A Conversation around the Integration of Sex and Gender When Modeling Aspects of Fear, Anxiety, and PTSD in Animals

Introduction

The prevailing practice of using only male animals in preclinical neuroscience research is about to change. ¹ Due to a mandate from the National Health Institute (NIH) in 2015, researchers are urged to use equal number of females and males in experiments. In this conversational paper, we adopt a feminist perspective and explore the benefits and problems of using female animals, with a focus on animal models of fear, anxiety, and post-traumatic stress disorder (PTSD), a condition known to disproportionately affect women in the western hemisphere (for Europe, see Alonso et al. 2004; Wittchen et al. 2011; for the United States, see Kessler et al. 2012). On one hand, the experimental flexibility and methodological advantages of animal models provide indispensable information on the brain mechanisms that underlie negative emotional states. Thus, to ignore female animals in preclinical research could be to overlook the nuances in female animals’ brains, which could lead to the development of mismatched clinical interventions. On the other hand, to uncritically accept that female animals are better models for women does not come problem-free. Mostly importantly, to cursory eyes, sex differences among male and female animals might be seen as “pure sex differences”; a demonstration of differences that would exist between women and men if culture had not refined them. In this context, we believe that the theory and the practice of using female animals should be subjected to feminist criticism.

First, we posit that a thorough understanding of fear, anxiety, and PTSD requires a multidisciplinary approach in which the contribution of gender-stratified and intersectional cultures in shaping the clinical realities of women and men should not be overlooked. Second, we highlight that translation of sex difference findings in animals requires significant caution. Given that sex differences in nonhuman animal species might reflect species-specific adaptations to their unique ecologies, and might not be shared by humans, we propose that the existence of commonalities between female animals and women should be treated not as a priori, but as an empirical question. Last, we explore the common logical leaps in translating sex differences from animals to humans. We argue that these logical fallacies originate from the axiom that sex is comprised of essential qualities that do not change across history, culture, and even species. Overall, we conclude that the inclusion of female animals does not need to be used as a means to create better models for women per se, but instead as a means to develop a more comprehensive understanding of how biological systems work by treating sex as a variable.

Terminology

Gender is an organizing social structure that refers to how societies manage feminine and masculine identities and that shapes our perceptions and actions around any kind of sex-related category. In this paper, we define gender narrowly as the culturally shaped psychological, behavioral, and physiological characteristics of women and men. We posit that gender does not primarily develop from genes and hormones (conventionally labeled “sex hormones” due to different circulating levels in men and reproduction-age women). Instead, masculine and feminine qualities result from the interwoven contribution of genetic, hormonal, and environmental factors in humans. Since these factors start to shape brain and behavior even before birth, it is impossible to discern their independent contributions to molding brain and behavior, and so we use the term “sex/gender” when we discuss humans. Thus, we use gender to signify cultural shaping and avoid it when we refer to nonhuman animals.

Background

In 1993, in order to reduce disparities in treatment responsiveness for women and men, the NIH enacted policies to include women subjects in NIH-funded clinical research. In 2015, it took a further step by publishing a notice policy entitled “Consideration for Sex as Biological Variable in NIH-Funded Research,” which recommends that the sex of the human participants and animal subjects is considered at all stages of preclinical research: “Women now account for roughly half of all participants in NIH-supported clinical research, which is subject to NIH’s Policy on the Inclusion of Women in Clinical Research. However, more often than not, basic and preclinical biomedical research has focused on male animals and cells. An over-reliance on male animals and cells may obscure understanding of key sex influences on health processes and outcomes” (NIH 2015a). While the notice does not require that researchers investigate sex differences specifically, it recommends: “For studies using both sexes, [researchers should] develop a data analysis plan prospectively that, at a minimum, provides for the collection of data disaggregated by sex (NIH 2015b, 2).

In response to the call to incorporate sex as a biological variable (SABV), important work has analyzed the advantages (Joel et al. 2015; Maney 2016; Shanksy and Woolley 2017; Joel and McCarthy 2017) and disadvantages (Joel et al. 2015; Maney 2016; Eliot and Richardson 2017; Joel and McCarthy 2017) of this policy. Recognizing sex as a meaningful and largely unattended variable in preclinical research is generally applauded. However, many scholars have raised concerns about how the SABV policy might disregard feminist contributions on sex/gender in health and biology. As a matter of fact, feminist scholars have been bringing to light the biases that strengthen the dichotomous perspective of sex/gender by emphasizing differences rather than similarities, especially in neuroscience (Fausto-Sterling 1992; Einstein 2007; Fine et al. 2013; Joel and Tarrasch 2014). Thus, many have expressed concerns that, because of the pre-existing biases embedded in the field, the NIH’s decision might reinforce gender binaries by further systematizing the prevailing emphasis on differences.

Our goal in this paper is to evaluate how SABV policy would improve behavioral neuroscience research and its translational implications through analyzing why female animals are needed. First, we evaluate whether their inclusion is required because they make better models for women; many research agendas that focus on sex differences are justified and promoted because of the different prevalence rates of mental disorders among women and men. Thus, we analyze whether and how female animals could be used as direct models of women due to their ability to embody “the common female biology.” Second, we concentrate on how incorporating female animals can indirectly improve the translational value of animal research by expanding the variables we consider when modeling the organization of biological systems. We also aim to analyze whether the incorporation of female animals might lead to damaging or misleading conclusions. Specifically, we concentrate on how the inclusion of female animals might improve our understanding of a disorder that affects women more than men in the western hemisphere: post-traumatic stress disorder (PTSD). Until the most recent edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-V), PTSD was classified as an anxiety disorder. Neural mechanisms underlying fear and anxiety have been studied extensively by decades of animal research and have been the basis of the current neurobiological models of PTSD. However, to this date, most of our current neurobiological understanding of PTSD and derived pharmacological treatments are based on research conducted with male animals (Beery and Zucker 2011).

Adopting a conversational format, this paper includes the perspective of three scholars. Dr. Annie Duchesne is a human behavioral neuroendocrinologist who conducts research to understand sex/gender differences in stress processes over the lifespan and has a background in animal research. Dr. Nur Zeynep Gungor is an animal electrophysiologist who works on the neural circuits of fear and anxiety and has a background in sociology. Dr. Robyn Bluhm is a philosopher of science and bioethicist with research interests in medicine and psychiatry who has conducted functional neuroimaging research on psychiatric conditions, including PTSD, and has published work critical of sex difference research in neuroimaging.

Roadmap

We break up our conversation into six themes that build upon each other. In theme 1, we explore the causes of the higher PTSD rates in women. Reviewing the current evidence and theories, we emphasize the need to study how the culturally shaped intersectional gender systems might render women prone to PTSD and anxiety disorders. In the remaining themes, we turn to animal models. In theme 2, we step back from the discussion of sex/gender to familiarize readers with the basic concepts and methods of animal research on fear, anxiety, and PTSD. In theme 3, we examine the current evidence on rodent sex differences in fear and anxiety. In theme 4, we consider how to draw parallels between sex in humans and sex in animals, particularly emphasizing the likelihood that species-specific adaptations might overpower the commonalities between species. We argue that to uncritically assume that female animals are better models of women will lead to erroneous conclusions and will misplace efforts to ameliorate women’s health problems. Instead, we discuss ways in which similarities across species in sex/gender can be experimentally explored and theoretically validated. In theme 5, we explore the benefits of using female animals, and focus on two topics: estradiol, a hormone family with high circulating levels in the females of both species, and variability between experimental subjects. In theme 6, we discuss the effects of culturally predominant sex/gender perceptions on scientific thinking.

Theme 1: Higher Prevalence of PTSD in Women: Main Explanatory Factors in the Human Literature, and Their Limitations

Zeynep: What are the core symptoms of PTSD? How is someone diagnosed with it?

Annie: The characterizing symptoms of PTSD have changed over time. Originally termed the “gross stress reaction” to “severe physical demands or extreme emotional stress” in the first edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM I; APA 1952); PTSD was characterized as a transient anxiety disorder in the DSM III (APA 1980). Commonly diagnosed in US veterans returning from the Vietnam War, PTSD was thought to originate in response to “stressors that would evoke significant symptoms of distress in almost everyone” (238), leading to re-experiencing of trauma by recurrent recollection; numbing of responsiveness to external world; hyperalertness; disturbances in sleep, attention, and memory; and avoidant behaviors (First et al. 2010).

PTSD’s diagnostic criteria became more refined in the consequent versions of DSM, and the definition of “trauma” was also revised. Feminist clinicians took a central role in reforming the definition of trauma within the different DSM versions such that it included violence against women and girls (Herman 1997). In DSM III-R (APA 1987), trauma was characterized as an event “outside the range of usual human experience” and was exemplified as threat to one’s or loved ones’ physical integrity; witnessing natural, accidental, or intentional disasters; and seeing other people getting seriously injured or killed (First et al. 2010). In DSM-IV (APA 1994), however, it was accepted that traumatic events, defined as events that would remind humans of their mortality, are quite prevalent. Instead, the traumatic event was required to involve “intense fear, helplessness or horror” (428) in the exposed individual (First et al. 2010). In the current edition of the DSM (DSM-V; APA 2013), the reference to trauma-evoked subjective responses (like intense fear) has been dropped. Instead, the event must lead to the core sets of PTSD symptoms (Maercker et al. 2013): the intrusive (e.g., vivid nightmares, flashbacks, or thoughts), the avoidant (e.g., avoiding situations or activities proximally or distally related to the trauma), the hyperarousal symptom clusters (e.g., frequent startles, nervousness, problems sleeping), and the negative alterations in cognitions and mood (e.g., problems in memory, guilt related to trauma, detachment from others). Furthermore, in the DSM-V, PTSD is under the “Trauma and Stress Related Disorders” category, although a general consensus on PTSD being a fear and anxiety disorder remains.

Robyn: But not everyone who experiences a traumatic event develops PTSD. How common are these events and what is the proportion of people who experience trauma and who have PTSD?

