Annotated Bibliography

Current Biology

Magazine

Comparative t hanatology

James R. Anderson

What is comparative thanatology? The scientifi c study of death and dying — thanatology — is an interdisciplinary fi eld of research that includes biological, medical, forensic, psychological and social perspectives on the end of individuals’ lives. Although some defi nitions restrict thanatology to the study of humans, there has long been interest in the effect of the death of a conspecifi c on the emotions and behavior of other species. How do other animals respond behaviorally and psychologically to dead or dying individuals? How are their responses infl uenced by the identity of the individual — for example, a same group-member or a relative — or by the cause of death? To what extent do other species share aspects of humans’ concept of death? Do they grieve? Such questions fall within the realm of comparative thanatology, a term that was used as early as 1926 in a review of the role of environmental factors in high death rates in marine organisms.

It is clear that not all topics in human thanatology apply to the study of responses to death and dying in other species. For example, there is no evidence that any species other than humans engage in post-death rituals that include washing and

Quick guide

Figure 1. An adult female chimpanzee inspects Photo credit: Katherine Cronin and Edwin van Le

adornment of deceased individuals, and religious observances do not exist in other species. A commonly held view is that only humans have a concept of death, but advances in knowledge about cognitive abilities in other species challenge this categorical position. Human adults’ concept of death typically entails an understanding of four components: inevitability, that all living organisms will eventually die; irreversibility, that a dead individual cannot come back to life; non-functionality, that a dead individual cannot perceive, feel, think, or act; and causality, that death follows a breakdown of one or more vital bodily functions. Research on children indicates age-related differences in their understanding of these four components, with general agreement that an adult-like concept of death is attained by around 10 years of age.

One obstacle to comparing humans’ and nonhumans’ understanding of death is that studies on humans typically rely on language, with participants explaining, responding to questions or (for children) selecting pictures in response to questions. There have been a few attempts to communicate linguistically about death with great apes trained to use American Sign Language. The chimpanzee Washoe’s baby was removed after an ultimately fatal illness. When researcher Roger Fouts returned, Washoe looked at him and signed: “Baby?” to which Fouts replied: “Baby dead, baby gone, baby fi nished”. According to Fouts: “Washoe dropped her cradled arms to her lap. She moved over to a far corner

her dead infant (left panel), and leaves the dead euwen at the Chimfunshi Wildlife Orphanage Trus

Current Biology 26, R543–R5

and looked away, her eyes vacant” (p. 224). Most scientists would rightly hesitate to accept the above interaction as strong evidence that great apes had a human-like concept of death; the possibilities of overly rich interpretation appear obvious. Although issues surrounding interpretation also arise in cases of more naturalistic behavioral observations of responses to death and dying individuals, such observations provide a more solid basis for a comparative thanatology.

Recent years have seen a growing literature on responses to the death of strangers or familiar individuals in a wide range of species. Observational and experimental studies of how social insects and birds respond to dead colony members have revealed several fascinating response patterns, including ‘necrophoresis’ in insects (see below). Furthermore, observational accounts of behavior shown towards ailing and dead individuals in free-ranging great apes, elephants, and cetaceans raise the distinct possibility that grief in response to bereavement is not uniquely human. I will focus here on studies of death and dying in insects, birds, and nonhuman primates.

What is necrophoresis? Since ancient times, it has been known that social insects including bees, ants and termites systematically remove dead colony members from the communal nest, a process termed necrophoresis. Dead conspecifi cs encountered outside of the nest may also elicit a reaction, for example ants returning to their nest may briefl y investigate a recently killed

infant on the ground while she eats (right panel). t, Zambia.

76, July 11, 2016 © 2016 Elsevier Ltd. R553

Current Biology

Magazine

ant and then run away. Individuals that have died within the nest are quickly removed and deposited elsewhere. Early anthropomorphic accounts of ant ‘funerals’ and ‘cemeteries’ eventually gave way to scientifi c studies of necrophoric and related behaviors, which in turn led to the discovery of specialized ‘undertakers’ within colonies. For example, although dead colony members may be explored and manipulated by many individuals inside the nest, certain workers — undertakers — appear specialized in quick and effi cient removal of corpses. An undertaker bee will seize the corpse in its mandibles, drag it to the exit of the hive, and then fl y up to 100 metres before dropping it or until the extra weight forces the undertaker to the ground, where the corpse may be abandoned.

