- Email: [email protected]
Dietary wariness influences the response of foraging birds to competitors
Animal Behaviour 89 (2014) 63e69
Contents lists available at ScienceDirect
Animal Behaviour journal homepage: www.elsevier.com/locate/anbehav
Dietary wariness influences the response of foraging birds to competitors Keith McMahon a, b, Allison Conboy a, Elise O’Byrne-White a, Robert J. Thomas c, Nicola M. Marples a, b, * a b c
School of Natural Sciences, Trinity College Dublin, Dublin, Ireland Trinity Centre for Biodiversity Research, Trinity College Dublin, Dublin, Ireland School of Biosciences, Cardiff University, Cardiff, U.K.
a r t i c l e i n f o Article history: Received 30 November 2013 Final acceptance 4 December 2013 Available online 19 January 2014 MS. number: 13-00988 Keywords: bird competition conspicuousness dietary wariness domestic chick foraging strategy novelty prey choice social learning
Foraging animals must choose between familiar prey and novel prey of uncertain profitability and potential toxicity. Owing to a healthy aversion to potentially dangerous prey, foragers show an initial transient wariness of novel food (neophobia). In addition, some individuals display a much longer period of avoidance before incorporating the novel food into their diet (termed dietary conservatism). There are two stable foraging strategies found within forager populations: (1) adventurous consumers (AC) which rapidly accept novel foods and (2) foragers showing dietary conservatism (DC). The expression of these two strategies may also vary with environmental conditions. We measured the effect of competition on the plasticity of foraging strategies when domestic chicks, Gallus gallus domesticus, foraged for familiar and novel coloured crumbs with or without competitor chicks. In addition we investigated the effect of prey detectability on the response of foragers to a competitor, by making the familiar food cryptic or conspicuous. AC birds responded to competition by accepting the novel prey more quickly than when foraging alone, regardless of how hard familiar food was to find. In contrast, DC birds failed to reduce their wariness in response to competition when the competitor’s food choice was obscured. The foraging strategies of the birds were thus found to be plastic in their expression, but this plasticity differed between inherently AC and DC individuals. The implications of these results are discussed in relation to the foraging strategies of wild and domestic birds. Crown Copyright Ó 2013 Published on behalf of The Association for the Study of Animal Behaviour by Elsevier Ltd. All rights reserved.
Foragers must decide whether to spend their time and energy searching for known profitable prey, or whether to risk dietary expansion to include novel prey of unknown palatability and profitability. Most animals (Barnett, 1958; Brigham & Sibley, 1999) show some degree of ‘dietary wariness’, causing them to hesitate before approaching and consuming unfamiliar food. Dietary wariness comprises two distinct behavioural processes, neophobia and dietary conservatism (DC). These operate over different timescales and differ in important respects. The initial transient fear of novelty, termed neophobia, was first described in rats encountering new objects (Barnett, 1958). Neophobia has since been observed in many animal groups (including birds, fish, mammals; reviews in Brigham & Sibley, 1999; Kelly & Marples, 2004; Mappes, Marples, & Endler, 2005; Marples, Kelly, & Thomas, 2005), in response not only to objects, but also to novel foods. This hesitation in approach is typically brief, lasting only a few minutes in most animals (Marples * Correspondence: N. M. Marples, School of Natural Sciences, Trinity College Dublin, Dublin 2, Ireland. E-mail address: [email protected] (N. M. Marples).
& Kelly, 1999), and is followed by investigation of the novel food or object (Coppinger, 1969). DC is a secondary, more lasting refusal to accept novel food into their diets by some (but not all) members of a population (Marples, Roper, & Harper, 1998; Thomas, Bartlett, Marples, Kelly, & Cuthill, 2004; Thomas, Marples, Cuthill, Takahashi, & Gibson, 2003). It is thus the time between first contact with the novel food (the end of neophobia) and the consumption of the novel food whenever it is encountered. The combined durations of neophobia plus DC can together be described as dietary wariness, incorporating the entire process of novel food acceptance. DC has been demonstrated both in birds (eight species; Marples et al., 2005) and in fish (five species; Thomas et al., 2010; Richards, Thomas, Marples, Snellgrove, & Cable, 2011; Richards et al. 2014). Marples and Kelly (1999) argued that neophobia and DC are two distinct processes, not only because of the large differences in duration of the avoidance, but also because neophobia is much easier to deactivate with experience of novel prey (Marples, Quinlan, Thomas, & Kelly, 2007). In addition, DC is a much more complex behaviour, comprising several steps (Marples & Kelly,
0003-3472/$38.00 Crown Copyright Ó 2013 Published on behalf of The Association for the Study of Animal Behaviour by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.anbehav.2013.12.025
64
K. McMahon et al. / Animal Behaviour 89 (2014) 63e69