Annie: In developed countries, a large number of people experience at least one traumatic event, most commonly the unexpected death of a loved one, motor vehicle accidents, or robbery, resulting in trauma exposure rates above 70% (Benjet et al. 2016). However, only a fraction of such develop PTSD. Based on the World Health Organization World Mental Health Survey, the lifetime prevalence of PTSD is found to be 5.6% among the trauma-exposed respondents, and half of respondents with PTSD report persistent symptoms (Koenen et al. 2017; diagnosed using DSM-IV criteria). The low rate of PTSD development is attributed both to the nature of the trauma and to pretrauma risk factors (also called vulnerability factors). Trauma type, severity, and intensity were shown to influence PTSD development (Ogle et al. 2014). In addition, multiple socioeconomic factors (lower education, unemployment, lower household income) and demographic factors (unmarried, young, and female) are found to be significant risk factors (Koenen et al. 2017). Indeed, in the United States (Kessler et al. 2012) and Europe (Alonso et al. 2004; Wittchen et al. 2011), women are twice as likely as men to develop PTSD.

Zeynep: The survey-based data collection methods of trauma prevalence studies have been criticized because they cannot capture the nature of the traumatic experience (Dohrenwend 2010). This is a problem, because PTSD development is related to the trauma’s severity. An interesting example involves American war prisoners in World War II. The incidence of PTSD among soldiers who returned from Japanese camps was higher than those who were captive in Germany, because prison conditions in Japan were harsher (Engdahl et al. 1997). This raises the question of whether the higher PTSD rates in women are related to the nature and the severity of trauma, rather than a pre-existing vulnerability. I would guess that sexual traumas can be perceived as more severe because of the stigma and guilt related to them, not to mention probable adverse personal and social consequences.

Annie: Trauma type and severity absolutely explain some of the variance between women and men. For example, compared to men, women are more frequently exposed to sexual trauma such as rape and intimate partner violence and, generally, more to premeditated traumas than to accidental ones (Silove et al. 2017). Among these, sexual traumas are associated with an increased risk of PTSD (Tolin and Foa 2006; Lassemo et al. 2017) and worse symptom severity (Guina et al. 2016) irrespective of gender. In pediatric sexual trauma, the severity of the abuse was related to worse PTSD symptoms (Maikovich et al. 2009). Pediatric sexual trauma is also experienced more by girls (Silove et al. 2017) and leads to high PTSD rates regardless of gender (Tolin and Foa 2006; Maikovich et al. 2009; but also see Walker et al. 2004). Altogether, sexual traumas contribute to girls’ and women’s increased PTSD risk – though note that there is also evidence for increased PTSD risk in women to “nongendered” trauma such as a severe car accident (Ryb et al. 2009).

Zeynep: Then, it is likely that pretrauma conditions also contribute to women’s vulnerability to PTSD. Why are women more vulnerable?

Annie: Cognitive and emotional differences between women and men have been found to contribute to the development of PTSD (reviewed in Olff et al. 2007). For example, research suggests that women tend to evaluate traumatic events as more threatening and uncontrollable; to have greater emotional responses at the time of the trauma, and to be more likely to experience dissociation (which involves limited or distorted awareness). Women are also more likely to adopt avoidant coping strategies and to engage in more counterfactual and ruminative thinking (e.g., imagining alternative outcomes). All of these factors are suggested to increase the detrimental effects of traumatic events (Olff et al. 2007; Pineles et al. 2017). Cyclical variations in ovarian hormones have also been proposed to influence women’s PTSD symptoms: women exposed to a traumatic experience during the menstrual phase characterized by high circulating levels of estrogens and progesterone reported greater PTSD-related symptoms (Bryant et al. 2011). However, there is a clear lack of evidence linking ovarian hormones fluctuations to women’s greater PTSD prevalence compared to men, a risk that, for example, does not decrease with menopause (Ditlevsen and Elklit 2010).

Zeynep: The differences in trauma evaluation or coping strategies can be related to socialization.

Robyn: This is a common theme in feminist criticisms of neuroscience. When we discuss biological propensities and vulnerabilities, it is very easy to assume that these vulnerabilities are innate (and therefore natural and unchangeable). But, obviously, women and men have different experiences of the social world from birth, and these experiences shape their biology. So, for example, speaking of differences in emotion regulation, there is research that indicates that parents talk very differently to their young daughters and sons about emotion (Adams et al. 1995; Fivush et al. 2000), and so it is no wonder that they grow up to have different responses to emotional events.

Annie: Definitely. Gender socialization shapes emotions. Compared with girls, boys are usually encouraged to confront their fears and dissuaded from avoiding fearful situations (reviewed in McLean and Anderson 2009). Importantly, there is evidence that gender role identification in women (i.e., embracing qualities that are stereotypically ascribed to that gender) correlates positively with fearfulness and anxiety (Casey 1993, reviewed in McLean and Anderson 2009).

Furthermore, rarely discussed in the biomedical context of PTSD research is that women and men are not walking in the world with the same rights and privileges. According to the US Census Bureau Population Survey (2008 Annual Social and Economic Supplement, as cited in www.americanprogress.org), women are poorer than men, and more women work in low-paying jobs. When women have the same qualifications as men, they receive less compensation. Furthermore, a quarter of women are single parents and, overall, having and raising children are still primarily women’s responsibilities. These gendered inequalities most likely contribute to women’s higher chronic stress. For example, women and men in the same occupations show similar stress levels during work (assessed by blood pressure and norepinephrine levels). However, after work, women’s stress levels stay high, while men’s decrease (Frankenhaeuser et al. 1989). Extending on those finding, another study demonstrated that women with children at home showed the highest stress levels after work, a difference probably due to the stress of household and parenting chores (reviewed in Lundberg 2005).

Robyn: Societal and cultural factors shape not only women’s biology, but also the differences among women, which can only be explained using an intersectional model of social differences. Crenshaw (1989) first developed the concept of intersectionality to describe problems that arise when race and sex are treated as distinct categories for social analysis. In particular, Crenshaw showed that, at that time, many feminists assumed that white women’s experiences were shared by all women, while antiracist work tended to focus on the experiences of Black men. As a result, Black women’s experiences were distorted or even erased. “Intersectionality” has since expanded to recognize other aspects of social location (e.g., socioeconomic status, age, sexual orientation, ability, etc.).

Annie: Fortunately, some research recognizes that women’s multiple and intersecting social statuses are at play in PTSD’s development and persistence. Recognizing that structural oppressions and opportunities can hinder or facilitate health, Samuels-Dennis et al. (2010a, 2010b, 2011) proposed the intersectionality model of trauma and PTSD (IMT-PTSD). IMT-PTSD differs from traditional models, which focus solely on the psychological and physiological responses of the trauma-exposed individual. Instead, it integrates social factors that increase the risk of violence and limit women’s ability to mobilize resources to manage PTSD symptoms. The authors investigated how the availability of personal and social resources among income-assisted single mothers mediate the relation between cumulative trauma and PTSD symptoms. They found that cumulative trauma increases PTSD symptoms not only directly, but also indirectly through the mediation of reduced personal and social resources (Samuels-Dennis et al. 2010a). In a subsequent study, they demonstrated the direct effects of interpersonal conflicts on PTSD symptoms in single mothers who have been exposed to childhood abuse and intimate partner violence. Importantly, empowerment through social support reduced violence-related PTSD symptoms (Samuels-Dennis et al. 2010b). These important studies, attentive to inequalities in power and privilege, show how intersecting social structures (in this case, economic class, family structure, and gender) situate women’s PTSD (Samuels-Dennis et al. 2011).

Robyn: Our culture is stratified by gender, race, class, and other social categories that create social hierarchies and systems of oppressions that limit women’s health, social capital, and personal resources. Starting from childhood, culture also creates systems of perceptions, cognitions, and human interactions that shape an individual’s life experiences in a gender-based manner, though these effects may be subtler and harder to measure. Thus, a lack of consideration of how the social structures affect biology undermines our understanding of women’s clinical picture. Therefore, research on the causes, prevention, and treatment of mental disorders should extend beyond relying solely on biomedical models and encompass how gendered societal factors contribute to etiology.

In summary, from a feminist perspective, a thorough understanding of mental disorders is bound to be a multidisciplinary effort, one in which women’s high PTSD rates should be investigated with an approach cognizant of our gender-stratified culture. First, women are exposed to more “gendered” violence. Second, women’s vulnerability to developing PTSD after trauma can be shaped by socialization and societal hierarchies. Obviously, as animals do not have our cultures, animal models cannot incorporate the intersectional perspective needed to understand PTSD.

Zeynep: However, animal models present other advantages in understanding emotion and associated mental disorders. Most of what we know about the neural mechanisms of fear and anxiety has been accumulated thanks to animal models. Before we discuss sex differences in animals, in the next section, we will explore animal models of fear, anxiety, and PTSD.

Theme 2: Animal Models: Finding Parallels between Humans and Other Animals on Fear, Anxiety, and PTSD

Robyn: We need to ask ourselves what researchers actually aim to accomplish when they use animal models of PTSD. Many philosophers of science think that the best way to describe the goals of biology research is to say that scientists aim to discover “mechanisms” and explain how they work. These mechanisms are made up of biological entities (e.g., tissues, cells, neural circuits) that engage in specific kinds of activities. The mechanisms as a whole respond in a regular fashion to specific inputs in order to produce particular outputs (e.g., Machamer et al. 2000). For example, islet cells in the pancreas produce insulin via a biological mechanism that science has investigated. And when insulin circulates in the body, it causes other physiological effects by serving as an input to different mechanisms. Similarly, in the case of diseases, such as diabetes, scientists can investigate how the normal functioning of mechanisms is disrupted (e.g., the mechanism that produces insulin does not work properly, or the mechanisms that insulin triggers throughout the body do not respond properly).

Given this understanding of what scientists do, we can think about the assumptions that go into the use of animal models: it seems that researchers have to be assuming that the biological and the neural mechanisms of PTSD are the same in these animals and in humans. But in the case of PTSD, in particular, neither the events that trigger these mechanisms nor the experiences and behaviors that are the result of the triggering events are the same in humans and in animals. So, here, researchers would have to bet that, with different inputs and outputs, the mechanisms are still comparable. Furthermore, they assume that the details of the mechanism’s functioning are similar enough across species that what we learn about neurobiology in an animal model can be extrapolated to human biology.

