyrot et al., Reference Peyrot, Van der Auwera, Milaneschi, Dolan, Madden, Sullivan, Strohmaier, Ripke, Rietschel, Nivard, Mullins, Montgomery, Henders, Heat, Fisher, Dunn, Byrne, Air, Wray and Penninx2018) and schizophrenia (Hess et al., Reference Hess, Mattheisen, Greenwood, Tsuang, Edenberg, Holmans, Faraone and Glatt2024). Unfortunately, the results are mixed. Specifically, studies on depression do not support the presence of interaction between depression PGS and the environment (Elbau et al., Reference Elbau, Cruceanu and Binder2019). On the contrary, the outlook for schizophrenia appears to be promising. There has also been an attempt to consider trauma exposure as a ubiquitous transdiagnostic risk factor for multiple negative outcomes.
In the spirit of Garmezy’s work with families with high-risk individuals who do not manifest the disorder, researchers (Hess et al., Reference Hess, Mattheisen, Greenwood, Tsuang, Edenberg, Holmans, Faraone and Glatt2024; Hess et al., Reference Hess, Tylee, Mattheisen, Adolfsson, Agartz, Agerbo, Albus, Alexander, Amin, Andreassen, Arranz, Bacanu, Bakker, Band, Barroso, Begemann, Bellenguez, Belliveau and Bender2021) developed a framework to hoard GWAS data and then excavate them for common genetic variants that protect high-risk individuals from schizophrenia; this work resulted in the derivation of the first-ever “polygenic resilience score” for schizophrenia. This reinforces the assumption that common variants that are not in linkage disequilibrium with known schizophrenia risk alleles might exert a protective effect. If so, this work can turn the table for genetic researchers who, instead of searching for and investigating risk alleles, will do so for protective alleles (Hess et al., Reference Hess, Tylee, Mattheisen, Adolfsson, Agartz, Agerbo, Albus, Alexander, Amin, Andreassen, Arranz, Bacanu, Bakker, Band, Barroso, Begemann, Bellenguez, Belliveau and Bender2021). This, in turn, can inspire new approaches to intervention.
Importantly, there are additional indicators of engagement at the molecular level, specifically through the human methylome, which reflects the dynamic response of the genome to the environment. Recent research has pinpointed genetic impacts on DNA methylation, shedding light on the regulatory processes underlying gene expression and disease risk. Importantly, approximately 34.2% of CpGs, the foundational unit of the methylome, are affected by SNPs. These genetic variants act either directly (cis-acting) or within 1 megabase of the tested CpG (Villicaña et al., Reference Villicaña, Castillo-Fernandez, Hannon, Christiansen, Tsai, Maddock, Kuh, Suderman, Power, Relton, Ploubidis, Wong, Hardy, Goodman, Ong and Bell2023). Importantly, it has been demonstrated that it is possible to construct a risk resilience score based on epigenetic markers (Magwai et al., Reference Magwai, Shangase, Oginga, Chiliza, Mpofana and Xulu2021). Thus, specific CpG sites exhibited significant correlations with resilience and were predominantly enriched in genes pivotal to neural plasticity, stress response, and immune function. Alterations in the methylation patterns of these genes potentially impact an individual’s coping mechanisms in the face of stressors. Moreover, the constructed methylation risk resilience score demonstrated efficacy in distinguishing between low- and high-resilience individuals, indicating that methylation signatures can be used for such differentiation (Lu et al., Reference Lu, Hsieh, Yang, Wang and Lin2023). Interestingly, the final model included three methylation probes (cg18565204, cg17682313, and cg07167608) in the genes (AARS, FBXW7, and LINC01107, respectively) that have not been flagged in any other study as candidate genes for resilience before. Yet, similar to the research on structural variance described above, there are candidate genes featured in numerous epigenetic studies. DNA methylation of the MHC, DNMT3A, DNMT3B, NR3C1, and FKBP5 genes has been reported to be significantly associated with posttraumatic stress disorder and resilience (Mehta et al., Reference Mehta, Miller, Bruenig, David and Shakespeare-Finch2020; Miller et al., Reference Miller, Shakespeare-Finch, Bruenig and Mehta2020). It has also been hypothesized that epigenetic mechanisms that substantiate the etiology of anxiety disorders and, possibly, conference to resilience, can be either shared or overlapping (Schiele & Domschke, Reference Schiele and Domschke2018). Nonetheless, unlike the case with research into the structural variation associated with resilience that can be investigated with any source of DNA, there is a serious concern regarding the source of DNA for epigenetic studies. To explain, as specialized cells respond to physiological and environmental stimuli differently, the modulation experienced and exerted by neurons can be specific to brain function (Moore et al., Reference Moore, Le and Fan2013) and, thus, not generalizable to other cell types (e.g., blood and saliva). Although there is a correlation between methylation profiles of different cell types, its value is far from one (Braun et al., Reference Braun, Han, Hing, Nagahama, Gaul, Heinzman, Grossbach, Close, Dlouhy, Howard, Kawasaki, Potash and Shinozaki2019; Chen et al., Reference Chen, Meng, Pei, Zheng and Leng2017; Magwai et al., Reference Magwai, Shangase, Oginga, Chiliza, Mpofana and Xulu2021; Thompson et al., Reference Thompson, Sharfi, Lee, Yrigollen, Naumova and Grigorenko2013). The degree to which altered methylation in the peripheral blood or saliva may reflect biomarkers of resilience is still an open question.
