SEARCH

SEARCH BY CITATION

Keywords:

  • cooperative breeding;
  • nestling predation;
  • parental care;
  • pied babbler;
  • synchrony

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Meterials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

1. Behavioural synchrony typically involves trade-offs. In the context of foraging, for example, synchrony may be suboptimal when individuals have different energy requirements but yield net benefits in terms of increased foraging success or decreased predation risk.

2. Behavioural synchrony may also be advantageous when individuals collaborate to achieve a common goal, such as raising young. For example, in several bird species, provisioners synchronize nest-feeding visits. However, despite the apparent prevalence of provisioning synchrony, it is not known whether it is adaptive or what its function might be.

3. Here, we propose a novel explanation for provisioning synchrony: it increases brood survival by decreasing the number of temporally separate nest visits and accordingly the chance that the nest will be detected by predators. Using cooperatively breeding pied babblers, we showed experimentally that provisioners synchronized nest visits by waiting for another provisioner before returning to the nest. Brood survival increased with provisioning synchrony. Provisioners were more likely to synchronize feeding visits for older nestlings as they were louder and possibly more conspicuous to predators. Finally, provisioners in large groups were more likely to wait for other provisioners and synchronized a higher proportion of all visits than those in smaller groups. Thus, provisioning synchrony may be one mechanism by which large groups increase brood survival in this species.

4. This study highlights a novel strategy that birds use to increase the survival of young and demonstrates the advantages of coordinated behaviour in social species.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Meterials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Animals often benefit from synchronizing their behaviour with other individuals. However, they also face a trade-off between reaping the benefits of behavioural synchrony and the costs of departing from their own optimal activity schedule (Ruckstuhl 1999; Dostalkova & Spinka 2007). Synchronized behaviour will therefore tend to arise where the benefits of synchrony outweigh the costs associated with either delaying or bringing forward a change in behaviour (Dostalkova & Spinka 2007; Meldrum & Ruckstuhl 2009). For example, in bighorn sheep Ovis canadensis (Shaw), subadult males have different energy requirements to adults and may therefore prefer to forage and rest at different times. However, foraging with conspecifics allows individuals to increase their foraging efficiency and decrease predator vigilance in this species. As a result, subadult males typically synchronize foraging with older sheep, despite having to forego their own optimal foraging schedule to do so (Ruckstuhl 1999).

Synchronized behaviour may also be advantageous where individuals cooperate to achieve a common goal. For example, young subsocial spiders Amaurobius ferox (Walckenaer) synchronize hunting efforts to capture prey (Kim, Krafft & Choe 2005a, b) and African wild dogs Lycaon pictus (Temminck) that hunt together are more successful and incur fewer costs (Creel & Creel 1995). One of the most overt and readily studied forms of cooperation in nature is cooperative breeding, where individuals collaborate to raise young (Brown 1987; Cockburn 1998). A prominent yet poorly understood example of synchronized behaviour occurs in some cooperatively breeding birds, where provisioners temporally synchronize feeding visits at the nest [pinyon jays Gymnorhinus cyanocephalus (Wied-Neuwied), Marzluff & Balda 1990; sociable weavers Philetairus socius (Latham), Doutrelant & Covas 2007; bell miners Manorina melanophrys (Latham), McDonald et al. 2008a; apostlebirds Struthidea cinerea (Gould) N. Raihani, personal observation]. However, it is not yet clear whether this phenomenon has an adaptive function or whether it simply arises as a consequence of coordinated activities outside of provisioning contexts (e.g. synchronized predator mobbing –McDonald et al. 2008a).

Synchronizing nest visits may allow provisioners to increase either the direct or indirect benefits of feeding young. Provisioners may accrue direct benefits from synchronizing nest visits if the benefits of helping are contingent on being observed by other group members. This may be the case where helpers have to provide a certain level of assistance in return for group membership (‘pay-to-stay’: Gaston 1978; Kokko, Johnstone & Wright 2002; Bruintjes & Taborsky 2008). Under these circumstances, subordinate individuals are expected to advertise their contributions to dominant individuals. Alternatively, provisioners might increase their attractiveness to potential breeding or dispersal partners when they are seen performing costly behaviours (‘social prestige’: Zahavi 1990, 1995). As such, benefits from increases in social prestige are often expected to vary according to individual sex and/or dominance status (Wright 1997). A recent study on P. socius suggested that provisioners benefited by advertising their contributions to other colony members (Doutrelant & Covas 2007). Non-breeding provisioners preferentially fed nestlings in the presence of an audience and were more likely to wait for an audience when bringing larger food items to the nest. However, the proposed advantages of signalling contributions (increasing mating success/establishing coalitions/inducing reciprocal altruism) were not measured and so it is unclear whether provisioners gained benefits from advertising their contributions to colony members. It is worth noting that other empirical studies that have explicitly addressed whether help has signalling characteristics in other candidate species have found that it does not (e.g. Wright 1997; McDonald et al. 2008a,b).

An alternative explanation for synchronous provisioning is that it allows individuals to reduce the costs associated with feeding young. For example, provisioners may prefer to travel to the nest with others to reduce the risk of being preyed upon en route to the nest (‘dilution effect’; Foster & Treherne 1981). Alternatively, if provisioners risk encountering predators when they arrive at the nest they may dilute this risk by travelling to the nest site together but waiting for another individual to feed the nestlings before they approach the nest themselves. This might be most important in species where the nest cavity is concealed and predators in the nest are not visible from the exterior, as is the case in P. socius colonies (Maclean 1973; Doutrelant & Covas 2007). In fact, rather than advertising their contributions to other colony members, the provisioning patterns observed in P. socius are also consistent with such an anti-predator strategy as acknowledged by Doutrelant & Covas (2007). As helpers are typically less motivated than parents to pay costs associated with investing in young, they may be more reluctant than parents to enter the potentially dangerous nest cavity and may therefore tend to wait for another provisioner to arrive before entering. Similarly, the finding that helpers waited longer with larger food items may also be explained in terms of increasing survival if larger items reduce manoeuvrability or obscure vision, thereby rendering provisioners more susceptible to predators lurking in the nest.

