In the year 2371, the android robot Data from Star Trek: The Next Generation has an “emotion chip” installed in his brain, to aid in his quest to understand humanity. The experiment is successful to a fault: Data become overwhelmed with wildly positive, negative, and erratic feelings in response to pleasurable or stressful events. The lesson is clear and supported much further by the empirical evidence described below: Resilience trades off with sensitivity, even at the level of the brain. Why, then, are cognitive trade-offs virtually unstudied in psychology and psychiatry?

Trade-offs can be considered as balances between two beneficial but incompatible phenotypes. Resilience, as conceptualized by Kalisch et al., can be favorable, because it reduces cognitive and emotional sensitivity and vulnerability to stressful events; however, it may also reduce sensitivity to beneficial opportunities. In turn, sensitivity, as a state that trades off with resilience, increases both gains from opportunity and losses from threat. These two opposite concepts form the core of the “vantage sensitivity” model developed by Pluess and Belsky (Reference Pluess and Belsky2013), whereby some individuals are relatively more sensitive than others with regard to psychological responses to environmental events, whether those events are negative (as in diathesis-stress models) or positive (Fig. 1). Under this paradigm, less-sensitive individuals are simply more resilient. In contrast, the cognitive-resilience model of Kalisch et al., in its reliance on “any mechanism that helps the organism fine-tune stress responses to optimal levels … and remain flexible” (sect. 1.3), implicitly denies the existence of such cognitive trade-offs. So: How important are they?

Figure 1. Cognitive trade-offs under a vantage sensitivity model, whereby resilience engenders benefits in poor environments but costs in good ones.

At the genetic level, cognitive trade-offs are strongly supported by evidence showing that certain genotypes increase liability to psychopathology for individuals in poor environments but confer benefits to individuals in good environments (review in Pluess & Belsky Reference Pluess and Belsky2013). By contrast, individuals with alternative, “resilience” genotypes at these loci exhibit neither the costs of adversity nor the benefits of advantage. The well-known COMT Val¹⁵⁸Met polymorphism provides another case of trade-offs: The Met allele mediates lower flexibility, but increased stability, compared with Val (e.g., Markant et al. Reference Markant, Cicchetti, Hetzel and Thomas2014); strong trade-offs also have been demonstrated from these alleles for executive compared with emotional tasks (Mier et al. Reference Mier, Kirsch and Meyer-Lindenberg2010). Comparable results obtain from studies of human polymorphisms in mice: For example, mice bearing the autism-associated R⁴⁵¹C mutation exhibit impaired social interactions, but enhanced spatial learning (Tabuchi et al. Reference Tabuchi, Blundell, Etherton, Hammer, Liu, Powell and Südhof2007).

At the level of physiology, trade-offs are controlled by condition-dependent effects of hormones, and for some hormones, these influences extend to the brain. For example, intranasal oxytocin administration leads to reduced analytic thinking, but also increased “holistic processing, divergent thinking and creative performance” (De Dreu et al. Reference De Dreu, Baas, Roskes, Sligte, Ebstein, Chew, Tong, Jiang, Mayseless and Shamay-Tsoory2014, p. 1). Similarly, serum estradiol relative to testosterone exhibits a negative relationship with spatial ability, but a positive association with verbal fluency (Kocoska-Maras et al. Reference Kocoska-Maras, Rådestad, Carlström, Bäckström, von Schoultz and Hirschberg2013).

