On the cost of begging vocalization: implications of vigilance

Begging vocalization is thought to have evolved as a consequence of the parent-offspring conflict over parental investment (Mock and Parker, 1997). Under this conflict, parents are reluctant to provide all the resources requested by current offspring because they are saving resources for future broods (Trivers, 1974). In this scenario, begging has evolved as an honest signal of need with the most hungry offspring begging most conspicuously. This information allows parents to optimally adjust reproductive investment, since they can accurately assess offspring food requirement (Godfray, 1991). Given the conflict of interest over resources between parents and offspring, the evolutionary stability of begging would be ensured by its cost for two reasons. First, the benefit of obtaining additional resources from the parents increases with need, and hence the benefit to be fed outweighs the cost of begging only when hungry. Second, if all offspring, independent of need, were to beg at the same level, parents would be rapidly selected to ignore begging solicitation (Godfray, 1991; MacNair and Parker, 1979). The hypothesis that begging is a costly signal of need has prompted numerous experiments to test its predictions. Empiricists have found that parents increase feeding rate when begging level is amplified via play-back experiments (Burford et al., 1998; Davies et al., 1998; Ottosson et al., 1997; Price, 1998; but see Clark and Lee, 1998) and that parents preferentially allocate food to the offspring begging most vigorously (e.g., Hofstetter and Ritchison, 1998; Kilner, 1995; Roulin et al., 2000). Another avenue of experimentation has focused on the cost of begging, and three different costs have been detected so far. These are reviewed below. Loud begging vocalization attracts predators (Briskie et al., 1999; Dearborn, 1999, Haskell, 1994, 1999; Leech and Leonard, 1997; Redondo and Castro, 1992; but see Halupka, 1998). As a consequence, predation imposes an intense selective pressure on begging calls to be less easy to locate (Haskell, 1999; Redondo and Arias de Reyna, 1988); on larger broods to be less noisy (Harper, 1986); on nestlings to reduce escalation of begging calls when competing with siblings (Briskie et al., 1999), and on parents to alert offspring to be silent when a predator is close to the nest (Halupka, 1998; Nuechterlein, 1988). Although the predation cost of begging is often substantial, in some predatory species (e.g., eagles) there is no such cost, since nestlings do not suffer predation despite begging intensely (Cramp and Simmons, 1980). In some species, begging elicits aggressive behavior from parents (Leonard et al., 1988, 1991) and siblings (Braun and Hunt, 1983; Drummond and Chavelas, 1989; Nuechterlein, 1981). For instance, in the black-capped chickadee (Parus atricapillus), parents are aggressive towards fledglings that persistently beg for food. This parental behavior may force offspring to forage independently (Leonard et al., 1991). In the western grebe (Aechmophorus occidentalis), dominant nestlings peck younger siblings that beg for food perhaps because such intimidative behavior facilitates the monopolization of a larger than equal share of the resources provided by the parents (Nuechterlein, 1981). Although punishment costs may play a role in the maintenance of begging as an honest signal of need in some species, few actual cases of punishment have been reported in the literature. This suggests that such costs may not be particularly widespread. Begging involves conspicuous vocalization and stretching of the neck and legs. The observation that the intensity of these activities predicts which nestling is fed first (e.g., Dearborn, 1998; Hofstetter and Ritchison, 1998; Lichtenstein and Sealy, 1998; Teather, 1992) motivated researchers to investigate the cost of begging imposed by physical effort. To date, only five studies have examined the energetic cost of begging, by comparing the quantity of oxygen consumed by begging and nonbegging nestlings under laboratory conditions. Using closedchamber respirometry, nestlings of seven bird species were tested one at a time (Bachman and Chappell, 1998; Chappell and Bachman, 1998; Leech and Leonard, 1996; McCarty, 1996). Nestlings were removed from their nest 30-120 min beforehand and tested over a 15 min period. Begging was found to elevate the metabolic rate of chicks by only 1.05-1.28 times (Leech and Leonard, 1996; MacCarty, 1996), in stark contrast to song and sexual displays that raise metabolic rate by up to 15.6 times (Leech and Leonard, 1996). Soler et al. (1999) used the doubly labeled water technique in the great spotted cuckoo (Clamator glandarius) and magpie (Pica pica), a method that allowed them to measure begging effort over a complete day. Two cuckoo and two magpie nestlings were removed from different nests a day before being tested together in a same artificial nest. Under laboratory conditions, they were stimulated four times during 20 min every 2 h (i.e., one feeding session). A total of 7-8 such feeding sessions were performed during the day of the experiment. Begging nestlings consumed a nonsignificantly higher quantity of oxygen than nonbegging nestlings, suggesting that begging is not extremely costly. The apparently low energetic expenditure found in all five studies may be due to the fact that: (a) measurement of oxygen consumption disregards anaerobic metabolism, a possible