Finally, although still preliminary, the extant research provides the foundation for further exploration of such targets as neuropeptide Y (see below), glutamate and gamma-aminobutyric-acid (as mentioned above), and a class of potassium channels family Q (KCNQ channels), which are responsible for the muscarinic currents in neurons (Tan et al., Reference Tan, Costi, Morris, Van Dam, Kautz, Whitton, Friedman, Collins, Ahle and Chadha2020).
Thus, given the profile of the results so far, it seems plausible that there are specific sources of individual differences in the genome that substantiate, either through structural or through functional variation, individual differences in resilience. Yet, the pattern results for the genome are even more “quilt-ish” than it is for the brain. Thus, the ensemble of candidate genes, although reasonable in theory, does not get consistently implicated in practice, questioning the robustness of individual results. The completed GWAS do not engage the hypothesized candidate genes and do not replicate each other’s findings. The only gene that has been independently implicated in genetic and epigenetic studies of resilience is FKBP5, but, as discussed above, its encoded protein lacks brain or behavior specificity as it has a generic role in immunoregulation and basic cellular processes involving protein folding and trafficking. To conclude, in general, more data are needed to clarify the current footprint of the involvement of the genome in the emergence and manifestation of resilience. As a practical consideration, it is recommended to sample wholistically, both through the variation in the genome and methylome, to generate the needed unbiased data while readiness to perform is documented under different stressogenic situations.
Resilience in the body
The large body of literature associating various physiological indicators of bodily functioning and circulating biomarkers (e.g., hormones, neuropeptides, neurotransmitters) with resilience vs. vulnerability to psychological distress has already been discussed in a number of reviews of different types (Charney, Reference Charney2004; McEwen, Reference McEwen2016; Osório et al., Reference Osório, Probert, Jones, Young and Robbins2017; Watanabe & Takeda, Reference Watanabe and Takeda2022); the review of this literature is outside of the range of the present essay. Here, only selected indicators and biomarkers are mentioned in the context of the discussion above.
