Original Article
Spiders are special: fear and disgust evoked by pictures of arthropods

https://doi.org/10.1016/j.evolhumbehav.2008.08.005Get rights and content

Abstract

Because all spiders are predators and most subdue their prey with poison, it has been suggested that fear of spiders is an evolutionary adaptation. However, it has not been sufficiently examined whether other arthropods similarly elicit fear or disgust. Our aim was to examine if all arthropods are rated similarly, if only potentially dangerous arthropods (spiders, bees/wasps) elicit comparable responses, or if spiders are rated in a unique way. We presented pictures of arthropods (15 spiders, 15 beetles, 15 bees/wasps, and 15 butterflies/moths) to 76 students who rated each picture for fear, disgust, and how dangerous they thought the animal is. They also categorized each animal into one of the four animal groups. In addition, we assessed the participants' fear of spiders and estimates for trait anxiety. The ratings showed that spiders elicit significantly greater fear and disgust than any other arthropod group, and spiders were rated as more dangerous. Fear and disgust ratings of spider pictures significantly predicted the questionnaire scores for fear of spiders, whereas dangerousness ratings of spiders and ratings of other arthropods do not provide any predictive power. Thus, spider fear is in fact spider specific. Our results demonstrate that potential harmfulness alone cannot explain why spiders are feared so frequently.

Introduction

For humans, spiders are among the top five most feared animals; in the UK, about 30% of women and 20% of men are anxious, nervous, or frightened when confronted with a spider (Davey, 1994a). A specific phobia (see Hofmann, Alpers, & Pauli, 2009) of spiders has been documented to be the most prevalent phobia related to animals, with a prevalence rate of 3.5% (Jacobi et al., 2004). Moreover, spiders are preferentially processed in the visual system by those who fear them (e.g., Alpers et al., in press, Gerdes et al., 2008).

The observation that some stimuli, such as spiders and snakes, are feared with higher probability than other animal groups or objects is often explained in terms of biological preparedness (Seligman, 1971). This hypothesis is based on the fact that most spiders and many snakes are predatory animals and possess venom to immobilize their prey. They are thus thought to have been potentially hazardous for our pretechnological ancestors. Seligman defined prepared fear learning by the following criteria: learning is specific to selective cues, is noncognitive, is highly resistant to extinction, and can be acquired in one trial. For example, laboratory-raised rhesus monkeys are less afraid of snakes compared with those raised in the wild (Mineka, Keir, & Price, 1980), but they acquire this fear easily (Mineka, Davidson, Cook, & Keir, 1984) and retain it (Mineka & Keir, 1983, Mineka et al., 1980).

Further support for the preparedness hypothesis comes from conditioning experiments (Öhman, Erixon, & Lofberg, 1975). When pictures of spiders and snakes served as conditioned stimuli that predicted mildly aversive shock, participants showed stronger and lasting skin conductance responses (an index of emotional activation) than when flower or mushroom pictures were paired with the shock. In addition, the conditioned fear to these stimuli was significantly more resistant to extinction (Öhman et al., 1975). Based on the resistance to extinction documented for experimentally acquired responses (McNally, 1987), the preparedness hypothesis has been widely accepted and entered almost every textbook on biological psychology and abnormal psychology, although evidence for other characteristics of preparedness (ease of acquisition, irrationality, and belongingness) is much more limited (McNally, 1987).

From an evolutionary perspective (for overviews, see Nesse, 1990), readily associating spiders with fight-or-flight responses should have a selective advantage for humans because spiders are potentially hazardous. Indeed, most arthropods, such as arachnids (e.g., spiders and scorpions) and hymenoptera (bees, wasps, ants), are venomous. However, although arthropods make up 75% of the world's animal species, only few come in direct contact with humans and even fewer cause significant medical problems. As to spiders, of all known species (about 38,000), very few (approximately 0.1%–0.3%) are considered to cause significant morbidity or mortality (Foelix, 1996, Maretić, 1987, Steen et al., 2004). Since spiders generally prey upon insects or other spiders, their venom has not evolved to harm large vertebrates such as humans. Spiders rarely use venom in response to vertebrates for defense and generally do so as a last resort (Foelix, 1996). Moreover, most studies of spider bites have been retrospective and bites have not been confirmed by eyewitnesses (Diaz, 2004). For example, 80% of suspected cases of spider bites in Southern California were caused by other arthropods, mostly ticks and reduviid bugs (Russel & Gertsch, 1982).

