“… The basic gist of the theory is that humans are motivated to quell the potential for terror inherent in the human awareness of vulnerability and mortality by investing in cultural belief systems (or worldviews) that imbue life with meaning, and the individuals who subscribe to them with significance (or self-esteem). Since its inception, the theory has generated empirical research into not just the nature of self-esteem motivation and prejudice, but also a host of other forms of human social behavior. To date, over 300 studies conducted in over a dozen countries have explored such topics as aggression, stereotyping, needs for structure and meaning, depression and psychopathology (e.g., phobias), political preferences, creativity, sexuality and attraction, romantic and interpersonal attachment, self-awareness, unconscious cognition, martyrdom, religion, group identification, disgust, human-nature relations, physical health, risk taking, and legal judgments…”


“… Terror Management Theory (TMT) was developed in 1986 by social psychologists Jeff Greenberg, Tom Pyszczynski, and Sheldon Solomon based upon Ernest Becker’s ideas. TMT posits that while humans share with all life-forms a biological predisposition toward self-preservation in the service of reproduction, we are unique in our capacity for symbolic thought, which fosters self-awareness and the ability to reflect on the past and ponder the future. This spawns the realization that death is inevitable and can occur at any time for reasons that cannot be anticipated or controlled.

The awareness of death engenders potentially debilitating terror that is “managed” by the development and maintenance of cultural worldviews: humanly constructed beliefs about reality shared by individuals that minimize existential dread by conferring meaning and value. All cultures provide a sense that life is meaningful by offering an account of the origin of the universe, prescriptions for appropriate behavior, and assurance of immortality for those who behave in accordance with cultural dictates. Literal immortality is afforded by souls, heavens, afterlives, and reincarnations associated with all major religions. Symbolic immortality is obtained by being part of a great nation, amassing great fortunes, noteworthy accomplishments, and having children. Psychological equanimity also requires that individuals perceive themselves as persons of value in a world of meaning. This is accomplished through social roles with associated standards. Self-esteem is the sense of personal significance that results from meeting or exceeding such standards.

Three lines of research provide empirical support for TMT:

  • The anxiety-buffering function of self-esteem is established by studies where momentarily elevated self-esteem results in lower self-reported anxiety and physiological arousal.
  • Making death salient by asking people to think about themselves dying (or viewing graphic depictions of death, being interviewed in front of a funeral parlor, or subliminal exposure to the word “dead” or “death”) intensifies strivings to defend their cultural worldviews by increasing positive reactions to similar others, and negative reactions toward those who are different.
  • Research verifies the existential function of cultural worldviews and self-esteem by demonstrating that non-conscious death thoughts come more readily to mind when cherished cultural beliefs or self-esteem is threatened.

TMT has generated empirical research (currently more than 500 studies) examining a host of other forms of human social behavior, including aggression, stereotyping, needs for structure and meaning, depression and psychopathology, political preferences, creativity, sexuality, romantic and interpersonal attachment, self-awareness, unconscious cognition, martyrdom, religion, group identification, disgust, human-nature relations, physical health, risk taking, and legal judgments. In 2015, Greenberg, Pyszczynski and Solomon published The Worm at the Core, which reviews this vast body of research supporting Becker’s central claim that the fear of death is “the mainspring of human activity…”

Fear of the unknown: One fear to rule them all?
by R. Nicholas Carleton  /  June 2016

Abstract: “The current review and synthesis was designed to provocatively develop and evaluate the proposition that “fear of the unknown may be a, or possibly the, fundamental fear” (Carleton, 2016) underlying anxiety and therein neuroticism. Identifying fundamental transdiagnostic elements is a priority for clinical theory and practice. Historical criteria for identifying fundamental components of anxiety are described and revised criteria are offered. The revised criteria are based on logical rhetorical arguments using a constituent reductionist postpositivist approach supported by the available empirical data. The revised criteria are then used to assess several fears posited as fundamental, including fear of the unknown. The review and synthesis concludes with brief recommendations for future theoretical discourse as well as clinical and non-clinical research. Fear of the unknown (FOTU) will be defined herein as, “an individual’s propensity to experience fear caused by the perceived absence of information at any level of consciousness or point of processing”; relatedly, intolerance of uncertainty (IU) will be defined as, “an individual’s dispositional incapacity to endure the aversive response triggered by the perceived absence of salient, key, or sufficient information, and sustained by the associated perception of uncertainty.” (Carleton, 2016; p. 31). The paper begins with a brief historical overview of anxiety determinants and hierarchical models for anxiety and fear. The original criteria for fundamental fears are presented as foundations for contemporary revisions. The revised criteria are then used to assess several fears posited as fundamental, including fear of the unknown. The paper concludes with brief recommendations for future theoretical discourse as well as clinical and non-clinical research.