source of energy during short-term begging activities that require intense muscular activities (Chappell and Bachman, 1998; Weathers et al., 1997); (b) nestlings were not begging constantly during the tests (Kilner and Johnstone, 1997); and (c) low energetic expenditure does not necessarily imply that begging does not impose fitness penalties (Verhulst and Wiersma, 1997). In the following section, I point out an overlooked aspect of begging, namely vigilance, that may entail energetic costs and other more intricate costs. I refer to this cost as “the vigilance cost of begging.” In the context of begging studies, I define vigilance as the process by which offspring look out for parents' arrival in order to start to beg as quickly as possible. Vigilance requires selective attention, that is the ability to distinguish relevant (e.g., the noise made by an arriving parent) from nonrelevant stimuli (e.g., the noise made by wind), but also the ability to react quickly once the relevant stimuli occurred. The assessment of interindividual variation in the level of vigilance and of within individual alteration in vigilance performance can be done at two levels. From a behavioral perspective, one can measure the time required to beg once a parent is detected, that is the rapidity to resume vocalization behavior (particularly important in the context of parent-offspring interactions) or to move body (particularly important in the context of sibling competition). From a physiological perspective, vigilance can be assessed via electrophysiology using electroencephalogram (EEG) and via cardiovascular activity using electrocardiogram (ECG). In human, it has been shown that more vigilant individuals differ from others in the form taken by the alpha-, beta- and delta-waves on a EEG (Schulz et al., 1996; Schwarz-Ottersbach and Goldberg, 1986) but also in interbeat interval and heart rate (Beh, 1990). Thus, it may be worth developing these physiological methods to assess the level of vigilance in nonhuman organisms. Chicks should be vigilant by watching out for their parents' return, if a rapid begging reaction increases the likelihood to be fed before nest-mates. This may be so either because parents feed them in priority or because rapid chicks have advantages in sibling competition. Correlational evidence that a nestling increases the likelihood of receiving a food item from its parents when it starts to beg before its nest-mates has been given in indigo buntings (Passerina cyanea) (Dearborn, 1998), red-winged blackbirds (Agelaius phoeniceus) (Teather, 1992), and Eastern screech-owls (Otus asio) (Hofstetter and Ritchison, 1998). Apparently, in the latter study the difference in the rapidity with which fed and unfed chicks started to beg once a parent arrived was relatively larger than the differences in calling rate and volume. To further examine whether in the face of sibling competition nestlings benefit from being vigilant, I recorded begging activity of barn owls (Tyto alba) located in western Switzerland. This species is especially interesting as nestlings vocalize both in the absence and presence of the parents. Calls produced in their absence have been shown to facilitate sibling negotiation over resources next delivered by the parents (Roulin et al., 2000). I created 65 two-chick broods (five of them were tested twice using different chicks) by temporarily removing other siblings out of their nest, randomly chosen. These broods were created in the evening, and from 2130 to 2330 h. I filmed them using an infra-red sensitive camera and a microphone. On the videos, I counted the number of calls produced by the two siblings during 15 min before a parent arrived with the first prey item of the night. I also counted calls in the presence of the parent, that is just before it delivered its prey item to one of the two offspring identified with a ring placed on a different leg. The more vigilant nestling was defined as the one that reacted first (i.e., that made a body movement) after a parent landed on the perch. In some cases, I could not determine which nestling moved first, and both siblings were considered as similarly vigilant. In the absence of the parent, the more vigilant nestling vocalized more intensely than its less vigilant sibling (Wilcox-on matched-pair signed-rank test, z = 2.86, p =.004, n = 47; Figure 1), and tended to do so in the presence of parents (z = 1.61, p =.11, n = 37; note that sample size is lower since nestlings were sometimes not visible in the presence of the parents precluding any assignment of begging calls to one of the two siblings). When the nestling that monopolized the item vocalized more intensely than its sibling, it was more vigilant in 27 cases (85%), as vigilant in two cases (6%), and less vigilant in three cases (9%). To determine whether vigilance may provide competitive advantages independently of begging intensity, I considered 21 broods for which the nestling that received the first prey item delivered of the night vocalized less intensely than its sibling. In this sample of broods, the less vocal chick was more vigilant than its sibling in 11 cases (52%), as vigilant in four cases (19%), and less vigilant in the last six cases (29%). Thus, a high level of vigilance appears to be correlated to the probability of being fed first despite low begging level (in 71% of the 21 broods, binomial test, p =.039, one-tailed). Note that vigilance level was not related to hatching asynchrony given that in 25 out of 47 