A final possibility for synchronized provisioning that has not yet been considered is that it increases brood survival by decreasing the risk of predation. The chance that predators will detect a nest increases with parent visit rate (Skutch 1949; Martin, Scott & Menge 2000; Muchai & DuPlessis 2005). Thus, provisioners face a trade-off between provisioning nestlings and revealing the nest location to eavesdropping predators. This may be especially pertinent in cooperatively breeding species where increased numbers of nest attendants may increase the conspicuousness of the nest (Poiani & Pagel 1997; Strickland & Waite 2001). Previous work has shown that this trade-off can be resolved by excluding alloparents from the nest or by bringing larger food loads to reduce the total visit rate (Strickland & Waite 2001). Provisioners might also resolve this trade-off by synchronizing feeding visits, as this would allow fewer temporally separate nest visits for a given rate of food delivery. Where provisioning synchrony reduces brood predation risk it might be most apparent when nestlings are older and the increased amplitude of their begging vocalizations (Macgregor & Cockburn 2002; Leonard & Horn 2006) renders them more conspicuous to predators (Onnebrink & Curio 1991; Haskell 1994; Dearborn 1999).

In this study, we present observational and experimental data on nestling provisioning in cooperatively breeding pied babblers Turdoides bicolor (Jardine). Pied babblers are medium-sized (65–85 g) passerines from the southern Kalahari. Groups comprise a dominant breeding pair (breeders) and up to eight non-reproductive helpers of both sexes (Raihani 2008). Offspring are fed by breeders and helpers in the nest and for an extended period post-fledging (Ridley & Raihani 2007a). Pied babblers often synchronize nest-feeding visits, apparently by waiting for another individual to accompany them before they approach the nest. Here we ask (i) do provisioners actively delay returning to the nest to synchronize feeding visits; (ii) when does provisioning synchrony occur; and (iii) what is the likely function of provisioning synchrony in pied babblers?

Meterials and methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Meterials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Study site and population

Data were collected from October 2005–February 2008 at the Kuruman River Reserve, South Africa (26º58′ S; 20º49′ E) (see Raihani & Ridley 2007a, for information on climate and vegetation). The babblers at this site are wild but habituated to the close presence (<2 m) of human observers, enabling detailed behavioural data to be collected (see Ridley & Raihani 2007b, for information on habituation techniques). All individuals are identifiable by a unique combination of colour rings. Pied babblers build a new nest for each breeding attempt. Nests are open-cup and are typically built in Acacia erioloba (Mey) trees (Raihani & Ridley 2007b). The breeding male and female produce >96% of all the offspring in a group (M. J. Nelson-Flower, unpublished data) and are identified through observations of dominance assertions and breeding activity (Raihani 2008). The term ‘provisioners’ is used throughout to mean both breeders and helpers (non-breeding individuals > 12 months old). Individuals were sexed using a DNA test (Griffiths et al. 1998). Accurate ages (to the nearest day) for nestlings were determined by checking nests daily for hatching. Nests were monitored throughout the nestling period and predation was inferred if the nest was deserted. Although small nestlings occasionally starved, this never resulted in the desertion of the entire brood (N. Raihani, personal observation). Based on observations of mobbing behaviour at the nest (N. Raihani, unpublished data), the most common diurnal nest predators at this site were yellow and slender mongooses Cynictis penicillata (Ruppell) and Galerella sanguinea (Cuvier) and pale chanting goshawks Melierax canorus (Thunberg). It is harder to infer nocturnal nest predators from mobbing behaviour as they were seldom spotted during the day. However, likely candidates include spotted and Verreaux’s eagle owls Bubo africanus and Bubo lacteus (Temminck), small spotted genets Genetta genetta (Linnaeus) and African wild cats Felis silvestris (Schreber).

Statistical methods

All statistics were performed using r version 2.8.1 (R Development Core Team 2008). Not all data were available for all individuals so sample sizes vary across analyses. All statistical tests were two-tailed. Data were checked to ensure that they conformed to the assumptions of the tests used and were transformed where necessary. We used paired t-tests to analyse experimental data. For all other analyses, we used general/generalized linear mixed models (LMMs/GLMMs), which allow the effects of multiple explanatory terms – including repeated measures such as group or individual identity – to be estimated. Where appropriate random terms were hierarchically nested, we excluded random terms that did not significantly improve the explanatory power of the model.

We used the statistical concept of model selection using maximum-likelihood estimation. Using this approach, a series of models are tested, with each model representing a biological hypothesis. Akaike’s Information Criteria (AIC) values were compared between different models. AIC values are calculated by the formula: ‘deviance + 2k’ where deviance is −2 log likelihood and k is the number of parameters in the model. A decrease in AIC corresponds to an increase in the fit of the model to the data or a reduction in the number of parameters fitted. Thus, comparing AIC values from different models allows you to pick the model that explains most of the variance in the data set using the least number of explanatory terms (the ‘best’ model). To investigate the effects of explanatory terms, we used stepwise-deletion (Crawley 2007). Initially, all potential explanatory terms and all two-way interactions were included in a ‘maximal’ model. In subsequent models, we sequentially removed terms, retaining only those whose removal considerably decreased the fit of the model to the data (defined as an increase of 2 or more AIC units when the term was removed). In model output tables, we present the maximal models, models with significant first-order interactions and models containing fixed effects.