1.1. Determining anxiety
Psychologists have sought fundamental determinants of cognitions, behaviours, personalities, and pathologies since Wundt. Contemporary textbooks describe psychology as a science dedicated to understanding the multiple determinants of cognitions, behaviours, and emotions (e.g., Gerrig, Zimbardo, Desmarais, & Ivanco, 2011; Lilienfeld et al., 2015). Many psychologists rely on a constituent reductionist postpositivist philosophy of science in such efforts; oversimplified, the approach identifies increasingly fundamental components of a construct (i.e., constituent reductionism), without ignoring the emergent properties, and while accepting an underlying reality that can be measured or understood by progressive but imperfect approximations (i.e., postpositivism). Identifying fundamental components has broadly benefited many scientific fields (e.g., chemistry, biology, physics). Such practices fit well with structuralism and functionalism (Munger, 2003), and Gestalt researchers acknowledge progressive reductionism as beneficial, caveating costs to understanding complex whole systems (Humphrey, 1924a, Humphrey, 1924b). Accordingly, many psychologists have applied similar practices to understanding and modifying cognitions, emotions, and behaviours. Theorists from Freud (Breuer & Freud, 1974; Freud, 1924), to Spielberger (1975), to Barlow, 2000, Barlow, 2002, have suggested individual differences make some people more likely to experience fear or anxiety. The tendency to respond with fear or to experience pervasive anxiety was called neuroticism by Freud and trait anxiety by Spielberger. Spielberger (1972) described anxiety as “psychobiological” (p.489) and “extremely complex and involve[ing] a number of different measurable components” (Spielberger, 1975; p. 139). He posited an interaction between exposure to stressors, appraisals of stressors (in line with Scherer, 2009), and a recursive interactivity between experiences of state and trait anxiety (in line with Scherer & Brosch, 2009). Spielberger (1975) further suggested components of anxiety “should be independently conceptualized and operationally measured as critical variables in a state-trait theory of anxiety” (p. 139). In other words, anxiety was not fundamental1 unto itself, and Spielberger underscored the differentiation of state and trait anxiety was only the beginning of understanding anxiety. Theorists have since supported a constituent reductionist postpositivist approach to understanding trait anxiety (Barlow, Ellard, Sauer-Zavala, Bullis, & Carl, 2014; Barlow, Sauer-Zavala, Carl, Bullis, & Ellard, 2014; Brown and Barlow, 2009, Cuijpers et al., 2010, Lahey, 2009), particularly when used synonymously with constructs like neuroticism (Ormel, Rosmalen, & Farmer, 2004).

Such synonymous usage has led prominent theorists to argue the functional equivalency of several constructs related to trait anxiety. For example, Kagan and Snidman (2004) clarify that inhibited and uninhibited to the unfamiliar temperament categories refer to introverted and extraverted, respectively. Barlow and colleagues (Barlow, Sauer-Zavala et al., 2014) argue the same two categories respectively refer to trait anxiety and negative affect, as well as positive affect. Clark and Beck (2010) argued trait anxiety is “…so closely related to negative emotionality (i.e., Neuroticism or Negative Affect) that the two are considered almost synonymous” (p.104). If (1) neuroticism by any name is a higher-order biopsychosocial construct that is facilitated by fear of potentially recurring aversive experiences (Barlow, 2000); (2) fear is distinguished from anxiety by being present-oriented and relatively certain, rather than future-oriented and relatively uncertain (Barlow, 2000); (3) prolific anxiety defines neuroticism (Barlow, Sauer-Zavala et al., 2014); and (4) fear “is the basic cognitive process underlying all anxiety disorders” (Clark & Beck, 2010, p. 29), then a hierarchical structure identifying increasingly fundamental fears offers several potential advantages; for example, informing and being informed by psychological theories of evolution, emotion, cognition, development, personality, and decision-making.