Peripheral biomarkers
There are numerous physiological indicators that are used to understand how individuals respond to stress and gauge their resilience (Chen et al., Reference Chen, Kelly, Sengupta, Heydendael, Nicholas, Beltrami, Luz, Peixoto, Abel and Bhatnagar2015; Daskalakis et al., Reference Daskalakis, Cohen, Nievergelt, Baker, Buxbaum, Russo and Yehuda2016; Palmfeldt et al., Reference Palmfeldt, Henningsen, Eriksen, Müller and Wiborg2016; Walker et al., Reference Walker, Pfingst, Carnevali, Sgoifo and Nalivaiko2017). The main indicators are autonomic measures, like heart rate variability (HRV), which reflects the body’s stress adaptation by showing how the vagal control of heart rate changes in response to environmental changes. Vagally mediated HRV is employed as an index used to evaluate the extent of top-down appraisals, mediated by cortical-subcortical pathways, shape brainstem activity and autonomic responses in the periphery of the organism (Gillie & Thayer, Reference Gillie and Thayer2014; Thayer et al., Reference Thayer, Åhs, Fredrikson, Sollers and Wager2012). Higher scores on trait resilience psychometric scales have been observed in individuals with high vagally mediated HRV at rest (Souza et al., Reference Souza, Magalhães, Da Cruz, Mendonça-De-Souza, Duarte, Fischer, Souza, Coutinho, Vila, Gleiser, Figueira and Volchan2013). Conversely, chronic reductions in vagal activity, as indicated by HRV, have been consistently associated with psychopathology (Clamor et al., Reference Clamor, Lincoln, Thayer and Koenig2016; Gillie & Thayer, Reference Gillie and Thayer2014). It is worth noting that HRV has been connected to individual differences in brain morphology, particularly ACC (Carnevali et al., Reference Carnevali, Koenig, Sgoifo and Ottaviani2018). Moreover, there have been reports on the associations between regional brain morphometric characteristics, specifically cortical thickness, and resting state vagally mediated HRV (Winkelmann et al., Reference Winkelmann, Thayer, Pohlack, Nees, Grimm and Flor2017). Some studies have used the dexamethasone suppression test (DST), where the DST suppression rate indicated stress resilience (Ma et al., Reference Ma, Chang, Chi, Tsai, Yang and Chen2016). The DST serves to assess the activation of the HPA axis, an integral system in physiological arousal and physiological stress (Fink, Reference Fink2017). However, there is a lack of compelling evidence that supports the DST as a distinct measure of resilience rather than simply physiological arousal (O’Donohue et al., Reference O’Donohue, Mesagno and O’Brien2021).
The discussion regarding the correlation between these biomarkers and stress resilience is limited, particularly in studies that incorporate multiple biomarkers without a defined total stress resilience score or an examination of the individual relationship between each marker and resilience (Carlson et al., Reference Carlson, Dikecligil, Greenberg and Mujica-Parodi2012; Hoge et al., Reference Hoge, Bui, Palitz, Schwarz, Owens, Johnston, Pollack and Simon2018; Schneider et al., Reference Schneider, Lyons and Khazon2013; Smeets, Reference Smeets2010). Importantly, many psychobiological factors interact to promote resilience (Feder et al., Reference Feder, Nestler and Charney2009). Because of this interaction, it is unclear whether a few of these indicators, analyzed separately in response to one stressor, can cumulatively quantify resilience. Similarly, it is unclear whether, when sampled across multiple contexts, situations, and tasks, they provide a convergent indicator of resilience.
Circulating biomarkers
Circulating biomarkers (CB) are biomarkers that circulate cell-free in plasma/serum and include nucleic acids, extracellular vesicles, proteins, and metabolites. CB may provide early indicators of maladaptive responses to external and internal stressors and can be used to monitor (neuro)physiological status ongoingly. Specifically, there are four types of CB, which are essential in understanding the dynamic response to stress and resilience. Recent reviews have provided a comprehensive account of the role of CB in resilience (Beckner et al., Reference Beckner, Main, Tait, Martin, Conkright and Nindl2022; Charney, Reference Charney2004; O’Donohue et al., Reference O’Donohue, Mesagno and O’Brien2021); here, the main points of these reviews are summarized and expanded to include the most recent data.