Zeynep: Animal researchers agree that the basic behavioral responses to danger and associated brain mechanisms are so essential for survival that they must have evolved early in the development of the species and are conserved in humans (LeDoux 2015). Of course, the neural mechanisms of negative emotions in humans are not limited to the conserved mechanisms of fear and anxiety but are also shaped by our higher cognitive and emotional skills. Similarly, in the rodents, large areas of fear-related regions in the brain are dedicated to processing olfactory information, rodents’ most acute sense. Yet, despite species-specific characteristics, both rodents and humans are equipped to scan the environment for threats, evaluate and manifest defensive responses. Obviously, the input conditions for threat detection and the outputs of the response selection are different between rodents and humans, as Robyn states. But still, we deduce that there is enough similarity in the mechanisms that studying rodents will inform us about human biology.

Annie: What kinds of experiments/tests are used to probe the common mechanisms among rodents and humans?

Zeynep: To tap into the similarities in mechanisms, we use simple behavioral tests that are reproducible across species. For instance, classical fear conditioning is used across species with minor modifications. In this paradigm, an initially neutral stimulus, such as a tone, is paired with a noxious stimulus, usually a mild shock. As a result, both animals and humans develop fear responses to the tone, even when it is presented without a shock. Some of the easy-to-measure fear responses are behavioral freezing in rodents and changes in skin conductance (due to sweat) in humans. Later in the training, if the tone is presented repetitively without the mild shock, both humans and animals show a reduction in the fear responses. This phenomenon is called extinction.

We use a couple of anxiety assays with animals. Note that here anxiety is defined as a motivational state when facing diffuse and undefined threats – not existential angst. Anxiety can be drastically different and complex in humans, but we think that, in its most simple definition, anxiety is a negative motivational state aroused by uncertainty. Anxiety-like behaviors can be assessed in rodents with environment-exploration tests (Lister 1990; Steimer 2011). Rodents have a natural tendency to explore new environments, but, at the same time, open and brightly-lit areas (where the predation risks in nature are higher) provoke anxiety. Avoidance of high-risk areas at the expense of exploration is thought to be anxiety-like behavior. Notice that these tests probe “normal” fear learning and anxiety and are not models of mental disorders (although aspects of these tests are used in the disorder models as well).

Robyn: Is there any evidence that similar mechanisms do control fear and anxiety in rodents and humans?

Zeynep: So far, the research has shown that similar brain networks (interconnected brain regions) underlie fear and anxiety in humans and rodents. For example, inactivation and electrophysiological studies in animals, along with imaging studies in humans (mostly using functional magnetic resonance imaging) show that amygdala and the bed nucleus of the stria terminalis are active during negative emotions (Phelps and LeDoux 2005; Gungor and Pare 2016; Shackman and Fox 2016). Similarly, the ventromedial prefrontal cortex is involved in controlling and reducing fear in both humans and animals (reviewed in Milad and Quirk 2012; Milad et al. 2014). For example, hypofunction in this region is observed in humans with anxiety disorders (Milad and Quirk 2012; Milad et al. 2014). Furthermore, lesions in this brain region in animals are associated with persistent fear even when the threat stimuli have become safe during extinction training (Milad and Quirk 2012; Milad et al. 2014).

Annie: What about PTSD models?

Zeynep: Most PTSD models expose the animals to potentially life-threatening situations either by exposure to brief bodily harm (shock or immobilization stress) or psychogenic stressors like predator sight or odor (Goswami et al. 2013; Flandreau and Toth 2017). They follow a simple logic; exposure to a life-threatening event will induce changes in the fear and anxiety circuitry.

Robyn: But basic fear and anxiety behaviors common among mammals can only account for limited aspects of PTSD. PTSD also involves problems with emotion regulation, and in some cases even the inability to identify the emotions you are feeling, as well as problems with body awareness (e.g., Messman-Moore and Bhuptani 2017). There does not seem to be any clear way to develop animal models of these capacities.

Furthermore, even when used to their best advantage, animal models are inherently limited in that they cannot incorporate important social factors that contribute to inter-individual differences between people, including socioeconomic status, race, ability, and the relationships among such variables. We previously discussed the importance of an intersectional analysis of trauma and PTSD. As Annie presents, there are research approaches in humans that try to develop an intersectional framework. But, of course, there is no way to address these issues in animal research.

Zeynep: PTSD is a complex disorder with many manifestations unique to humans. As in any human condition, PTSD is deeply shaped by personal narrative, reasoning, and emotional strategies. These are not the aspects that animal models of PTSD target. Instead, they target the changes in fear learning, rudimentary anxiety, and arousal, as they are known to be impacted by PTSD in humans. Acknowledging the fact that only limited aspects of PTSD are modeled, we use the term “PTSD-like animals” instead of “animals with PTSD.” And there are similarities between the PTSD patients and PTSD-like animals. A meta-analysis confirms that people with anxiety disorders show stronger fear conditioning and extinction deficits (Lissek et al. 2005). Going back to the tone-shock pairing example, first, it is easier for people with anxiety disorders to pair a tone with a shock; and second, when the tone is repetitively presented alone and no longer predicts the shock, it is harder for them to extinguish fear response to the tone. Similarly, PTSD-like animals show enhanced anxiety and fear learning (Rau et al. 2005) as well as extinction deficits (Goswami et al. 2010). Furthermore, other PTSD symptoms like hyperarousal and increased alcohol consumption are also replicated with the animal models (Perusini et al. 2016). Exploring the brain mechanisms responsible for these changes with rodents might illuminate the dysfunctions in humans with PSTD and suggest venues for future medical intervention.

Annie: Animal models have undeniable advantages over human models. The question, then, is how animal and human research complement each other.

Zeynep: When it comes to animal studies, there is always a trade-off: On one hand, they are intrinsically limited in explaining complex human behaviors. On the other hand, they enable a great deal of experimental flexibility. ² Experiments conducted with human subjects are the opposite. Although human studies allow us to observe all behavioral aspects of fear and anxiety, experimental designs to investigate brain mechanisms or to establish causal relations in brain and behavioral processes are limited. For instance, the techniques used to study the human brain have to be noninvasive. Thus, common imaging techniques provide only narrow and correlational information about brain function, and naturally occurring lesions in humans do not provide the control of the experimental interventions that can be done in animals. Likewise, behavioral manipulations like random assignment to certain experimental conditions, which allow us to deduce the causes of observed phenomena, are not acceptable in humans in most cases.

In summary, we rely on the similarities in the brain mechanisms between humans and animals, and so we use rodents in order to overcome the limitation of human studies. We can study evolutionarily preserved mechanisms underlying fear and anxiety in nonhuman animals and even model aspects of some disorders like PTSD. In the case of PTSD models, this gives us great advantages like controlling the timing and intensity of the trauma, so we can study both the pre-existing and the trauma-dependent changes in brain and behavior in depth, and so can identify susceptibility dispositions as well as plausible targets for intervention.

Theme 3: Rodent Literature: Sex Differences in Fear, Anxiety, and PTSD-Like Symptoms

Robyn: Using animal models to understand PTSD requires, first, that humans and rodents share an underlying mechanism; second, that this mechanism can be “activated” in the lab using fear conditioning or other tests; and third, that it functions more or less in the same way in all fear-inducing environments. But, at the same time, the NIH policy on SABV says that whether an animal is female or male might influence how this mechanism functions, so researchers need to test whether sex influences their findings. And if this research might be used to draw conclusions about the mechanisms of PTSD in humans, the NIH policy seems to assume that sex is a straightforward biological variable that influences the mechanism in the same way in both animals and humans. (After all, for reasons we discussed above, researchers cannot actually check to see whether sex influences the mechanism in the same way in humans as in rodents.) But that seems like an odd supposition: surely differences across species (other than strictly reproductive differences) are greater than differences between the sexes.

Annie: I agree with Robyn that there seems to be a paradox in translational research. In order to produce valuable translational knowledge, on the one end, the discipline aims to discover mechanisms that are similar across species. On the other end, it emphasizes the importance of sex differences (or individual differences, more broadly) within species; i.e., how one rodent differs from another. The assumption is that mechanisms regulating sex differences are conserved as well; for example, the X chromosomes among female mammals would operate comparable biological and behavioral outcomes. Thus, taking such differences into account should increase the translational value of any given animal model.

Conserved commonalities (e.g., chromosomal, gonadal, and hormonal) exist among the sexed organisms of many different mammalian species and affect brain and behavior (Joel and McCarthy 2017). It follows that studying effects of sexed components on threat processing across species might have translational value.

Robyn: Is there any evidence that such conserved commonalities actually produce common mechanisms or behaviors among females or males of different species?

Zeynep: Keeping in mind that the data coming from various laboratories using different strains of rodents are not always consistent, generally females seem to be the less anxious sex (reviewed in Donner and Lowry 2013; Kokras and Dalla 2014). Thus, we have an obvious problem from the start. If female rodents do not show the same heightened anxiety as women do, a physiological mechanism showing a particularity in rodent females cannot be proposed as a contributing factor to women’s anxiety vulnerability.

Annie: But the discrepancy between the human and the rodent data might be attributed to a few factors. For one, fear and anxiety behavioral tests we discussed in theme 2 were developed with male animals.

Zeynep: Yes, our ability to measure fear and anxiety behaviors is determined by, first, the animals’ perception of the test conditions and, second, the behavioral measure itself. Sex differences in fear and anxiety might be confounded by the fact that females and males differ in one of these two. Let’s look at behavioral measures first. I mentioned in theme 2 that most anxiety tests probe animals’ avoidance of open spaces and fear expression is measured with freezing levels. There is evidence that activity levels are higher in females (Archer 1975; Valle and Gorzalka 1980). Thus, during an anxiety test, female rodents might simply move more, which may or may not indicate a willingness to explore while overcoming the anxiety that open spaces induce. Likewise, females are much better at learning any task that involves escaping (Beatty 1979). Moving increases their opportunity to find routes to avoid negative events, though some studies also indicate that females are better at suppressing a rewarding movement when presented simultaneously with an aversive event (Archer 1975; but see also Beatty 1979). In the end, it is difficult to know whether the differences reflect real anxiety levels (or learning levels, as in the latter example), or just differences in performance (Shors 2006). Furthermore, typical fear responses in a fear conditioning test also differ (Archer 1975); there is suggestive evidence that while females escape more, males freeze more. Currently, the picture is incomplete – but we are starting to see some encouraging work that investigates escape behavior as a fear response (Gruene et al. 2015).

