7
Stress in Mind:
A Stress Response Hypothesis of Cognitive Evolution
Staying alive is a tricky business for any organism, no matter how simple. The basic job involves securing matter and energy to keep the show on the road—to keep the system churning out the parts (and networks of parts) that convert matter and energy into chemical processes, which enable the system to secure yet more matter and energy and thus to persist long enough to reproduce. That is certainly not all there is to life, but it is the bottom line (Bertalanffy 1950; Corning and Kline 1998). The trouble is, the world is full of a staggering variety of phenomena, including other beings of diverse kinds, and all sorts of things can happen. Many (probably most) of these things are unpredictable. Some (rather a lot) are perilous if keeping the show on the road is the goal. The requisite matter and energy aren’t always available when needed, or they are present but conditions are otherwise adverse: the waterhole is contaminated, predators or competitors loom. Add to this fire, flood, tsunami, drought, blizzards, gales, cyclones, earthquakes, landslides, volcanic eruptions, extremes of heat and cold, parasites, disease, and conflict between individuals and groups (including war)—to say nothing of asteroid impact—and it is easy to suspect that as many conditions threaten life as sustain it.
In this chapter I will argue that the response to existential challenge provided the scaffold, the basic framework or skeleton, upon which cognition evolved as a biological function, the evolution of which, in turn, facilitated further elaboration of the noncognitive aspects of this very response. Just as the evolution and development of bone and muscle scaffolded one another, so too the response to existential challenge supported the evolution and development of cognition, and the evolving cognitive repertoire supported the further evolution and development of the stress response. Supporting a stress response hypothesis of cognitive evolution is a robust body of emerging evidence concerning the unexpectedly intimate functional linkage of the immune system, the body’s frontline defense against danger, invasion, and damage (Matzinger 1994), and the mammalian repertoire for sensing, evaluating, and acting on stimuli. Hundreds of studies over the past four decades have shown that psychological challenge is “capable of modifying various features of the immune response” (Segerstrom and Miller 2004, 601). What is now also abundantly clear (if still poorly understood) is the critical role the immune system plays in normal cognitive function, an idea considered impossible only twenty-five years ago but a fact now “firmly established” (Yirmiya and Goshen 2011).
The idea that cognition is important for meeting the challenges of a “hostile” world teeming with predators, often deceptive competitors, and parasites is not new (Sterelny 2003). Arguably, all animals (including humans) have “one essential judgment” to make “about nearly all situations, stimuli, and signals: the degree to which they indicate a threat or are safe” (Gilbert 2002, 275). Nevertheless, the strong claim I am making—that the cognitive capacities of an organism function as they do in large part due to a long evolutionary history of responding to predictable and unpredictable stressors in partnership with the immune system—is, as far as I can ascertain, unusual and this for two reasons. First, the stress response hypothesis of cognitive evolution presented here rests on two theoretical pillars, one of which is novel, the other of which is still relatively new. The first pillar is a construct I call the continuum of tolerance, which offers a new way of conceptualizing animal responses based on the tolerability of current conditions within the context of overall organic functioning. The second pillar is the idea that the (“cognitive”) brain-based stress response and the (“noncognitive”) immune system effectively constitute an integrated network for organism defense that shares a common evolutionary origin and a history of continuing coevolution. I call this the immune–brain coevolution thesis.
The second claim to novelty concerns the principal lines of empirical evidence used to support the hypothesis. They include discoveries of immune system involvement in neural plasticity, the most basic processes involved in memory and learning (McAfoose and Baune 2009; Yirmiya and Goshen 2011), and in the close similarities in the mechanisms that enable both cognitive memory and “pain memory” in the spinal cord (Ji et al. 2003). Another line of evidence is the emerging, as-yet-unspecified role of the immune system in the generation of new cells in the adult brain (Carpentier and Palmer 2009), and the sensitivity of neurogenetic processes to psychological and/or physiological stress (Dranovsky et al. 2011).
The concept of scaffold is used here in a biological sense, as a framework for the construction or support of something else, which is conceptually and physically separable from the constructed thing but may remain functionally integral to it. Biological examples include the vertebrate skeleton, and the “cuttlebone” or chitin “pen” of some cephalopods. The scuttling of old ships to create new coral reefs is an artificial case. I believe that the ever-present requirement to respond to predictable and, especially, unpredictable existential challenge drove the evolution of cognition to an initial degree of complexity that would have been difficult, if not impossible, to achieve outside the context of defense, for the simple reason that the cognitive apparatus would have been far too energetically costly to maintain for other purposes. This stands in contrast to standard approaches that emphasize the reproductive imperative or food seeking as the main engines of cognitive evolution (see, e.g., Masse et al. 2009). Put simply, an animal must survive threats to its existence before it can reproduce. Many organisms, from bacteria to mammals, have developed strategies for surviving critical shortages of vital nutrients, such as sporulation, diapause, estivation, and hibernation.
It is important to emphasize that a scaffold is not a ratchet, which pulls in one direction only. The same basic skeleton can give rise to many different types of structure (a dolphin’s flipper, a bat’s wing, a human hand) with radically different capacities and functions (swimming, flying, grasping). So, too, the need for effective defense shaped cognition in different ways in different organisms under different circumstances. Of course, getting “smart” by expanding the behavioral repertoire isn’t the only defensive option. Armor, camouflage, mimicry, and onboard weaponry (e.g., toxins, stingers) are also effective forms of protection. Nevertheless, I claim that the cognitive armamentarium (sensing, memory, valuing, learning, marshaling behavior) is always a critical part of the stress response, even in its most sophisticated forms. Just as the sleek, well-formed muscles of an animal only hint at the nature of the bones that scaffolded their development, and without which they cannot function, so too the defensive origins of cognition are largely obscured by the myriad unrelated tasks of which this most fascinating of biological functions is capable.
The chapter is divided into two parts. Part 1 lays the groundwork, terminologically and theoretically, for the stress response hypothesis of cognitive evolution. Part 2 examines recent empirical data that shows how the noncognitive elements of the stress response and the cognitive repertoire develop and operate codependently.
Part 1: Conceptual and Theoretical Foundations
What We Talk about When We Talk about Stress and Cognition
Although no consensus currently exists regarding its “satisfactory” definition (Segerstrom and Miller 2004), stress, a concept physiologist Hans Selye (1950) borrowed from physics, originally was used to describe any perturbation to which an organism must adapt. On this definition stress is a state of affairs perceived by the organism to impinge upon the state of dynamic balance, known as homeostasis, toward which a normally functioning physiology tends (Cannon 1932). Any state of affairs perceived to challenge homeostasis therefore presents a stress stimulus, or stressor, to an organism (Cannon 1935). Stressors may be states of affairs that are purely physiological, as in the case of tissue damage, or purely cognitive, as in the apprehension of a threat, or some combination of the physiological and psychological (Chrousos and Gold 1992). The dynamic process by which an organism adjusts to changing conditions to maintain homeostatic balance over time is called allostasis (Sterling and Eyer 1988), which also has physiological and psychological dimensions. Allostatic load refers to the cumulative burden of such changes over longer periods. High allostatic load is associated with prolonged stress, and is implicated in many disorders that carry high social and personal costs, including cardiovascular disease, depression, posttraumatic stress disorder, and Type 2 diabetes (McEwen 2000).