1999), and is not correlated with neophobia (K. McMahon, unpublished data). All populations tested in either laboratory or field experiments have shown that only some individuals express DC (Marples et al., 1998), while other members of the same foraging population are quick to sample and incorporate new food items into their diets, and are known as adventurous consumers (AC; Marples et al., 2007; Thomas et al., 2010). This stable polymorphism of two foraging strategies within the population has been shown to be a genetically mediated trait in Japanese quail, Coturnix coturnix japonicus (see Marples & Brakefield, 1995). Because neophobia lasts only a few minutes in domestic chicks, Gallus gallus domesticus, the behavioural strategy of each individual can be identified using a relatively brief DC test (Marples & Kelly, 1999). While all individuals initially exhibit neophobia, DC individuals refuse to eat novel prey even after they have approached and handled the item (i.e. overcome their neophobia) or observed conspecifics eating it. Rather than a fear of approaching or touching the novel prey, DC is expressed as a persistent unwillingness of these individuals to broaden their diets to include the novel prey (Marples et al., 1998). It is thus possible to measure the two aspects of dietary wariness separately, with neophobia being the length of time until an individual first touches or handles a novel prey item, and DC being the remaining time before the novel prey is eaten consistently (i.e. eaten each time it is encountered; Kelly & Marples, 2004). The expansion of an individual forager’s diet therefore requires both its neophobia and its DC to be overcome (Marples et al., 2005, 2007). The fact that DC has been documented in two very divergent taxa (birds and fish), suggests that it may be a widespread foraging strategy (Marples et al., 2005; Marples & Kelly, 1999; Thomas et al., 2010). For any species with a limited food supply, DC individuals in the population may be at a foraging disadvantage if they continue to avoid novel foods, but reducing their conservatism would also expose them to greater risks. A conservative approach to dietary sampling in individuals within a generalist species would help to minimize negative or unprofitable experiences, such as consumption of toxic prey (Lee, Marples, & Speed, 2009). Therefore, despite the reduction in potential food options, conservative feeders that foraged only on familiar prey of known value would avoid the risk of injury, illness or death through dietary indiscretion. These DC individuals might maintain their fitness levels alongside AC individuals, as the AC foragers may fall victim to prey defences at a greater rate, negating some of the fitness advantages conferred by a larger menu (Thomas et al., 2010). In addition, it is possible that DC foragers become more skilled at finding and handling their few favoured prey types, while AC foragers, with their wider diet, are likely to be less skilled at detection and handling of their many types of prey (Sherry & McDade, 1982). Investigating such choices in foraging strategies is crucial to understand the ecology and evolution both of foraging species and of their prey. Despite the relevance of understanding the foraging strategies of individuals, limited attention has been given to the plasticity of expression of these strategies under differing foraging conditions. It might be expected that individual experience and ecological circumstances may modify the expression of the underlying genetic propensity of an individual to be AC or DC. Indeed, there is evidence that the expression of AC or DC in individuals can be modified by experience. For example, Marples et al. (2007) showed that the expression of DC can be reduced in domestic chicks through extensive experience with benign novel foods, although full reversion to DC will occur after only one experience with an unpalatable food. Further evidence (Barnett, 2007; Sherratt, 2002; Skelhorn & Rowe, 2006a, 2006b) suggests that increased hunger levels may temporarily decrease DC expression towards aposematic prey. To understand the degree of plasticity in DC expression, we aimed in this study to evaluate the effects of conspecific
competition as an external factor affecting the costs and benefits of different foraging strategies. Birds foraging in a flock are likely to compete with other flock members for food, through both interference and exploitation competition (Sih, 1993) which together reduce the amount of time that an individual can spend making foraging decisions. Optimal foraging theory would predict individuals experiencing high levels of competition would broaden their diets to include less valuable or possibly less familiar foods, so as to avoid starvation (Stephens & Krebs, 1986). Additionally, social learning may further encourage broadening of the diet to incorporate unfamiliar prey on which conspecifics are observed to feed (Fryday & Greig-Smith, 1994). The combination of competition and social learning may therefore favour a relaxation of DC expression, but it is also possible that birds may ignore conspecific foraging behaviours and remain inflexible in their level of DC (cf. Richards et al., 2011). We used generalist foragers, domestic chicks, to test the hypothesis that DC expression is reduced when a forager is in competition with a conspecific that readily eats novel coloured food. A further experiment then evaluated DC expression in the same system, but in relation to the relative conspicuousness of novel and familiar foods. We hypothesized that when familiar food is relatively cryptic, increased search and attention costs will encourage foragers to reduce their expression of DC, and sample the more conspicuous novel prey (Jones, Krebs, & Whittingham, 2006). This study is the first to explore the degree of plasticity of AC and DC foraging strategies by birds in a complex foraging environment. METHODS Outline of Experiments To classify individual chicks as DC or AC, before each experiment we first tested each chick for its baseline level of DC using a ‘DC test’. We carried out three experiments to investigate: (1) the effects of testing for DC on the subsequent behaviour of the chicks; (2) the effects of a competitor on the expression of DC when the novel and familiar foods were equally conspicuous; and (3) the effects of a competitor on the expression of DC when the novel prey was more conspicuousness than the familiar prey. Chick Husbandry We carried out the three experiments using three separate batches of male ‘Cobb 500’ strain chicks, which were obtained at 1 day old from a commercial hatchery (Annyalla Chicks Ltd., Castlebury, Co. Monaghan, Ireland). The experiments