Neuroendocrine biomarkers are key to both stress adaptation – cortisol, epinephrine, and norepinephrine – and to countering stress-induced effects of the HPA axis and regulation of synaptic plasticity – neuropeptide Y and BDNF. These markers can be sampled through a variety of cell types. For example, cortisol can be sampled through blood, saliva, urine, and hair, among other cells. It has been demonstrated that serum cortisol concentrations can increase by more than 250% under severe stress, both physical (Morgan, Wang, Southwick, et al., Reference Morgan, Wang, Southwick, Rasmusson, Hazlett, Hauger and Charney2000) and psychological (Morgan, Wang, Mason, et al., Reference Morgan, Wang, Mason, Southwick, Fox, Hazlett, Charney and Greenfield2000) duress. Higher cortisol concentration while under stress was associated with poorer cognitive performance (Lieberman et al., Reference Lieberman, Bathalon, Falco, Kramer, Morgan and Niro2005). Similarly, higher cortisol concentration at baseline was a significant predictor of dropout from short survival training (Vaara et al., Reference Vaara, Eränen, Ojanen, Pihlainen, Nykänen, Kallinen, Heikkinen and Kyröläinen2020). However, higher baseline cortisol was predictive of successful selection for performance in a long, high-demand selection course and positively correlated with self-reported grit and resilience (Farina et al., Reference Farina, Thompson, Knapik, Pasiakos, McClung and Lieberman2019). Salivary cortisol also has been shown to be a useful biomarker of resilience (Nishimi et al., Reference Nishimi, Koenen, Coull, Segerstrom, Austin and Kubzansky2022). It is important to note that both blood and saliva cortisol are marked by significant variability between and within people (Hruschka et al., Reference Hruschka, Kohrt and Worthman2005), across and within studies (Kudielka et al., Reference Kudielka, Hellhammer and Wüst2009), and across time (Kudielka et al., Reference Kudielka, Hellhammer and Wüst2009) limiting the interpretability and generalizability of results (Hayes et al., Reference Hayes, Sculthorpe, Cunniffe and Grace2016). This massive variability potentially limits the utilization of cortisol both cross-sectionally and longitudinally. Importantly, research has reported the modulating role of physical fitness, where individuals with higher, compared to lower, fitness exhibited lower norepinephrine and higher neuropeptide Y 24-hr post-stress (Szivak et al., Reference Szivak, Lee, Saenz, Flanagan, Focht, Volek, Maresh and Kraemer2018). In addition, stress has been reported to exert a negative impact on circulating BDNF, decreasing its amount (Beckner et al., Reference Beckner, Conkright, Eagle, Martin, Sinnott, LaGoy, Proessl, Lovalekar, Jabloner, Roma, Basner, Ferrarelli, Germain, Flanagan, Connaboy and Nindl2021; Henning et al., Reference Henning, Scofield, Spiering, Staab, Matheny, Smith, Bhasin and Nindl2014; Suzuki et al., Reference Suzuki, Tokuno, Nibuya, Ishida, Yamamoto, Mukai, Mitani, Tsumatori, Scott and Shimizu2014) and dampening cognitive performance (Beckner et al., Reference Beckner, Conkright, Eagle, Martin, Sinnott, LaGoy, Proessl, Lovalekar, Jabloner, Roma, Basner, Ferrarelli, Germain, Flanagan, Connaboy and Nindl2021; Gepner et al., Reference Gepner, Hoffman, Hoffman, Zelicha, Cohen and Ostfeld2018).
Inflammatory cytokines are soluble protein messenger molecules secreted by immune cells, adipose tissue, and a number of other organs. Pro-inflammatory cytokines (interleukin 6 – IL-6, IL-1β, and tumor necrosis factor TNF-α) trigger or heighten inflammation by relaying messages coordinating an immune response. Anti-inflammatory cytokines (IL-4 and IL-10) stop or lessen inflammation by relaying messages that prevent an excessive immune response that can lead to tissue damage. Previous research has indicated that prolonged exercise, inadequate training recovery, or excessive training stress can lead to an increase in circulating levels of IL-6 and TNF-α (Jürimäe et al., Reference Jürimäe, Mäestu, Jürimäe, Mangus and von Duvillard2011; Main et al., Reference Main, Dawson, Heel, Grove, Landers and Goodman2010). They are commonly observed in conjunction with exercise-induced muscle damage (Smith, Reference Smith2000) and have a detrimental effect on mood state (Booth et al., Reference Booth, Probert, Forbes-Ewan and Coad2006). As such, modifications in circulating inflammatory levels could potentially offer a means of tracking physiological and psychological pressure and indirectly evaluating physiological resilience. Moreover, the profile of the inflammatory response, as reviewed recently, at least at this point, is inconsistent (Chester et al., Reference Chester, Edwards, Crowe and Quirk2013; Li et al., Reference Li, Wilder-Smith, Kan, Lu, Cao and Wong2014) and indicates the sensitivity of the response not only to external stressogenic characteristics but also to individual characteristics. This inconsistency calls for more research (Beckner et al., Reference Beckner, Main, Tait, Martin, Conkright and Nindl2022).