Defining stress in a way that captures fine-grained distinctions is difficult, in part because “[e]verything that happens in life is stressful, and everything that you do is coping” (Contrada and Baum 2011, xxi), and in part because what constitutes a stressor depends significantly on the individual as well as on the type of organism. This is as true of bacteria, which are much more individually variable than previously recognized (Storz and Hengge-Aronis 2000), as it is of humans. However, the Cannon/Selye definition is adequate for our purposes precisely because of its generality. It is one thing to claim that the physiological response to homeostatic perturbation, which depends crucially on the detection of internal and external change, scaffolded the evolution and development of cognition, but quite another to specify what the nature of those stresses might have been and how evolution proceeded in response to such stressors. The latter leads to realms of speculation that can only complicate matters.
Similarly, because our target is the evolution of a biological function, not the special capacities of a single mammalian species (i.e., Homo sapiens), we will enlist a definition from comparative psychology that is also fairly general. Sara Shettleworth (1998, 5) defines cognition as “the mechanisms by which animals acquire, process, store, and act on information from the environment. These include perception, learning, memory, and decision making.” Although Shettleworth’s definition can be applied to phyla she may not have had in mind (e.g., bacteria), it is uncontroversial. In this case, cognition is comprised of the total suite of mechanisms that underwrite information acquisition, storage, processing, and use.
To underscore the function of cognition in the biological economy, without which organism defense makes no sense, I extend the definition as follows: Biological cognition is the complex of sensory and other information-processing mechanisms an organism has for becoming familiar with, valuing, and exploiting its environment in order to meet existential goals, the most basic of which are survival, well-being, and reproduction. On this expanded definition cognition necessarily includes—in addition to perception, memory, learning, and decision making—the processes by which stimuli are appraised and valued, traditionally the domain of affect and motivation. While the classical “trilogy of mind” (cognition, emotion, motivation) has provided a useful heuristic for psychological theorizing and research since the mid-nineteenth century (Hilgard 1980), relative to normal biological functioning as currently understood these distinctions are, as Lazarus (1999, 3) observes, “more or less fictions of scientific analysis, whose independence doesn’t truly exist in nature.” However dissociable they may appear to be under experimental and/or pathological conditions, under normal circumstances cognition, emotion, and motivation are mutually interdependent processes.
Moreover, the family of theories of emotion currently dominant holds that affect is significantly the product of appraisal (Scherer et al. 2001), which is generally regarded as the domain of cognition. The importance of appraisal to stress responsiveness in particular cannot be overestimated. A modern clinical guide to human stress responses lists “cognitive appraisal and affective integration” as the second step in a six-step process, in which the stressful event is the first, “neurological triggering mechanisms” the third, and “coping behavior” the last step in the sequence (Everly and Lating 2002, 23).
The Continuum of Tolerance (OK)
To persist, a living system must have some means of ascertaining three general kinds of states, which fall along a continuum. First, the system must be able to determine that its overall functioning in the current circumstances is adequate or tolerable: it is okay (OK). This means that conditions are such that homeostasis can be maintained; no major adjustments need be made. OK is the default setting, the course upon which the system will continue until impinged upon by some countervailing force. Second, the system must be able to recognize countervailing forces when they arise. Foremost, the system must be able to discern that its external circumstances and/or its internal functioning threaten persistence in some way: circumstances are intolerable, or not okay (OK–). They may damage or kill, if not responded to (e.g., flight, fight, freeze, submit, faint, sporulate, estivate, hibernate). Finally, to complete the continuum of tolerance, there must be a means of moving the system from a low baseline of acceptability (e.g., recovery from insult) to a more optimal level of functioning. Conditions perceived as having the potential to enhance functioning are more than okay (OK+). Such conditions are associated with reward, or pleasure. Here I roughly adopt Damasio’s (1999) homeostatic schema of (normal) pleasure and pain in which pleasure (OK+ in my terminology) is the satisfaction of seeking behavior arising from perception of a homeostatic need whereas pain/discomfort (OK– states) signal that something threatens homeostasis and corrective action is needed.
The set point for adequate functioning is always organism specific, not simply species specific. Assessment of a state of affairs as OK, OK–, or OK+ is a function of the current state of the individual, interpreted by the total biological organization of that individual, given the myriad factors contributing to that state. The main advantage of the continuum-of-tolerance approach proposed here, which provides the foundation for a stress response hypothesis of cognitive evolution, is the implicit recognition that what constitutes OK, OK–, and OK+ is significantly shaped by factors specific to an individual, in a particular context. These include genetic, developmental, social, and cognitive factors, in addition to homeostatic parameters and environmental conditions.
Some examples of how the continuum of tolerance works might be useful. Amoebae of the species Dictyostelium discoideum live a more or less solitary life, foraging and reproducing as single cells, so long as necessary nutrients are present. Such default conditions, broadly speaking, are OK (keeping in mind that in nature conditions fluctuate). When nutrients become scarce and starvation looms (OK–), amoebae aggregate in the thousands to form, ultimately, complex three-dimensional structures called fruiting bodies. Cells in the head of the fruiting body transform into long-lived spores (the new OK). When conditions capable of sustaining life (OK+) are present once again, the spores transform into amoebae. Another example is hibernation among certain mammals. During summer and early autumn in the Northern Hemisphere, bears eat plentifully and mate (OK+, which becomes the new OK). As the days shorten with approaching winter, when temperatures plunge and food becomes scarce (OK–), the animals prepare shelter where they will “sleep” through the dire months, slowly metabolizing their fat stores (the new OK). By the arrival of spring, the animal is a fraction of his or her former weight (particularly if she has given birth) and desperately needs food (OK–). The first feeding opportunities thus have high reward value (OK+), as does the arrival of abundant sources of special food (e.g., spawning salmon), until a new default state (OK) develops. Similarly, while much of the experimental literature tends to regard addictive behavior almost entirely in terms of reward or pleasure seeking (OK+), if and when physical or psychological dependence is established as the animal’s default state (OK), absence of the locus of dependence, whatever it may be, becomes a stressor (OK–) requiring remedial action.
Immunohistochemical studies of human fetal tissue provide molecular support for a tripartite continuum of tolerance. The first complex neuromodulators to appear in the developing embryo—acetylcholine, substance P, and methionine-enkephalin (Luo et al. 1990)—could be seen, highly abstractly, as molecular markers for OK, OK–, and OK+, respectively. Acetylcholine is a neurotransmitter found in all of the body’s nervous systems and is associated with “normalizing” various homeostatic processes, including counteracting systemic inflammation (Tracey 2007). Substance P is a potent proinflammatory neuropeptide involved in a stunningly wide variety of stress responses in the central and peripheral nervous systems (Lyon et al. 2011). Met-enkephalin is the body’s natural ligand for the receptor that binds morphine (Hirota et al. 1985). All three strongly regulate one another in brain tissue (Sastry 1995) and are highly conserved across the animal kingdom.