5.3. Fear of death as fundamental
Benjamin Franklin is often referenced as saying nothing can be said to be certain, except death and taxes. Death is the conclusion of the process that is dying. Fearing death is not evidenced by reflexes to preserve life (e.g., breathing, Moro5), because reflexes are not emotions (Leventhal and Scherer, 1987, Moors, 2009; Moors, Ellsworth, Scherer, & Frijda, 2013). Reflexes could be appraised percepts causally associated with fear pending appropriate learning (Mulligan & Scherer, 2012); however, reflexes would not evidence fear of death as inherent any more than salivation evidences fear of starvation as inherent. For example, Moro following a sudden drop could cause fear of vestibular motor sensations associated with falling or fear of pain after impact; conversely, it could cause joy if caught repeatedly being caught by a parent was perceived as certain. A person who fears death may not fear death, per se, instead fearing (1) the process of dying, which is reducible to other fears (e.g., AS, FNE, IIS, fear of pain); (2) that death will be unpleasant (e.g., the afterlife may be painful); (3) unknowns about death (e.g., death may or may not involve an afterlife), making the unknown the object of fear; or (4) the concept of oblivion, an unequivocal unknown, which requires substantial a priori learning. Neither dying nor death are inherent concepts; both must be learned (Balk, 2010). “Absence of fear of death among children under ten [years of age] is in keeping with Anthony’s study reported in The Childs Discovery of Death (Anthony, 1940)” (Bowlby, 1973, p.117). As detailed in Carleton’s review (Carleton, 2016), children fear unknowns, darkness, and snakes–among other things–long before demonstrating understanding or fear of death (Bowlby, 1973; Kagan & Snidman, 2004). Furthermore, death is not necessarily aversive or avoided. Many religions posit death as involving a fantastic and appetitive afterlife (Segal, 2010). Where the afterlife is potentially appetitive, any associated fear must result from (1) insufficient certainty in the promise of that appetitive afterlife or (2) the process of dying as a transition to that afterlife. Notional development of an appetitive afterlife may have been to cope with unknowns related to death. “For this fear of death is indeed the pretence of wisdom, and not real wisdom, being the appearance of knowing the unknown; since no one knows whether death, which they in their fear apprehend to be the greatest evil, may not be the greatest good” (Socrates, 399 BC). In any case, fear of death, while arguably pervasive and poignant, does not appear fundamental.

5.4. Fear of pain as fundamental
Haruki Murakami claimed “Pain is inevitable. Suffering is optional” (Murakami, 2009; p.vii); relatedly, fear of pain may be fundamental (Carleton, 2012; Carleton, Abrams, Asmundson et al., 2009). Pain is a sensory and emotional experience associated with perceived or potential tissue damage (Loeser & Treede, 2008; p. 475). Nociception is the neuronal processing of noxious stimuli, beginning with the activation of nociceptors (Loeser & Treede, 2008; p. 475). Nociception is inherently aversive (Pear, 2014, Rolls, 2013), but evolutionarily derivative for mitigating injury and death (Broom, 2001; Fields, 2006) and not necessarily fear provoking (Asmundson & Carleton, 2008; Carleton & Asmundson, 2012). Nociception is neither an emotion (Leventhal and Scherer, 1987, Moors, 2009) nor synonymous with the subjective and complex experience of pain, which involves nociception, appraisals, attributions of meaning (Asmundson, Norton, & Norton, 1999; International Association for the Study of Pain, 1986; Loeser and Treede, 2008, Melzack and Katz, 2004). The distinction complicates assessing fear of pain as fundamental because fear of pain typically refers to the experience of pain rather than nociception (Asmundson & Carleton, 2008; Carleton & Asmundson, 2012). Nociception evokes appraisals that may produce the experience of pain (Asmundson & Wright, 2004) and is not required to experience pain (e.g., phantom limb pain; Diers, Christmann, Koeppe, Ruf, & Flor, 2010; Melzack, 1990). Fear of pain references the unpleasant sensory and emotional experience that may be associated with the nociception, but is the result of substantial and complex learning based on attributions of meaning to nociception (Asmundson et al., 1999; International Association for the Study of Pain, 1986; Melzack & Katz, 2004), making it neither inherent nor non-derivative. Nociception might be aversive but not fear-provoking if the following are certain: (1) short length; (2) manageable intensity; (3) no consequential sequelae (e.g. not harmful); or (4) consequential sequelae are appetitive (e.g., goal completion; pleasure). For example, consuming spicy foods, long-distance running, or elective skin puncturing may be sought because the nociception is considered certainly short and consequentially pleasurable. In contrast, uncertainty regarding the duration, intensity, and injuriousness associated with nociception could dramatically increase fear and anxiety. Increased FOTU appears to increase fear of pain in response to nociception (Helsen, Goubert, & Vlaeyen, 2013), therein supporting FOTU as a critical moderator or determinant for fear and anxiety-related nociception. Despite the aforementioned challenges to fear of pain as fundamental, it appears normally and continuously distributed (Asmundson & Carleton, 2008; Asmundson, Vlaeyen, & Crombez, 2004). Fear of pain is also a logical reduction of pain-related anxiety (Carleton & Asmundson, 2009; Powell, Honey, & Symbaluk, 2012), which accounts for variance in higher-order constructs (e.g., chronic pain; Carleton & Asmundson, 2012) and appears ubiquitous (Carleton, Abrams, Asmundson et al., 2009). Pain-related fear and anxiety are related, but factorially distinct from each other (Carleton & Asmundson, 2009), and from AS, IIS, and FNE (Carleton, Abrams, Asmundson et al., 2009; Carleton, Thibodeau et al., 2014); however, meta-analytic results support a robust relationship with AS such that pain-related fear and anxiety may be derivatives of, or reflexively related to, AS (Ocanez, McHugh, & Otto, 2010). In any case, fear of pain appears well-supported as a lower-order construct that meets some criteria for being fundamental; however, fear of pain requires learned appraisals and attributions, and appears logically reducible, or reflexively related, to AS, IIS, FNE, death, and FOTU. As such, fear of pain, while critically important, may not be fundamental.