Furthermore, as summarized by Beckner et al. (Reference Beckner, Main, Tait, Martin, Conkright and Nindl2022), stress influences the biological activity of hormones. Importantly, it has been observed that both strenuous physical effort and caloric deficit can lead to alterations in IGF-I binding proteins and sex-hormone binding globulin. These proteins are essential to regulate the bioavailability of IGF-I and testosterone, respectively (Hamarsland et al., Reference Hamarsland, Paulsen, Solberg, Slaathaug and Raastad2018; Henning et al., Reference Henning, Scofield, Spiering, Staab, Matheny, Smith, Bhasin and Nindl2014). A 70% decline in testosterone concentration coupled with a 46% increase in sex-hormone binding globulin was observed during strenuous training (Henning et al., Reference Henning, Scofield, Spiering, Staab, Matheny, Smith, Bhasin and Nindl2014). Dehydroepiandrosterone (DHEA) is an endogenous hormone and a precursor to testosterone. DHEA’s role is to modulate the adverse effects of elevated cortisol, thereby providing beneficial behavioral and neurotrophic effects (Morgan et al., Reference Morgan, Rasmusson, Pietrzak, Coric and Southwick2009; Morgan et al., Reference Morgan, Southwick, Hazlett, Rasmusson, Hoyt, Zimolo and Charney2004; Taylor et al., Reference Taylor, Sausen, Potterat, Mujica-Parodi, Reis, Markham, Padilla and Taylor2007). The adrenal cortex secretes DHEA, which can be converted into dehydroepiandrosterone sulfate (DHEA-S) by sulfotransferase in the adrenals, liver, and small intestine, which accounts for the majority of DHEA in circulation as a result of its longer biological half-life (15–30 mins vs. 7–10 h, respectively) (Morgan et al., Reference Morgan, Rasmusson, Pietrzak, Coric and Southwick2009). These hormones are commonly known as DHEA(s) collectively, unless otherwise specified, due to their ability to produce similar physiological effects (Morgan et al., Reference Morgan, Rasmusson, Pietrzak, Coric and Southwick2009). Given that DHEA-S can counter some of the catabolic effects of cortisol, examining the ratio of these two hormones rather than absolute abundance has been used as an assessment of hormonal imbalance or vulnerability to stress (Wu et al., Reference Wu, Feder, Cohen, Kim, Calderon, Charney and Mathé2013). It has been demonstrated that baseline DHEA concentration was a predictor of human performance (Morgan et al., Reference Morgan, Rasmusson, Pietrzak, Coric and Southwick2009). Additionally, Morgan et al. (Reference Morgan, Southwick, Hazlett, Rasmusson, Hoyt, Zimolo and Charney2004) reported a substantial elevation in DHEA and DHEA-S concentrations from baseline in response to survival training, which remained elevated at 24-h post-training. Trainees exhibiting higher DHEA(S)-salivary cortisol ratios during stress achieved higher performance scores (Morgan et al., Reference Morgan, Rasmusson, Pietrzak, Coric and Southwick2009).
Similarly to the assortment of physiological indicators tied to resilience, there is a multiplicity of circulating biomarkers relevant for measuring resilience. It is crucial to comprehend this multitude and apply appropriate data reduction measures. For example, Handley et al. (Reference Handley, Rogosch, Duprey, Russotti and Cicchetti2023) demonstrated the practicality of using latent profile analysis to capture heterogeneity in diurnal cortisol and diurnal DHEA, potentially as a protective mechanism against cortisol levels (Charney, Reference Charney2004). Additionally, research has demonstrated that a childhood neuroendocrine profile characterized by high diurnal cortisol alongside low diurnal DHEA was specifically linked to improved adaptive functioning during the transition to adulthood (Handley et al., Reference Handley, Duprey, Russotti, Levin and Warmingham2024).
In lieu of conclusion
As exemplified at the opening of this section, the field has not yet converged on the definition of resilience. So, why not offer one more? This proposed definition might open additional opportunities to study resilience, in Cicchetti’s words (2020, p. 7), “as multidimensional spanning psychosocial and neurobiological factors.” To remind the reader where this essay started, resilience is viewed here as an integration over a dynamic sampling of readiness. For clarification, readiness is the state of being physically, cognitively, emotionally, and behaviorally ready to perform optimally during a task. During development, a child encounters various tasks, gains experience in dealing with various situations, and trains to perform in different contexts, albeit environments, events, and circumstances are never wholly predictable. Resilience is then the capacity to exhibit readiness near-continuously across multiple contexts, situations, and tasks by withstanding or quickly recovering from physical and cognitive challenges. As readiness can be sampled through a variety of different indicators for a given task, in a given situation, and in a particular context, observing and assessing readiness throughout development (naturalistically or as repeatedly simulated experimentally) via integration over these repeated samplings can generate an index of resilience, which, in turn, can dynamically fluctuate and impact readiness as dictated by the notion of the adaptive cycle model describing the dynamics of complex systems (Holling, Reference Holling, Clark and Munn1986).