Of these three states of affairs—OK, OK–, and OK+—those determined as OK– are normally the most salient, very likely because they require immediate remedial action. A wide-ranging review by Baumeister and colleagues (2001) found that negatively valenced events have a greater impact on individual affect, cognitive style, and behavior than do positively valenced events in terms of major life events, the outcome of close relationships, patterns of social interaction, interpersonal interactions, and learning processes.
Bad emotions, bad parents, and bad feedback have more impact than good ones, and bad information is processed more thoroughly than good. The self is more motivated to avoid bad self-definitions than to pursue good ones. Bad impressions and stereotypes are quicker to form and more resistant to disconfirmation than good ones.… Hardly any exceptions (indicating greater power of good) can be found (Baumeister et al. 2001, 323).
Integrated patterns of physiological and behavioral response that facilitate an organism’s adaptation to homeostatic challenge are now widely known as stress responses (Sternberg et al. 1992). Stress response activation can be as subtle as the release into the human bloodstream of signaling molecules that herald the beginning of inflammation but which rarely correlate with subjective reports of distress (Segerstrom and Miller 2004). Or they can involve quite dramatic alterations in form and function, involving coordinated changes in expression of hundreds of genes (Lengeler and Postma 1999). A stress response can be viewed as the organism’s pattern of reaction to a perception that something, either within itself or in its surrounding milieu, is OK–. The stimulus may be life threatening or merely perturbing; it may be a lack of or the presence of something, but it is sensed as challenging the organism’s current set point for adequate functioning in such a way that organizational integrity may be threatened, and something must be done, now.
The continuum-of-tolerance approach highlights the need for cognitive processes at a very early stage of evolution. Ascertaining the three types of state requires the capacity to (1) sense existentially relevant features (perception), (2) retain information for a nonzero period and compare what’s happening now with what was happening some time ago (memory), (3) determine the salience of sensory input given the system’s current state (valence/affect), (4) integrate sensory input from multiple sources because (in extant organisms at least) there is always more than one sensory modality, and (5) select between alternative behaviors given conflicting sensory input (action selection).
A third advantage is the theoretical leverage the continuum of tolerance provides on the dissolution of the traditional distinction between the immune system and the brain-based (neuroendocrine) stress response, the focus of the next section. Until relatively recently the brain was regarded as “immune privileged”; the immune system was thought not to operate there (Schwartz and Shechter 2010). Roughly, the neuroendocrine system was thought to respond to challenging “cognitive” stimuli (e.g., a predator) whereas the immune system responded to “noncognitive” threats (pathogens) (Blalock 1989). The continuum-of-tolerance approach assumes that mechanisms for meeting OK– states will be functionally interdependent. From an evolutionary standpoint maintenance of separate subsystems for communicating different kinds of threat (e.g., pathogens, predators) seems energetically profligate; however, the details for addressing the threat (infection, flight) may vary. The genomic age has revealed just how economical evolution tends to be when something works.
The Immune–Brain Coevolution Thesis
In vertebrates the key mediator of adaptation to environmental change in general, and homeostatic challenge in particular, is the hypothalamic–pituitary–adrenal (HPA) axis, sometimes referred to simply as the “stress axis” (Vedder 2008). The core of the neuroendocrine system, the HPA axis is comprised, in broad outline, of the hypothalamus, an almond-sized organ located deep in the brain just above the brainstem; the pituitary gland, a pea-sized organ that protrudes from the bottom of the hypothalamus; and the adrenal glands, which sit atop the kidneys. These three organs are linked via a variety of blood-borne molecules that enable their rapid communication (see in particular Dunn 2008; Vedder 2008).
The main molecular driver of the HPA-based stress system appears to be corticotropin-releasing hormone (CRH), secreted by the hypothalamus and other tissues, including peripheral tissues, in response to stress. CRH signals the pituitary gland to release adrenocorticotropic hormone (ACTH), which in turn stimulates the adrenals to release the classic “stress hormones”: glucocorticosteroids (cortisol and corticosterone), which modulate glucose metabolism and provide fuel for action, and catecholamines (i.e., noradrenaline, adrenaline, dopamine), which modulate the central and sympathetic nervous systems (CNS, SNS) to induce behavior. These hormones and neurotransmitters in turn modulate the brain areas that initiate the cycle, further activating or inhibiting activity depending on the circumstances, stimulating the release of other messengers and thereby “creating a brain-to-body-to-brain reverberating feedback loop” (Yirmiya and Goshen 2011, 182).
In actual fact the HPA axis continually cycles through periods of intrinsic activation and quiescence depending upon evolved, species-specific behavioral patterns and individual-specific realizations of those patterns shaped by genetic endowment, developmental history, and current state. As Vedder (2008) points out, the HPA axis pathway “is not a simple structural and uniform line of functioning in one direction “from hypothalamus to pituitary to the adrenals (21; author’s italics). Rather, it consists of a variety of elements, which include classic immune mediators, such as cytokines (“cell movers”); lipid compounds, such as prostaglandins; neuropeptides, such as substance P and enkephalins; and the cell-surface receptors that bind these and other molecules.
Arguably the most radical revision in thinking about the stress response has been the recognition of the extensive interaction between, and reciprocal comodulation of, the neuroendocrine system and the immune system. The blood–brain barrier was believed to be an unbreachable wall between the brain and the “noncognitive” immune system, presumed to function wholly autonomously (Maier 2003). Although anomalies were well documented—aversive conditioning was shown to increase antibody production in mammals since the 1920s, for example—the prevailing dogma ensured that new discoveries were marginalized or neglected (Ader 2000) until the evidence could no longer be ignored (Blalock 1989).
Key to acceptance of immune–brain communication was the discovery by Edwin Blalock and colleagues that mammalian immune cells secrete peptides and hormones thought to be exclusive to the brain (Blalock and Smith 1980), which hinted that the immune and neuroendocrine systems regulate one another (Blalock 1989). Today three classes of cytokine in particular are strongly associated with modulating the HPA axis: the interleukin (IL)-1 family (especially IL-1β), IL-6, and tumor necrosis factor-alpha (TNF-α) (Dunn 2008). All three are capable of stimulating systemic inflammation (proinflammatory), are associated with several stress-related inflammatory disorders (Webster Marketon and Sternberg 2008), and stimulate, directly or indirectly, each of the major components of the HPA axis (Turnbull and Rivier 1995).