were approved by the ethics committee of Trinity College Dublin, and the chicks were held under licence from the Department of Health and Children, Ireland number B100/2756 held by N. Marples. Batch 1 (used for experiment 1) contained 29 chicks, batch 2 (experiment 2) contained 55 chicks and batch 3 (experiment 3) contained 63 chicks. For each batch, individuals were housed together in a wooden holding pen (200 cm long 60 cm wide) with wood shavings covering the floor. The chicks were maintained on a 12:12 h light:dark regime, under standard fluorescent lights in addition to natural lighting. Ambient temperature was 24 4 C, but chicks could find warmer areas by standing under two infrared heat lamps suspended above their holding pen. From their arrival at the laboratory, the chicks were fed chick starter crumbs (William Connolly & Sons, Red Mills, Goresbridge, Co. Kilkenny, Ireland), dyed green (O’Brien’s liquid green 90; 1 ml dye: 30 ml water), which resulted in their subsequent treatment of this colour as familiar. Food and water were available ad libitum for the duration of the study, except
K. McMahon et al. / Animal Behaviour 89 (2014) 63e69
for the hour prior to the trials when food (but not water) was removed, to encourage active foraging during the tests. Each chick was marked with a unique colour combination on the head to distinguish between individuals. As in our previous studies, the chicks did not respond to their own marks or those of their conspecifics (Marples et al., 2007). At the end of the experiment, as the chicks were male and could not be rehoused on a farm or reused by any other researchers, in accordance with Irish welfare legislation they were euthanized. This was carried out by a qualified technician by means of intraperitoneal injection of 0.5 ml Euthatal in accordance with the licence B100/2756. Experimental Foods The chick starter crumbs (originally light brown) used in the study were dyed unique colours to be visually distinct and easily noticeable to the chicks (Marples et al., 2007). Crumbs were mixed with the dye solution and stirred for 3 min until all the crumbs had taken up the dye; they were then left to air dry completely before being used. Green was used as the familiar colour throughout the study, and green crumbs were provided ad libitum throughout, except where specified in the experimental methods below. Blue, yellow, orange and red coloured crumbs served as novel prey items during the experiments (coloured using O’Brien’s liquid blue (FCF/ 0.5KGS), O’Brien’s lemon yellow (T/0.5K), O’Brien’s liquid orange (SY/0.5K), O’Brien’s Christmas red; in each case diluting 1 ml of dye in 30 ml of water). Pretraining When the chicks were 2 days old, familiarization with foraging in the test arena began. The arena was a wooden square enclosure measuring 84 cm 84 cm, with 21 cm high walls. The wooden floor was covered with 13 13 rows of evenly spaced circular depressions spaced 3 cm apart, into each of which a single chick crumb could be placed. Since solitary chicks tend to become agitated and refuse to forage, two ‘buddy chicks’ were present in the arena to prevent this. The buddy chicks were allowed to eat to satiation before the trial, then kept in a ‘buddy area’ to prevent them from interfering with the foraging of the test chick (Marples et al., 2007). This buddy area square (A1 in Fig. 1) was fenced off from the rest of the arena with chicken wire to allow for visual and auditory contact between the test and buddy chicks. The buddy chicks were drawn from a pool of additional chicks not used as subjects in the experiments. New buddy chicks were randomly selected for each trial. Each test chick was familiarized with foraging in the arena by being offered familiar green crumbs during five successive
A
B
C
D
1
2
84 cm
3
4 Figure 1. Diagram of the floor of the foraging arena showing dimensions. Square A1 was the ‘buddy area’ and was unavailable to foraging chicks.
65
‘pretraining’ sessions over one and a half days. In the first pretraining session, the test chicks were allowed to forage in the training arena in a group of five chicks. Each subsequent session contained progressively fewer foraging companions, so that the final session involved a lone chick foraging in the arena but with the two buddy chicks added to the buddy area. All of the experimental chicks were watched to confirm that they consumed the familiar food in all of the pretraining sessions. Measuring DC Expression (DC Test) After the pretraining sessions, each of the chicks in batches 2 and 3 were tested individually for their levels of neophobia (latency to peck at the novel food) and DC (latency to eat three pieces of novel food once neophobia had been overcome). Our previous studies have shown that the latency to eat three food items signifies the acceptance of that particular novel food item into the diet, and thus an overcoming of DC (Marples & Kelly, 1999). In these ‘DC tests’, chicks were given a choice between a 50:50 ratio of familiar (green) and novel (orange) food. Each chick was allowed to forage individually for a 10 min period. Latency to eat three orange crumbs gave their ‘DC score’ with higher scores indicating more DC birds, as they had taken longer to eat the novel food. Chicks that did not eat any of the novel food within the 10 min were given the highest possible score (600 s) and classified as DC. Chicks that ate the novel food within this period (i.e. a latency of less than 600 s) were classified as AC individuals. In both batches of chicks the four most AC individuals were chosen as the competitors for the experiment and were housed together (but separately from all other chicks). All groups of chicks were held in identical pens in the same room under the same conditions. Pretraining of Competitor Chicks All competitor chicks were trained to eat both familiar and novel coloured crumbs, both as their maintenance diet in their housing pen and in the foraging arena prior to subsequent trials. Once all of the competitor