Second, differences can also arise in animals’ perception of the testing conditions, meaning that females and males can experience the experimental stressors differently. For instance, results of any sex-difference experiment that induces pain are confounded if one sex has higher pain sensitivity (for meta-analysis, see Mogil and Chanda 2005).

Annie: Also, the studies that find female rodents less anxious are conducted in animals that do not present any maladaptive fear processing as in mental disorders. The well-replicated findings about differences between women and men are in regards to PTSD and anxiety disorders.

Robyn: It sounds like we need to keep in mind (at least) three factors: “baseline” fear and anxiety, “healthy” responses to stressors, and pathological responses to stressors. Any of these may or may not reflect the same sex differences found in humans.

Zeynep: Yes, as an alternative to comparing baseline levels of anxiety, some studies have investigated the effects of stress. Consistent with the baseline levels of anxiety – keeping in mind the confounding variables in behavioral tests – maternal separation stress and immune challenges early in life result in higher anxiety-like behaviors in males, even when measured by less problematic measures like ultrasonic vocalizations that are used as distress signals. In contrast, social isolation stress seems to affect adolescent females more (reviewed in Donner and Lowry 2013).

However, in order to draw more accurate parallels to mental disorders in humans, we need to study the sex differences in the subpopulation of rodents that display extreme manifestations of fear and anxiety (Shanksy 2015). For PTSD models, this means that we need studies that compare females and males only among the animals that have developed PTSD-like symptoms (Shanksy 2015). However, we need to be careful: anxiety disorders or PTSD symptoms disturb normal flow of life and have debilitating effects. Even “extreme manifestations of fear and anxiety” in rodents probably falls short of that characterization.

Annie: As well, would you agree that another possible clinical benefit to the inclusion of females to animal research on fear and PTSD might be related to development of tailored therapeutic approaches? Can we not study the sex components common to females of both species, even when we do not see behavioral commonalities?

Zeynep: This is an interesting perspective. As we will discuss in part 5, there can be commonalities in physiology among females of both species that impact on fear and anxiety processing in similar ways. If there are enough experimentally backed-up parallels, studying these parallels will surely help to develop a better understanding of the nuances in female biologies and could open the door for tailored therapeutic avenues in women and men. The existence of differences in the responses to threat between female and male rodents does not necessarily imply that these are at source of women’s vulnerability, but instead that they might be viewed as opportunities through which treatments could be developed to relieve disorder symptoms. This is an important distinction, where we can see a complementary potential of combining animal and human research.

Annie: What would be the next steps for animal researchers?

Zeynep: We need to be aware of the differences in performance, such as activity levels or preferred fear responses, which do not necessarily reflect the differences in fear and anxiety levels. Having said that, I do not believe any of these differences requires a major restructuring of the current behavioral paradigms, but little tweaks. We need to remember that sex very rarely creates categorical differences. Instead, the distribution of values that measure most traits shows significant variability and the overlaps between the sexes (Beatty 1979; Rippon et al. 2014; Maney 2016), even for behaviors that are thought to be characteristically sex-specific such as mounting (behavioral response of male rodents during intercourse, Beach 1942; as cited in Fausto-Sterling 2000, 207, 352).

In summary, studying fear and anxiety with both females and males demands attention to measures that might show variation between sexes but are not directly related to emotion. Furthermore, although conserved chromosomal, gonadal, and hormonal commonalities exist between females of different species and have effects on brain and behavior; limited research to this date does not indicate that rodent females exhibit anxiety-proneness. Whether such a parallel exists between female animals and women is still an open research question. However, if female animals do not exhibit the same fear and anxiety behavioral tendencies as women, it is not possible to say that they could be studied as direct models of women to reveal why women are more vulnerable to PTSD.

Theme 4: Thinking about Species-Specific Characteristics and Translational Missteps

Annie: Uncritically assuming that females of various species are going to show similarities in all kinds of behavior leads to great risks of translational missteps. We can define translational missteps as the faulty interpretations of research findings in animals onto a human condition or phenomenon. The findings in animal studies might reflect species-specific conditions, and these findings might not always apply to humans.

Zeynep: I think there are implicit assumptions in fear and anxiety research that lay the groundwork for translational missteps in this field. As Annie said, there are conserved commonalities between females and males of different mammals like sex chromosomes, gonads, and differentiating pattern of circulating gonadal hormones. But do these conserved sex-related commonalities affect fear and anxiety mechanisms in the same manner in all species? Not necessarily. The faulty reasoning leading to translational missteps works in two steps:

1. An argument that women’s susceptibility to anxiety must be evolutionary. That is, an anxious temperament must have enhanced women’s reproductive success (or alternatively, was an inconsequential outcome of other successful survival strategies). This argument is teleological; existence of a present sex/gender difference does not necessitate an evolutionary origin.

2. An argument that sex differences making women more prone to anxiety are preserved in species, thus, are common in rodents and humans. Since females are the primary caregivers in mammalian species, and invest more in the offspring, they developed a more anxious temperament for the better protection of the offspring. But not necessarily. Parenting practices, social hierarchies, and mating preferences differ vastly among mammalian species (Roughgarden 2004). Different ecologies have led to the development of different sex roles. Indeed, although females are the primary caregivers in most mammalian species, biparental, male uniparental, and group parenting strategies exist (Kohl et al. 2017). Furthermore, other stress-inducing events such as mate search and maintaining a standing in social hierarchies can also create anxiety-prone temperaments in mammals. For example, in some species, high stress levels are observed in dominant animals who must constantly assert their dominance by physical means, or dominate in unstable social hierarchies. Alternatively, high stress levels are observed in subordinate animals when resources or social support are scarce and subordination is hereditary, leading to lifelong harassment (Sapolsky 2005).

Robyn: Not only that, but when we appeal to an evolutionary explanation for the similarities in sexual differences across species, we are saying that the evolutionary precursors of human behaviors are the same across species; that is, humans, rodents, and our shared ancestors all faced the same evolutionary pressures. But this is only true at a very abstract level of description. All species need to be reproductively successful, but the strategies different species use may be very different.

Zeynep: Yes, unfortunately, when we think about evolutionary influences shaping brain and behavior, the distinction between conserved mechanisms (qualities that appeared early in the evolution of species and were preserved) and species-specific adaptations (qualities that appeared in a species to adapt to its own ecology) can get blurred. Theoretically, sex differences observed in rodents can arise from an ancient conserved mechanism, a species-specific adaptation, or the animals’ current environment (the rearing and housing conditions in the laboratory). Likewise, sex/gender differences in humans might have similar roots (laboratory conditions would be replaced with culture in this case).

Annie: More generally, the blurring between the conserved mechanisms and species-specific adaptations poses a challenge to the translational value of animal research – not only when studying sex differences, but also when studying fear and anxiety.

Zeynep: Yes, the complexity of the subject matter has often prevented the animal research on fear and anxiety from overstating its conclusions: Our goal in studying the neural basis of fear conditioning or open field exploration is not to create complete models of fear learning or anxiety regulation in humans. Yes, similarities in behavior and brain mechanisms between human and rodents, arising from the evolutionarily conserved characteristics, is the premise for the translational value. But we also know that these commonalities exist in interaction with other mechanisms that enable species-specific characteristics. For example, as we discussed in theme 2, animals and humans are equipped to scan the environment for threats and engage in defensive behaviors to reduce harm. Although all mammals evaluate threat information (e.g., calculate the probability of occurrence, assess controllability), humans have distinct abilities such as cognitive reappraisal (e.g., changing the response to an event by reinterpreting the emotional stimulus). Thus, when we model negative emotions or disorders like PTSD, we are modeling aspects of these motivational or mental states, focusing on the similarities between the species. As a matter of fact, to emphasize the distinctions between the rodent states and the human states, we limit our terminology to “anxiety-like behaviors” or “PTSD-like animals.” Even a proposal to abandon the term “fear” for “threat processing” has been made, especially to prevent the public misperceptions caused by using the same vocabulary (LeDoux 2014).

Robyn: I wonder whether even our experiences of fear and anxiety are different than those that other animals have. We not only experience fear and anxiety, but we think about and try to make sense of those experiences, which in turn shapes our subsequent experiences.

Zeynep: That is an example of reappraisal. Humans’ complex cognitive and emotional abilities interact with the conserved mechanisms of fear and anxiety. Likewise, these abilities should interact with the mechanisms underlying sex-influenced behaviors. Human sex-influenced behaviors (even before the effects of culture) emerge from not only conserved mechanisms, but also species-specific qualities. Human sex is a complex system that interacts with the abstract and elaborate cognitive-emotional systems.

Robyn: I want to go even further and say that we cannot make sense of the idea that the effects of biology (including sex) can be separated from the effects of culture. Conceptually, we understand that biology and culture are different phenomena, but when we think about specific behaviors or capacities – which are the result of both – we cannot tease apart the influence of each.

Zeynep: I agree. Using the cognitive abilities developed in response to the evolutionary pressures, humans created cultures, and in turn cultures induced “environmental pressures” that further shaped the most optimal strategies they adopted in a dynamic manner. This is not a deterministic process, however, because “cache” genetic repertoire can be unrevealed with a change in the environment and lead to development of alternative strategies (Fine et al. 2017).

Regardless, humans have the abilities to abstract, analyze, and reinterpret themselves and their surroundings, as well as the means of socially transmitting their practices and contemplations. Likewise, quite complex mating and parenting strategies were formed as adaptive reactions to physical and social environments. These strategies constitute a part of humans’ sex (and inseparably gender)-influenced behaviors and physiology repertoire (which may or may not overlap with other animals). Then if this is the case, when studying sex with animals we need to consider what aspects of sex-influenced behaviors we are trying to model: What are similar among species and can be considered to reflect a conserved common female physiology? Female rodents cannot be complete “models of women” just as PTSD-animals cannot be complete models of PTSD patients.