Data collection

After provisioners caught a food item, they immediately ingested it; immediately flew to the nest to feed young; or waited on a perch in a tree before approaching the nest. Waiting provisioners behaved like sentinels [see Ridley & Raihani (2007b) and Hollen, Bell & Radford (2008)], perching on branches either above the foraging group or in the nest tree itself (N. Raihani, personal observation). In this study, feeds were termed ‘synchronous’ if two provisioning visits occurred within 5 s of one another.

Did provisioners wait to synchronize feeding visits?

Synchronized provisioning might have been coincidental if provisioners found food at the same time and then immediately returned to the nest. However, our observations that pied babblers delayed returning to the nest after finding food items raised the possibility that babblers actively waited to synchronize provisioning. We tested whether this was the case using an experimental approach with two treatments. We fed a focal bird one mealworm (i) alone (fed alone) and (ii) at the same time as another, randomly chosen, individual in the group (fed together) and recorded the delay between the focal bird receiving the mealworm and taking it to the brood. We started timing as soon as we gave the mealworm to the focal bird so the ‘delay’ included handling time for the mealworm as well as time spent waiting and journeying to the nest. We ensured that focal birds were the same approximate distance (to the nearest 5 m) from the nest in both treatments so that journey time could not account for differences in the delay to feed nestlings. Although experiments were carried out at random intervals throughout the nestling period, each provisioner we tested received both treatments on the same day to control for any potential effects of nestling age on handling time (since pied babblers tend to spend longer manipulating food items when feeding very young nestlings). The order in which each provisioner received ‘fed alone’ and ‘fed together’ treatments was randomized and there was an interval of at least 15 min between each treatment. Only two focal birds did not feed their mealworms to nestlings so data from these individuals were excluded. Data from 17 provisioners (10 females, 7 males) from eight groups were log-transformed and analysed using a paired t-test. We hypothesized that if provisioners delayed provisioning to synchronize feeding visits they would delay for longer in the ‘fed alone’ treatment than in the ‘fed together’ treatment.

When did provisioning synchrony occur?

We investigated the factors affecting the proportion of all nest-feeding visits that were synchronized using the lmer function in r (Bates, Maechler & Dai 2008). Data were collected ad libitum (Altmann 1974) on provisioning to 44 broods in 13 groups over a total of 105 observation days. The number of synchronized feeding visits on each observation day was the response term in a GLMM with a binomial distribution of errors and logit-link function. The binomial total was the total number of feeds at that nest during the observation day. ‘Group’ and ‘brood’ identities were initially included as random terms in the model (with ‘brood’ nested within ‘group’); however, ‘group’ did not improve the fit of the model to the data (likelihood ratio test χ2‘group’ ≈ 0,  1; LRT χ2‘brood’ ≈ 4·3,  0·04) and was subsequently excluded. The following terms, plus all two-way interactions were included in the maximal model: ‘nestling age’ (days), ‘total provisioning rate’ (feeds h−1) and ‘group size’.

Why did provisioners synchronize feeding visits?

We tested two hypotheses for provisioning synchrony in pied babblers: (i) it allowed provisioners to advertise their contributions or (ii) it increased brood survival.

Did synchronized feeding allow provisioners to advertise contributions?

Provisioning synchrony might have allowed individuals to advertise their contributions to other group members. However, simply measuring individual variation in provisioning synchrony was unlikely to tease apart the factors that motivated individuals to synchronize feeding visits. This is because ‘signallers’ would be expected to wait for ‘receivers’ before returning to the nest, meaning that both parties would display similar levels of provisioning synchrony despite the asymmetry in motivation to synchronize with others. We felt that a better approach was to measure the delay between finding a food item and returning to the nest to feed young. If provisioners did not benefit from signalling contributions, then they should have returned to the nest immediately after finding a food item. Conversely, if provisioners could benefit from advertising their contributions to others, then they should have waited until another individual also found a food item before returning to the nest. We predicted that provisioners with more to gain from signalling would wait longer after finding a food item than individuals that were either the intended audience for the signal or that stood to benefit less from signalling contributions. For example, if helpers were advertising contributions to breeders to pay rent on the territory, then helpers should have been more likely to delay returning to the nest as they waited for breeders to also find food items and visit the nest. Alternatively, if individuals were signalling to attract potential mating/dispersal partners, then delay to feed should have varied with sex. Males would have been expected to advertise to potential breeding partners, whereas females (the dispersing sex, Raihani 2008) would have been expected to advertise to potential dispersal partners. Under the signalling hypothesis for synchronized provisioning, we therefore expected feeding delay to vary consistently with either sex and/or dominance status depending on the benefits individuals stood to gain from advertising their contributions.