chicks were readily eating both colours of crumbs, the experimental trials began. Experiment 1: Effects of Testing for DC on Behaviour Marples et al. (2007) reported that dietary wariness could be deactivated through experience of novel foods. Thus, by taking part in the initial DC test described above (which was necessary to establish a DC score, and therefore identify each test individual’s underlying foraging strategy), it was possible that chicks might have reduced their dietary wariness as a result of the exposure to novel food. If so, this would have affected their responses to novelty in subsequent experiments. However, Marples et al. (2007) showed that DC was only reduced by long-term continuous exposure to the novel food in a social setting. Short-term exposures such as those experienced in the DC tests should not have been enough to reduce DC. Nevertheless, to ensure that the DC tests were not confounding our results, we tested whether this brief exposure to novelty was sufficient to reduce DC. We used the first batch of chicks for this purpose. One group (16 chicks) received a DC test in which they were presented with three pieces of familiar green food and three pieces of novel red food. They were exposed to the novel food for three trials, each of 3 min, with at least 2 h between each trial. The other 13 chicks were used as controls, receiving the same trials, but only familiar green food. This protocol differed from the one used in experiments 2 and 3 because it represented a more robust test of the effect of exposure to novelty. Chicks tend to lose interest in tasks longer than 3 min, so while keeping the length of exposure
K. McMahon et al. / Animal Behaviour 89 (2014) 63e69
similar to experiments 2 and 3 (below) we divided the DC test into three trials. The day after the first trial, all 29 chicks were tested for DC in exactly the same manner, but this time all chicks were given a choice between familiar green food and novel blue food. The total latency to eat three pieces of novel food across the three trials was recorded for each test, allowing us to detect any effect of the first DC test on the chicks’ latency to eat a new novel colour of food in the second test. Experiment 2: Conspicuous Foods Here, we assessed the effect of the presence of a competitor on an individual’s expression of dietary wariness. The second batch of chicks was first tested for DC (as described above for the first experiment) before progressing into trials with and without a competing forager. The foraging arena was prepared with two familiar green crumbs and two novel blue crumbs within each of the 15 test squares of the arena. Single crumbs were placed into randomly chosen depressions in the test square (Fig. 1). This gave a total of 60 available crumbs (30 familiar coloured and 30 novel coloured) per trial. Each trial lasted 8 min, as chicks tended to abandon foraging and focus on the buddy chicks or go to sleep after that time. We recorded the latency to peck a novel crumb, the number and time of subsequent pecks and the latency to eat a novel crumb. If no novel crumbs were eaten, the chick was noted as having taken 480 s, the full duration of the trial. Ten AC chicks and 10 DC chicks served as control groups and were each tested foraging alone in the arena (i.e. without competition from another chick, but with buddies present). The two experimental groups, 10 AC and 10 DC chicks, were observed foraging in competition with one chick from the pool of four competitor chicks. The competitor chicks were used in rotation, so as to maintain sufficient hunger levels to ensure that they foraged during the trials. In each competition trial, both the test chick and the competitor chick were placed in the arena at the same time, at opposite ends of the front row (A4 and D4) and allowed to forage together for 8 min. As with the control group, latencies to eat familiar green crumbs, to peck novel blue crumbs and to eat novel blue crumbs were recorded. These trials took place over 2 days. Experiment 3: Cryptic Familiar versus Conspicuous Novel Food The third experiment was carried out using the third batch of chicks, which were divided into two treatment groups exactly as in experiment 2. The familiar green crumbs were made cryptic by covering the foraging arena floor with a thin layer of wood shavings which had been dyed green with the same food colour as the familiar food. These ‘distractors’ served to obscure the green crumbs and so impair foraging. Prior to each trial the arena was set up with two familiar green crumbs and two novel blue crumbs in each square sector, as in the previous experiment. Trials lasted 8 min and latencies to eat green crumbs, to peck blue crumbs and to eat blue crumbs were recorded, with a maximum possible latency of 480 s. The same variables were recorded during these trials as in experiment 2. Data Analysis The time taken for each chick to overcome neophobia (latency to peck at a novel food item) and dietary wariness (latency to eat three novel food items, including both neophobia and DC components) was measured for AC and DC chicks in each treatment group across experiments 2 and 3 (no competition þ no distractors, competition þ no distractors, no competition þ distractors, competition þ distractors). Analyses of data were carried out using the statistical software R (version 2.13.2, R Development Core Team, 2009). Generalized