In summary, we define the faulty projections of research findings in animals onto a human condition or phenomenon as translational missteps. These occur when we disregard species-specific adaptations that organisms might have developed. All sex-influenced behavior and physiology of an organism are products of an interaction between the common conserved mechanisms and species-specific mechanisms and processes. Thus, to determine whether female animals can directly model an aspect of women’s physiology, we should critically evaluate similarities between species in that domain. If convincing parallels in behavior and physiology are found, sufficient to warrant the assumption that female rodents and women share conserved brain mechanisms, further research targeting these parallels might be helpful in addressing women’s health problems. If no parallels are found, the use of female rodents over male rodents will not have any direct advantages. (We will discuss indirect advantages in theme 5.3)

Annie: Consulting ethological studies in order to find a species in which females share women’s vulnerability to anxiety disorders and PTSD might be a promising possibility – not overlooking the fact that the higher disorder rates in women may be so tightly related to our gender-stratified cultures that this line of research might prove void of any translational value.

Theme 5: Benefits of Incorporating Females into Animal Research

Subtheme 5.1: How can female animals help? The effects of hormones on fear processing as an aspect of sex that shows similarities between female rodents and women.

Annie: Maybe estrogens could be considered an aspect of sex that we can study, since their effects show similarities between species? Studies that focus on the effects of estrogens on female animals’ fear processing have shown some translational parallels with women (Maeng and Milad 2015). Specifically, the positive role of circulating estrogen levels during fear extinction recall, the capacity to learn that a stimulus is no longer paired with a threat, have been observed in rats and in women (reviewed below). In theme 2, we discussed how both PTSD patients and PTSD-like animals suffer from extinction recall deficits. Therefore, the role of circulating estrogens levels could be relevant in understanding PTSD in women, even though they do not help to explain the differences in PTSD prevalence between women and men.

In female rats, one-time administration of estradiol (the most common type of estrogen) increases fear extinction recall (Graham and Milad 2013). Moreover, female rats with high circulating levels of estrogens show higher extinction recall compared to those with low levels (Milad et al. 2009) and males (Chang et al. 2009; Rey et al. 2014). That being said, blocking the production of estrogens in male rats also lead to deficit in fear extinction, which demonstrates that estrogens irrespective of the rat’s sex can influence fear processing (Graham and Milad 2014).

Zeynep: Estrogens are commonly called women’s “sex hormones,” despite the fact that they are involved in various non-sex-related functions in the body and that women and men produce and require estrogens for healthy functioning (Fausto-Sterling 2000). What are the effects of estrogens on fear extinction in men?

Annie: In human studies, the effects of estrogens on fear processing have only been assessed in women. Although it is well known that reproductive age women have higher levels of estrogens, the bias of studying these hormones exclusively in women is overwhelming and, unfortunately, creates an incomplete picture. Thus, the effects of estrogens on men’s cognitive and affective processing are mostly still unknown. However, a series of studies have shown positive effects of higher levels of estrogens on fear extinction recall in women, paralleling the findings in rats. Separating women based on circulating levels of estrogens, Milad et al. (2010) revealed that women with higher estrogens levels showed greater extinction recall compared to women with lower levels; in addition, estrogens levels moderately correlated with better extinction recall (Milad et al. 2010). Furthermore, one-time administration of estradiol increases fear extinction recall in women (Graham and Milad 2013). Finally, compared to trauma-exposed women without PTSD, women with PTSD are reported to have lower estrogens levels and reduced extinction recall (Glover et al. 2012).

While the above-mentioned studies were carefully designed and executed, the findings need to be replicated by other laboratories in order to be conclusive. Currently, some human and rodent studies failed to observe estrogens’ reducing effects on fear responses (Lonsdorf et al. 2015; Lebron-Milad et al. 2012; Hoffman et al. 2010) or even reported opposite effects (Milad et al. 2006; Toufexis et al. 2007).

Zeynep: Is there any evidence to show that women’s greater risk to PTSD is related to hormones?

Annie: No, but this proposition is commonly and misleadingly promoted in the literature (see, e.g., Li and Graham 2017). Instead, the current data seem to indicate that estrogens might explain the variation among women, not variation between women and men. Finally, while the use of female animals can be helpful in studying the effects of hormones, researchers should be aware of the scientific caveats. The research involving hormones is just as vulnerable to translational missteps discussed in theme 4 as any other animal research agenda.

In summary, more research on the positive effects of estrogens levels on fear processing, using both male and female subjects, is required to further ascertain parallels between rodents and humans. Although hormones do not explain women’s greater PTSD risk compared to men, they should be regarded as a fruitful venue in understanding the neurobiology of fear processing in both women and men.

Subtheme 5.2: How can female animals help? Modeling gendered factors.

Annie: Another possible avenue to understand women’s higher PTSD risk could be to step out of the sex difference framework. Instead, we can model gendered factors using the advantages of animal models like increased experimental flexibility and the ability to establish causal relationships. Gendered factors can be defined as the socioenvironmental factors that create the disadvantaged positions of women in society, such as oppression through social hierarchy, the cumulative experiences of trauma when lacking social support, and chronic stress, as we discussed in theme 1. Women’s experiences cannot fully be captured when these variables are translated to the animal realm. Still, we could directly and individually measure whether these gendered risk factors increase vulnerability to PTSD-like symptomatology by altering brains and behaviors of both male and female animals.

Subtheme 5.3: How can female animals help? Increasing variability.

Zeynep: Even if female rodents cannot serve as direct models of women, including female animals in research might increase variability, which would in turn improve the generalizability and translational value of our research models.

Robyn: What kind of variability do you mean? I thought that part of the appeal of animal models is that it makes it possible to standardize experimental protocols, as well as subjects.

Zeynep: That is correct. As a matter of fact, males were the preferred sex to study to avoid possible variance in physiology and behavior due to estrous cycle in females. However, a recent meta-analysis showed that female mice (tested regardless of phase) showed less within group variance than males on a variety of measures, for example pertaining to morphology and metabolism (Prendergast et al. 2014). Interestingly, males showed higher variance even in hormonal measurements. Thus, to incorporate female animals without explicitly testing for estrous cycle should not be a problem in most fields of study, and can be the regular practice in laboratories that do not specialize in studying the effects of estrous cycle.

I argue that we should seek female incorporation as an opportunity to better understand the biological system we are studying, whether that biological system be a species or a brain circuitry. It should be useful to observe all members of that species, regardless of their reproductive abilities. Female rodents can be experimental subjects not as direct models for women but to further our understanding of where and how sex as a biological variable influences and organizes a biological system. This information will not have direct explanatory power in understanding sex/gender differences in humans. Instead, we would learn more about the myriad strategies that evolution has taken to create overlaps and diversity between sexes. Indeed, not all animal work is undertaken to model human physiology directly. We do study various systems in animals even though they do not have immediate resemblance to humans. For example, brain circuitry underlying pheromone processing in rodents is studied not because we want to model a corresponding system in humans (the existence of a pheromone system in humans is questionable; Meredith 2001). Instead, we study these systems because we want to understand how brain networks work as information processors. We want to extract the general principles of how information is coded when animals learn, make decisions, and generate behavior.

As I explained above, for most traits, the distribution of values shows significant variability and great overlap between the sexes (Rippon et al. 2014). For example, above we discussed how in threatening situations males freeze more often and females escape more often. But it is not that males do not escape or females do not freeze. Introducing females into research might give us a more complete fear behavior repertoire by drawing attention to behaviors that are displayed by most females and some males. Whenever we assume a behavior is categorically female or male, we ignore the subpopulations that exhibit behaviors more typical of the other sex.

Robyn: Yes, variability and overlap are important. For example, if recommendations regarding drug dosages are based on sexes’ average responses, a treatment could be ineffective – or even dangerous – for individuals whose responses are more like that of the “average” member of the other sex. But if the results reported in the experiments with female animals just talk about statistically significant sex differences between females and males, then they are not paying attention to variability within the sexes or to overlap between them.

Zeynep: That is a big problem. When we compare females and males, as Maney (2016) advises, “we need to ask how much the sexes differ, not whether or not they do.” Unfortunately, sex is usually articulated in a dichotomous manner and difference-oriented research enhances that. One useful tool to overcome this problem is introduced in the author’s same work, sexdifference.org, which allows you to visualize the overlaps between two groups with mean and standard deviation values with simple and clear graphs.

In summary, the sole goal of animal research is not to directly model aspects of human behavior and physiology. We study biological systems to derive the general principles of how the brain encodes and decodes information. Including sex as a biological variable will help to build a more complete understanding of these principles by incorporating the variability in the functioning of the biological system.

Theme 6: Female Incorporation and the Risk of a Search to Find “Pure Sex Differences”

Annie: The standardization of experiments and subjects that Robyn mentioned might have distorted our view of fear and anxiety systems. On the one hand, paradigms like fear conditioning have the undeniable strength of being highly controlled tests that are employed in a vast number of species. On the other hand, these “clean” translational models come at the cost of being removed from the social and environmental context in which animals actually live, and so they provide a restricted and somewhat distorted view of fear and anxiety systems and associated behavioral repertoires. As Pellman and Kim (2016, 6) point out: “Fear conditioning research likely provides an incomplete picture of the brain’s fear system because it is based on assessing the magnitude of a specific response (e.g., freezing) in a small experimental chamber that restricts the animal’s repertoire of behavior … Ethologically-relevant paradigms provide a greater match to the real-world threat situations that the brain’s fear system evolved to solve, ³ and thus present means to critically evaluate the utility of fear conditioning models” (also see Parsons et al. 2017). Furthermore, non-natural environments might restrict our understanding in a sex-specific manner. For example, our understanding of rats’ sexual behaviors completely changed when females were able to express them in more naturalistic settings, instead of only being subject to a male’s approach in a small experimental box (Mendelson and Gorzalka 1987).

Then, when distancing animals (and humans in a lab) from their natural or ecological referents, researchers have to deal with the fact that the observed phenomenon is the result of the experimental environment interacting with the phenomenon you aim to study (a risk of which every experimental researcher is well aware). The daunting question is: How much does the former impact the latter?

Zeynep: That is an important point. A goal of using standard tests is to reveal the most fundamental aspects of fear and anxiety by attempting to minimize the species-specific conditions and contexts, as well as variability in the laboratory environments. This, as Annie points out, probably gives us an incomplete and maybe even distorted view of the fear and anxiety systems.