Inspired by Blalock’s pioneering work in mammals and assuming an evolutionary approach, invertebrate biologist Enzo Ottaviani and colleagues looked to mollusks for clues to the historical development of the immune and neuroendocrine “stress” systems. They found that “despite the variety of stressors in different species and taxa,” the stress response in invertebrates “is remarkably similar” to that of mammals, employing fundamentally similar “basic mechanisms and molecules” (Ottaviani and Franceschi 1996, 436)—including CRH, ACTH, glucocorticoids, catecholamines, and the three major proinflammatory cytokines. The fundamental difference between the stress–immune system of invertebrates and those of mammals (and many jawed vertebrates) appears to be that the latter have two systems of immunity (innate and acquired) as well as a neuroendocrine system distributed across numerous tissues and organs. The invertebrate stress response, on the other hand, appears to rely on innate immunity alone, and the entire neuroendocrine–immune apparatus is concentrated in a single immunocyte (Ottaviani and Franceschi 1997). Ottaviani and Franceschi (1996) concluded that, whether the mechanisms are relatively simple (invertebrates) or highly elaborated (mammals), the neuroendocrine stress response, immune response, and all mechanisms of inflammation are best regarded as “an integrated network of adaptive mechanisms” evolved to meet endogenous and exogenous challenges to homeostasis. Neuroscientists Robert Dantzer and Keith Kelley (1989) came to a comparable conclusion based on different evidence.
Meanwhile, Steven F. Maier and Linda R. Watkins were pursuing a similar idea from the perspective of neuropsychology, based in part on “sickness behavior” in fevered animals (Hart 1988). Sickness behavior is a package of symptoms—including lethargy, reduced appetite, increased sleep, and reduced social activity, including mating activity—mediated by cytokines (Dantzer and Kelley 2007). Maier and Watkins were trying to make sense of sickness behavior in animals displaying “learned helplessness,” a conditioning syndrome that manifests in animals exposed to inescapable adversity, such as electric shock (Seligman and Maier 1967). Variations of the learned helplessness paradigm are now among the leading animal models of depression (Yan et al. 2010), a mental disorder strongly associated with inflammation (Dantzer et al. 2008). Sickness behavior, which has since been documented in diverse vertebrates, today is known to be induced by inflammation arising from acute or chronic psychological stress as well as by injury and infection (Maier and Watkins 1998). Impairments in memory, learning, and mood are also established as part of the package.
Maier and Watkins combined the idea of bidirectional immune–neuroendocrine communication, evidence of cytokine involvement in many psychological disorders, and the idea of a tightly integrated, phylogenetically ancient defensive network to arrive at a novel evolutionary hypothesis about immune–brain coevolution (Maier and Watkins 1998; Watkins and Maier 1999; Maier 2003; Maier and Watkins 2003). Drawing on invertebrate evidence as well as neuroscientific and immunological data, Maier and Watkins argued that the “immune” elements of the defensive network probably evolved first, to combat pathogens and promote tissue repair following injury, and were later co-opted by evolving neural mechanisms. When more elaborate behavioral defenses evolved—as the result, for example, of more complex sensorimotor systems—Watkins and Maier plausibly speculate that these new behavioral defenses co-opted existing immune–neural mechanisms. The evidence for this is that “many of the component processes that operate to control infection, inflammation, and injury are either behavioral adjustments or physiological adjustments that are mediated by the central nervous system” (Maier and Watkins 1998, 95). Such adjustments necessarily involve motivational, affective, and cognitive processes.
While Maier and Watkins embrace the idea that, in evolution, it is easier to renovate than to create, such adaptationist thinking obscures the significant degree to which “integrated developmental blocks and pervasive constraints of history and architecture” not only channel evolution but also provide a stable basis for the emergence of novel mechanisms and functions (Gould and Lewontin 1979, 163). William Wimsatt (1999) calls the phenomenon of creative constraint generative entrenchment. Generative entrenchment would explain how successful neuroimmune mechanisms provided a stable (“entrenched”) basis for the generation of novel behavioral processes relating to organism defense. Such a basis, while canalizing the possibilities for evolutionary novelty, does not necessarily limit to organism defense the functioning of the new emergent processes. If successful, the “subsidiary” functions may become traits upon which natural selection can act, even in such elaborate ways that (over time) they may appear more or less independent.
In sum, life’s delicate balancing act requires that the means for ascertaining external threats (cognition, neuroendocrine system) be functionally linked to the organism’s internal defensive repertoire (immunity). Phylogenetic comparisons suggest that these adaptive defense mechanisms operate as an integrated network, are evolutionarily ancient, possibly of a common origin, and are highly conserved.
Part 2: How the Stress Response and Cognition Intersect
Structural/Functional Overlap in the Brain
Although clinicians still distinguish between physiological and psychological causes of stress and disease, it is becoming increasingly clear that the body does not. Stress response activation requires neither tissue damage nor pain nor actual existential threat; cognitive appraisal of existential challenge alone is sufficient. “[H]umans reacting to stressors, which are not life-threatening but are ‘perceived’ as such, mount similar stress/inflammatory responses” (Black and Garbutt 2002, 1). Studies in mice have shown that dangerous intestinal lesions can develop entirely as a result of psychological stress (Zheng et al. 2009). A look at the structural overlap and functional linkage between cognitive processes and the body’s paradigm “stress system” makes clear why this might be the case.
The hypothalamus is where the stress system interacts with brain regions critically involved in cognition and behavior, including the hippocampus, amygdala, nucleus accumbens, and prefrontal cortex. The hypothalamus is densely connected to these and other brain regions involved in cognition. Speaking simplistically, the hippocampus is the brain’s main center for laying down, consolidating, and retrieving memories of stimuli and thus is critical for learning (Wais et al. 2006). The two nuclei of the amygdala, which sit astride the two lobes of the hippocampus, are the main organs for valuing stimuli, particularly for determining whether something is OK– (Feinstein et al. 2011) or OK+ (Waraczynski 2006). The nucleus accumbens is involved in coordinating reward-seeking (OK+) and fear-related (OK–) behavior based on amygdalar input (Gill and Grace 2011). The prefrontal cortex is where current circumstances and memories are integrated to determine the course of action (Bechara 2004). Lesions in or pharmacological blockade of these structures frequently results in problems with memory, valuing, learning, and decision making. Together with the hypothalamus, these brain centers are among the key players in the “emotional nervous system,” which integrates inputs from the body’s exteroceptive and interoceptive senses to provide an ongoing, dynamic picture of “how things are” (Dallman and Hellhammer 2011, 13), which determines what the organism will do next.
To get an idea of just how critical the hypothalamus is to the maintenance of vertebrate life, we need to remind ourselves of life’s bottom line. It’s all about energy: getting it, using it as efficiently as possible, and storing it for future need. The hypothalamus is involved in a wide range of critical physiological processes, both voluntary and involuntary, which regulate the organism’s energetic economy, including food intake and metabolism; the sleep–wake cycle; circadian and ultradian regulation of myriad hormones, neuropeptides and neurotransmitters; cardiovascular, respiratory, renal, and muscular activity; sexual receptivity and related behavior; mood; and social cognition, including territorial aggression (Dallman and Hellhammer 2011).
Noteworthy is the colocalization in the hypothalamus of central coordination of both the HPA-based stress response and the master pacemaker of the organism’s various circadian clocks, the suprachiasmic nucleus. Circadian rhythmicity synchronizes energy intake and expenditure (e.g., sleeping and waking, feeding regime, some social cues) to temporally relevant signals over twenty-four hours (e.g., light–dark cycle, temperature changes) (Cagampang et al. 2011). Circadian alterations thus are cued to regular or predictable changes in the milieu in which the organism evolved and makes its living. By contrast, the stress system is geared to meeting irregular or unpredictable changes, and includes mechanisms for liberating metabolic fuel to enable physiological and behavioral adaptations.