linear models (GLMs) were used to examine the role of different fixed factors in influencing the different measures of dietary wariness (neophobia and DC) recorded in the experiment. The factors considered were: (1) whether the focal chick had been classified as an AC or DC individual, (2) competition (i.e. presence of a competitor chick) or no competition, (3) distractors (i.e. green coloured wood shavings rendering the familiar coloured prey less conspicuous) or no distractors. Model selection and validation procedures followed Crawley (2007) and Zuur, Ieno, Walker, Saveliev, and Smith (2009). Neophobia latency data were analysed using Gaussian GLMs (with identity or log-link functions), whereas dietary wariness latency data were analysed using Poisson GLMs (with log-link functions). The data from experiment 1 were analysed using gamma GLMs (with inverse-link functions). The results of experiments 2 and 3 were compared in the analyses below, using a gamma GLM (with inverse-link functions) to investigate whether differences in latencies to eat three pieces of novel food existed between chicks from batches 2 and 3. All analyses used two-tailed hypothesis testing. RESULTS Experiment 1: Effects of Testing for DC on Behaviour The results of the first experiment revealed that there was no difference in DC expressed in the second trial of this experiment between the control group, which had not previously received a DC test, and the treatment group which had received a DC test in the first trial of the experiment (gamma GLM, inverse link: t1,28 ¼ 0.537, P ¼ 0.596; Fig. 2). Thus the exposure to novelty in the context of a DC test was not enough to reduce subsequent DC and therefore did not confound the results of the following experiments. DC Tests for Experiments 2 and 3 The initial DC test scores for chicks in batches 2 and 3 based on the latency to eat three novel coloured crumbs when foraging alone allowed us to define 42 chicks as AC, as they accepted the novel food within 10 min. The other 66 chicks were classified as DC as they refused to accept the novel food within the 10 min trial (Fig. 3). The difference in DC scores between chicks in the two batches was marginally nonsignificant (gamma GLM, inverse link: F1,116 ¼ 1.307, P ¼ 0.054). Factors Influencing Neophobia and Dietary Wariness As there were no differences in methodology between experiments 2 and 3, they were analysed together to explore simultaneously the roles of competition pressure and prey crypticity. The
300 DC Score (s)
66
200 100 0 DC Test
No DC Test Treatment group
Figure 2. Box plot of the dietary wariness of chicks for both treatment groups of experiment 1: chicks that received a DC test and those that did not. The box plots show the median and 25th and 75th percentiles; the whiskers indicate the values within 1.5 times the interquartile range and the circle is an outlier.
K. McMahon et al. / Animal Behaviour 89 (2014) 63e69
identity link, square root-transformed dependent variable: F1,85 ¼ 4.1274, P ¼ 0.045). Similarly, in the model explaining the time taken to eat three novel prey (dietary wariness), the three-way interaction term was also significant (Poisson GLM, log-link function, likelihood ratio test: LRT1,85 ¼ 1826.8, P < 0.001). In other words, the manner in which prey conspicuousness, and the competitive environment, influenced both neophobia and dietary wariness differed significantly between AC and DC focal birds. We therefore analysed the AC birds separately from the DC birds in our further analyses.
AC Chicks | DC Chicks
600 Latency (s)
500 400 300 200 100 0 0
20
40
60
80
100
120
Chick ID
Foraging Behaviour of AC Birds
Figure 3. Scatter plot depicting the initial DC test scores for all chicks, used in their assignment as AC or DC birds. DC test scores represent the latencies (s) for each chick to eat three novel coloured crumbs when foraging alone.
For the AC birds (Fig. 4a), the model explaining neophobia did not show a significant two-way interaction between the presence of a competitor and the presence of distractors (Gaussian GLM, log link, square root-transformed dependent variable: F1,34 ¼ 2.156, P ¼ 0.158). In other words, there was no significant difference in the effect of a competitor between experiment 2, when distractors were absent, and experiment 3, when distractors were present. When the two-way interaction was excluded from the model, there was a significant effect of competition (Gaussian GLM, log link, square root-transformed dependent variable: F1,35 ¼ 7.4322, P ¼ 0.009), with the presence of a competitor chick being associated with a reduction in neophobia, whereas there was no significant effect of the presence/absence of distractors (Gaussian GLM, log link, square root-transformed dependent variable: F1,35 ¼ 0.7325, P ¼ 0.398). In other words, the AC chicks reduced their neophobia in response to a competitor, but distractors did not affect this response. When the dietary wariness in the AC birds was analysed, a similar pattern as for neophobia was evident. However, in this case
results revealed distinct differences between AC and DC foragers in terms of the plasticity of expression of both the time to peck at the novel prey (i.e. their neophobia) and the time to eat three novel prey (i.e. their dietary wariness, incorporating both neophobia and DC). These two responses were measured in the presence or absence of competition pressure and under two levels of prey crypticity (with and without distractors resembling the familiar food). We first tested for the three-way interaction between (1) whether the focal chick had initially exhibited an AC or a DC foraging strategy, (2) the presence of a competitor chick and (3) prey conspicuousness (presence of distractors). In the model explaining the time taken to peck at the novel prey (neophobia), the three-way interaction term was significant (Gaussian GLM,
Neophobia 600
67
|
Dietary wariness
(a)
500 400 300 200
Latency (s)
100 0 NC−ND C−ND NC−D 600
C−D
NC−ND C−ND NC−D
C−D
C−D NC−ND C−ND NC−D Treatment groups
C−D
(b)
500 400 300 200 100 0 NC−ND C−ND NC−D
Figure 4. Box plot of the median latency for (a) AC and (b) DC designated chicks to end neophobia and dietary wariness, under the conditions of the four treatments, as indicated below each box plot. NC ¼ no competitor present; C ¼ competitor present; ND ¼ no distractors present; D ¼ distractors present, making the familiar food less conspicuous than the novel food. The box plots show the median and 25th and 75th percentiles; the whiskers indicate the values within 1.5 times the interquartile range and the circles are outliers.