Corpses are removed from the nest faster than other kinds of waste materials in both bees and ants; depending on species the latter may discard dead conspecifi cs on refuse piles or in special chambers; they may even eat them (termites are also known to engage in such cannibalism). In some cases ants and termites cover corpses with soil and vegetation fragments, a kind of burial behavior that is an effective strategy for avoiding disease from a pathogen outbreak in the nest. Honeybees, ants and termites all show some degree of fl exibility in their responses to dead conspecifi cs, with avoidance, speed of removal, cannibalism and burial infl uenced by cause of death and time elapsed since death.

Despite this fl exibility, however, removal of dead individuals from the hive or nest is largely a stereotyped behavior. The response is triggered by fatty acids, notably oleic acid, released by decaying corpses. That this substance acts as a classical ethological releasing stimulus is shown by the demonstration that a spot of oleic acid applied to any object — even a healthy nest mate — will result in the latter being ejected from the hive or nest as if it were dead. In many cases, however, corpses are removed before signifi cant decomposition occurs, suggesting that there is an alternative releasing stimulus. In one study, chemical compounds present in live ant cuticle were virtually absent in dead ants after one hour, suggesting that chemical signals from live workers inhibit

R554 Current Biology 26, R543–R576, July 1

inappropriate ejection from the nest. Degradation of these chemicals shortly after death ensures rapid disposal of corpses and thus further reduces the risk of pathogen infection. In another study, ‘corpse-carrying’ ants tended to remain outside the nest or congregated near the entrance when inside, which may be another mechanism to limit intra-nest pathogen transmission. There is no evidence that necrophoric activities are accompanied by anything that would be recognized as grieving.

Do birds hold ‘funerals’? The Merriam- Webster online dictionary defi nes a ‘funeral’ as “the observances held for a dead person usually before burial or cremation”. ‘Observance’ may be “a customary practice, rite, or ceremony”, or simply “an act or instance of watching”. The word ‘funeral’ is sometimes used to describe responses towards dead conspecifi cs in social insects and in birds, particularly corvids. There is little doubt that corpses elicit special attention in some species. Ravens and crows may gather around a dead conspecifi c and call loudly, attracting others to the vicinity. In one study, western scrub jays’ calls on detecting a dead jay lying near a feeding station attracted neighboring jays, resulting in ‘cacophonous aggregations’.

Cacophonous aggregations were also elicited by a stuffed owl, which was mobbed. A dead, prostrate jay always elicited cacophonous aggregations, whereas a stuffed, upright jay elicited fewer such reactions, but it was also mobbed; the prostrate jay remains were never mobbed. In that study and another in which American crows were exposed to dead conspecifi cs, foraging near the corpse was disrupted. Furthermore, crows responded in much milder fashion to a dead pigeon than to a dead conspecifi c. Unlike crows, pigeons paid little attention to dead conspecifi cs.

It seems clear that for members of the corvid family, discovering a dead conspecifi c elicits behavior that could result in obtaining information about potential risks in the immediate environment. Conceivably they might also try to identify the dead individual, as such information could have implications for mating and territorial opportunities. In support of the information-gathering hypothesis, increased hippocampal and cerebellar

1, 2016

activity occurs in crows watching a person holding a dead conspecifi c. By contrast, what has not been shown in these studies is that the birds are mourning — feeling or expressing sorrow for the dead individual. Nor have observations been made of birds disposing of dead individuals either by burial or some other solution, which means that ‘funeral’ applies only in a loose sense to corvids’ reactions to dead conspecifi cs. Studies of responses to death of a partner in pair-bonded species might clarify the issue of grief in birds. But it is in the nonhuman primates that we see the strongest evidence for this emotional reaction to death.

Do bereaved primates grieve? There appears little reason not to attribute to monkeys and apes the capacity to grieve following the permanent loss of a friend or family member. Experimental studies carried out on young monkeys several decades ago demonstrated a typical two-phase reaction to prolonged, involuntary separation from a primary attachment fi gure — usually the mother. First, the youngsters showed intense agitation, moving around and vocalizing, behavior that normally would result in the mother and separated offspring regaining contact. But in the case of prolonged separations, after several hours of what was termed ‘protest’ behavior, the young monkeys’ behavior changed dramatically, from high activity levels to silent immobility, sitting in a hunched posture, with reduced interest in the environment. This latter phase was interpreted as the onset of a depressive state, or ‘despair’, an interpretation that received support from studies that included physiological measures and effective treatment with anti-depressant drugs.