Animal experiments provide “substantial evidence” that the homeostatic set points of the circadian clock and the HPA axis are established early in development, which affect lifelong patterns of energy intake and expenditure, sleep and arousal, and sensitivity to stressors (Cagampang et al. 2011, 215). A study in cichlid fish, for example, found that a single change in early development in food availability—for better or worse—affected learning ability for the rest of the animal’s life (Kotrschal and Taborsky 2010). Fish whose feeding regime was altered in early life (even just once) learned novel objects and tasks faster than those whose feeding regimes remained regular. One interpretation of these results is that early perturbation of environmental quality signals to the developing individual that the world can change unpredictably in ways that affect basic biological needs, which enhances vigilance and alertness to change. Change, as perturbation, is always first a stressor. Greater awareness of novelty enhances memory and learning. This may seem somewhat paradoxical, given the well-known deleterious effects of severe and chronic stress on memory, learning, and decision making (Carpentier and Palmer 2009). However, support comes from a recent mammalian study in which greater reactivity to stress, as against the experience of stress, is correlated with better category learning (Ell et al. 2011).
Proinflammatory Cytokines in Memory and Learning
The idea that proinflammatory cytokines might have a role in normal brain function was first seriously advanced just over a decade ago, when Vitkovic and colleagues (2000) published a prescient review of evidence concerning the presence of IL-1β and TNF-α in the normal adult brain in the absence of infection or other pathology. To the authors the evidence pointed to “constitutive cytokine synthesis by neural cells” (467), which suggested cytokines must perform neural as well as immune functions. Given that 50%–60% of the neurons in the fetal brain die before birth, and that neuronal cell death as well as dendritic arborization and pruning continue throughout life, the suspicion that an immune-like system performed at least a janitorial function in the brain had been long entertained. Microglia proved to be the “immune cells” of the brain and spinal cord; they are ubiquitous, highly ramified throughout the CNS, and phagocytic (Frank et al. 2007). Combining features of both innate and adaptive immune functions, microglia are potent synthesizers of proinflammatory cytokines and antigen-presenting molecules.
As Vitkovic predicted, proinflammatory cytokines are now known to play important roles in the brain under “immunologically unchallenged conditions” (McAfoose and Baune 2009, 355), specifically, in the neural plasticity that makes memory and learning possible and in the generation of new brain cells. Without memory and the simplest kind of learning (e.g., habituation), cognition is impossible, and there is arguably no basis for maintenance of “self,” much less defense against “nonself.” Immunomodulation of neural plasticity thus provides potent support for a stress response hypothesis of cognitive evolution. Given that proinflammatory cytokines are now known to be involved in remodeling tissues as vital as bone, muscle, and fat, as well as in reproductive organs (Yirmiya and Goshen 2011), it makes sense that these molecular actors also play a role in adapting the structure and function of tissues in the body’s central controller, the brain.
Although relatively young, a fairly extensive literature has grown on the association of cytokines with the cellular remodeling processes that underlie memory and learning. For detail, I recommend two excellent recent reviews that approach the material from different vantage points: McAfoose and Baune (2009), who begin with an inventory of processes involved in memory and learning and present a “cytokine model of cognitive function,” while Yirmiya and Goshen (2011) show how cytokines and other immune elements modulate cognitive processes, first in the normal brain and then under pathological conditions.
From invertebrates to mammals, memory and learning involve neural plasticity, the capacity of neurons to adapt to changing conditions by altering their function, structure, and/or chemical profile. Most of the action involved in neural plasticity is believed to take place at the neural synapse, and the genes that influence what takes place there (Bailey et al. 1996). Two different processes are involved: synaptic plasticity and synaptic scaling. Synaptic plasticity can involve increasing the signaling efficiency of the cell (long-term potentiation, or LTP) or reducing it (long-term depression). Synaptic scaling refers to the processes by which the strengths of all synapses on a cell are dynamically adjusted up or down “in response to prolonged changes in the cell’s electrical activity” (Yirmiya and Goshen 2011, 186). It appears to be a kind of homeostatic averaging mechanism that compensates for too much or too little stimulation of some synapses relative to all, so that the cell’s overall weight is not dramatically altered over short periods of time by stimulation that proves relatively transient. Synaptic scaling is believed to be necessary for “maintaining efficient encoding of information across neural networks,” and may be important for memory consolidation (Shepherd et al. 2006, 475).
Proinflammatory cytokines have been implicated in both synaptic plasticity and synaptic scaling in cognitive-behavioral studies using rodents (McAfoose and Baune 2009; see chart on p. 359) although their exact role is still poorly understood. Interesting from an evolutionary standpoint is the recent discovery that long-term memory in fruit flies is also mediated by cytokines through a highly conserved signaling pathway (the Janus kinase and its cognate signal transducer and transcription activator, also known as JAK/STAT), which is found in animals from slime molds to humans (Copf et al. 2011). However, the most dramatic demonstration of the immune system’s critical role in normal cognition involves mice genetically engineered to be immune deficient. These mice do poorly on memory and learning tests, compared to healthy mice. Their cognitive performance improves markedly, however, when their immune competence is restored by bone marrow transplant (Kipnis et al. 2008).
In sum, strong support for the critical role of organism defense in the evolution of cognition is provided by the involvement of immune mediators in processes critical to memory and learning, the foundation upon which all other cognitive functions arguably evolved.
Cognitive Memory and Pain Memory
LTP is a kind of sensitization; the more a neural connection is stimulated, the stronger (more sensitized or active) it becomes. LTP in various parts of the brain (i.e., hippocampus, amygdala, prefrontal cortex) has been the focus of research into cognitive memory virtually since the phenomenon was discovered (Craver 2007). A form of LTP that takes place in the spinal cord, rather than the hippocampus, is believed to underlie chronic widespread pain. Pain is a major feature of the body’s stress response system. Pain induces the animal to protect itself, reduce general activity to enable necessary repair, and learn to avoid the noxious stimulus that caused it in future (Vlaeyen and Linton 2000).
Hyperalgesia is an exaggerated experience of pain relative to the precipitating stimulus and can be induced experimentally by repeated noxious stimulation. This results in long-term changes in the excitability of neurons in the dorsal horn of the spinal cord (Woolf 1983). This phenomenon, called central sensitization, is defined as “enhancement in the function of neurons and circuits in nociceptive pathways caused by increases in membrane excitability and synaptic efficiency as well as to reduced inhibition” (Latremoliere and Woolf 2009, 895). Central sensitization is LTP gone haywire.
Many of the more than 100 different molecules implicated in hippocampal LTP are also involved in central sensitization (Ji et al. 2003). Whether in the spinal cord or the hippocampus, LTP involves the movement of calcium ions across the neuronal membrane at the synapse. From the standpoint of the thesis of this chapter (that the response to stress provided a scaffold for the evolution of cognition) it is significant that both “cognitive” memory and “pain” memory are subserved by similar mechanisms. Given the importance to survival of reacting appropriately to noxious stimuli, Woolf and Salter (2000) propose that development of “the capacity to detect and remember danger” via neuronal plasticity very likely was the result of “a major evolutionary drive” (Woolf and Salter 2000, 1765), whether in cognitive memory or in processes such as central sensitization.