68
K. McMahon et al. / Animal Behaviour 89 (2014) 63e69
the two-way interaction between competition and prey crypticity was highly significant (Poisson GLM, log link: LRT1,34 ¼ 1453, P < 0.0001). In other words, when competition was present there was a reduction in wariness, and this reduction was more pronounced in the experiment with distractors present compared to the experiment with no distractors (Fig. 4a). Foraging Behaviour of DC Birds In the DC birds (Fig. 4b), the model explaining neophobia included a nonsignificant interaction between the presence of a competitor and the presence of distractors (Gaussian GLM, log link, square root-transformed dependent variable: F1,52 ¼ 1.0682, P ¼ 0.306). When the two-way interaction was excluded from the model, there was a significant independent effect of competition on neophobia (Gaussian GLM, log link, square root-transformed dependent variable: F1,52 ¼ 4.391, P ¼ 0.041). There was no significant independent effect of distractors (F1,52 ¼ 1.068, P ¼ 0.306). In other words, competition reduced neophobia, but prey conspicuousness did not. In contrast, the model explaining dietary wariness included a highly significant interaction between the presence of a competitor and the presence of distractors (Poisson GLM, log link: LRT1,51 ¼ 391.11, P < 0.0001). In other words, wariness was reduced in response to competition when there were no distractors present, but no such effect was found when familiar prey was more cryptic than novel prey (Fig. 4b). In summary, Fig. 4b shows that competition reduced the expression of both neophobia and dietary wariness in DC birds only in experiment 2, that is, when the chicks could clearly see what their companion was eating. In experiment 3 where the familiar crumbs were cryptic, the presence of a competitor had no significant effect on the latency to peck at, and eat, the novel crumbs. This contrasts with the AC chicks, which significantly reduced their latency in these circumstances. DISCUSSION The findings of this study highlight the contrast between AC and DC foraging strategies. While it has previously been shown that individuals have a genetic propensity towards being either AC or DC foragers (Marples & Brakefield, 1995), this study demonstrates for the first time that both competition and prey conspicuousness may modulate the expression of these foraging strategies. Furthermore, this modulation differed depending on whether the individual was inherently an AC or a DC forager. This demonstrates that the underlying foraging strategies of these birds are plastic to some degree, as birds that were strongly DC (in their DC test) acted as far less conservative in some experimental circumstances. When faced with a competitor that readily ate both types of food, the AC chicks increased their acceptance of novel food, but did so significantly only when distractors made the familiar food cryptic. In other words, when the AC birds had a rival competing for the food, and had to work harder to find the familiar food, they gave up any preference for it and ate the novel food (which was more conspicuous). In contrast, the DC birds accepted the novel food more quickly in the presence of a rival forager only when there were no distractors to disguise the prey. They maintained their avoidance of the novel prey when a rival was present, even though distractors made the familiar prey more cryptic. Thus, under competition, DC birds remained highly conservative in their foraging under circumstances when we would expect DC foraging to be more costly, because familiar prey was more difficult to find. This finding contrasts with previous work which has shown a reduction in dietary wariness in circumstances where a
conservative foraging strategy would be expected to be costly (Barnett, Bateson, & Rowe, 2007; Lee et al., 2009; Mappes et al., 2005; Richards et al., 2011). However, in our study, DC foragers may have benefitted from increased efficiency in foraging if they were able to form a search image for the cryptic familiar food (Tinbergen, 1960). AC foragers, in contrast, expanded their diet to include novel prey instead of specializing on the familiar food. The differences between AC and DC birds in their responses to the rival forager suggest that they may have perceived the rival in different ways. The AC birds responded to a competitor as predicted, by increasing their intake of the novel food when it was more conspicuous than the alternative food source. Competition pressure reduces time available for viewing prey (Gamberale-Stille, 2001; Guilford, 1986) decreasing attack latencies of competing chicks feeding on both palatable and aposematic prey (GamberaleStille, 2001). This is consistent with the observed decreases in both neophobia and DC expression in AC birds demonstrated in our study. In contrast, the DC birds reduced their wariness in the presence of a competitor only when they could see the food clearly (treatments without distractors), and remained fully averse to novel food when there were distractors present. This result could be interpreted as the DC birds considering the companion forager as a demonstrator of food choice in a social context, in that they appeared to copy the choice of novel food more readily when both foods were clearly visible. Such social learning based on observation can increase feeding rate (Fryday & Greig-Smith, 1994; Gentle, 1985; Wauters, Richard-Yris, & Talec, 2002) as chicks can learn to identify profitable foods by watching and