To investigate this, we conducted 20 focal watches (Altmann 1974) on 17 provisioners (8 females, 9 males; 11 breeders, 6 helpers) from six groups (mean number of focals per provisioner ± SE = 1·2 ± 0·1). Each focal watch was 30-min long and during each focal all behaviours were recorded to the nearest second. We recorded the size of any food items found (see Raihani & Ridley 2007a) and whether or not the adult subsequently ate the item or provisioned nestlings. ‘Foraging efficiency’ (g caught per foraging min) was calculated per focal for each provisioner using pre-determined values of biomass for food item size (see Raihani & Ridley 2007a). During these focals, provisioners fed nestlings a total of 41 times. For each visit, we calculated the ‘delay to feed’ (s) by subtracting the time at which provisioners fed the chicks from the time at which they found the food item. This value was transformed using a Box–Cox Powers transformation to normality and set as the response term in a LMM with normal error distribution and identity-link function using the lme function in r (Pinheiro et al. 2008). As with the experimental procedure, the delay to feed also incorporated the time that was spent manipulating the food item. To reduce as much as possible the effect that handling time could have on delay to feed, we (i) included ‘food item size’ as a potential explanatory variable in the model and (ii) conducted focal watches only while nestlings were 6–10 days old (total nestling period <19 days; Raihani & Ridley 2007b), as provisioners spent longer manipulating food items for newly hatched chicks. The following terms and all two-way interactions were fitted to the maximal model: ‘group size’, ‘food item size’ (a three-level factor, small, medium or large), ‘sex’, ‘dominance status’ and ‘foraging efficiency’. Because the sample size was relatively small with respect to the number of explanatory terms being fitted, we calculated AIC values corrected for small sample sizes (AICc; Burnham & Anderson 2002). ‘Group’ and ‘provisioner’ identities were initially fitted as random terms (‘provisioner’ nested within ‘group’) but did not improve the fit of the model to the data (LRT χ2‘group’ ≈ 0,  1; LRT χ2‘provisioner’ ≈ 0,  1) and were subsequently excluded. Thus, the model output shows the predictions from a GLM rather than LMM.

Did synchronized provisioning increase brood survival?

We investigated whether brood survival increased with provisioning synchrony using a GLMM with binomial error distribution and logit-link function. For this analysis, we used the lmer function in r (Bates et al. 2008). ‘Brood survival’ (1 = survived; 0 = depredated) was the response term, with ‘provisioning synchrony’, ‘total provisioning rate’, ‘group size’ and all two-way interactions included as explanatory terms. Data from 42 breeding attempts from 16 groups were available. Each nest was observed at least twice for a 90-min period. During observation sessions, all provisioning visits were recorded and were categorized as either synchronous (if two individuals arrived within <5 s of one another) or asynchronous. ‘Provisioning synchrony’ was calculated as number of synchronous feeds/total number of feeds. ‘Provisioning rate’ per nest was calculated by dividing the total number of feeds by the total observation time. ‘Group identity’ was initially included as a random term but did not improve the fit of the model to the data (ΔAIC when ‘group’ was included as a random term in the model = 2) and was subsequently excluded. Thus, the model output shows the predictions from a GLM rather than GLMM.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Meterials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Did provisioners wait to synchronize feeding visits?

Feeding experiments suggested that individuals waited after catching food items to synchronize nest visits with other group members. Provisioners delayed visiting the nest for longer when they were fed alone (mean delay ± SE alone = 2·70 ± 0·8; range = 0·29–14·0 min) than when they were fed together (1·45 ± 0·3; range = 0·3–5·23 min) (paired t-test: t16 = 2·53; = 0·022). When fed alone 8/17 provisioners (47·1%) waited until another group member found a food item before returning to the nest. When fed together 16/17 (94·1%) provisioners returned to the nest with the other provisioner. Thus, provisioners synchronized nest visits more often in the ‘fed together’ treatment than in the ‘fed alone’ treatment (two proportion test, χ2 = 6·94. d.f. = 1, = 0·008).

When did provisioning synchrony occur?

The mean provisioning rate per nest (±SE) was 10·3 ± 0·5 (range = 2–23) feeds h−1 and 41·2 ± 0·02% of these visits were synchronized. Provisioning rate increased with nestling age (correlation = 0·43, t = 4·7, d.f. = 99, < 0·001) and, to a lesser extent, with group size (correlation = 0·33, t = 3·6, d.f. = 103, < 0·001). Despite this correlation, we found independent positive effects of both nestling age and group size on provisioning synchrony (Table 1; Figs 1 and 2). The best model (model 1a, Table 1) indicated that an interaction between group size and provisioning rate influenced provisioning synchrony. Further examination revealed that when provisioning at high rates (>20 feeds h−1) large groups synchronized provisioning more often than smaller groups. However, since the ΔAIC between model 1a and model 1b (Table 1) was <2 they can be considered broadly equivalent (Burnham & Anderson 2002), meaning that the significance of the interaction should be treated with caution.

Table 1.   Output of GLMM investigating the factors affecting provisioning synchrony
Model rankModel nameResponse = synchronized feeds; binomial total = total number of feeds; random term = clutch idDev KAICΔAIC
4/6Maximal modelAge + feeds h−1 + group size + all first-order interactions172·38188·33·7
1/6Model 1aAge + feeds h−1 + group size + feeds h−1*group size 172·66184·60
2/6Model 1bAge + feeds h−1 + group size 175·55185·50·9
5/6Model 1cFeeds h−1 + group size 196·14204·119·5
6/6Model 1dAge + group size 206·84214·830·2
3/6Model 1eAge + feeds h−1178·54186·51·9
Estimated effectsExplanatory termEstimateSE
  1. Age = nestling age (days), feeds h−1 = nest provisioning rate.

  2. The number of synchronized feeds was set as the response term and the total number of feeds as the binomial total in a GLMM with binomial error distribution and logit-link function. Explanatory terms fitted to the model are listed. Interactions between terms are denoted by the ‘*’ symbol. Dev (deviance) is the −2 log likelihood; K is the total number of parameters (n explanatory terms + n random terms + residual variance), AIC is the Akaike’s Information Criteria and ΔAIC is the difference between the AIC value for that model and the AIC value of the best model. Models are ranked according to their AIC values – the model with the lowest AIC value is the ‘best’ model. Data come from 105 observation days (at 44 nests in 13 groups).