Proinflammatory Cytokines in Neurogenesis
Just as the mammalian brain was once considered an immune-free zone, so it also was thought incapable of regeneration in adulthood. The doctrine of “no new adult neurons,” too, is rapidly becoming an historical curiosity. In humans, two locations are now well established as sites where new brain cells (of all types, not simply neurons) are generated in adulthood: (1) the subgranular zone of the dentate gyrus in the hippocampus, which as we have seen is critical for cognitive memory and learning; and (2) the subventricular zone, which feeds neural stem cells (NSCs) and neural progenitor cells into the olfactory bulb via the rostral migratory stream (Curtis et al. 2011).
Although debate continues about the degree to which new neurons become functional, and the mechanisms by which they are integrated into existing neural networks, Yirmiya and Goshen (2011) provide several lines of evidence indicating that generation of new neurons is important to cognitive function. First, neurogenesis has been shown to increase following certain forms of learning and memory formation. Second, the rate of hippocampus-dependent memory formation is positively correlated with neurogenesis. Third, conditions that improve memory, such as exercise and environmental enrichment, also have been shown to increase neurogenesis. Finally, memory and learning are impaired if neurogenesis is blocked by drugs.
A variety of immune elements have been implicated in neurogenesis both positively and negatively, including the cardinal proinflammatory cytokines, microglia, T-cells, and prostaglandins (Montgomery and Bowers 2012; Yirmiya and Goshen 2011). Much remains to be known, but if a “central dogma” of neurogenesis could be said to exist at this early stage of discovery it is that “acute activation” of proinflammatory pathways, local or systemic, “has a profound negative effect” on neurogenesis (Carpentier and Palmer 2009, 81). Generally speaking, bacterial infection and other means of activating innate immunity depress the generation of new neurons, both in infants and adults. It remains unclear to what extent the generation of NSCs and neural progenitor cells are similarly affected, however. The few new neurons that do grow during inflammatory conditions are morphologically similar but functionally different from those grown under unstressed conditions, displaying accentuated inhibitory or excitatory responses at different stages of maturity (Jakubs et al. 2008).
Like the central stress response, neurogenesis appears to be sensitive to developmental and environmental influences, not only infection. Systemic bacterial infection in early life has been shown to affect neurogenesis in adulthood (Bland et al. 2010). Rats infected in infancy produce substantially more new microglia than new neurons in the hippocampus in adulthood. Another recent study also suggests that hippocampal NSC populations are highly reactive to environmental stress (Dranovsky et al. 2011). Contrary to researchers’ expectations that neurogenesis would cease under psychological stress, NSCs continued to accumulate in the dentate gyrus of socially isolated mice, but they did not become neurons. When the mice were returned to the company of other mice, the NSCs differentiated into neurons. This suggests a type of cellular plasticity in which the brain responds to changing conditions “by shifting the dynamics of both stem cell differentiation and survival” (Dranovsky et al. 2011, 920), rather than altering the dynamics of neurogenesis. Whether this pattern is repeated in all cases of stress, or simply in social stress, is unknown. However, NSCs express a particular class of receptors that play a key role in innate immunity (Carpentier and Palmer 2009) and so may be sensitive to a variety of existential challenges. For our purposes it is enough to show that the brain’s capacity to renew itself in key cognitive centers, however extensive or limited it may prove to be, is positively as well as negatively influenced by the systems involved in the organism’s response to stress.
Conclusion
The aim of this chapter was to show that the response to stress substantially shaped the evolution of cognition as we now know it. The organism’s need to respond appropriately to predictable and unpredictable existential threat ensured that the processes that keep the show on the road and the processes by which the organism knows the world were intimately bound, such that they interact as a complex. In other words, the stress response scaffolded the evolution of cognition.
Supporting a stress response hypothesis of cognitive evolution are compelling correlations between the cognitive and noncognitive processes involved in organism defense in several areas: substantial overlap of brain centers involved in cognition and those involved in the stress response; the immunomodulation of neural plasticity and neurogenesis; similarity of the mechanisms involved in cognitive memory in the brain and pain memory in the spinal cord; and strong indications of an ancient common origin and subsequent coevolution. Correlations, heuristics, and logic tell us nothing about mechanism or causation, however. Nevertheless, strong correlation is the necessary prelude to discovery of mechanism and, finally, development of causal explanation. Further, the scaffold concept provides a useful heuristic for conceptualizing the mutually supporting, coevolutionary relationship between two systems long believed to operate autonomously.
If correct, a stress response hypothesis of cognitive evolution would have potentially disturbing consequences for the epidemic of stress-related affective, cognitive, and physiological disorders that has come to characterize developed modernity. But that is another story.
Acknowledgments
Thanks to John Quintner and Milton Cohen, who gave me a reason to pursue an amateur interest more systematically; to the Australian Research Council, whose postdoctoral fellowship (DP0880559) enabled me to do so; to Linnda Caporael and the other Konrad Lorenz Institute for Evolution and Cognition Research workshop organizers and participants, for providing me a scaffold upon which to evolve these ideas; to Mark Daniel, for giving me the opportunity to pursue the quarry further; and to Richard Bradshaw, always.
References
Ader, R. 2000. On the development of psychoneuroimmunology. European Journal of Pharmacology 405: 167–176.
Bailey, C. H., D. Bartsch, and E. R. Kandel. 1996. Toward a molecular definition of long-term memory storage. Proceedings of the National Academy of Sciences of the United States of America 93:13445–13452.
Baumeister, R. F., E. Bratslavsky, C. Finkenauer, and K. D. Vohs. 2001. Bad is stronger than good. Review of General Psychology 5:323–370.
Bechara, A. 2004. The role of emotion in decision-making: Evidence from neurological patients with orbitofrontal damage. Brain and Cognition 55:30–40.
Bertalanffy, L. v. 1950. The theory of open systems in physics and biology. Science 111:23–29.
Black, P. H., and L. D. Garbutt. 2002. Stress, inflammation and cardiovascular disease. Journal of Psychosomatic Research 52:1–23.
Blalock, J. E. 1989. A molecular basis for bidirectional communication between the immune and neuroendocrine systems. Physiological Reviews 69:1–32.
Blalock, J. E., and E. M. Smith. 1980. Human leukocyte interferon: Structural and biological relatedness to adrenocorticotropic hormone and endorphins. Proceedings of the National Academy of Sciences of the United States of America 77:5972–5974.
Bland, S. T., J. T. Beckley, S. Young, V. Tsang, L. R. Watkins, S. F. Maier, and S. D. Bilbo. 2010. Enduring consequences of early-life infection on glial and neural cell genesis within cognitive regions of the brain. Brain, Behavior, and Immunity 24:329–338.
