A long-term experimental study demonstrates the costs of begging that were not found over the short term

Abstract

Parent-offspring conflict theory predicts that begging behaviour could escalate continuously over evolutionary time if it is not prevented by costliness of begging displays. Three main potential physiological costs have been proposed: growth, immunological and metabolic costs. However, empirical evidence on this subject remains elusive because published results are often contradictory. In this study, we test for the existence of these three potential physiological costs of begging in house sparrow (Passer domesticus) nestlings by stimulating a group of nestlings to beg for longer and another group for shorter periods than in natural conditions. All nestlings were fed with the same quantity of food. Our study involves a long-term experimental treatment for begging studies (five consecutive days). Long-term studies frequently provide clearer results than short-term studies and, sometimes, relevant information not reported by the latter ones. Our long-term experiment shows (i) a clear effect on the immune response even since the first measurement (6 hours), but it was higher during the second (long-term) than during the first (short-term) test; (ii) evidence of a growth cost of begging in house sparrow nestlings not previously found by other studies; (iii) body condition was affected by our experimental manipulation only after 48 hour; (iv) a metabolic cost of begging never previously shown in any species, and (v) for the first time, it has shown a simultaneous effect of the three potential physiological costs of begging: immunocompetence, growth, and metabolism. This implies first, that a multilevel trade-off can occur between begging and all physiological costs and, second, that a lack of support in a short-term experiment for the existence of a tested cost of begging does not mean absence of that cost, because it can be found in a long-term experiment.

Figure 1. Effect of the experimental treatment on immune response calculated from RM-ANOVAs performed for each experimental session.

P-values associated with differences in each experimental session are indicated as ns: P>0.05; st: 0.1≤P≥0.05; *: P<0.05; **: P<0.001 and ***: P<0.0001. Numbers of nests (i.e. those with measurements from the four nestlings) used in each of the comparisons are also shown.

Figure 2. Effects of the experimental treatment on the immune response according to nestling rank calculated from RM-ANOVAs in the successive experimental sessions.

The significance of interactions between experimental treatment and nestling rank are also shown. P-values associated to LSD post hoc tests (i.e. treatment effect within nestling rank) are indicated as ns: P>0.05; st: 0.1≤P≥0.05; *: P<0.05; **: P<0.001 and ***: P<0.0001.

Figure 3. Effect of the experimental treatment on body condition calculated from RM-ANOVAs over nestling development.

P-values associated with differences in each experimental session are indicated as ns: P>0.05; st: 0.1≤P≥0.05; *: P<0.05; **: P<0.001 and ***: P<0.0001. Numbers of nests (i.e. those with measurements from the four nestlings) used in each comparisons are also shown.

Figure 4. Effect of the experimental treatment on mass gained by nestlings and calculated from RM-ANOVAs in each experimental session.

P-values are indicated as ns: P>0.05; st: 0.1≤P≥0.05; *: P<0.05; **: P<0.001 and ***: P<0.0001. Numbers of nests (i.e. those with measurements from the four nestlings) used in each one of comparisons are also shown.

Figure 5. Effect of the experimental treatment on mass excreted by nestlings based on the RM-ANOVAs for each experimental session.

P-values associated to LSD post hoc test are indicated as ns: P>0.05; st: 0.1≤P≥0.05; *: P<0.05; **: P<0.001 and ***: P<0.0001. Numbers of nests (i.e. those with measurements from the four nestlings) used in each comparison are also shown.

Figure 6. Effects of the experimental treatment on mass excreted by nestlings according to nestling rank calculated from RM-ANOVAs in the experimental sessions where the interaction effects between treatment and nestling rank resulted significant (or almost).

The significance values of interactions between experimental treatment and nestling rank are also shown. P-values associated to LSD post hoc (i.e. treatment effect within nestling rank) tests are indicated as ns: P>0.05; st: 0.1≤P≥0.05; *: P<0.05; **: P<0.001 and ***: P<0.0001.

Figure 7. Effect of the experimental treatment on metabolic costs of nestlings based on the RM-ANOVAs performed in each experimental session.

P-values are indicated as ns: P>0.05; st: 0.1≤P≥0.05; *: P<0.05; **: P<0.001 and ***: P<0.0001. Numbers of nests (i.e. those with measurements from the four nestlings) used in each comparison are also shown.