copying others (Wauters et al., 2002). It is possible that seeing the competitor chick eat the novel food without harm encouraged the DC foraging chick to sample and accept the novel food as palatable and profitable, a behavioural modification previously seen in house sparrows, Passer domesticus (see Fryday & Greig-Smith, 1994). The differences between AC and DC foragers in their response to a competitor supports the hypothesis that these are two distinct processes (Marples & Kelly, 1999) even though the precise mechanism through which they act is not yet known. In addition, the existence of AC and DC foraging strategies as a dichotomy, rather than as a continuum of latencies to attack novel prey, is a matter of some debate which has yet to be resolved. The genetic basis and heritability of the trait (Marples & Brakefield, 1995) support the view that these strategies are distinct. Individuals show consistent behaviour towards novel foods across a range of foraging contexts. Differences in behaviours other than latency to accept novel foods that correlate with the two foraging strategies, such as those found between AC and DC individuals in the present experiment, add support to the view that they are distinct strategies. In this study, the expression of AC and DC foraging strategies was modified by both competition and food conspicuousness in different ways. This study is the first to demonstrate plasticity in the expression of DC in response to competition. These insights into the plasticity of DC in both AC and DC individuals have several important applications. For instance, the domestic poultry industry may benefit by allowing birds to observe each other foraging whenever food types need to be changed, reducing the drop in growth rate that is typical of dietary changes (Gentle, 1985; Jones, 1986; Marples et al., 2007). Furthermore, reintroduction programmes for endangered socially foraging species may be more successful if natural levels of DC expression are enabled to be expressed by offering social groups of individuals a variety of foods prior to release. Ecosystem changes may result in an alteration in prey available to predators (Møller, Fiedler, & Berthold, 2010; Schaefer, Jetz, & Böhning-Gaese, 2008; Walther et al., 2002). As a result of any of these changes in prey availability, increased competition between foragers could also
K. McMahon et al. / Animal Behaviour 89 (2014) 63e69
change the relative costs and benefits of the two strategies, and hence change the balance between AC and DC foragers in the population. Understanding how AC and DC foragers respond to such changes will help to predict shifts in ecosystem function, and to assist in prioritizing conservation actions. Our results demonstrate that birds using the AC and DC foraging strategies differ markedly in their response to the combination of competition and prey crypticity. Determining both the drivers and effects of DC expression in foragers will help to uncover evolutionary relationships between predators and prey, and to discover which ecological conditions maintain the observed polymorphism in foraging strategies. Acknowledgments We thank Peter Stafford and Alison Boyce for help with bird husbandry throughout this project. We are also grateful to the journal editors and anonymous referees for improvements to the manuscript. This work was part funded by the IRCSET Embark Initiative. References Barnett, C. A., Bateson, M., & Rowe, C. (2007). State-dependent decision making: educated predators strategically trade off the costs and benefits of consuming aposematic prey. Behavioral Ecology, 18, 645e651. Barnett, S. A. (1958). Experiments on ‘neophobia’ in wild and laboratory rats. British Journal of Psychology, 49, 195e201. Brigham, A. J., & Sibley, R. M. (1999). A review of the phenomenon of neophobia. In P. D. Cowan, & C. J. Feare (Eds.), Advances in vertebrate pest management (pp. 67e 84). Furth, Germany: Filander Verlag. Coppinger, R. P. (1969). The effect of experience and novelty on avian feeding behavior with reference to the evolution of warning coloration in butterflies. Part I: reactions of wild-caught adult blue jays to novel insects. Behaviour, 35, 45e60. Crawley, M. J. (2007). The R book. Chichester, U.K.: J. Wiley. Fryday, S. L., & Greig-Smith, P. W. (1994). The effects of social learning on the food choice of the house sparrow (Passer domesticus). Behaviour, 128, 281e300. Gamberale-Stille, G. (2001). Benefit by contrast: an experiment with live aposematic prey. Behavioral Ecology, 12, 768e772. Gentle, M. J. (1985). Sensory involvement in the control of food intake in poultry. Proceedings of the Nutrition Society, 44, 313e321. Guilford, T. (1986). How do ‘warning colours’ work? Conspicuousness may reduce recognition errors in experienced predators. Animal Behaviour, 34, 286e288. Jones, K. A., Krebs, J. R., & Whittingham, M. J. (2006). Interaction between seed crypsis and habitat structure influence patch choice in a granivorous bird, the chaffinch Fringilla coelebs. Journal of Avian Biology, 37, 413e418. Jones, R. B. (1986). Responses of domestic chicks to novel food as a function of sex, strain and previous experience. Behavioural Processes, 12, 261e271. Kelly, D. J., & Marples, N. M. (2004). The effects of novel odour and colour cues on food acceptance by the zebra finch. Animal Behaviour, 68, 1049e1054. Lee, T. J., Marples, N. M., & Speed, M. P. (2009). Can dietary conservatism explain the primary evolution of aposematism? Animal Behaviour, 79, 63e74.