cases (53%, binomial test, p >.50) the younger chick was more vigilant than its older sibling. Taken together, the three above mentioned studies (Dearborn, 1998; Hofstetter and Ritchison, 1998; Teather, 1992) and my observations suggest that a chick is more likely to be fed first when it reacts more quickly than siblings at the arrival of a parent. Calling rate of two siblings in the absence of the parents. The most vigilant individual was defined as the one that moved body most rapidly when a parent landed on the perch of the nest-box. Sample size is 47 experimental two-chick broods (10 in 1997, two in 1998, and 35 in 2000). Histograms represent median and bars interquartiles. Although a nestling increases the likelihood of receiving a food item from its parents when it starts to beg before its nestmates (Dearborn, 1998; Hofstetter and Ritchison, 1998; Lichtenstein and Sealy, 1998; Teather, 1992; present study), the rapidity with which nestlings commence begging has not yet been considered as potentially costly. Vigilance may be expensive as nestlings that start to beg rapidly at the arrival of a parent often resume begging for irrelevant stimuli such as the noise of wind or rain (Dearborn, 1998). Thus a high level of vigilance may lead nestlings to beg on inappropriate occasions. However, this may not be the only cost of vigilance. The competitive advantage provided by a rapid begging reaction surely requires a high level of vigilance that involves extra activity in the brain, one of the most metabolically active organs in the body (Roland, 1993). In human, vigilance for attention and decision making is energetically demanding and is sustained over a long time period only with difficulty (Åhsberg et al., 2000; Davies and Parasuraman, 1982). Vigilance tasks involve mental processes such as information processing, requiring glucose (Benton et al., 1994) and oxygen (Moss et al., 1998). The energy released by the oxidation of glucose enhances the synthesis of acetylcholine (among other neurotransmitters; Buchanan, 2000), a crucial molecule for maintaining a high level of vigilance (Wesnes and Warburton, 1984) and locomotion (Sholomenko et al., 1991). Depletion in energy due to vigilance thus leads to depletion in acetylcholine levels. This, in turn, leads to a further reduction in vigilance and a lowered motivation to invest in expensive activities (Åhsberg et al., 2000). Since vigilance appears to provide advantages in sibling competition, nestlings would be expected to invest a substantial amount of energy in vigilant behavior, perhaps at the expense of the intensity of begging (since both activities rely on energy). However, the trade-off here may not be simple. In humans, a slight physical effort enhances vigilance, where intense physical effort sustained over a longer time period impairs vigilance (Davey, 1973). A physiological cost of vigilance may not solely lie in energetic terms, but also in the capacity to mount an efficient immune response against invading parasites and pathogens. The nervous and immune systems are tightly linked (Maier et al., 1994). Acetylcholine, the previously described neurotransmitter that is important for sustaining vigilance tasks, has been shown to have immunosuppressive effects and, conversely, the activation of the immune system partly suppresses the synthesis of acetylcholine (Qiu et al., 1996). Although these interactions may be adaptive, the requirement for a high level of vigilance may trade against the of the immune a high level of vigilance nestlings from and thus from energetically expensive activities is and leads to vigilance 1973). Nestlings may be so that when they are they may being vigilant and This may only a of the nestlings in a beg when the parents at the nest, individuals that were all and and This cost of vigilance may be particularly relevant in species in which size is and feeding rate is Although a nestling has to be vigilant when parents food the probability of being fed on a given parental is low when the number of is when food is not a nestling that is fed last in a may not a high level of vigilance the time its nest-mates are being The hypothesis of a vigilance cost of begging has for the cost of begging should be First, two chicks may the same effort in physical begging, the chick begging first may a cost of vigilance. experiments (Bachman and Chappell, 1998; Chappell and Bachman, 1998; Leech and Leonard, 1996; McCarty, have the amount of energy consumed by a nestling by it from its nest-mates and it to However, since high of vigilance are most when sibling competition is such may the cost of begging under conditions. Second, since vigilance can be sustained only over a time period (Åhsberg et al., 2000; Davies and Parasuraman, the energy nestlings invest in begging should be over a long time period. This may be crucial if the cost of vigilance increases with that where nestlings do not have time to and due to the of one to assess the cost of begging may be to the to which nestlings are to begging and vigilance, but also the time required for from Another may be to provide extra energy (e.g., of to the chicks and to their begging begging provided chicks should beg more intensely than and be more vigilant, chicks may be less hungry due to the consumption of one could assess whether the ability to siblings in increases when vigilance is sustained during a time period but when nestlings are to be vigilant during a longer period. As a consequence, the of nestlings from their