 Constant−1·640·22
 Age0·030·01
 Feeds h−10·060·01
 Group size0·070·03
image

Figure 1.  The proportion of all feeds that were synchronized as a function of nestling age (days). The line represents the predicted mean generated from the output of model 1b (Table 1) controlling for other terms in the model. Points represent mean values (±SE) generated from the raw data.

Download figure to PowerPoint

image

Figure 2.  The proportion of all feeds that were synchronized as a function of group size. The line represents the predicted mean generated from the output of model 1b (Table 1). Points represent mean values (±SE) generated from the raw data.

Download figure to PowerPoint

Why did provisioners synchronize feeding visits?

Did synchronized provisioning allow provisioners to advertise contributions?

Data from focal watches showed that provisioners waited 0·4–6·1 (2·2 ± 0·2) min before returning to the nest with food. Provisioners in larger groups apparently waited longer before returning to the nest than provisioners in smaller groups (Table 2). However, the ΔAICc between the model containing group size as an explanatory term and the model containing no explanatory terms was <2, meaning that the effect of group size on delay to feed was tenuous. We hypothesized that if provisioners synchronized nest-feeding to advertise contributions to other group members then individuals of different sex and/or status should differ in their propensity to synchronize provisioning visits. However, we found no significant effect of sex, status or the two-way interaction on delay to return to the nest with food (Table 2).

Table 2.   Output of GLM investigating factors associated with delay between finding a food item and returning to the nest to feed young
Model rankModel nameResponse = delayDevKAICcΔAICc
7/7Maximal modelfe + sex + status + group size + item size + all first-order interactions−80·616−25·937·4
6/7Model 2afe + sex + status + group size + item size −68·86−55·18·2
5/7Model 2bfe + sex + status + group size −68·65−57·55·8
4/7Model 2cfe + status + group size −68·64−60·03·3
3/7Model 2dfe + group size −67·63−61·32·0
1/7Model 2eGroup size−67·42−63·30·0
2/7Model 2fn/a−63·41−61·41·9
Estimated effectsExplanatory termEstimateSE
  1. fe = foraging efficiency (g caught per minute foraging), status = breeder/helper, group size = small/large, item size = small/medium/large.

  2. The delay (seconds) between finding a food item and provisioning nestlings was set as the response term in a GLM with normal error distribution and identity-link function.

  3. Group and provisioner identities were initially included as random terms but did not improve the fit of the model to the data and were subsequently excluded. Data come from 41 feeding events (17 provisioners in six groups).

 Constant1·530·05
 Group size0·070·03
Did synchronized provisioning reduce nestling predation?

Of the 41 broods studied, 13 were depredated (31·7%), whereas the remaining 28 broods fledged successfully. Provisioning synchrony was positively associated with brood survival (Fig. 3) and the model containing this term alone received most support (Table 3). Provisioning rate and group size also positively influenced brood survival but to a lesser extent (Table 3). There was no correlation between provisioning rate and provisioning synchrony (correlation = 0·06, t = 0·41, d.f. = 40, = 0·69), although there was a positive correlation between group size and provisioning synchrony (correlation = 0·30, t = 2·0, d.f. = 40, = 0·05).

image

Figure 3.  Nestling survival probability as a function of the proportion of all feeding visits that were synchronized. The solid line represents the mean survival probability and was generated from the predictions of the GLM (Table 3) using mean values for feeding rate (feeds h−1 = 11) and group size (group size = 4·7). The points are the mean observed levels of provisioning synchrony for each brood.

Download figure to PowerPoint

Table 3.   Output from GLM investigating the factors affecting clutch survival
Model rankModel nameResponse = survived (1); depredated (0)DevKAICΔAIC
4/4Maximal modelFeeds h−1 + synch + group size + all first-order interactions43·4757·57·2
3/4Model 3aFeeds h−1 + synch + group size 43·6451·61·3
2/4Model 3bSynch + group size44·9350·90·6
1/4Model 3cSynch46·3250·30
Estimated effectsExplanatory termEstimateSE
  1. Feeds h−1 = feeding rate, synch = provisioning synchrony.

  2. Nestling survival (1 = survived; 0 = did not survive) was set as the response term. Group identity was initially included as a random term but did not improve the fit of the model to the data and was subsequently excluded. Data come from 42 nests from 16 groups.

 Constant−1·190·90
 Synch4·622·1
 Feeds h−10·090·08
 Group size0·290·26

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Meterials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

In most birds, including pied babblers (Raihani & Ridley 2007b), reproductive failure can be primarily attributed to nestling predation (Ricklefs 1969; Martin 1995). In some cases, nestling predation favours life-history strategies that minimize investment in young, for example reductions in brood sizes (Fontaine & Martin 2006). However, selection might also act on provisioners to adjust their behaviour to reduce nestling predation (Conway & Martin 2000; Strickland & Waite 2001; Eggers, Griesser & Ekman 2005, 2008; Fontaine & Martin 2006; Raihani & Ridley 2007b). Here, we describe a novel strategy that provisioners use to increase brood survival. By synchronising provisioning visits, pied babblers decreased the number of separate nest visits for a given rate of food delivery. We suggest that this reduced the chance that the nest would be detected by predators and increased brood survival as a result.

The relatively low provisioning rates we observed suggest that provisioning synchrony was unlikely to have occurred by chance alone. Each nest was visited c. 10 times per hour, yet over 40% of all visits were synchronized, which is much higher than would be expected by chance. Furthermore, our experiments showed that provisioners actively delayed returning to the nest to synchronize nest visits with other group members. Although our experimental results showed that the mean delay to return to the nest was over 1 min even when two provisioners were fed together, at least part of this delay may be attributed to time spent handling the mealworm and travelling to the nest rather than waiting for another individual.