Moreover, by seeking the “fundamentals,” we risk falling into the fallacy that we are in a “species-neutral zone” with the laboratory mice as a “template” organism. Unfortunately, thinking of rodents’ brains as template brains can easily – and wrongly – lead to the assumption that the rodent sex differences reflect “pure,” essential differences. It is as if the differences between female and male rodents provide a window to what differences would characterize humans if we were not “distorted” by cultural effects, and we forget that both rodents and humans manifest species-specific adaptations. Scientists and nonscientists alike commonly perceive SABV to reveal the “noncultural and nonsocial biological effects,” even though finding “pure sex differences” is not the goal of the NIH’s new mandate.

Robyn: The idea that rodents represent template brains also suggests that human culture adds extra complexity to the innate template. It does not really surprise me that this idea is implicit in science. It is implicit in our culture as a whole. People think it is common sense that women and men are different, underneath all of the similarities that societies try to impose.

Scientists are part of their culture, and so then researchers interested in sex differences also tend to think that it is common sense that women and men are different. Moreover, I think they also see scientific studies that show sex differences in some characteristic as “showing us the truth,” value-neutral, and objective – unlike criticisms of this research, which are seen as “political,” opinion-dependent, and subjective. But plenty of feminist work shows the way that beliefs about gender and gender differences shape science (e.g., Fausto-Sterling 1992, 2000; Kaiser et al. 2009; Jordan-Young 2010; Bluhm 2014).

Zeynep: Yes, philosophers of science argue that scientists (like all other people) are constrained by the socioeconomic structures of their time and are affected by the dominant patterns of thoughts in their culture. Perceptions, opinions, and ideologies around gender impact scientist’s hypotheses and the way they interpret data (van den Wijngaard 1997; Haraway 1996; Harding 1996). Most feminist schools argue that “essentialism” is the dominant framework that shapes how we think about sex/gender.

Annie: Can we define essentialism?

Zeynep: Essentialism can be defined (in an elementary way) as the belief that women and men have different underlying and unchanging “essences” that shape their bodies, behaviors, and minds. Feminine essence allegedly makes women, for example, more nurturing, cooperative, and emotional (as opposed to rational). According to this scheme, all women share these essential qualities throughout the history and cultures. Essentialists often use the terms “innate” and “hard-wired” to suggest that there are certain biological differences between the sexes that have limited room for variability and, more importantly, change. I would argue that essentialist thought is so embedded in our cultures that it arranges our perceptions, cognitions, and opinions about gender in both overt and subtle ways.

In the same manner, essentialism causes us scientists to extrapolate human sex/gender differences to animals: We expect female rodents to be more anxiety vulnerable, as anxiety vulnerability is part of the “emotional female essence.” Because of essentialism, we think of sex as a fixed phenomenon that emerges in different organisms in the same manner: Thus, the possibility of similarity in the sex differences of rodents and humans becomes an axiom. Essentialism makes us forget that, even without the effects of human culture, human sex differences in fear and anxiety could be drastically different than sex differences of rodent species because every species has evolved to adapt its own ecology. Species exhibit a stunningly diverse array of mating, parenting, foraging, and cooperating strategies (to name a few; Roughgarden 2004), all of which could create sex differences and overlaps in fear and anxiety behaviors and underlying brain mechanisms.

However, essentialism leads us to think that animals will present us with the pure sex differences. We can reveal the absurdity of this presupposition with a simple thought exercise. By studying the conserved fear and anxiety mechanisms in rodents, we do not seek to discover the pure fear and anxiety mechanisms in humans. We do not try to discover how humans would detect threat, feel the negative emotions, and select appropriate responses without the influence of culture. Then, why would studying rodent sex differences help us to uncover the pure sex differences in humans?

In summary, standardization of behavioral paradigms and subjects in animal experiments could misguide the understanding of fear and anxiety systems and create the illusion that we are studying template organisms. In turn, this could create a further illusion that animals give us an opportunity to discover the pure differences between women and men, so-called noncultural and nonsocial but biological effects. The assumption that the same fixed and unchanging sex characteristics define all sex-influenced behaviors across species is a derivation of essentialist thinking, which is pervasive in shaping how we think about gender. While species differences in systems like fear and anxiety seem obvious to most of us, essentialism renders species differences in sex invisible by resorting to a formulation of sex that is unchanging across history, cultures, and even species.

Summary

Since there are numerous gendered factors that are not linked to evolutionarily conserved sexed variables, to assume that rodent sex differences in fear, anxiety, and PTSD models reflect the pure differences and are directly informative about women’s physiology will not help to improve women’s health. Instead, these assumptions might impede it. To uncritically accept that essential differences between women and men render women more prone to anxiety can make us, as researchers and as a society, blind to the gender inequalities and the role they might play in women’s vulnerability to PTSD.

Instead, incorporating female animals can indirectly improve the translational value of animal research by expanding the variables we consider when we model the organization of biological systems. Including sex as a biological variable will help to build a more complete understanding of the general principles of how the brain encodes and decodes information.

Bibliography

Alonso, J., M.C. Angermeyer, S. Bernert, R. Bruffaerts, T.S. Brugha, H. Bryson, et al. 2004. “Prevalence of Mental Disorders in Europe: Results from the European Study of the Epidemiology of Mental Disorders (ESEMeD) Project.” Acta Psychiatr Scand 109, no. 420: 21–7.

American Psychiatric Association. 1952. Diagnostic and Statistical Manual of Mental Disorders. Washington, DC: American Psychiatric Publishing.
–. 1980. Diagnostic and Statistical Manual of Mental Disorders, 3rd ed. Washington, DC: American Psychiatric Publishing.
–. 1987. Diagnostic and Statistical Manual of Mental Disorders, 3rd ed., revised. Washington, DC: American Psychiatric Publishing.
–. 1994. Diagnostic and Statistical Manual of Mental Disorders, 4th ed. Washington, DC: American Psychiatric Publishing
–. 2013. Diagnostic and Statistical Manual of Mental Disorders, 5th ed. Washington, DC: American Psychiatric Publishing.

Archer, J. 1975. “Rodent Sex Differences in Emotional and Related Behavior.” Behavioral Biology 14: 451–79.

Adams, S., J. Kuebli, P.A. Boyle, and R. Fivush. 1995. “Gender Differences in Parent-Child Conversations about Past Emotions: A Longitudinal Investigation.” Sex Roles 33, no. 5–6: 309–23.

Beatty, W.W. 1979. “Gonodal Hormones and Sex Differences in Nonreproductive Behaviors in Rodents: Organizational and Activational Influences.” Hormones and Behavior 12: 112–63.

Beery A.K., and I. Zucker. 2011. “Sex Bias in Neuroscience and Biomedical Research.” Neurosci Biobehav Rev 35, no. 3: 565–72.

Benjet, C., E. Bromet, E.G. Karam, R.C. Kessler, K.A. McLaughlin, A.M. Ruscio, et al. 2016. “The Epidemiology of Traumatic Event Exposure Worldwide: Results from the World Mental Health Survey Consortium.” Psychological Medicine 46, no. 2: 327–43.

Bluhm, R. 2013. “Self-Fulfilling Prophecies: The Influence of Gender Stereotypes on Functional Neuroimaging Studies of Emotion.” Hypatia 28, no. 4: 870–86.

Bryant, R.A., K.L. Felmingham, D. Silove, M. Creamer, M. O’Donnell, and A.C. McFarlane. 2011. “The Association between Menstrual Cycle and Traumatic Memories.” J Affect Disor 131, no. 1–3: 398–401.

Casey, R. 1993. “Children’s Emotional Experience: Relations among Expression, Self-Report, and Understanding.” Dev Psychol 29, no. 1: 119–29.

Chang, Y.-J., C.-H. Yang, Y.-C. Liang, C.-M. Yeh, C.-C. Huang, and K.-S. Hsu. 2009. “Estrogen Modulates Sexually Dimorphic Contextual Fear Extinction in Rats through Estrogen Receptor Beta.” Hippocampus 11: 1142–50.

Clayton, J.A., and F.S. Collins. 2014. “Policy: NIH to Balance Sex in Cell and Animal Studies.” Nature 509, no. 7500: 282–3.

Crenshaw, K. 1989. “Demarginalizing the Intersection of Race and Sex: A Black Feminist Critique of Antidiscrimination Doctrine, Feminist Theory and Antiracist Politics.” University of Chicago Legal Forum 1: 1, article 8.

Ditlevsen, D.L., and A. Elklit. 2010. “The Combined Effect of Gender and Age on Post Traumatic Stress Disorder: Do Men and Women Show Differences in the Lifespan Distribution of the Disorder? Ann Gen Psychiatry 9: 32.

Dohrenwend, B.P. 2010. “The Stressor Criterion A in Posttraumatic Stress Disorder: Issues, Evidence, and Implications.” In Simpson et al., Anxiety Disorders Theory, Research and Clinical Perspectives, 216–27.

Donner, N.C., and C.A. Lowry. 2013. “Sex Differences in Anxiety and Emotional Behavior.” Pflugers Arch 465. no. 5: 601–26.

Eliot, L., and S.S. Richardson. 2016. “Sex in Context: Limitations of Animal Studies for Addressing Human Sex/Gender Neurobehavioral Health Disparities.” J Neurosci 36, no. 47: 11823–30.

Engdahl, B., M.A. Dikel, R. Eberly, and A. Blank. 1997. “Posttraumatic Stress Disorder in a Community Group of Former Prisoners of War: A Normative Response to Severe Stress.” American Journal of Psychiatry 154: 1576–81.

Einstein, G. 2007. Sex and the Brain: A Reader. Cambridge: MIT Press.

Fausto-Sterling, A. 1992. Myths of Gender: Biological Theories about Women and Men, revised ed. New York: Basic Books.
-–. 2000. Mem. New York: Basic Books.

Fine, C., R. Jordan-Young, A. Kaiser, and G. Rippon. 2013. “Plasticity, Plasticity, Plasticity… and the Rigid Problem of Sex.” Trends Cogn Sci 17, no. 11: 550¬1.

Fine, C., J. Dupre, and D. Joel. 2017. “Sex-Linked Behavior: Evolution, Stability and Variability.” Trends in Cogn Sci 21, no. 9: 666–73.

First, M.B., D.K. Caban, and R. Lewis-Fernandez. 2010. “Development of Nosology of Anxiety Disorders.” In Simpson et al., Anxiety Disorders Theory, Research and Clinical Perspectives, 20–39.

Fivush, R., M.A. Brotman, J.P. Buckner, and S.H. Goodman. 2000. “Gender Differences in Parent-Child Emotion Narratives.” Sex Roles 42, no. 3–4: 233–53.