Cagampang, F. R., K. R. Poore, and M. A. Hanson. 2011. Developmental origins of the metabolic syndrome: Body clocks and stress responses. Brain, Behavior, and Immunity 25:214–220.
Cannon, W. B. 1932. The Wisdom of the Body. New York: Norton.
Cannon, W. B. 1935. Stresses and strains of homeostasis. American Journal of the Medical Sciences 189: 1–14.
Carpentier, P. A., and T. D. Palmer. 2009. Immune influence on adult neural stem cell regulation and function. Neuron 64:79–92.
Chrousos, G. P., and P. W. Gold. 1992. The concepts of stress and stress system disorders: Overview of physical and behavioral homeostasis. Journal of the American Medical Association 267:1244–1252.
Contrada, R. J., and A. Baum. 2011. Preface. In The Handbook of Stress Science: Biology, Psychology, and Health, edited by R. J. Contrada and A. Baum, xxi–xxii. New York: Springer.
Copf, T., V. R. Goguel, A. Lampin-Saint-Amaux, N. Scaplehorn, and T. Preat. 2011. Cytokine signaling through the JAK/STAT pathway is required for long-term memory in Drosophila. Proceedings of the National Academy of Sciences of the United States of America 108:8059–8064.
Corning, P. A., and S. J. Kline. 1998. Thermodynamics, information and life revisited: II. Thermoeconomics and control information. Systems Research and Behavioral Science 15:453–482.
Craver, C. F. 2007. Explaining the Brain: Mechanisms and the Mosaic Unity of Neuroscience. New York: Oxford University Press.
Curtis, M. A., M. Kam, and R. L. M. Faull. 2011. Neurogenesis in humans. European Journal of Neuroscience 33:1170–1174.
Dallman, M. F., and D. Hellhammer. 2011. Regulation of the Hypothalamo–Pituitary–Adrenal Axis, Chronic Stress, and Energy. In The Handbook of Stress Science: Biology, Psychology, and Health, edited by R. J. Contrada and A. Baum, 11–36. New York: Springer.
Damasio, A. R. 1999. The Feeling of What Happens: Body and Emotion in the Making of Consciousness. New York: Harcourt Brace.
Dantzer, R., and K. W. Kelley. 1989. Stress and immunity: An integrated view of relationships between the brain and the immune system. Life Sciences 44:1995–2008.
Dantzer, R., and K. W. Kelley. 2007. Twenty years of research on cytokine-induced sickness behavior. Brain, Behavior, and Immunity 21:153–160.
Dantzer, R., J. C. O’Connor, G. G. Freund, R. W. Johnson, and K. W. Kelley. 2008. From inflammation to sickness and depression: When the immune system subjugates the brain. Nature Reviews Neuroscience 9:46–56.
Dranovsky, A., A. M. Picchini, T. Moadel, A. C. Sisti, A. Yamada, S. Kimura, E. D. Leonardo, and R. Hen. 2011. Experience dictates stem cell fate in the adult hippocampus. Neuron 70:908–923.
Dunn, A. J. 2008. The HPA Axis and the Immune System: A Perspective. In NeuroImmune Biology, vol. 7, edited by A. D. Rey, G. P. Chrousos, and H. O. Besedovsky, 3–15. Amsterdam: Elsevier.
Ell, S. W., B. Cosley, and S. K. McCoy. 2011. When bad stress goes good: Increased threat reactivity predicts improved category learning performance. Psychonomic Bulletin & Review 18:96–102.
Everly, G. S., and J. M. Lating. 2002. A Clinical Guide to the Treatment of the Human Stress Response. New York: Kluwer Academic.
Feinstein, J. S., R. Adolphs, A. Damasio, and D. Tranel. 2011. The human amygdala and the induction and experience of fear. Current Biology 21:34–38.
Frank, M. G., M. V. Baratta, D. B. Sprunger, L. R. Watkins, and S. F. Maier. 2007. Microglia serve as a neuroimmune substrate for stress-induced potentiation of CNS pro-inflammatory cytokine responses. Brain, Behavior, and Immunity 21:47–59.
Gilbert, P. 2002. Evolutionary approaches to psychopathology and cognitive therapy. Journal of Cognitive Psychotherapy 16:263.
Gill, K. M., and A. A. Grace. 2011. Heterogeneous processing of amygdala and hippocampal inputs in the rostral and caudal subregions of the nucleus accumbens. International Journal of Neuropsychopharmacology 14: 1301–1314.
Gould, S. J., and R. C. Lewontin. 1979. The spandrels of San Marco and the Panglossian paradigm: A critique of the adaptationist programme. Proceedings of the Royal Society of London. Series B, Biological Sciences 205:581–598.
Hart, B. L. 1988. Biological basis of the behavior of sick animals. Neuroscience and Biobehavioral Reviews 12:123–137.
Hilgard, E. R. 1980. The trilogy of mind: Cognition, affection, and conation. Journal of the History of the Behavioral Sciences 16:107–117.
Hirota, N., Y. Kuraishi, Y. Hino, Y. Sato, M. Satoh, and H. Takagi. 1985. Met-enkephalin and morphine but not dynorphin inhibit noxious stimuli-induced release of substance P from rabbit dorsal horn. Neuropharmacology 24:567–570.
Jakubs, K., S. Bonde, R. E. Iosif, C. T. Ekdahl, Z. Kokaia, and O. Lindvall. 2008. Inflammation regulates functional integration of neurons born in adult brain. Journal of Neuroscience 28:12477–12488.
Ji, R.-R., T. Kohno, K. A. Moore, and C. J. Woolf. 2003. Central sensitization and LTP: Do pain and memory share similar mechanisms? Trends in Neurosciences 26:696–705.
Kipnis, J., N. C. Derecki, C. Yang, and H. Scrable. 2008. Immunity and cognition: What do age-related dementia, HIV-dementia and “chemo-brain” have in common? Trends in Immunology 29:455–463.
Kotrschal, A., and B. Taborsky. 2010. Environmental change enhances cognitive abilities in fish. PLoS Biology 8 (4): e1000351.
Latremoliere, A., and C. J. Woolf. 2009. Central sensitization: A generator of pain hypersensitivity by central neural plasticity. Journal of Pain 10:895–926.
Lazarus, R. S. 1999. The Cognition–Emotion Debate: A Bit of History. In Handbook of Cognition and Emotion, edited by T. Dalgleish and M. J. Power, 3–19. New York: Wiley.
Lengeler, J. W., and P. W. Postma. 1999. Global Regulatory Networks and Signal Transduction Pathways. In Biology of the Prokaryotes, edited by J. W. Lengeler, G. Drews, and H. G. Schlegel, 491–523. New York: Blackwell Science.
Luo, C. B., D. R. Zheng, Y. L. Guan, W. Z. Shen, Y. G. Liu, and D. T. Yew. 1990. Localization of acetylcholinesterase positive neurons and substance P and enkephalin positive fibers by histochemistry and immunohistochemistry in the sympathetic intermediate zone of the developing human spinal cord. Neuroscience 39 (1): 97–102.