Individuals that delayed to synchronize provisioning with others were likely to pay an opportunity cost in terms of lost foraging time. Given this cost it makes sense to question the benefit of synchronized provisioning. A previous study on P. socius suggested that individuals synchronized provisioning to advertise their quality to other group members (Doutrelant & Covas 2007). However, we argue that this is unlikely to be a plausible explanation for provisioning synchrony in pied babblers for two reasons. First, the breeding pair are the parents of over 96% of the offspring born (M. J. Nelson-Flower, unpublished data) meaning that there is little potential for helpers to gain breeding opportunities within their group by advertising their contributions. Second, females (the dispersing sex) usually disperse alone which may preclude advertising to prospective coalition partners (Raihani 2008). Alternatively, provisioning may represent a form of rent paid by helpers to breeders (e.g. Bruintjes & Taborsky 2008). If so then helpers should target their advertisement to breeders by delaying provisioning to synchronize with breeders. Conversely, as breeders do not need to pay rent to helpers, they should not wait to synchronize provisioning. Our results did not support this hypothesis as there was no effect of status (breeder/helper) on the time provisioners delayed before returning to the nest with food. Rent payment theory also predicts that dominants will increase aggression towards, and perhaps evict, ‘lazy’ subordinates (Mulder & Langmore 1993; Bruintjes & Taborsky 2008). However, in pied babblers overt aggression between group members is extremely rare and in over five years of detailed observations just five instances of a breeder evicting a helper were witnessed (Raihani 2008), despite considerable variation in helper contributions (Ridley & Raihani, unpublished data). To sum up, although we cannot refute the possibility that synchronized provisioning has signalling characteristics in this species; given the current data this does not seem to be the most parsimonious explanation.

Instead our data suggest that provisioning synchrony increased brood survival. Several other studies have shown that brood survival decreases with parental visit rate (Skutch 1949; Martin et al. 2000; Ghalambor & Martin 2001), perhaps because several separate visits to the nest result in several separate bouts of nestling begging. Synchronized provisioning, on the other hand, enables two food items to be delivered to the nest for one bout of nestling begging. As several of the common nest predators at this site (with the exception of snakes) probably use begging as a cue to detect nests, reducing the number of begging bouts may be the mechanism by which provisioning synchrony decreases the risk of brood predation. Synchronized provisioning may therefore allow provisioners to resolve the trade-off between feeding nestlings and revealing the nest location to predators. Strategic responses to predation risk have also been shown in Gray jays Perisoreus canadensis (Linnaeus), where parents prevent non-breeders from feeding nestlings despite allowing them to feed fledglings (Strickland & Waite 2001); and in Siberian jays Perisoreus infaustus (Linnaeus), where parents avoid visiting the nest at times when the predation risk is highest (Eggers et al. 2005) and visit poorly concealed nests less often than well-hidden nests (Eggers et al. 2008). Our finding that provisioners synchronized nest visits more often when feeding older nestlings is also consistent with the hypothesis that provisioning synchrony increases brood survival as older nestlings are typically louder (Raihani & Ridley 2007a) and therefore more vulnerable to predators (Onnebrink & Curio 1991). Our observations that provisioners occasionally arrived at the nest tree alone but did not approach the nest until another individual arrived may initially seem to be inconsistent with the idea that provisioning synchrony increases nestling survival. However, it is likely to be nestling begging rather than the arrival of provisioners at the nest tree that reveals the nest’s location to predators (Haskell 1994). Nestlings do not commence begging until provisioners approach close to the nest or give food-associated ‘purr’ calls (Raihani & Ridley 2007a). As such, provisioners that wait in the nest tree are unlikely to increase the risk of predators detecting the nest.

In contrast to a previous study on this species (Raihani & Ridley 2007b), we did not find a strong effect of group size on brood survival. However, the earlier study did not control for the effects of provisioning synchrony. Provisioning synchrony increased with group size and we found some evidence to suggest that provisioners in large groups delayed longer before feeding nestlings. Bearing in mind that the effect sizes associated with our observed results are relatively small, we suggest that increased provisioning synchrony may be one mechanism by which larger groups improve brood survival in this species. If so we would not expect to find an additional effect of group size on brood survival once the effect of provisioning synchrony has been accounted for.

Two potential hypotheses that we did not test were that provisioning synchrony allowed provisioners to dilute their chances of predation during the journey to the nest tree or at the nest itself. Although we cannot quantitatively examine these hypotheses, other evidence suggests that they are unlikely explanations for synchronized provisioning in this species. Adult predation, which can be measured by monitoring groups over long periods, is extremely infrequent (Ridley & Raihani 2007a, b; unpublished data). Although it has been argued that a paucity of observed predation events may be attributable to the presence of a human observer (Isbell & Young 1993), we did occasionally witness predation events in the field. Despite this, we never witnessed provisioners being preyed upon en route to or at the nest. Although provisioners synchronized feeds, journeys to the nest tree were not always synchronized as provisioners sometimes travelled to the nest tree alone and then waited there for another provisioner to arrive before approaching and feeding the brood (N. Raihani, personal observation). In these cases, provisioning synchrony did not dilute the predation risk en route to the nest tree. Synchronous provisioning might have reduced the risk of encountering predators concealed within the nest (e.g. snakes), if provisioners waited to allow other individuals to approach the nest before them. This also seems unlikely as in contrast to species where the nest cavity is concealed from the exterior, babbler nests are open-cup so provisioners can see into the nest before they approach it to feed young.