nest by for oxygen consumption 30-120 min (Bachman and Chappell, 1998; Chappell and Bachman, 1998; Leech and Leonard, 1996; McCarty, may to the under which a vigilance cost could be Nestlings were no longer under the of sibling competition and have time to from the costs by begging nestlings that were stimulated to beg in the laboratory were in In this one could the of parental feeding are nestlings when they can start to beg and they may not require a high level of vigilance. if parental are chicks may need to be more vigilant to commence begging as rapidly as possible when a food item is In other the cost of begging may be when feeding are not due to the requirement of a high level of vigilance. However, note that the may as when feeding are nestlings may start to invest in vigilance a time before the expected arrival of a parent, a larger vigilance cost than in broods for which feeding are not (e.g., et al., 1990). because vigilance on the nervous of a vigilance cost may be by the researchers when it would be worth experiments of laboratory experiments to assess the cost of begging, since may given the immunosuppressive of acetylcholine of other Buchanan, 2000), it would be interesting to measure immune response an in where a high of vigilance has to be should also be to distinguish between the of begging intensity and vigilance on A experimental is to increase the level of noise (e.g., the noise made by between parental to more the ability to determine the when a parent at the nest, and to force nestlings to be more vigilant (e.g., and during parental should not be to parent-offspring The parent-offspring conflict over parental investment may be begging behavior (Godfray, Begging involves and the of is likely to in vocalization may be the of sibling competition to the attention of parents (Harper, MacNair and Parker, et al., 1996), but also to signal level (Godfray, may an aspect of such as et al., 2000), this does not the that this signal may also 1997). stretching of the neck and may size and parents if they preferentially feed larger chicks 1997). begging has on which can evidence that the vigilance of begging is or has been to is provided by the studies of and Lichtenstein and These researchers a species that in the nests of other species, in their indigo other do not the and This an to study parent-offspring conflict at the level, because (a) parents are related to the they are for the nestlings, and (b) nestlings are not related to their nest-mates. The latter that a nestling does not any fitness from its nest-mates to be fed first, and hence it should be under to its nest-mates by rapidly to the arrival of the parent. The monopolization of a share of the resources by chicks (Dearborn, 1998; Lichtenstein and Sealy, 1998) may for the cost by their more rapid begging response when the parent at the nest. This that the of a higher of vigilance in may have occurred. In other may have the of a ability to an parent to process this and to body rapid In the offspring fitness by resources with their and they may be under a pressure to a high of vigilance. by is is to be to nestlings to This suggests that begging escalation from sibling competition may not be in of the and of begging but also in the period before offspring start to vocalize once parents are at the nest. in vigilance may not only from sibling competition, but may signal food requirement to parents. I provide two where the hypothesis of a vigilance of begging may to the may with the of honest of First, species that are under intense predation pressure have evolved begging calls of different (Haskell, 1999), and may be in the they can signal food is a signal of food but also attracts begging calls may to be less conspicuous so that parents have to rely on other to accurately assess level 1997). such may be the with which nestlings react to their arrival or to parental calls that begging behavior (e.g., and 1997; et al., 1997). Since the arrival of a parent may not be parents can be that the nestling that begging most rapidly is the most vigilant. may also benefit from a rapid offspring since they may to forage more this is the the vigilance of begging may also be selected in species a As a have the that parents may not accurately assess of For instance, noise may be so high that the difficulty in begging may for more conspicuous and hence more costly 1998). For instance, of the can number up to and the noise created by so individuals may have selected for begging that are from chicks to parents and 1998). However, as out by if the level of noise or by the parents is so great that the costs of an begging signal be by the offspring, the system is no longer In such cases, the species may other that are not by such may be a high level of vigilance. on begging in the context of parent-offspring conflict have (Godfray, These are on the that begging fitness costs. To date, evidence for such costs are and raise the of whether costs at all (e.g., Leech and Leonard, 1996). have also begging in of and 1998). the evidence for costs of begging, I have some of the from the that begging may entail a vigilance Although vigilance appears to entail these may be different from so considered and physical and for honest of need may independently be on of begging, on the This out that apparently may entail costs that are not of a vigilance cost of begging up of in the context of I to the which a during which I this and two provided on a of this I I created two-chick broods under the of the de no