It seems logical to expect that similar behavioral and emotional outcomes would occur following the loss of an attachment fi gure in wild primates. Reports suggest that this is indeed the case. But if suitable alternative social partners are available, both captive and wild primates can be buffered from the emotional upheaval caused by the death of friends or family members. Although the outlook for unweaned orphans in the wild is not good, some orphans may survive following adoption by older individuals; the latter are often but not always kin.

Current Biology

Magazine

Psychologists Dorothy Cheney and Robert Seyfarth have described how wild adult female chacma baboons in Botswana experience signifi cantly elevated glucocorticoid levels — an indicator of stress — following the loss of a close relative to a predator; females in the same group with no equivalent loss show no such increased stress. Shortly after the predation event, bereaved females (‘bereaved’ is those authors’ description) engage in more social grooming with a greater number of partners than before. The authors suggest that by extending their social network, bereaved females compensate for the traumatic loss of a signifi cant grooming partner. This extended social network may facilitate emotional readjustment: by the second month following the predation event the females’ stress hormone levels return to normal.

There are few accounts of successful predatory attacks on great apes and their effects on survivors. In the Taï Forest, following a fatal attack by a leopard on an adolescent female chimpanzee, other chimpanzees present vocalized loudly and males engaged in aggressive displays, with some males dragging the dead body over short distances. Although adults also gently contacted and groomed the corpse, infants were prevented from approaching it. Aggressive displays and dragging the corpse were also observed in Gombe chimpanzees when they encountered a dead adult female member of their community. Several individuals approached and briefl y investigated the dead female, including touching it; females and older males did this less frequently than younger males. The dead female’s six-year- old daughter was the only female to touch the body. By contrast, when an adult male died after falling from a tree, other Gombe chimpanzees vocalized loudly, mutually embraced, and closely inspected the body, but nobody touched it. In stark contrast to these cases, in a small captive group of chimpanzees in which an old female died following illness, apart from brief aggressive displays by an adult male that involved stamping on the body, no post-mortem contacts occurred. In the minutes leading up to the old female’s death, however, the same chimpanzees did gently contact and manipulate her.

Taken together, these kinds of observations indicate variability in the responses of adult chimpanzees to the death of a member of their community. Infl uencing factors may include age and sex of both survivors and the dead individual, cause of death, and social status. Although there is little in the above descriptions that would clearly indicate grief, extrapolation from the baboon data would support the idea that close relatives of the deceased at least would experience a period of stress that in humans would be described as grief. Following the peaceful death of the old captive female chimpanzee, the other group members’ sleep was disturbed, two of them showed increased social grooming, and the dead female’s adult daughter spent the night sleeping near to her mother’s corpse; in the cases of wild chimpanzees described above, the corpse was abandoned after several hours. Intriguingly, before they leave the dead body of conspecifi cs, chimpanzees in the Taï Forest sometimes cover them with leafy branches that they detach and drop. The precise meaning of this behavior remains to be elucidated. Chimpanzees share neuro-endocrinological circuits that in humans are activated in emotional states including grief, and research has shown that they have mental representations of other individuals as well as self- recognition. Combined with descriptions of their responses to death, these facts strongly support the view that these primates can and probably do grieve.

Do adult females grieve for dead infants? One of the most widespread and reported death-related phenomena in nonhuman primates is the continued transportation of and caring for dead infants by those infants’ mothers. These behaviors have been described in many species: although less common in prosimians, they are often seen in monkeys (especially Old World species) and apes (Figure 1). Dead infant-directed maternal behavior, including transport, grooming, and protection, can last from a few hours to several months; in the latter case the remains of the infant may bear little resemblance to a live individual.

Despite hundreds of cases having been observed, the mechanisms underlying continued maternal investment in dead infants and its

Current Bi

emotional correlates remain to be elucidated. While the hormonal state of the mother undoubtedly contributes — she is in a physiological and mental state of readiness to behave maternally and she still has the target object, though it is no longer animate — it cannot explain all facets. For example, in some cases the dead infant will be abandoned well before the mother’s hormonal state changes signifi cantly as a result of her loss, whereas in other cases maternal behavior may persist beyond the resumption of menstrual cycles.