To summarize, we suggest that synchronized provisioning is a strategy that birds can use to increase brood survival and suggest that this may be of particular importance in species where several individuals feed at a single nest. Several studies have quantified the positive effects of helpers on offspring condition and survival in cooperatively breeding species (e.g. Russell et al. 2002; Brouwer, Heg & Taborsky 2005; Hodge 2005; Ridley & Raihani 2007a). However, our results show that helper effects may amount to more than just the amount of help given: by working together provisioners may augment the benefits they confer on offspring. Behavioural synchrony may therefore represent an important, yet understudied, predictor of group productivity in cooperatively breeding species.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Meterials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Thanks to Tim Clutton-Brock for supervision. The Northern Cape Conservation Authority permitted research on pied babblers and Kotze and de Bruin families allowed us access to babbler groups on their land. Matt Bell, Krystyna Golabek, Andy Radford, Rebecca Rose, Sarah Knowles and Helen Wade helped with maintaining habituation of babbler groups. Thanks to Sinead English, Sarah Hodge, Alex Thornton and Jon Wright for useful comments and discussion. Thanks also to the Editors and two anonymous referees. This research was funded by NERC, University of Cambridge and the South African DST/NRF.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Meterials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Altmann, J. (1974) Observational study of behaviour: sampling methods. Behaviour, 49, 227267.
  • Bates, D., Maechler, M. & Dai, B. (2008) lme4: Linear Mixed-Effects Models Using S4 Classes. R Package Version 0.999375-28. R Development Core Team, Vienna, Austria, .
  • Brouwer, L., Heg, D. & Taborsky, M. (2005) Experimental evidence for helper effects in a cooperatively breeding cichlid. Behavioral Ecology, 16, 667673.
  • Brown, J.L. (1987) Helping and Communal Breeding in Birds. Princeton University Press, Princeton, NJ.
  • Bruintjes, R. & Taborsky, M. (2008) Helpers in a cooperative breeder pay a high price to stay: effects of demand, helper size and sex. Animal Behaviour, 75, 18431850.
  • Burnham, K.P. & Anderson, D.R. (2002) Model Selection and Multimodel Inference: A Practical Information-Theoretic Approach. Springer-Verlag New York Inc, New York, USA.
  • Cockburn, A. (1998) Evolution of helping behaviour in cooperatively breeding birds. Annual Reviews in Ecology and Systematics, 29, 141177.
  • Conway, C.J. & Martin, T.E. (2000) Evolution of passerine incubation behavior: influence of food, temperature, and nest predation. Evolution, 54, 670685.
  • Crawley, M.J. (2007) The R Book. John Wiley & Sons Ltd, West Sussex, UK.
  • Creel, S. & Creel, N.M. (1995) Communal hunting and pack size in African wild dogs, Lycaon pictus. Animal Behaviour, 50, 13251339.
  • Dearborn, D.C. (1999) Brown-headed cowbird nestling vocalizations and the risk of nest predation. Auk, 116, 448457.
  • Dostalkova, I. & Spinka, M. (2007) Synchronization of behaviour in pairs: the role of communication and consequences in timing. Animal Behaviour, 74, 17351742.
  • Doutrelant, C. & Covas, R. (2007) Helping has signalling characteristics in a cooperatively breeding bird. Animal Behaviour, 74, 739747.
  • Eggers, S., Griesser, M. & Ekman, J. (2005) Predator-induced plasticity in nest visitation rates in the Siberian jay (Perisoreus infaustus). Behavioral Ecology, 16, 309315.
  • Eggers, S., Griesser, M. & Ekman, J. (2008) Predator-induced reductions in nest visitation rates are modified by forest cover and food availability. Behavioral Ecology, 19, 10561062.
  • Fontaine, J.J. & Martin, T.E. (2006) Parent birds assess nest predation risk and adjust their reproductive strategies. Ecology Letters, 9, 428434.
  • Foster, W.J. & Treherne, J.E. (1981) Evidence for the dilution effect in the selfish herd from fish predation on a marine insect. Nature, 293, 466467.
  • Gaston, A.J. (1978) The evolution of group territorial behaviour and cooperative breeding. The American Naturalist, 112, 10911100.
  • Ghalambor, C.K. & Martin, T.E. (2001) Fecundity-survival trade-offs and parental risk-taking in birds. Science, 292, 494497.
  • Griffiths, R., Double, M.C., Orr, K. & Dawson, R.J.G. (1998) A DNA test to sex most birds. Molecular Ecology, 7, 10711075.
  • Haskell, D. (1994) Experimental evidence that nestling begging behaviour incurs a cost due to nest predation. Proceedings of the Royal Society of London. Series B: Biological Sciences, 257, 161164.
  • Hodge, S.J. (2005) Helpers benefit offspring in both the short and long-term in the cooperatively breeding banded mongoose. Proceedings of the Royal Society of London. Series B: Biological Sciences, 272, 24792484.
  • Hollen, L.I., Bell, M.B.V. & Radford, A.N. (2008) Cooperative sentinel calling? Foragers gain increased biomass intake. Current Biology, 18, 576579.
  • Isbell, L.A. & Young, T.P. (1993) Human presence reduces leopard predation in a population of vervet monkeys Cercopithecus aethiops. Animal Behaviour, 45, 12331235.