6. Fear of the unknown 6.1. Inherent
Perhaps the earliest direct written reference to FOTU as the fundamental fear came from Lovecraft in 1927: “The oldest and strongest emotion of mankind is fear, and the oldest and strongest kind of fear is fear of the unknown” (as cited in Joshi & Schultz, 2001; p. 255). The quote remains relevant, with colloquialisms referencing not knowing as the worst part of distressing situations. FOTU tautologically does not require a priori learning; indeed, the first thing that could be feared would be “the perceived absence of information at any level of consciousness” (Carleton, 2016; p. 31). All other fears appear to require learning involving the perceived presence of perceived or recalled information (Bandura, 1965, Bandura, 1971, Mowrer, 1947, Muris et al., 2002, Pinker, 1997; Powell et al., 2012, Skinner, 1953). Broad theoretical congruence across psychological disciplines suggests unknowns are inherently appraised as aversive (e.g., emotion, development, attachment, neurobiology; see for review and synthesis, Carleton, 2016). Contemporary emotion theories place checks for unknowns at the earliest stage of emotion processing (i.e., low-level neural substrates), underscore pervasive influences thereafter (e.g., conceptual cortical areas; Mulligan and Scherer, 2012, Scherer, 2009, Scherer, 2013), and implicate aggregate responses to unknowns as defining affective dispositions (Scherer & Brosch, 2009). Models of attachment and temperament explicitly discuss FOTU as inherent and intrinsic to the development of neuroticism (e.g., Ainsworth & Bell, 1970; Bowlby, 1973; Colonnesi et al., 2011, Degnan and Fox, 2007, Kagan and Snidman, 2004; Lewis-Morrarty et al., 2016, in press; Madigan, Atkinson, Laurin, & Benoit, 2013; Moehler et al., 2008). Finally, the Uncertainty and Anticipation Model of Anxiety (UAMA; Grupe & Nitschke, 2013) posits negative interactions with uncertainty as facilitating maladaptive responses including “inflated estimates of threat cost improbability, hypervigilance, division safety learning, behavioural and cognitive avoidance and heightened reactivity to threat uncertainty” (Grupe & Nitschke, 2013; p. 490). Neurobiological researchers have emphasized FOTU as a cornerstone for BIS activation and therein fear and anxiety (Bach and Dolan, 2012, Gray and McNaughton, 2003, Grupe and Nitschke, 2013, Herry et al., 2007; Jackson, Nelson, & Proudfit, 2015; Kagan & Snidman, 2004; Nelson, Kessel, Jackson, & Hajcak, 2016; Thayer, Ahs, Fredrikson, Sollers, & Wager, 2012). A recent meta-analysis evidenced unknowns as increasing heart rate variability and activation of the amygdala and ventromedial prefrontal cortex (Thayer et al., 2012). There is also “distinct neural encoding (including summary statistic-type representations) of uncertainty [that] occurs in distinct neural systems” (Bach & Dolan, 2012, p. 572). Furthermore, the earliest cognitive processing of stimuli automatically classifies a stimulus as threatening or not based on knowns and unknowns (Bradley, Mogg, Millar, & White, 1995; Cisler and Koster, 2010, Mathews and MacLeod, 1994, Mathews and MacLeod, 2005, McNally, 1995), with unknown stimuli consistently categorized as threatening. Unknowns incrementally activate the BIS (Gray & McNaughton, 2003), increase error-related negativity based on event-related potentials (Jackson, Nelson, & Hajcak, 2016; Jackson et al., 2015), potentiate startle (Nelson & Shankman, 2011), and produce sustained amygdala and hippocampal increases (Herry et al., 2007, Jackson et al., 2015).