Flandreau, E.I., and M. Toth. 2017. “Animal Models of PTSD: A Critical Review.” Curr Top Behav Neurosci doi: 10.1007/7854_2016_65.

Frankenhaeuser, M., U. Lundberg, M. Fredrikson, B. Melin, M. Tuomisto, and A.L. Myrsten, et al. 1989. “Stress on and off the Job as Related to Sex and Occupational Status in White-Collar Workers.” J Organ Behav 10, no. 4: 321–46.

Glover, E.M., T. Jovanovic, K.B. Mercer, K. Kerley, B. Bradley, and K.J. Ressler, et al. 2012. “Estrogen Levels Are Associated with Extinction Deficits in Women with Posttraumatic Stress Disorder.” Biol Psychiatr 72, no. 1: 19–24.

Glover, E.M., T. Jovanovic, and S.D. Norrholm. 2015. “Estrogen and Extinction of Fear Memories: Implications for Posttraumatic Stress Disorder Treatment.” Biol Psychiatry 78, no. 3: 178–85.

Goswami, S., M. Cascardi, O.E. Rodríguez-Sierra, S. Duvarci, and D. Paré. 2010. “Impact of Predatory Threat on Fear Extinction in Lewis Rats.” Learning and Memory 17, no. 10: 494–501.

Goswami, S., O. Rodríguez-Sierra, M. Cascardi, and D. Paré. 2013. “Animal Models of Post-Traumatic Stress Disorder: Face Validity.” Front Neurosci 7: 89.

Graham, B.M., and M.R. Milad. 2013. “Blockade of Estrogen by Hormonal Contraceptives Impairs Fear Extinction in Female Rats and Women.” Biol Psychiatry 73, no. 4: 371–8.

–. 2014. “Inhibition of Estradiol Synthesis Impairs Fear Extinction in Male Rats.” Learn Mem 21, no. 7: 347–50.

Graham, B.M., C. Ash, and M.L. Den. 2017. “High Endogenous Estradiol Is Associated with Enhanced Cognitive Emotion Regulation of Physiological Conditioned Fear Responses in Women.” Psychoneuroendocrinology 80: 7–14.

Gruene, T.M., K. Flick, A. Stefano, S.D. Shea, and R.M. Shansky. 2015. “Sexually Divergent Expression of Active and Passive Conditioned Fear Responses in Rats.” Elife 4: e11352.

Guina, J., R.W. Nahhas, K. Kawalec, and S. Farnsworth. 2016. “Are Gender Differences in DSM-5 PTSD Symptomatology Explained by Sexual Trauma?” J Interpers Violence pii: 0886260516677290.

Gungor, N.Z., and D. Pare. 2016. “Functional Heterogeneity in the Bed Nucleus of the Stria Terminalis.” J Neuro 36, no. 31: 8038–49.
Haraway, D. 1996. “Situated Knowledges: The Science Question in Feminism and the Privilege of Partial Perspective.” In Keller and Longino, Feminism & Science, 249–64.

Harding, S. 1996. “Rethinking Standpoint Epistemology: What Is ‘Strong Objectivity’?” In Keller and Longino, Feminism & Science, 235–49.

Herman, J.L. 1997. Trauma and Recovery: The Aftermath of Violence – From Domestic Abuse to Political Terror. New York: Basic Books.

Hoffman, A.N., C.E. Armstrong, J.J. Hanna, and C.D. Conrad. 2010. “Chronic Stress, Cyclic 17β-Estradiol, and Daily Handling Influences on Fear Conditioning in the Female Rat.” Neurobiol Learn Mem 94, no. 3: 422–33.

Joel, D., A. Kaiser, S.S. Richardson, S.A. Ritz, D. Roy, and B. Subramaniam. 2015. “Lab Meeting: A Discussion on Experiments and Experimentation: NIH to Balance Sex in Cell and Animal Studies.” Catalyst 1.

Joel, D., and M.M. McCarthy. 2017. “Incorporating Sex as a Biological Variable in Neuropsychiatric Research: Where Are We Now and Where Should We Be?” Neuropsychopharmacology 42, no. 2: 379–85.

Joel, D., and R. Tarrasch. 2014. “On the Mis-presentation and Misinterpretation of Gender-Related Data: The Case of Ingalhalikar’s Human Connectome Study.” Proc Natl Acad Sci USA 111, no. 6: E637.

Jordan-Young, R.M. 2010. Brain Storm: The Flaws in the Science of Sex Difference. Cambridge: Harvard University Press.

Kaiser, A., S. Haller, S. Schmitz, and C. Nitsch. 2009. “On Sex/Gender Related Similarities and Differences in fMRI Language Research.” Brain Research Review 61, no. 2: 49–59.

Keller, E.F., and H.E. Longio. 1996. Feminism & Science. New York: Oxford University Press

Kessler, R.C., M Petukhova, N.A. Sampson, A.M. Zaslavsky, and H.U. Wittchen. 2012. “Twelve-Month and Lifetime Prevalence and Lifetime Morbid Risk of Anxiety and Mood Disorders in the United States.” Int J Methods Psychiatr Res 21, no. 3: 169–84.

Koenen, K.C., A. Ratanatharathorn, L. Ng, K.A. McLaughlin, E.J. Bromet, and D.L. Stein, et al. 2017. “Posttraumatic Stress Disorder in the World Mental Health Surveys.” Psychol Med doi: 10.1017/S0033291717000708.

Kohl, J., E.A. Autry, and C. Dulac. 2017. “The Neurobiology of Parenting: A Neural Circuit Perspective.” Bioessays 39, no. 1: 1–11.

Kokras, N., and C. Dalla. 2014. “Sex Differences in Animal Models of Psychiatric Disorders.” Br J Pharmacol 171, no. 20: 4595–619.

Lassemo, E., I. Sandanger, J.F. Nygård, and K.W. Sørgaard. 2017. “The Epidemiology of Post-traumatic Stress Disorder in Norway: Trauma Characteristics and Pre-existing Psychiatric Disorders.” Soc Psychiatry Psychiatr Epidemiol 52, no. 1: 11–19.

Lebron-Milad, K., and M.R. Milad. 2012. “Sex Differences, Gonadal Hormones and the Fear Extinction Network: Implications for Anxiety Disorders.” Biol Mood Anxiety Disord 2, no. 1: 3.

LeDoux, J.E. 2014. “Coming to Terms with Fear.” PNAS 111, no. 8: 2871–8.
–. 2015. Anxious: Using the Brain to Understand and Threat Fear and Anxiety. New York: Viking.

Li, S., and B.M. Graham. 2016. “Estradiol Is Associated with Altered Cognitive and Physiological Responses During Fear Conditioning and Extinction in Healthy and Spider Phobic Women.” Behav Neurosci 130, no. 6: 614–23.
–. 2017. “Why Are Women so Wulnerable to Anxiety, Trauma-Related and Stress-Related Disorders? The Potential Role of Sex Hormones.” Lancet Psychiatry 4, no. 1: 73–82.

Liberzon, I., and J.L. Abelson. 2016. “Context Processing and the Neurobiology of Post-Traumatic Stress Disorder.” Neuron 92, no. 1: 14–30.

Lissack, S., A.S. Powers, E.B. McClure, E.A. Phelps, G.W. Hawariat, C. Grillon, and D.S. Pine. 2005. “Classical Fear Conditioning in the Anxiety Disorders: A Meta-Analysis.” Behaviour Research and Therapy 43: 1391–424.

Lister, R.G. 1990. “Ethologically-Based Animal Models of Anxiety Disorders.” Pharmac. Ther 46: 321–40.

Lonsdorf, T.B., J. Haaker, D. Schümann, T. Sommer, J. Bayer, and S. Brassen, et al. 2015. “Sex Differences in Conditioned Stimulus Discrimination during Context-Dependent Fear Learning and Its Retrieval in Humans: The Role of Biological Sex, Contraceptives and Menstrual Cycle Phases.” J Psychiatry Neurosci 40, no. 6: 368–75.

Lundberg, U. 2005. “Stress Hormones in Health and Illness: The Roles of Work and Gender.” Psychoneuroendocrinology 30, no. 10: 1017–21.

Machamer, P., L. Darden, and C.F. Craver. 2000. “Thinking about Mechanisms.” Philosophy of Science 67, no. 1: 1–25.

Maeng, L.Y., and M.R. Milad. 2015. “Sex Differences in Anxiety Disorders: Interactions Between Fear, Stress, and Gonadal Hormones.” Hormones and Behavior 76: 106–17.

Maercker, A., C.R. Brewin, R.A. Bryant, M. Cloitre, M. van Ommeren, and L.M. Jones, et al. 2013. “Diagnosis and Classification of Disorders Specifically Associated with Stress: Proposals for ICD-11.” World Psychiatry 12, no. 3: 198–206.

Maikovich, A.K., and K.C. Koenen. 2009. “Posttrauamatic Stress Symptoms and Trajectories in Child Sexual Abuse Victims: An Analysis of Sex Differences using the National Survey of Child and Adolescent Well-Being.” J Abnorm Child Psychol 37, no. 5: 727–37.

McLean, C.P., and E.R. Anderson. “Brave Men and Timid Women? A Review of the Gender Differences in Fear and Anxiety.” Clinical Psychology Review 29: 496–505.

Maney, D.L. 2016. “Perils and Pitfalls of Reporting Sex Differences.” Philos Trans R Soc Lond B Biol Sci 371, no. 1688: 20150119.

Marder, E., M.L. Goeritz, and A.G. Otopalik. 2015. “Robust Circuit Rhythms in Small Circuits Arise from Variable Circuit Components and Mechanisms.” Current Opinion in Neurobiology 31: 156–63.

Mendelson, S.D., and B.B. Gorzslka. 1987. “An Improved Chamber for the Observation and Analysis of the Sexual Behavior of the Female Rat.” Physiology and Behavior 39, no. 1: 67–71.

Meredith, M. 2001. “Human Vomeronasal Organ Function: A Critical Review of Best and Worst Cases.” Chemical Senses 26, no. 4: 433–45.