There remain many unanswered questions regarding this intriguing phenomenon, including how the age of the infant and indeed the age or parity of the mother infl uence the likelihood of the behavior. One possibility is that strength of pre-death mother– offspring attachment infl uences the intensity of the mother’s emotional response to the offspring’s death and hence the probability of continued maternal behavior. Some reports have indicated that mothers who retain dead offspring show behavioral signs of depression, so might more strongly attached mothers show more evidence of depression, or at least a greater negative effect? Again, physiological monitoring of such females throughout their period of maternal caretaking and transport would be valuable in helping to confi rm the existence of grieving.

Conclusion. From the literature reviewed above, it is possible to posit a spectrum of responses to death in animals. At one end of the spectrum are mechanistic, hard-wired, functional responses that are likely performed in the absence of any conscious emotional components, as exemplifi ed by necrophoresis in social insects. At the other end of the spectrum we see complex, socially malleable patterns that are likely to incorporate emotional states including sadness and grief. The latter patterns are a result of activation of psycho-neuro-endocrine systems that are largely shared with our nearest evolutionary neighbors, the anthropoid primates; it is therefore not surprising that the strongest case for bereavement-induced grief may be seen in monkeys and apes.

Other authors have made strong arguments for extending the capacity to grieve to other large-brained, socially complex mammals, notably

ology 26, R543–R576, July 11, 2016 R555

Current Biology

Magazine

elephants and cetaceans. But what of the four components of human adults’ understanding of death that were mentioned earlier: inevitability, irreversibility, non-functionality, and causality? Evidence appears strong enough to suggest that for large-brained animals that have some capacity for mental attribution, with repeated experience of death in their society, the second and third components apply. However, it remains unknown whether they are capable of understanding that all creatures die — including themselves — and understanding the biological causes of death.

Where can I fi nd out more? Anderson, J.R. (2011). A primatological perspective

on death. Am. J. Primatol. 73, 410–414. Anderson, J.R., Gillies, A., and Lock, L.C. (2010).

Pan thanatology. Curr. Biol. 20, R349–R351. Archer, J. (1999). The Nature of Grief. (London:

Routledge.) Biro, D., Humle, T., Koops, K., Sousa, C., Hayashi, M.,

and Matsuzawa, T. (2010). Chimpanzee mothers at Bossou, Guinea carry the mummifi ed remains of their dead infants. Curr. Biol. 20, R351–R352.

Boesch, C. (2012). Wild Cultures: A Comparison Between Chimpanzee and Human Cultures. (Cambridge: Cambridge University Press.)

Cheney, D.L., and Seyfarth, R.M. (2007). Baboon Metaphysics: The Evolution of a Social Mind. (Chicago: University of Chicago Press.)

Choe, D.-H., Millar, J.G., and Rust, M.K. (2009). Chemical signals associated with life inhibit necrophoresis in Argentine ants. Proc. Natl. Acad. Sci. USA 106, 8251–8255.

Cronin, K.A., van Leeuwen, E.J.C, Mulenga, I.C., and Bodamer, M.D. (2011). Behavioral response of a chimpanzee mother toward her dead infant. Am. J. Primatol. 73, 415–421.

Cross, D.J., Marzluff, J.M., Palquist, I., Minoshima, S, Shimizu, T., and Miyaoka, R. (2013). Distinct neural circuits underlie assessment of a diversity of natural dangers by American crows. Proc. R. Soc. Lond B 280, 20131046.

Fouts, R. (1997). Next of Kin. (London: Michael Joseph.)

Iglesias, T.L., McElreath, R., and Patricelli, G.L. (2012). Western scrub-jay funerals: cacophonous aggregations in response to dead conspecifi cs. Anim. Behav. 84, 1103–1111.

King, B.J. (2013). How Animals Grieve. (Chicago: University of Chicago Press.)

López-Riquelme, G.A., and Fanjul-Moles, M.L. (2013). The funeral ways of social insects. Social strategies for corpse disposal. Trends Entomol. 9, 71–129.

Stewart, F.A., Piel, A.K., and O’Malley, R.C. (2012). Responses of chimpanzees to a recently dead community member at Gombe National Park, Tanzania. Am. J. Primatol. 74, 1–7.