Lyon, P., M. Cohen, and J. Quintner. 2011. An evolutionary stress-response hypothesis for chronic widespread pain (fibromyalgia syndrome). Pain Medicine 12:1167–1178.
Maier, S. F. 2003. Bi-directional immune–brain communication: Implications for understanding stress, pain, and cognition. Brain, Behavior, and Immunity 17:69–85.
Maier, S. F., and L. R. Watkins. 1998. Cytokines for psychologists: Implications of bidirectional immune–brain communication for understanding behavior, mood, and cognition. Psychological Review 105:83–107.
Maier, S. F., and L. R. Watkins. 2003. Immune-to-central nervous system communication and its role in modulating pain and cognition: Implications for cancer and cancer treatment. Brain, Behavior, and Immunity 17 (Supp 1): 125–131.
Masse, N. Y., G. C. Turner, and G. S. X. E. Jefferis. 2009. Olfactory information processing in Drosophila. Current Biology 19:R700–R713.
Matzinger, P. 1994. Tolerance, danger, and the extended family. Annual Review of Immunology 12:991–1045.
McAfoose, J., and B. T. Baune. 2009. Evidence for a cytokine model of cognitive function. Neuroscience and Biobehavioral Reviews 33:355–366.
McEwen, B. S. 2000. Allostasis and allostatic load: Implications for neuropsychopharmacology. Neuropsychopharmacology 22:108–124.
Montgomery, S. L., and W. J. Bowers. 2012. Tumor necrosis factor-alpha and the roles it plays in homeostatic and degenerative processes within the central nervous system. Journal of Neuroimmune Pharmacology 7: 42–59.
Ottaviani, E., and C. Franceschi. 1997. The invertebrate phagocytic immunocyte: Clues to a common evolution of immune and endocrine systems. Immunology Today 18:169–174.
Ottaviani, E., and C. Franceschi. 1996. The neuroimmunology of stress from invertebrates to man. Progress in Neurobiology 48:421–440.
Sastry, B. 1995. Neuropharmacology of nicotine: Effects on the autoregulation of acetylcholine release by substance P and methionine enkephalin in rodent cerebral slices and toxicological implications. Clinical and Experimental Pharmacology & Physiology 22:288–290.
Scherer, K. R., A. Shorr, and T. Johnstone, eds. 2001. Appraisal Processes in Emotion: Theory, Methods, Research. New York: Oxford University Press.
Schwartz, M., and R. Shechter. 2010. Protective autoimmunity functions by intracranial immunosurveillance to support the mind: The missing link between health and disease. Molecular Psychiatry 15:342–354.
Segerstrom, S. C., and G. E. Miller. 2004. Psychological stress and the human immune system: A meta-analytic study of 30 years of inquiry. Psychological Bulletin 130:601–630.
Seligman, M. E., and S. F. Maier. 1967. Failure to escape traumatic shock. Journal of Experimental Psychology 74:1–9.
Selye, H. 1950. The Physiology and Pathology of Exposure to Stress: A Treatise Based on the Concepts of the General-Adaptation-Syndrome and the Diseases of Adaptation. Montreal: Acta Inc.
Shepherd, J. D., G. Rumbaugh, J. Wu, S. Chowdhury, N. Plath, D. Kuhl, R. L. Huganir, and P. F. Worley. 2006. Arc/Arg3.1 mediates homeostatic synaptic scaling of AMPA receptors. Neuron 52:475–484.
Shettleworth, S. J. 1998. Cognition, Evolution and Behavior. New York: Oxford University Press.
Sterelny, K. 2003. Thought in a Hostile World: The Evolution of Human Cognition. Malden, MA: Blackwell.
Sterling, P., and J. Eyer. 1988. Allostasis: A New Paradigm to Explain Arousal Pathology. In Handbook of Life Stress, Cognition and Health, edited by S. Fisher and J. Reason, 629–649. Oxford: Wiley.
Sternberg, E. M., G. P. Chrousos, R. L. Wilder, and P. W. Gold. 1992. The stress response and the regulation of inflammatory disease. Annals of Internal Medicine 117:854–866.
Storz, G., and R. Hengge-Aronis, eds. 2000. Bacterial Stress Responses. Washington, DC: ASM Press.
Tracey, K. J. 2007. Physiology and immunology of the cholinergic antiinflammatory pathway. Journal of Clinical Investigation 117:289–296.
Turnbull, A. V., and C. Rivier. 1995. Regulation of the HPA axis by cytokines. Brain, Behavior, and Immunity 9:253–275.
Vedder, H. 2008. Physiology of the Hypothalamic–Pituitary–Adrenocortical Axis. In NeuroImmune Biology, vol. 7, edited by A. D. Rey, G. P. Chrousos, and H. O. Besedovsky, 17–31. Amsterdam: Elsevier.
Vitkovic, L., J. Bockaert, and C. Jacque. 2000. “Inflammatory” cytokines: Neuromodulators in normal brain? Journal of Neurochemistry 74:457–471.
Vlaeyen, J., and S. Linton. 2000. Fear-avoidance and its consequences in chronic musculoskeletal pain: A state of the art. Pain 85:317–332.
Wais, P. E., J. T. Wixted, R. O. Hopkins, and L. R. Squire. 2006. The hippocampus supports both the recollection and the familiarity components of recognition memory. Neuron 49:459–466.
Waraczynski, M. A. 2006. The central extended amygdala network as a proposed circuit underlying reward valuation. Neuroscience and Biobehavioral Reviews 30:472–496.
Watkins, L. R., and S. F. Maier. 1999. Implications of immune-to-brain communication for sickness and pain. Proceedings of the National Academy of Sciences of the United States of America 96:7710–7713.
Webster Marketon, J. I., and E. M. Sternberg. 2008. Neuroendocrinology of Inflammatory Disorders. In NeuroImmune Biology, vol. 7, edited by A. D. Rey, G. P. Chrousos, and H. O. Besedovsky, 319–348. Amsterdam: Elsevier.
Wimsatt, W. C. 1999. Generativity, Entrenchment, Evolution, and Innateness. In Biology Meets Psychology: Philosophical Essays, edited by V. G. Hardcastle, 139–179. Cambridge, MA: MIT Press.
Woolf, C. J. 1983. Evidence for a central component of post-injury pain hypersensitivity. Nature 306:686–688.
Woolf, C. J., and M. W. Salter. 2000. Neuronal plasticity: Increasing the gain in pain. Science 288:1765–1769.
Yan, H. C., X. Cao, M. Das, X.-H. Zhu, and T.-M. Gao. 2010. Behavioral animal models of depression. Neuroscience Bulletin 26:327–337.
Yirmiya, R., and I. Goshen. 2011. Immune modulation of learning, memory, neural plasticity and neurogenesis. Brain, Behavior, and Immunity 25:181–213.
Zheng, P.-Y., B.-S. Feng, C. Oluwole, S. Struiksma, X. Chen, P. Li, S.-G. Tang, and P.-C. Yang. 2009. Psychological stress induces eosinophils to produce corticotrophin releasing hormone in the intestine. Gut 58: 1473–1479.