Messman-Moore, T.L., and P.H. Bhuptani. 2017. “A Review of the Long-Term Impact of Childhood Maltreatment on Posttraumatic Stress Disorder and Its Comorbidities: An Emotion Dysregulation Perspective.” Clinical Psychology Science and Practice 24, no. 2: 154–69.

Milad, M.R., and G.H. Quirk. 2012. “Fear Extinction as a Model for Translational Neuroscience: Ten Years of Progress.” Annual Review of Psychology 63: 129–51.

Milad, M.R., B.L. Rosebaum, and N.M. Simon. 2014. “Neuroscience of Fear Extinction: Implications for Assessment and Treatment of Fear-Based and Anxiety-Related Disorders.” Behavioral Research and Therapy 62: 17–23.

Milad, M.R., M.A. Zeidan, A. Contero, R.K. Pitman, A. Klibansk, and S.L. Rauch, et al. 2014. “The Influence of Gonadal Hormones on Conditioned Fear Extinction in Healthy Humans.” Neuroscience 168, no. 3: 652–8.

Milad, M.R., J.M. Goldstein, S.P. Orr, M.M. Wedig, A. Klibanski, and R.K. Pitman, et al. 2006. “Fear Conditioning and Extinction: Influence of Sex and Menstrual Cycle in Healthy Humans.” Behav Neurosci 120, no. 6: 1196–203.

Mogil, J.S., and M.L. Chanda. 2005. “The Case for Inclusion of Female Subjects in Basic Science Studies of Pain.” Pain 117: 1–5.

National Institutes of Health. 2015a. “Consideration of Sex as a Biological Variable in NIH-Funded Research (Notice number: NOT-OD-15-102).” https://grants.nih.gov/grants/guide/notice-files/NOT-OD-15-102.html.
National Institutes of Health. 2015b. “Consideration of Sex as a Biological Variable in NIH-Funded Research (Guidance to notice number: NOT-OD-15-102).” https://orwh.od.nih.gov/sites/orwh/files/docs/NOT-OD-15-102_Guidance.pdf.

Norris, F., J. Perilla, and G. Ibañez. 2001. “Sex Differences in Symptoms of Posttraumatic Stress: Does Culture Play a Role?” J Traumatic Stress 14, no. 1: 7–28.

Ogle, C.M., D.C. Rubin, and I.C. Siegler. 2014. “Cumulative Exposure to Traumatic Events in Older Adults.” Aging Ment Health 18, no. 3: 316–25.

Olff, M., W. Langeland, and N. Draijer. 2007. “Gender Differences in Posttraumatic Stress Sisorder.” Psychol Bull 133, no. 2: 183–204.

Parsons, M.H., R. Apfelbach, P.B. Banks, E.Z. Cameron, C.R. Dickman, and A.S.K. Frank, et al. 2017. “Biologically Meaningful Scents: A Framework for Understanding Predator-Prey Research across Disciplines.” Biological Reviews doi: 10.1111/brv.12334.

Pellman, B.A., and J.J. Kim. 2016. “What Can Ethobehavioral Studies Tell Us about the Brain’s Fear System?” Trends in Neurosciences 39, no. 6: 420–31.

Perusini, J.N., E.M. Meyer, V.A. Long, V. Rau, N. Nocera, J. Avershal, J. Maksymetz, I. Spigelman, and M.S. Fanselow. 2016. “Induction and Expression of Fear Sensitization Caused by Acute Traumatic Stress.” Neuropsychopharmacology 41: 45–57.

Phelps, E.A., and J.E. LeDoux. 2005. “Contributions of the Amygdala to Emotion Processing: From Animal Models to Human Behavior.” Neuron 48, no. 2: 175–87.

Pineles, S.L., K.A. Arditte Hall, and A.M. Rasmusson. 2017. “Gender and PTSD: Different Pathways to a Similar Phenotype.” Curr Opin Psychol 14: 44–8.

Prendergast, B.J., K.G. Onishi, and I. Zucker. 2014. “Female Mice Liberated for Inclusion in Neuroscience and Biomedical Research.” Neuroscience and Biobehavioral Reviews 40: 1–5.

Rau, V., J.P. DeCola, and M.S. Fanselow. 2005. “Stress-Induced Enhancement of Fear Learning: An Animal Model of Posttraumatic Stress Disorder.” Neuroscience and Biobehavioral Reviews 29: 1207–23.

Rey, C.D., J. Lipps, and R.M. Shansky. 2014. “Dopamine D1 Receptor Activation Rescues Extinction Impairments in Low-Estrogen Female Rats and Induces Cortical Layer-Specific Activation Changes in Prefrontal-Amygdala Circuits.” Neuropsychopharmacology 39, no. 5: 1282¬–9.

Rippon, G., R. Jordan-Youn, A. Kaiser, and C. Fin. 2014. “Recommendations for Sex/Gender Neuroimaging Research: Key Principles and Implications for Research Design, Analysis, and Interpretation.” Frontiers in Human Neuroscience 8: 650.

Roughgarden, J. 2004. Evolution’s Rainbow: Diversity, Gender, and Sexuality in Nature and People. Berkeley: University of California Press.

Ryb, G.E., P.C. Dischinger, K.M. Read, and J.A. Kufera. 2009. “PTSD after Severe Vehicular Crashes.” Ann Adv Automot Med 53: 177–93.

Samuels-Dennis, J.A., A. Bailey, K. Killian, and S.L. Ray. 2010a. “The Mediating Effects of Empowerment, Interpersonal Conflict, and Social Support on the Violence PTSD Process Among Single Mothers.” Canadian Journal of Community Mental Health 32, no. 1: 109–24.

Samuels-Dennis, J.A., P.W. Ford-Gilboe, W.R. Avison, and S.L. Ray. 2010b. “Cumulative Trauma, Personal and Social Resources, and Post-traumatic Stress Symptoms among Income-Assisted Single Mothers.” Journal of Family Violence 25, no. 6: 603–17.

Samuels-Dennis, J.A., P.W. Ford-Gilboe, and A. Bailey. 2011. “The Intersectionality Model of Trauma and Post-traumatic Stress Disorder (IMT-PTSD).” In Health Inequities in Canada: Intersectional Frameworks and Practices, edited by Olena Hankivsky, 274–92. Vancouver: UBC Press.

Sapolsky, R.M. 2005. “The Influence of Social Hierarchy on Primate Health.” Science 308, no. 5722: 648–52.

Shackman, A.J., and A.S. Fox. 2016. “Contributions of the Central Extended Amygdala to Fear and Anxiety.” J Neuro 36, no. 31: 8050–63.

Shansky, R.M. 2015. “Sex Differences in PTSD Resilience and Susceptibility: Challenges for Animal Models of Fear Learning.” Neurobiol Stress 1: 60–5.

Shansky, R.M, and C.S. Woolley. 2016. “Considering Sex as a Biological Variable Will Be Valuable for Neuroscience Research.” J Neurosci 36, no. 47: 11817–22.

Shors, T.J. 2006. “Stressful Experience and Learning across the Lifespan.” Annual Reviews Psychology 57: 55–85.

Silove, D., J.R. Baker, M. Mohsin, M. Teesson, M. Creamer, and M. O’Donnell, et al. 2017. “The Contribution of Gender-Based Violence and Network Trauma to Gender Differences in Post-traumatic Stress Disorder. PLoS One 12, no. 2: e0171879.

Simpson, Helen B., Yuval Neria, Roberto Lewis-Fernandez, and Franklin Schnerier, eds. 2010. Anxiety Disorders Theory, Research and Clinical Perspectives. New York: Cambridge University Press.

Steimer, T. 2011. “Animal Models of Anxiety Disorders in Rats and Mice: Some Conceptual Issues.” Dialogues Clin Neurosci 13, no. 4: 495–506.

Tolin, D.F., and E.B. Foa. 2006. “Sex Differences in Trauma and Posttraumatic Stress Disorder: A Quantitative Review of 25 Years of Research.” Psychological Bulletin 132, no. 6: 959–92.

Toufexis, D.J., K.M. Myers, M.E. Bowser, and M. Davis. 2007. “Estrogen Disrupts the Inhibition of Fear in Female Rats, Possibly through the Antagonistic Effects of Estrogen Receptor Alpha (ERalpha) and ERbeta.” J Neurosci 27, no. 36: 9729–35.

Valle, F.P., and B.B. Gorzalka. 1980. “Open-Field Sex Differences Prior to Puberty in Rats.” Bulletin of the Psychonomic Society 16: 429–31.

Walker, J.L., P.D. Carey, N. Mohr, D.J. Stein, and S. Seedat. 2004. “Gender Differences in the Prevalence of Childhood Sexual Abuse and in the Development of PTSD.” Archives of Women’s Mental Health 7: 111–21.

van den Wijngaard, M. 1997. Reinventing the Sexes: The Biomedical Construction of Femininity and Masculinity. Indianapolis: Indiana University Press.

Wittchen, H.-U., S. Schönfeld, C. Thurau, S. Trautmann, M. Galle, and K. Mark, et al. 2012. “Prevalence, Incidence and Determinants of PTSD and Other Mental Disorders: Design and Methods of the PID-PTSD Study.” International Journal of Methods in Psychiatric Research 21, no. 2: 98–116.

Zeidan, M.A., S.A. Igoe, C. Linnman, A. Vitalo, J.B. Levine, and A. Klibanski, et al. 2011. “Estradiol Modulates Medial Prefrontal Cortex A\and Amygdala Activity during Fear Extinction in Women and Female Rats.” Biological Psychiatry 70, no. 10: 920–7.

1. We would like to thank our editors Giordana Grossi, Rebecca Young, Gina Rippon, and Carolyn Yates; our colleagues Gillian Einstein, Cordelia Fine, Drew Headley, and Pinelopi Kryazi; and the two anonymous reviewers for their constructive comments and helping us to improve this manuscript. Competing financial interests statement: The authors declare that they have no competing financial interests.[↑]
2. In developed countries, all research programs that use humans and animals as experimental subjects must be approved by ethical committees that specialize in reducing unnecessary harm and promoting optimal conditions for research subjects. For up to date information on animal research regulations, please consult the NIH’s Office of the Laboratory Animal Welfare guidelines.[↑]
3. Or, simply, the real-world threat situations that the brain’s fear system adapted to address. We want to clarify that evolution proceeds blindly without any specifics goals to establish or problems to solve. [↑]