6.2. Evolutionarily supported
FOTU logically fits within evolutionary psychology models (e.g., Buss, 1995). Indeed, there appears to be substantial theoretical and neurobiological evidence supporting the evolutionary basis of FOTU (Brosschot, Verkuil, & Thayer, 2016). Enough fear to approach unknowns with caution (i.e., treating unknowns as potential threats) would be adaptive, so long as the intensity did not compromise survival activities (e.g., seeking food, shelter, mates; Carleton, 2012). Accordingly, evolution should have produced a selection bias for assessing unknowns as likely threatening and therein BIS activating (Bach & Dolan, 2012; Brosschot et al., 2016; Cosmides and Tooby, 1996, Cosmides, 1989, Gigerenzer, 1991Thayer et al., 2012), but still potentially beneficial (Gray & McNaughton, 2003). That bias appears well reflected in neurobiology, temperament, and development research (Brosschot et al., 2016; Carleton, 2016). “Given the evolutionary advantage associated with the assumption of threat, the view that we and others have proposed is that the “default” response to uncertainty, novelty, and threat is the sympathoexcitatory preparation for action commonly known as the fight or flight response (Herry et al., 2007, Thayer and Lane, 2009). This default threat response may be related to the well-known ‘negativity bias,’ a phenomenon that describes the tendency to prioritize negative information over positive (Cacioppo, Gardner, & Berntson, 1999). From an evolutionary perspective this represents a system that errs on the side of caution—when in doubt prepare for the worst—thus maximizing survival and adaptive responses (LeDoux, 1996). (Thayer et al., 2012, p. 749)” Humans and closely related non-human animals (e.g., bonobos) prefer certainty to risk (Kacelnik and Bateson, 1996, Tversky and Kahneman, 1981), but adapt based on learning history (e.g., the aggregate of encounters with unknowns) and situational variables (e.g., the relative number of knowns and unknowns; Gilby & Wrangham, 2007; Heilbronner, Rosati, Stevens, Hare, & Hauser, 2008; Stevens et al., 2005 Stevens, Rosati, Ross, & Hauser, 2005). Non-human primate infants initially respond to unknowns with fear and anxiety, but rapidly habituate in the absence of aversive consequences (Kalin & Shelton, 1989; Kalin, Shelton, & Takahashi, 1991; Timmermans, Vochteloo, Vossen, Röder, & Duijghuisen, 1994). In non-primate mammals, “unknown objects evoking fear are at first avoided, increasing the immediate safety of the animal. However, if the level of fear is low or when it subsides with time or gained distance, animals start active exploration of the [unknown] object, territory, or context that evoked their anxiety or fear and information gathered in the process of exploration allows them to adapt to the novelty or environmental changes” (Pisula et al., 2012; p. 145; Timmermans et al., 1994). Wild animals treat unknowns as more aversive than animals raised in the relative safety of laboratories (Tanaś & Pisula, 2011). There is also substantial evidence that responses to unknowns vary based on genetic variation (Flagel, Waselus, Clinton, Watson, & Akil, 2014), with early exposure to reliable environments impacting phenotype (Weaver et al., 2004). For example, Tang and colleagues experimentally evidenced neonate rats as initially responding to unknowns with fear (Akers et al., 2008; Tang, Akers, Reeb, Romeo, & McEwen, 2006; Tang, Reeb-Sutherland, Romeo, & McEwen, 2012). The same studies demonstrated attenuation of the responses – for the duration of their lifespan – through repeated brief exposures to unknowns during early development. The attenuation was maximized when the neonates had a mother with high levels of self-stress regulation who, most importantly, provided reliable care. Poor self-stress regulation in the mother or unreliable provision of care produced neonates with higher FOTU; nevertheless, the exposures robustly attenuated those responses.

6.5. Non-Derivative (i.e., logically irreducible)
Each of AS, FNE, and IIS, as well as fears of death and pain, can be logically reduced to another feared element (e.g., being anxious about illness for fear of being negatively evaluated), form reflexively circular tautologies with each other, or are dependent on FOTU. In contrast, the current logical, rhetorical, and biological evidence suggests FOTU is non-derivative. The unknown itself has long been considered potentially threatening (e.g, Dugas, Letarte, Rhéaume, Freeston, & Ladouceur, 1995; Epstein, 1972; Freeston, Rhéaume, Letarte, Dugas, & Ladouceur, 1994). Unknowns exacerbate fear (e.g., Dugas, Hedayati et al., 2005; Gray and McNaughton, 2003, Grupe and Nitschke, 2013; Hock & Krohne, 2004) and that fear moderates responses to ambiguous stimuli (e.g., Hedayati, Dugas, Buhr, & Francis, 2003, November). As FOTU increases, people become more likely to accept negative consequences than tolerate uncertainty (e.g., Buhr and Dugas, 2002, Ghosh, 1997, Rassin and Muris, 2005). The appraisal models of emotion (Moors et al., 2013, Scherer, 2009), contemporary anxiety models (Barlow, Ellard et al., 2014; Clark & Beck, 2010; Gray & McNaughton, 2003), and the developmental literature (Ainsworth, Blehar, Waters, & Wall, 1978; Bowlby, 1973, Bowlby, 1989, Kagan and Snidman, 2004) all place fearful responses to unknowns as the cornerstone of anxiety and non-derivative (Carleton, 2016). Consider a person who reports FOTU because it might lead to an aversive outcome feared because of AS, FNE, IIS, fear of death, or fear of pain. Irrespective of the feared aversive outcome, the logical reduction process leads to asking what about that outcome is problematic. For example, I’m afraid of the unknown because I do not know if I will have a heart attack. If I have a heart attack I might die. If I die my deity might judge me negatively and punish me painfully. I’m afraid to experience such pain because I don’t know how long or intense it will be. I’m afraid because I don’t know. The iterative process of deriving fear-evoking elements appears to end, tautologically, with FOTU.