  • Kim, K.W., Krafft, B. & Choe, J.C. (2005a) Cooperative prey capture by young subsocial spiders: I. Functional value. Behavioral Ecology and Sociobiology, 59, 92100.
  • Kim, K.W., Krafft, B. & Choe, J.C. (2005b) Cooperative prey capture by young subsocial spiders: II. Behavioral mechanism. Behavioral Ecology and Sociobiology, 59, 101107.
  • Kokko, H., Johnstone, R.A. & Wright, J. (2002) The evolution of parental and alloparental effort in cooperatively breeding groups: when should helpers pay to stay? Behavioral Ecology, 13, 291300.
  • Leonard, M.L. & Horn, A.G. (2006) Age-related changes in signalling of need by nestling tree swallows (Tachycineta bicolor). Ethology, 112, 10201026.
  • Macgregor, N.A. & Cockburn, A. (2002) Sex differences in parental response to begging nestlings in superb fairy-wrens. Animal Behaviour, 63, 923932.
  • Maclean, G.L. (1973) The sociable weaver, part 1-5. Ostrich, 44, 176261.
  • Martin, T.E. (1995) Avian life history evolution in relation to nest sites, nest predation and food. Ecological Monographs, 65, 101127.
  • Martin, T.E., Scott, J. & Menge, C. (2000) Nest predation increases with parental activity: separating nest site and parental activity effects. Proceedings of the Royal Society of London. Series B: Biological Sciences, 267, 22872293.
  • Marzluff, J.M. & Balda, R.P. (1990) Pinyon jays: making the best of a bad job by helping. Cooperative Breeding in Birds: Long Term Studies of Ecology and Behavior (eds P.B.Stacey & W.D.Koenig), pp. 197237. Cambridge University Press, Cambridge.
  • McDonald, P.G., Marvelde, L.T., Kazem, A.J.N. & Wright, J. (2008a) Helping as a signal and the effect of a potential audience during provisioning visits in a cooperative bird. Animal Behaviour, 75, 13191330.
  • McDonald, P.G., Kazem, A.J.N., Clarke, M.F. & Wright, J. (2008b) Helping as a signal: does the removal of potential audiences alter helper behavior in the bell miner? Behavioral Ecology, 19, 10471055.
  • Meldrum, G.E. & Ruckstuhl, K.E. (2009) Mixed-sex group formation by bighorn sheep in winter: trading costs of synchrony for benefits of group living. Animal Behaviour, 77, 919929.
  • Muchai, M. & DuPlessis, M.A. (2005) Nest predation of grassland bird species increases with parental activity at the nest. Journal of Avian Biology, 36, 110116.
  • Mulder, R.A. & Langmore, N.E. (1993) Dominant males punish helpers for temporary defection in superb fairy-wrens. Animal Behaviour, 45, 830833.
  • Onnebrink, H. & Curio, E. (1991) Brood defense and age of young: a test of the vulnerability hypothesis. Behavioural Ecology and Sociobiology, 29, 6168.
  • Pinheiro, J., Bates, D., DebRoy, S. & Sarkar, D.; the Core R Team (2008) nlme: Linear and Nonlinear Mixed Effects Models. R Package Version 3. 1-89. R Development Core Team, Vienna, Austria.
  • Poiani, A. & Pagel, M. (1997) Evolution of avian cooperative breeding: comparative tests of the nest predation hypothesis. Evolution, 51, 226340.
  • R Development Core Team (2008) R: A Language and Environment for Statistical Computing. R Development Core Team, Vienna, Austria.
  • Raihani, N.J. (2008) Cooperation and conflict in pied babblers. PhD thesis, University of Cambridge, Cambridge.
  • Raihani, N.J. & Ridley, A.R. (2007a) Adult vocalizations during provisioning: offspring response and postfledging benefits in wild pied babblers. Animal Behaviour, 74, 13031309.
  • Raihani, N.J. & Ridley, A.R. (2007b) Variable fledging age according to group size: trade-offs in a cooperatively breeding bird. Biology Letters, 3, 624627.
  • Ricklefs, R.E. (1969) An analysis of nestling mortality in birds. Smithsonian Contributions to Zoology, 9, 148.
  • Ridley, A.R. & Raihani, N.J. (2007a) Variable post-fledging care in a cooperative bird: causes and consequences. Behavioral Ecology, 18, 9941000.
  • Ridley, A.R. & Raihani, N.J. (2007b) Facultative response to a kleptoparasite by the cooperatively breeding pied babbler. Behavioral Ecology, 18, 324330.
  • Ruckstuhl, K.E. (1999) To synchronise or not to synchronise: a dilemma for young bighorn males? Behaviour, 136, 805818.
  • Russell, A.F., Clutton-Brock, T.H., Brotherton, P.N.M., Sharpe, L.L., McIlrath, G.M., Dalerum, F.D., Cameron, E.Z. & Barnard, J.A. (2002) Factors affecting pup growth and survival in cooperatively breeding meerkats Suricata suricatta. Journal of Animal Ecology, 71, 700709.
  • Skutch, A.F. (1949) Do tropical birds rear as many young as they can nourish? Ibis, 91, 430455.
  • Strickland, D. & Waite, T.A. (2001) Does initial suppression of allofeeding in small jays help to conceal their nests? Canadian Journal of Zoology, 79, 21282146.
  • Wright, J. (1997) Helping-at-the-nest in Arabian babblers: signalling social status or sensible investment in chicks? Animal Behaviour, 54, 973979.
  • Zahavi, A. (1990) The quest for social status in a cooperative breeder. Cooperative Breeding in Birds: Long-Term Studies of Ecology and Behaviour (eds P.B.Stacey & W.D.Koenig), pp. 103130. Cambridge University Press, Cambridge.
  • Zahavi, A. (1995) Altruism as a handicap: the limitations of kin selection and reciprocity. Journal of Avian Biology, 26, 13.