New labels for old whines

Abstract

With characteristically colorful prose, Mock et al. (2011) critically review empirical work on nestling begging displays inspired by Godfray’s (1991) ‘‘signal of need’’ model. They contrast Signals of Need (SoN) with 2 ‘‘parsimonious alternative types of honest begging,’’ Signals of Hunger (SoH) and Signals of Quality (SoQ). Starting with SoH, these are supposed to provide a proximate indication of offspring fullness. This, Mock et al. assure us, makes them distinct from SoN, which on their interpretation can only provide information about offspring reproductive value. Indeed, they strongly criticize existing studies for interpreting evidence that begging reflects hunger as support for SoN; as Mock et al. define them, hunger and need are wholly distinct. This criticism seems to us to be misplaced and based on a false dichotomy. In models of begging, ‘‘need’’ simply refers to the benefits that offspring stand to gain from eliciting additional investment (e.g., Godfray 1991; Godfray and Johnstone 2000). These benefits may vary in relation to short-term need, that is, offspring hunger in Mock et al.’s terminology or in relation to long-term need, that is, offspring condition or reproductive value (e.g., Price et al. 1996; Lotem 1998). The logic of the handicap principle is applicable in the same way to both cases, and both have thus been interpreted as SoN; this is why so many authors have advanced evidence for the impact of short-term food deprivation on begging as support for SoN. It is of course interesting to explore how shortand long-term need respectively contribute to variation in begging intensity—see, for instance, Price et al. (1996), Lotem (1998), and Kilner et al. (1999). Nevertheless, in our view, it is confusing to treat signals of short-term need as an entirely separate category of display from signals of long-term need, and it is misleading to suggest that existing theory focuses only the latter and ignores the former. Turning to SoQ, Mock et al. suggest that begging may advertise merit rather than need and may serve to minimize parents’ wasting investment on dependents with dim prospects. In our view, there is indeed a real difference between SoQ and SoN, and, because most discussion of begging has indeed focused on SoN, we see no harm in Mock et al.’s effort to revive interest in SoQ. Nevertheless, Mock et al.’s suggestion is not entirely novel—Royle et al. 2002, for instance, have previously shown that models of begging can generate precisely the pattern Mock et al. predict in which parents provide more resources to chicks in better condition. Furthermore, we see 2 underlying problems with SoQ as a general hypothesis for the evolution of offspring begging. First, it is puzzling why parents would need to intervene actively in the process of food allocation in order to favor offspring in better condition, as SoQ suggests, when these offspring are superior competitors in any case. If the parents’ goal is to sacrifice weaker offspring with poorer prospects, they could simply allow sibling competition to take its course (e.g., Royle et al. 2002; Kolliker and Richner 2004). By contrast, SoN provides a stronger reason why parents should attempt to seize control of food allocation because this allows them to override differences in competitive ability and instead feed according to need. Second, although there are indeed phases of development during which infanticide is common, and offspring might benefit by dissuading parents from targeting them for ‘‘disinvestment,’’ this risk does not persist throughout chick growth (Mock and Parker 1997). SoQ might thus play a role at certain key points in development, when the risk of infanticide looms large, but looks unconvincing as a general explanation for begging that persists throughout the rearing of the young. As a final thought, we wonder how much is to be gained by setting up alternatives to Godfray’s (1991) original model of begging, which Mock et al. treat as the definitive statement of the SoN hypothesis. Godfray’s analysis has served its purpose admirably by forcing empiricists to think more clearly about the details of interactions within the family. As a result, we now know that real animal families are far more complicated and interesting than his model assumed, with provisioning responses determined by parental life history, prenatal signals of parental quality, and partner behavior as well as by begging intensity. Equally, Godfray’s model has stimulated many refinements of the theory, with more recent models attempting to incorporate some of the complexities revealed by empirical studies of parent–offspring interaction, from Godfray’s own 1995 model of competitive begging (Godfray 1995), to analyses of sibling negotiation, prenatal influences, and partial begging. In our view, future work should focus on incorporating these complexities into Godfray’s (1991) framework for thinking about family life rather than on replacing SoN with artificial alternatives.