6.6. Accounts for higher-order variance
Initial clinical research into FOTU, typically reflected by IU, has focused on worry and generalized anxiety disorder (GAD; Dugas et al., 1995; Dugas, Marchand, & Ladouceur, 2005; Freeston et al., 1994, Ladouceur et al., 1999). Freeston et al. postulated a causal relationship between IU and worry as constituents of GAD. Despite initial focus on worry and GAD, there is now overwhelming evidence that FOTU, most commonly measured as IU, represents a broad transdiagnostic construct accounting for variance in several higher-order constructs (see for review, Carleton, 2012, Carleton, 2016, Hong and Cheung, 2015). Fearing the unknown, as most commonly reflected by IU, has accounted for statistically significant variance in symptoms of panic disorder (Buhr & Dugas, 2009; Carleton, Duranceau et al., 2014; Carleton, Fetzner, Hackl, & McEvoy, 2013), social anxiety disorder (Boelen and Reijntjes, 2009, Boelen et al., 2010; Carleton, Collimore et al., 2010; Khawaja & McMahon, 2011; Teale Sapach, Carleton, Mulvogue, Weeks, & Heimberg, 2015; Whiting et al., 2014), obsessive compulsive disorder (Grayson, 2010; Jacoby, Fabricant, Leonard, Riemann, & Abramowitz, 2013; Khawaja and McMahon, 2011, Lind and Boschen, 2009; Tolin, Abramowitz, Brigidi, & Foa, 2003), posttraumatic stress disorder (Banducci, Bujarski, Bonn-Miller, Patel, & Connolly, 2016; Boelen, 2010; Boelen, Reijntjes, & Smid, 2016; Fetzner, Horswill, Boelen, & Carleton, 2013; Oglesby, Boffa, Short, Raines, & Schmidt, 2016; Otis, Keane, & Kerns, 2003; White & Gumley, 2009), substance use (Banducci et al., 2016), health anxiety (Boelen and Carleton, 2012, Fergus and Bardeen, 2013, Fergus and Valentiner, 2011, Fergus, 2013; Kurita, Garon, Stanton, & Meyerowitz, 2013; Wright, 2016), separation anxiety (Boelen, Reijntjes, & Carleton, 2014), and eating disorders (Konstantellou & Reynolds, 2010; Renjan, McEvoy, Handley, & Fursland, 2016; Sternheim, Startup, & Schmidt, 2011). FOTU has also been associated with depression (Boelen et al., 2016; Meeten, Dash, Scarlet, & Davey, 2012; Miranda, Fontes, & Marroquin, 2008; Miranda & Mennin, 2007; Nelson, Shankman, & Proudfit, 2014; Yook, Kim, Suh, & Lee, 2010), anger (Dugas, 2016; Fracalanza, Koerner, Deschenes, & Dugas, 2014), and personality disorder symptoms (Berenbaum, Bredemeier, & Thompson, 2008; Fergus & Rowatt, 2014). There have been several broad studies, including meta-analyses, supporting FOTU, reflected by IU, as a transdiagnostic factor that appears generally comparable across psychopathologies (Boswell et al., 2013; Carleton, Mulvogue et al., 2012; Freeston, 2016; Freeston et al., 2016; Gentes and Ruscio, 2011, Mahoney and McEvoy, 2012c, McEvoy and Erceg-Hurn, 2016, McEvoy and Mahoney, 2013, Norr et al., 2013). Furthermore, response patterns based on multigroup confirmatory factor analyses with questions assessing FOTU appear invariant across sex (ps = 0.06 to 0.43), while being significantly different when comparing clinical samples to community (p < 0.01) and undergraduate (p < 0.01) samples (Carleton, Mulvogue et al., 2012).