Finally, at the level of psychiatry, cognitive trade-offs can be analyzed by determining whether increased risks for one disorder coincide with decreased risks for another. For example, three well-documented factors confer protection from schizophrenia: large birth size (Byars et al. Reference Byars, Stearns and Boomsma2014), congenital blindness (Silverstein et al. Reference Silverstein, Wang and Keane2013), and duplications of the 22q11.2 copy number locus (Rees et al. Reference Rees, Kirov, Sanders, Walters, Chambert, Shi, Szatkiewicz, O'Dushlaine, Richards, Green, Jones, Davies, Legge, Moran, Pato, Pato, Genovese, Levinson, Duan, Moy, Göring, Morris, Cormican, Kendler, O'Neill, Riley, Gill, Corvin, Craddock, Sklar, Hultman, Sullivan, Gejman, McCarroll, O'Donovan and Owen2014). Each of these three factors that reduces schizophrenia risk also increases risk for autism (Byars et al. Reference Byars, Stearns and Boomsma2014; Hobson & Bishop Reference Hobson and Bishop2003; Rees et al. Reference Rees, Kirov, Sanders, Walters, Chambert, Shi, Szatkiewicz, O'Dushlaine, Richards, Green, Jones, Davies, Legge, Moran, Pato, Pato, Genovese, Levinson, Duan, Moy, Göring, Morris, Cormican, Kendler, O'Neill, Riley, Gill, Corvin, Craddock, Sklar, Hultman, Sullivan, Gejman, McCarroll, O'Donovan and Owen2014), providing evidence that these two disorders trade off in their causes and can be conceptualized as diametric (Crespi & Badcock Reference Crespi and Badcock2008). More generally, social abilities commonly trade off with spatial skills, in autism as well as neurotypical individuals (e.g., Keehn et al. Reference Keehn, Shih, Brenner, Townsend and Müller2013; Russell-Smith et al. Reference Russell-Smith, Maybery, Bayliss and Sng2012), and schizophrenia genetic risk is positively associated with higher verbal relative to spatial skills (Kravariti et al. Reference Kravariti, Toulopoulou, Mapua-Filbey, Schulze, Walshe, Sham, Murray and McDonald2006). Perhaps most important, these findings also suggest that some disorders themselves represent dysfunctions mediated by extremes of cognitive trade-offs, as between empathizing and systemizing in Baron-Cohen's (Reference Baron-Cohen2009) model for autism.

Bipolar disorder and depression represent paradigmatic disorders underlain by cognitive-affective sensitivities, in that overly positive appraisals of events and self-capabilities mediate the emergence of mania and hypomania, whereas overly negative appraisals mediate the onset and maintenance of depression (e.g., Beck Reference Beck2008; Lee et al. Reference Lee, Lam, Mansell and Farmer2010). In this framework, the positive-appraisal bases for resilience postulated by Kalisch et al. may, paradoxically, overlap with risks for mania and hypomania. Bipolar disorder can indeed be considered in terms of overly developed goal seeking, driven by high reward-sensitivity; however, it also is associated with pronounced enhancements, including elevated IQ (e.g., Koenen et al. Reference Koenen, Moffitt, Roberts, Martin, Kubzansky, Harrington, Poulton and Caspi2009) and high social and academic achievement (e.g., Johnson et al. Reference Johnson, Fulford and Carver2012; MacCabe et al. Reference MacCabe, Lambe, Cnattingius, Sham, David, Reichenberg, Murray and Hultman2010). Such benefits presumably accrue primarily to individuals who are highly motivated and sensitive to successes, but also fortunate enough to develop in a favorable environment.

The upshot of these considerations is that diverse evidence supports a model of resilience trading off with sensitivity, such that Kalisch et al.'s quest for purely beneficial neural resilience to stress-induced mental disorders becomes challenging at least, and at most, quixotic. Where do these considerations leave us, with regard to reducing risks for such disorders?

First, increased resilience certainly can be fostered among high-sensitivity individuals beset by environmental stress, once we know how. Determining the mechanisms for resilience, by comparing neural and cognitive phenotypes across resilient genotypes for multiple differential-susceptibility loci, offers a simple way forward.

Second, we must improve our understanding of cognitive trade-offs, by realizing that many psychological deficits are intrinsically linked with corresponding strengths, and that psychiatric risk genotypes of many genetic polymorphisms also should confer benefits. Such strengths, and benefits, will be overlooked if the study of psychopathology continues its usual litany of characterizing dysfunctions rather than testing for trade-offs.

With the help of the psychological counselor Deanna Troi, the android Data eventually develops some measure of control over his emotions. His emotion chip is later removed, however, because it renders him vulnerable to confusion, fear, depression, rage, and manipulation by others. He learns, as may we someday, that the costs of sensitivity, like the costs of resilience, can sometimes exceed the advantages.