Sugiyama, Y., Kurita, H., Matsui, T., Kimoto, S., and Shimomura, T. (2009). Carrying of dead infants by Japanese macaque (Macaca fuscata) mothers. Anthropol. Sci. 117, 113–119.

Sun, Q., and Zhou, X. (2013). Corpse management in social insects. Int. J. Biol. Sci. 9, 313–321.

Swift, K.N., and Marzluff, J.M. (2015). Wild American crows gather around their dead to learn about danger. Anim. Behav. 109, 187–197.

Department of Psychology, Kyoto University Graduate School of Letters, Yoshida- honmachi, Sakyo-ku, Kyoto 606-8501, Japan. E-mail: [email protected]

R556 Current Biology 26, R543–R576, July 1

Carrion fl owers

Steven D. Johnson

What are carrion fl owers? Carrion fl owers are those which mimic the smell and appearance of rotting animal carcasses. They attract necrophagous (carrion-feeding) insects that normally use dead animals as brood sites. These insects visit the carrion fl owers, mistaking them for a dead animal, and in some cases even lay their eggs on the fl owers. The duped insects transfer pollen between the fl owers and thus act as pollinators of the plants (Figure 1).

Insects attracted to carrion fl owers include fl esh fl ies (Sarcophagidae), blowfl ies (Calliphoridae), house fl ies (Muscidae) and some beetles (e.g., Dermestidae and Silphidae). It is typically the females which visit carrion fl owers while searching for brood sites, but males are sometimes also attracted. This is because many carrion insects mate on or in the vicinity of animal carcasses.

The evolution of carrion mimicry by fl owers occurs when the resulting deception of insects causes an increase in the reproductive success of the plant. Carrion fl owers have evolved in dozens of different plant families, but the milkweed (Apocynaceae), orchid (Orchidaceae), arum (Araceae) and pipevine (Aristolochiaceae) families have particularly large numbers of species with carrion fl owers. Arums, orchids and pipevines often have fl oral chambers in which insects are temporarily trapped, sometimes by a barrier of spines, before being released with a load of pollen. The released insect must be trapped again by another fl ower of the same species for cross-pollination to occur.

Why are some carrion fl owers so large? An intriguing feature of carrion fl owers is their tendency to be very large. Indeed, Raffl esia arnoldii, a parasitic plant, and Amorphophallus titanum, a type of arum lily, which produce the largest single fl ower and blossom (a fl ower-like infl orescence) in the world, respectively, are both

Quick guide

1, 2016 © 2016 Elsevier Ltd.

carrion fl owers. Evidence suggests that in the Raffl esiaceae morphological diversifi cation has occurred at an explosive rate, with some lineages exhibiting a more than three-fold increase in fl ower size in a little over one million years. A number of hypotheses have been put forward to explain this tendency to gigantism among carrion fl owers, including visual mimicry of large animal carcasses or the production of extra scent and heat, but none of these explanations has been tested rigorously.

What attracts insects to carrion fl owers? To the human nose, carrion fl owers smell remarkably like rotting animal fl esh. This has been confi rmed by analyses using gas chromatography that have shown that the blends of volatile compounds emitted by carrion fl owers are strikingly similar to those emitted by dead animals in early stages of decomposition. In particular, carrion fl owers tend to emit oligosulfi des (dimethyl mono-, di- and trisulfi des) that provide the characteristic odour of decaying fl esh. These sulfi des originate from degradation of protein, specifi cally the sulfur-containing amino acids methionine and cysteine.

Studies have shown that the antennae of many carrion fl ies and beetles respond to dimethyl disulfi de (DMDS) and dimethyl trisulfi de (DMTS), and that fl ies can be attracted to these compounds even in the absence of visual cues. In one experiment, it was shown that a blend of DMDS and DMTS placed near inconspicuous green–yellow fl owers was suffi cient to induce a switch from pollination of these fl owers by wasps to pollination by carrion fl ies. Some recent studies suggest that DMTS is the most important olfactory cue for the attraction of carrion fl ies. Interestingly, DMDS and DMTS are also emitted by many bat-pollinated fl owers, but these fl owers are situated in a context (high in the canopy) that fl ies do not associate with oviposition sites. Most carrion fl owers are situated close to or at ground level.

Many carrion fl owers are also remarkably visually similar to the carcasses of small animals, complete with a pink surface with purple blotches and long white