The other potentially hazardous arthropod group is the order Hymenoptera (bees, wasps, ants). Many have evolved highly specialized poison glands that are generally used for defense and, by some, to overcome prey. Overall, even potent honeybee or wasp venom did not evolve to kill but to repel large vertebrates trying to intrude onto nests (Schmidt, 1990). Hence, the response to the venom is very fast and partly caused by activated pain receptors that cause pain out of proportion to the wound inflicted (Schmidt, 1990). Although the dose that causes mortality in 50% of nonallergic human individuals (lethal dose=50) was estimated to be 500–1500 bee stings (Camazine, 1988, Michener, 1975), Mejia, Arbelaez, Henao, Sus, and Arango (1986) reported that receiving more than 1000 stings can indeed be lethal. Stings of bees and wasps are a significant hazard because of their high incidence and ability to produce fatal anaphylactic reactions or respiratory dysfunctions at least in hypersensitive humans (Habermann, 1974, Habermehl, 1987). Compared with spider stings, bee and wasp stings are aggravated by the facts that humans encounter bees and wasps more often, encounters usually occur in swarms (Maretić, 1987, Schmidt, 1990), and encounters often occur near food sources.

Despite the low mortality risk from stings, spiders and bees/wasps (Hymenoptera) still differ from other groups of small animals in terms of potential harmfulness, which may have resulted in a selective advantage for avoidance responses and increase in fear. Surprisingly, there hardly exists any information on fear of arthropods other than the highly prevalent fear of spiders. A small number of single case reports show that these fears (e.g., fear of wasps or other insects) exist but that their prevalence may be very low (Brown et al., 2003, Elsesser et al., 2006, Jones & Friman, 1999). Contrary to this observation, the underlying rationale of the preparedness hypothesis would lead us to expect a stronger overlap of fear of spiders and other venomous arthropods. Alternatively, disgust instead of fear may play a special role in the common aversive behavior toward spiders (Davey, 1992). The disgust hypothesis postulates that emotional responses to spiders are culturally transmitted (Davey, 1994b) because these animals were historically associated with disease and infection from medieval times onward. However, it is unclear why mainly spiders, and not other “creepy crawlies,” have been considered to be responsible for infections and disease.

Contrary to spiders, some beetles infest food items, and preparedness for aversive responses to them, specifically the experience of disgust, is plausible. Taken together, it has not been convincingly documented that specific spider cues should be prepared for conditioning of fear or disgust in humans. Other arthropods that are comparable in terms of venomousness, appearance, or behavior to spiders may elicit similar reactions, but cultural transmission may exert strong biases on verbal labeling. Individuals who report being afraid of spiders may stick with a cultural stereotype (“fear of spiders is common”), although their fears may be much less specific than commonly thought. A variety of arthropods may elicit fear or disgust (e.g., beetles), but “fear of spiders” may merely be a culturally accepted verbal label for a wide spectrum of animal fears (see Wenegrat, 2001).

A necessary first step is therefore to investigate if different kinds of arthropods are perceived similarly, if only a subset (i.e., poisonous ones such as spiders and bees/wasps) elicits comparable responses, or if spiders are special. To this end, we compared ratings of fear and disgust elicited by grayscale pictures of spiders (Araneae), bees and wasps (Hymenoptera), beetles (Coleoptera), as well as butterflies and moths (Lepidoptera). Because it is not known as to what extent individuals are aware of the degree of harmfulness of the depicted animals, we collected estimates of harmfulness for each animal. Possible discrepancies between high fear ratings and low estimates for harmfulness could be interpreted as evidence for the irrationality of a fear (Seligman, 1971). Because individuals with high levels of fear of spiders (i.e., phobic fear) may be particularly prone to generalize all creepy crawlies as fear evoking, we also assessed self-reported fear of spiders separately from the picture ratings.

Section snippets

Participants, materials, and rating procedure

Seventy-six entry-level psychology students (22.4% were male, n=17) participated in the investigation [age: mean±SD=21.44±3.20 years, range=19–41 years; one participant did not indicate his/her gender]. We projected 60 grayscale pictures of arthropod species [15 spiders (Araneae), 15 beetles (Coleoptera), 15 bees/wasps (Hymenoptera), and 15 butterflies/moths (Lepidoptera)] in a lecture hall (See Appendix A for a complete species list). The gray-scaled pictures depicted a variety of phenotypes

Classification

Almost all pictures were classified correctly (Table 1). There was no significant difference in correct classification between animal groups (Kruskal–Wallis test: χ2=1.84, N=304, p=.606). For the following analyses, we removed all misclassified items, resulting in a total number of 4339 ratings instead of 4560.