FOTU predicts variance in higher-order constructs in longitudinal genetic twin research. For example, AS increases in children and adults as a function of exposure to “stressful events (particularly those that are uncontrollable and unpredictable i.e., unknown; Carleton, 2016)” (Zavos, Gregory, & Eley, 2012, p. 205). There is also evidence suggesting FOTU appears transdiagnostic in children and adolescents (Freeston et al., 2016; Wright, 2016). FOTU has also accounted for variance beyond AS (Boelen & Reijntjes, 2009; Carleton, Collimore et al., 2010; Carleton, Sharpe et al., 2007; Dugas, Gosselin, & Landouceur, 2001), fear of anxiety (Buhr & Dugas, 2009), metabeliefs (de Bruin, Rassin, & Muris, 2007; Dugas et al., 2007), disorder-specific IU (Thibodeau et al., 2015), positive and negative affectivity (Carleton, Collimore et al., 2010; Thibodeau et al., 2012), and neuroticism (Boelen and Reijntjes, 2009, Mahoney and McEvoy, 2012b). That said, “neuroticism is not an explanatory concept in the aetiology of psychopathology, since it measures a person’s characteristic level of distress over a protracted period of time” (Ormel et al., 2004; p. 906). Hierarchical linear regression entry order recommendations suggest the most fundamental (i.e., lowest-order) construct or the temporally earliest construct should be accounted for first, with increasingly higher-order or temporally distal constructs entered thereafter (Petrocelli, 2003). As such, “controlling” for neuroticism or analogues thereof before controlling for the lower-order determinants (e.g., fundamental fears) of such higher-order constructs may be suboptimal; nevertheless, such assessments have stringently demonstrated that FOTU, as most commonly reflected by IU, accounts for variance in critical higher-order constructs. Intentional and targeted reductions in FOTU, as reflected by IU, have produced reductions in higher-order constructs. Perhaps the earliest and most recognized treatment protocol for reducing FOTU was designed by Dugas and Ladouceur (Dugas & Ladouceur, 2000; Ladouceur, Dugas et al., 2000). The treatment is specific to IU and effective for reducing GAD symptoms (Dugas and Ladouceur, 2000, Dugas and Robichaud, 2007, Ladouceur, Dugas et al., 2000), but also produces changes within increasingly higher-order constructs, such as worry or neuroticism (Buhr & Dugas, 2009; Dugas, Laugesen, & Bukowski, 2012; Ladouceur, Dugas et al., 2000; Ladouceur, Gosselin, & Dugas, 2000). As such, the evidence supports GAD as “the phenotypic expression of high levels of neuroticism” (Barlow, Ellard et al., 2014, p.488) and reducing FOTU may be a lynchpin for reducing neuroticism (Carleton, 2012, Carleton, 2016). Dugas and Ladouceur’s (2000) treatment protocol is the most established and prolific; however, there are alternatives. For example, a metacognitive therapy (van der Heiden, Muris, & van der Molen, 2012) has substantially reduced IU and symptoms in patients with GAD (ds = 0.94 to 2.39). Further, Dugas and Ladouceur’s delineated treatment (Dugas & Robichaud, 2007) formed the basis for adaptations beyond GAD. Reductions in IU appear to reduce obsessive compulsive disorder symptoms (Grayson, 2010, Wilhelm and Steketee, 2006); more recently, Hewitt and colleagues (Hewitt, Egan, & Rees, 2009) used Dugas and Robichaud’s (2007) IU section to reduce social anxiety symptoms in a case study (i.e., reliable change index scores > 1.96). Similarly, Mahoney and McEvoy (Mahoney & McEvoy, 2012a) provided a 7-week cognitive behavioural group therapy (CBGT) for persons with social anxiety disorder.

The therapy included “learning to tolerate uncertainty before, during, and after social situations was emphasized during treatment and integrated into core skills” (p. 851). Reductions in IU (r2 = 0.57) predicted subsequent reductions in depression (r2 = 0.37) and social anxiety symptoms (r2 = 0.61), implicating a causal relationship. Boswell et al. (2013) also supported Carleton’s (Carleton, 2012, Carleton, 2016) proposition that FOTU represents a critical transdiagnostic factor with evidence from the Unified Protocol for the Transdiagnostic Treatment of Emotional Disorders (Barlow, Ellard et al., 2011). The Unified Protocol includes an exposure element focused on interactions with unknowns. Patients with various disorders received the treatment in a 16-week randomized control trial wherein symptoms fell (i.e., Hedges g = 0.52 to 1.11; Farchione et al., 2012). Most recently, changes in IU have been associated with reductions in repetitive negative thinking (d = 1.25), depression (d = 1.09), social anxiety symptoms (d = 1.65 to 2.18), and GAD symptoms (d = 2.20; McEvoy & Erceg-Hurn, 2016), further underscoring the transdiagnostic importance of FOTU. Hong and Cheung (Hong & Cheung, 2015) further evidenced FOTU as a common factor for treating anxiety and mood disorders with a 73 article meta-analytic assessment of vulnerabilities for depression and anxiety. The meta-analyses examined relationships between “six cognitive vulnerabilities associated with depression (i.e., pessimistic inferential style, dysfunctional attitudes, and ruminative style) and anxiety (i.e., anxiety sensitivity, IU, and fear of negative evaluation)” (p.1). Their results supported FOTU as accounting for substantial variance in higher-order constructs (rs = 0.45 to 0.57) and underscored the primacy of FOTU as a vulnerability with the highest core factor loading of 0.81. “First, IU had the strongest factor loading—implying that a fundamental fear of the unknown (Carleton, 2012) may feature heavily in this common core. This element of unknown may encompass external environmental uncertainties and threats and an individual’s internally oriented uncertainty about his or her own resources to deal with such threats” (Hong & Cheung, 2015; p.13)