The proposed definition of resilience through readiness allows the establishment of a paradigm for hypothesis generation and data collection and interpretation. The proposed paradigm emerges from the availability of wearable and portable devices that permit registering readiness online, in real-time, across multiple contexts, situations, and tasks. Correspondingly, this review intended to restate, from a slightly different perspective, what Cicchetti and colleagues repeatedly stated (Cicchetti, Reference Cicchetti2013), namely that a multidisciplinary approach integrating genetic, brain-imaging, physiological, and behavioral sampling could offer novel insights into and robust predictions of pathways to resilience to psychological stress in the face of adversity.
First, with regard to the brain, neuroimaging studies indicate that neural markers of readiness and resilience can be obtained with EEG and/or fNIRS. Currently, the number of such studies in the field of resilience is limited (Jauny et al., Reference Jauny, Eustache and Hinault2022; Lawler et al., Reference Lawler, Hruschak, Aho, Liu, Ip, Lajiness-O’Neill, Rosenblum, Muzik and Fitzgerald2021). To illustrate, elements of readiness, such as alertness, can be detected in near real-time through EEG, potentially utilizing a small number of sensors (Jagannathan et al., Reference Jagannathan, Ezquerro-Nassar, Jachs, Pustovaya, Bareham and Bekinschtein2018; Jung et al., Reference Jung, Makeig, Stensmo and Sejnowski1997). Furthermore, EEG researchers have discovered readiness potentials, which encompass changes in EEG data that transpire roughly 2 seconds prior to a voluntary action. Variations in cognitive load give rise to differences in readiness potentials (Baker et al., Reference Baker, Mattingley, Chambers and Cunnington2011), making them a potential neural indicator to monitor a participant’s approach to their peak cognitive performance capacity. Moreover, using fNIRS has facilitated the exploration of neural correlates associated with alertness. Notably, this investigation primarily examined distinctions between task conditions post-task instead of real-time identification of alertness (Herrmann et al., Reference Herrmann, Woidich, Schreppel, Pauli and Fallgatter2008). Studies on resilience using EEG exist, but they are limited in both scope and number (LaGoy et al., Reference LaGoy, Cashmere, Beckner, Eagle, Sinnott, Conkright, Miller, Derrow, Dretsch, Flanagan, Nindl, Connaboy, Germain and Ferrarelli2022; Polusny et al., Reference Polusny, Marquardt, Campbell, Filetti, Noël, Disner, Schaefer, Davenport, Lissek, Noorbaloochi, Sponheim and Erbes2021; Watanabe & Takeda, Reference Watanabe and Takeda2022). Combining EEG and fNIRS in wearable devices, which have or are about to become commercially available, to capture both time and localization of the brain activities, recording in natural and staged experimental situations will generate discrete data sets across which the emergence of resilience can be derived.
Second, with regard to the genome, the indicators of low heritability of resilience, the quilt-like nature of the obtained findings on the role of structural genetic variation as opposed to the evidence that the emergence of resilience is substantiated by epigenetic mechanisms, also necessitates repeated and dynamic acquisition of the genetic data. Such data acquisition is now possible with Oxford Nanopore Technologies, ONT (Lin et al., Reference Lin, Hui and Mao2021), which has revolutionized DNA sequencing with their very compact long-read sequencers, offering access to longer DNA fragments compared to previous generations of sequencers. ONT’s nanopore sequencing technology enhances epigenetic methylation profiling in several ways. Firstly, it allows for the direct detection of modifications, eliminating the need for specialized library preparation steps like bisulfite conversion. Modifications such as 5mC, 5hmC, 6mA, BrdU in DNA, and m6A in RNA can be directly identified at single-nucleotide resolution. Moreover, training base-calling algorithms enable the detection of other natural or synthetic epigenetic modifications. Secondly, nanopore sequencing excels in 5mC detection, offering gold-standard calling with more even genomic coverage, less GC bias, and shorter analysis runtimes compared to traditional bisulfite sequencing. Lastly, the long reads and direct modification detection capabilities of nanopore sequencing enable the characterization of methylation in repeat-rich regions, including large repetitive arrays in the human genome previously unexplored with short-read sequencing. In summary, ONT’s nanopore sequencers empower researchers to delve deeper into epigenetic modifications with unprecedented accuracy, longer reads, and streamlined sample processing. Thus, each of these devices can contribute valuable data separately in understanding the (neuro)physiological bases of readiness and resilience.
