Conceptualization, context, and comparison are key to understanding the evolution of fear

Abstract

The fearful ape hypothesis proposes that heightened fearfulness in humans is adaptive. However, despite its attractive anthropocentric narrative, the evidence presented for greater fearfulness in humans versus other apes is not sufficient to support this claim. Conceptualization, context, and comparison are strongly lacking in Grossmann's proposal, but are key to understanding variation in the fear response among individuals and species.

Grossmann presents the fearful ape hypothesis, claiming that from infancy humans are more fearful than other apes, and that “in the context of cooperative caregiving and provisioning unique to human great ape group life, heightened fearfulness was adaptive” (target article, abstract). We consider Grossmann's hypothesis to represent a false premise. We question the evidence provided to demonstrate heightened fearfulness in humans, which is critical to the author's argument. In the absence of such evidence, the proposed hypothesis does not take us beyond adaptationist storytelling (Gould & Lewontin, 1979) and we believe it should currently be treated as such. We argue that conceptualization, context, and comparison are strongly lacking in Grossmann's proposal, but are key to understanding variation in the fear response among individuals and species.

Grossmann discusses heightened fearfulness in humans without any explicit conceptualization or definition of how we should measure this multifaceted and complex emotional state (Boissy, 1995). While subjective conscious experience of emotion may be asked verbally of humans, this is not possible in pre-verbal infants, humans without language, or other animal species. Therefore, proxy measurements of emotional processes (including fear) have been used, including behavioural, neurobiological, physiological, and cognitive responses of individuals to a variety of different stimuli (Paul, Harding, & Mendl, 2005). Such methods offer important insights into the evolution of emotion. However, the results can be very difficult to interpret and judgements concerning affective states are stronger if based on multiple techniques and/or correlated measures, with appropriate controls (Miller, Garner, & Mench, 2006; Safryghin, Hebesberger, & Wascher, 2019). Grossmann cites a single study (Herrmann, Hare, Cissewski, & Tomasello, 2011) as evidence for heightened fearfulness in humans. However, that study (Herrmann et al., 2011) uniquely examines the behavioural responses of infant humans and apes to specific novel objects as a measure of temperament (shyness–boldness, or neophobia). Interpretation of this methodological paradigm has recently been criticized (Takola, Krause, Müller, & Schielzeth, 2021), and the inference that affective states such as fear may be judged by the behavioural response to novel objects alone is not supported by the current literature (Crane, Brown, Chivers, & Ferrari, 2020; Greggor, Thornton, & Clayton, 2015). A single study, based on a small number of “WEIRD” (Henrich, Heine, & Norenzayan, 2010) and “STRANGE” (Webster & Rutz, 2020) subjects (24 bonobos, 24 chimpanzees, 24 orangutans, and 14 German children), and which uses a single behavioural measure, is simply not sufficient to establish the key evidence for Grossmann's claim.

Grossmann also characterizes fearfulness (measured as neophobia) as a species-level trait, which does not vary across individuals or contexts. However, responses to novelty are known to be highly context dependent, varying with a range of social, cultural, environmental, and developmental factors (Coleman & Mellgren, 1994; Forss, Koski, & van Schaik, 2017; Greggor, Clayton, Fulford, & Thornton, 2016; Miller, Bugnyar, Pölzl, & Schwab, 2015; Moretti, Hentrup, Kotrschal, & Range, 2015; Webster & Lefebvre, 2000). For example, food neophobia in children varies considerably across differing cultural environments (Birch, 1995). Furthermore, some species show high individual repeatability in their responses over time, and to different types of novelty, whereas other species are seemingly more flexible (Miller et al., 2022; Vernouillet & Kelly, 2020). More broadly, ecological factors such as resource availability (Greggor et al., 2016) and, in particular, predation pressure (Isbell, 2006; Kawai & Koda, 2016; Öhman, 2009) are known to play a significant role in both intra- and inter-specific variation in fear responses (Mobbs, Hagan, Dalgleish, Silston, & Prévost, 2015; Pellman & Kim, 2016); often conceptualized as “the ecology of fear” (e.g., Laundre, Hernandez, & Ripple, 2010). Context can define whether neophobia is adaptive for an individual in the evolutionary sense. For instance, fear responses stemming from animals' encounters with predators can be beneficial to an individual in terms of increasing its survival (Mobbs et al., 2015; Pellman & Kim, 2016), or may influence adaptive evolutionary phenomena such as the development of enhanced visual systems (Isbell, 2006). On the other hand, fear responses can also be costly in terms of leading to poor innovation, competitive ability, or increased stress-related mortality (Crane et al., 2020). We therefore consider that the links between socioecological context and fear response behaviours, and their (mal)adaptive effects on free-living animals, are key considerations before any conclusions about the adaptiveness of fearfulness can be drawn.

Finally, although Grossmann places a strong emphasis on “phylogenetic origins” as evidence to support the fearful ape hypothesis, the evidence presented is highly hominoid-centric and no formal comparative analyses are conducted. There is no reference to non-ape species, despite the large literature on neophobia and fear responses in many other clades (Crane et al., 2020; Laundre et al., 2010). As behavioural ecologists, we find this disappointing, as neither fearfulness, nor cooperative care (which is distinct from cooperative breeding, as Grossmann points out; Clutton-Brock, 2006; König, 1997), are unique to humans. There are many examples of fear (as measured by neophobia, shyness) and recognition of emotions, for example, in response to predators, in several other primate species. Likewise, there is a large body of evidence for cooperative care, sometimes even independently of kinship (Clutton-Brock, 2006; Koenig & Dickinson, 2016). The socioecological drivers of neophobia and the role of individual differences have also been well studied in corvids (Miller et al., 2022; Vernouillet & Kelly, 2020). Indeed, some species show high levels of both neophobia and cooperative care (Horn et al., 2020), which would seem to be highly relevant to the proposed hypothesis. Comparative analyses have also been reported for other families of birds, as well as for ungulates (Blumstein, 2006; Mettke-Hofmann, Winkler, & Leisler, 2002; Schaffer et al., 2021), but are not considered in Grossmann's review. There is a growing corpus of data on this topic from which to develop more robust hypotheses and more formal empirical analyses. Therefore, a broader comparative analysis, controlling for phylogenetic relatedness, would surely provide much greater insight into the evolution of fear responses.