Ratings

Mean ratings of fear, disgust, and danger for all four animal groups are shown in Table 1. The multivariate ANOVA revealed that there was a significant main effect of animal group on each

Discussion

The preparedness hypothesis has been used to explain why fear of certain animals (typically snakes and spiders) is very common. Although there is empirical evidence for this perspective from conditioning experiments comparing pictorial representations of these animals with benign objects, there are surprisingly few studies comparing responses with other animal groups. The ratings of subjective responses to a range of arthropods enabled us to examine some of the basic tenets of the preparedness

Acknowledgments

We thank the students who participated in this study. We also thank Mario Fix for his help in adjusting the pictures' characteristics.

References (53)

  • WoodyS.R. et al.

    Disgust as a motivator of avoidance of spiders

    Anxiety Disorders

    (2005)
  • Alpers, G. W., Gerdes, A. B. M., Lagarie, B., Tabbert, K., Vaitl, D., & Stark, R. (in press). Attention modulates...
  • ArmfieldJ.M.

    Understanding animal fears: A comparison of the cognitive vulnerability and harm-looming models

    BMC Psychiatry

    (2007)
  • BarrettH.C.

    Adaptations to predators and prey

  • Bennett-LevyJ. et al.

    Fear of animals: What is prepared?

    British Journal of Psychology

    (1984)
  • CamazineS.

    Hymenopteran stings: Reactions, mechanisms and medical treatment

    Bulletin of the Entomological Society of America

    (1988)
  • CoddingtonJ.A. et al.

    Systematics and evolution of spiders (Araneae)

    Annual Review of Ecology and Systematics

    (1991)
  • CordainL. et al.

    Origins and evolution of the Western diet: Health implications for the 21st century

    American Journal of Clinical Nutrition

    (2005)
  • CraskeM.G.

    Origins of phobias and anxiety disorders: why more women than men?

    (2003)
  • DaundererM.

    Lexikon der Pflanzen- und Tiergifte. Diagnostik und Therapie

    (1995)
  • DaveyG.C.

    Self-reported fears to common indigenous animals in an adult UK population: The role of disgust sensitivity

    British Journal of Psychology

    (1994)
  • DaveyG.C.L.

    The “disgusting” spider: The role of disease and illness in the perpetuation of fear of spiders

    Society & Animals

    (1994)
  • DiazJ.H.

    The global epidemiology, syndromic classification, management and prevention of spider bites

    American Journal of Tropical Medicine and Hygiene

    (2004)
  • ElsesserK. et al.

    Attentional bias and evoked heart-rate response in specific phobia

    Cognition and Emotion

    (2006)
  • FoelixR.F.

    Metabolism

  • HabermannE.

    Pharmakologisch bedeutsame Inhaltsstoffe von Bienen- und Wespengiften

    Pharmazie in unserer Zeit

    (1974)
  • Cited by (112)

    • An ethologically based view into human fear

      2023, Neuroscience and Biobehavioral Reviews
    • Fear one, fear them all: A systematic review and meta-analysis of fear generalization in pathological anxiety

      2022, Neuroscience and Biobehavioral Reviews
      Citation Excerpt :

      Although the terminology of specific phobia suggests fear to be limited to concrete and circumscribable stimuli or situations, the specificity of phobias has been questioned before as individuals often report suffering from more than one phobia (Hofmann et al., 1997). For example, in line with this, spider fearfulness is associated with increased fear and disgust of similar animals, such as wasps, beetles, and even with butterflies (Gerdes et al., 2009). Also, spider fearful individuals report more negative valence ratings and are hypervigilant in response to schematic pictures of spiders (Kolassa et al., 2006, 2007).

    View all citing articles on Scopus

    We acknowledge the support given by the Research Group Emotion and Behavior, which is sponsored by the German Research Society (DFG), and its speakers Fritz Strack and Paul Pauli.

    1

    A.B.M. Gerdes and G. Uhl contributed equally to this work.

    View full text