Finally, although reviewed in this essay only briefly, multiple studies indicate that peripheral physiological markers of response to stimuli may be predictive of individual resilience. Hence, resilience is closely linked to particular physiological indicators that serve as mediators during periods of stress (Maier et al., Reference Maier, Amat, Baratta, Paul and Watkins2006). As an example, (Tutunji et al., Reference Tutunji, Kogias, Kapteijns, Krentz, Krause, Vassena and Hermans2023) have shown that HRV and other metrics derived from wearable devices, including Empatica E4 and other biosensors, possess the capability to accurately predict long-term stress levels. This discovery underscores the significance of these markers in resilience studies. In addition, a “readiness score” has been established by leveraging the predictive power of physiological markers such as HRV, skin temperature (ST), and accelerometry (Carper et al., Reference Carper, McGowan, Miller, Nelson, Palombi, Romeo, Spigelman and Doryab2020). The significance of these markers in evaluating individual readiness is highlighted by this score, which is calculated based on body stress, sleep quality, and physical activity (Oura Team, 2024). Likewise, the study conducted by Lee and Chun (Reference Lee and Chun2021) found associations between ST and skin conductivity, as measured by the Empatica E4 device, and individual alertness levels, specifically among office workers who were drowsy compared to those who were not. The assessment of cognitive states, particularly in safety-critical scenarios like driving, greatly benefits from the use of physiological markers, such as blood volume pulse, skin conductivity, ST, and respiration. These markers were found to be instrumental in predicting states of alertness, emphasizing their importance (Riani et al., Reference Riani, Papakostas, Kokash, Abouelenien, Burzo and Mihalcea2020). Additionally, they also possess the ability to generalize across various situations, generating a comprehensive resilience index by sampling alertness and readiness.
Such an approach to resilience as a process that integrates readiness to everything, including adversity, across multiple contexts, situations, and tasks appears to be instrumental for a rapid generation of relevant data and now, at least instrumentation-wide, appears to be realistic. In addition to these dynamic studies, it is important to consider and implement longitudinal studies, which are greatly warranted as a possible window into temporal causality so it could be established whether resilience traits arise when the necessary neurobiological foundation is assembled, whether the needed biological system arises to follow the emergence of resilience, or whether these are coupled processes (Holz, Boecker-Schlier, et al., Reference Holz, Boecker-Schlier, Jennen-Steinmetz, Hohm, Buchmann, Blomeyer, Baumeister, Plichta, Esser, Schmidt, Meyer-Lindenberg, Banaschewski, Brandeis and Laucht2018; Laucht et al., Reference Laucht, Esser, Baving, Gerhold, Hoesch, Ihle, Steigleider, Stock, Stoehr, Weindrich and Schmidt2000; Moreno-López et al., Reference Moreno-López, Ioannidis, Askelund, Smith, Schueler and van Harmelen2020; Morgan et al., Reference Morgan, Shaw and Forbes2014).
Finally, citing (again!) Denckla et al. (Reference Denckla, Cicchetti, Kubzansky, Seedat, Teicher, Williams and Koenen2020, p. 7), “Research on resilience is rooted in the field of developmental psychopathology. Scientists adhering to a developmental psychopathology framework emphasize the importance of incorporating multiple levels of analysis into their research. This approach states that different systems contribute to development and that these systems bidirectionally influence each other to contribute to outcomes.” In many ways, the future of research on resilience is tightly connected to the future of developmental psychopathology. Both are relatively young concepts representing relatively young fields (or a unified field) of research. In their acceleration into the future, hand-in-hand, the “ordinary magic” (Masten, Reference Masten2001) of resilience will be, no doubt, better understood but never trivialized.
Author contributions
The author expresses her gratitude to Connor Cheek, Lisa Chinn, and Pavel Dobrynin for their contributions to this essay and to Lauren Elderton for her editorial assistance.
Funding statement
The preparation of this essay was supported by grants from the US National Institutes of Health (P50HD052117, P20HD091005, R01HD109307) and by the Ministry of Science and Higher Education of the Russian Federation (Agreement 075-10-2021-093, Project COG-RND-2105).
Competing interests
No conflict of interest to declare.