7. The undiscovered country—implications and future directions
Assuming acceptance of the postulates herein and before (e.g., Bach and Dolan, 2012, Carleton, 2012, Carleton, 2016, Hong and Cheung, 2015, Thayer et al., 2012), substantial research remains to be conducted. The extant literature base does not yet include direct empirical assessments of FOTU. Instead, such assessments have been made using related constructs including uncertainty, ambiguity, and novelty. Even then, much of the research has been based on cross-sectional designs with self-report as the primary dependent variable of interest. Future research should work towards experimental, longitudinal designs with multimodal dependent variables. A more comprehensive description of research limitations and future directions is available from Shihata, McEvoy, Mullan, and Carleton this issue. Furthermore, additional research can and should be conducted to test how well FOTU meets each of the propositional requirements (i.e., the eight criteria) for identifying a fundamental fear. There appears to be sufficient preliminary support, but each of the criteria included initial recommendations for producing stronger evidence to support a fear as fundamental (see Table 1). Despite the substantial neurobiological evidence (e.g., Balderston et al., 2013; Brosschot et al., 2016; Gray & McNaughton, 2003), much more remains to be conducted (see Brosschot et al., 2016; Grupe & Nitschke, 2013). Research exploring how FOTU interacts with automatic (Balderston et al., 2013; Cisler & Koster, 2010; Leyro, Zvolensky, & Bernstein, 2010; Mathews and MacLeod, 1994, Mathews and MacLeod, 2005) and elaborated (Grenier, Barrette, & Ladouceur, 2005) processing also appears well-warranted (Fergus, Bardeen, & Wu, 2013). Human research could also be augmented by animal studies exploring FOTU (e.g., Herry et al., 2007; Mishra, Logue, Abiola, & Cade, 2011). The research to date has also focused primarily on anxiety and related disorders, with less research exploring depression (e.g., Nelson et al., 2014, Yook et al., 2010); however, FOTU may impact all psychopathology as well as normative emotions and decision-making (Bach and Dolan, 2012, Carleton, 2012, Carleton, 2016, Hong and Cheung, 2015).

As such, researchers should further explore the influence of fearing the unknown on depression, somatic disorders (e.g., chronic pain; Carleton & Asmundson, 2012), eating disorders (Konstantellou & Reynolds, 2010; Sternheim et al., 2011), and personality disorders (Berenbaum et al., 2008, Fergus and Rowatt, 2014), as well as supporting normative and healthy responses. There may also be important differences in the impact of chronic small unknowns (e.g., those associated with daily hassles) and one or more major unknowns (e.g., whether a person will pass a critical exam). Explicit integration of FOTU into progressive updates of established models for psychopathologies (e.g., the shared vulnerability model; Asmundson, Coons, Taylor, & Katz, 2002) remains to be considered. Research needs to be conducted exploring associated neurological processes, memory, and self-awareness. There is also research to be done exploring interventions – transdiagnostic and otherwise – for FOTU, building on the seminal theory by Dugas and Ladouceur (2000), Freeston et al. (1994), Ladouceur, Dugas et al. (2000), as well as recent research and theory from Barlow, Sauer-Zavala et al. (2014), Boswell et al. (2013), Einstein (2014), Mahoney and McEvoy (2012a), and McEvoy and Erceg-Hurn (2016). For example, FOTU or associated constructs (e.g., IU) could be made explicit, rather than implicit, components of psychoeducation and explicit targets for exposures. The inevitability of unknowns can also be addressed, and patients can be provided tools to better manage interactions with unknowns (e.g., cognitive restructuring, improved probability assessments, mindfulness, acceptance). The current review and synthesis extends previous ones in several ways (Bach and Dolan, 2012, Carleton, 2012, Carleton, 2016, Hong and Cheung, 2015). First, a thorough rationale has been offered for identifying fundamental fears. Second, direct progressive comparisons have now been made with other candidate constructs. Third, the available empirical research supporting FOTU, much using IU, as meeting the criteria for a fundamental fear has been reviewed. Overall, the result is a robust theoretical foundation, replete with preliminary direct experimental evidence and references to extensive indirect experimental evidence with IU, for exploring FOTU as a fundamental fear underlying anxiety and neuroticism. In sum, FOTU may also be the fundamental fear – a fear that rules all other fears, brings them together producing anxiety, and binds them, facilitating anxiety and neuroticism….”