69
Mappes, J., Marples, N. M., & Endler, J. A. (2005). The complex business of survival by aposematism. Trends in Ecology & Evolution, 20, 598e603. Marples, N. M., & Brakefield, P. M. (1995). Genetic variation for the rate of recruitment of novel insect prey into the diet of a bird. Biological Journal of the Linnean Society, 55, 17e27. Marples, N. M., & Kelly, D. J. (1999). Neophobia and dietary conservatism: two distinct processes? Evolutionary Ecology, 13, 641e653. Marples, N. M., Kelly, D. J., & Thomas, R. J. (2005). Perspective: the evolution of warning colouration is not paradoxical. Evolution, 59, 933e940. Marples, N. M., Roper, T. J., & Harper, D. G. C. (1998). Responses of wild birds to novel prey: evidence of dietary conservatism. Oikos, 83, 161e165. Marples, N. M., Quinlan, M., Thomas, R. J., & Kelly, D. J. (2007). Deactivation of dietary wariness through experience of novel food. Behavioral Ecology, 18, 803e810. Møller, A. P., Fiedler, W., & Berthold, P. (2010). Effects of climate change on birds. Oxford, U.K.: Oxford University Press. R Development Core Team. (2009). R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. Richards, E. L., Thomas, R. J., Marples, N. M., Snellgrove, D. L., & Cable, J. (2011). The expression of dietary conservatism in solitary and shoaling 3-spined sticklebacks Gasterosteus aculeatus. Behavioral Ecology, 22, 738e744. Richards, E. L., Alexander, L. G., Snellgrove, D., Thomas, R. J., Marples, N. M., & Cable, J. (2014). Variation in the expression of dietary conservatism within and between fish species. Animal Behaviour, 88, 49e56. Schaefer, T. C., Jetz, W., & Böhning-Gaese, K. (2008). Impact of climate change on migratory birds: community reassembly vs. adaptation. Global Ecology and Biogeography, 17, 38e49. Sherratt, T. N. (2002). The coevolution of warning signals. Proceedings of the Royal Society B, 269, 741e746. Sherry, T. W., & McDade, L. A. (1982). Prey selection and handling in two neotropical hover-gleaning birds. Ecology, 63, 1016e1028. Sih, A. (1993). Effects of ecological interactions on forager diets: competition, predation risk, parasitism and prey behaviour. In R. N. Hughes (Ed.), Diet selection: An interdisciplinary approach to foraging behaviour (pp. 182e208). Oxford, U.K.: Blackwell Scientific Publications. Skelhorn, J., & Rowe, C. (2006a). Taste-rejection by predators and the evolution of unpalatability in prey. Behavioral Ecology and Sociobiology, 60, 550e555. Skelhorn, J., & Rowe, C. (2006b). Avian predators taste-reject aposematic prey on the basis of their chemical defence. Biology Letters, 2, 348e350. Stephens, D. W., & Krebs, J. R. (1986). Foraging theory. Chichester, U.K.: Princeton University Press. Thomas, R. J., Bartlett, L., Marples, N. M., Kelly, D. J., & Cuthill, I. C. (2004). Prey selection by wild birds can allow novel and conspicuous colour morphs to spread in prey populations. Oikos, 106, 285e294. Thomas, R. J., King, T. A., Forshaw, H. E., Marples, N. M., Speed, M. P., & Cable, J. (2010). The response of fish to novel prey: evidence that dietary conservatism is not restricted to birds. Behavioral Ecology, 21, 669e675. Thomas, R. J., Marples, N. M., Cuthill, I. C., Takahashi, M., & Gibson, E. (2003). Dietary conservatism may facilitate the initial evolution of aposematism. Oikos, 101, 458e466. Tinbergen, L. (1960). The natural control of insects in pine woods: I. Factors influencing the intensity of predation by songbirds. Archives Néerlandaises de Zoologie, 13, 265e343. Walther, G. R., Post, E., Convey, P., Menzel, A., Parmesan, C., Beebee, T. J. C., et al. (2002). Ecological responses to recent climate change. Nature, 416, 389e395. Wauters, A. M., Richard-Yris, M. A., & Talec, N. (2002). Maternal influences on feeding and general activity in domestic chicks. Ethology, 108, 529e540. Zuur, A. F., Ieno, E. N., Walker, N., Saveliev, A. A., & Smith, G. M. (2009). Mixed effects models & extensions in ecology with R. Secaucus, NJ: Springer.
Contents lists available at ScienceDirect
Animal Behaviour journal homepage: www.elsevier.com/locate/anbehav
Dietary wariness influences the response of foraging birds to competitors Keith McMahon a, b, Allison Conboy a, Elise O’Byrne-White a, Robert J. Thomas c, Nicola M. Marples a, b, * a b c
School of Natural Sciences, Trinity College Dublin, Dublin, Ireland Trinity Centre for Biodiversity Research, Trinity College Dublin, Dublin, Ireland School of Biosciences, Cardiff University, Cardiff, U.K.
a r t i c l e i n f o Article history: Received 30 November 2013 Final acceptance 4 December 2013 Available online 19 January 2014 MS. number: 13-00988 Keywords: bird competition conspicuousness dietary wariness domestic chick foraging strategy novelty prey choice social learning
Foraging animals must choose between familiar prey and novel prey of uncertain profitability and potential toxicity. Owing to a healthy aversion to potentially dangerous prey, foragers show an initial transient wariness of novel food (neophobia). In addition, some individuals display a much longer period of avoidance before incorporating the novel food into their diet (termed dietary conservatism). There are two stable foraging strategies found within forager populations: (1) adventurous consumers (AC) which rapidly accept novel foods and (2) foragers showing dietary conservatism (DC). The expression of these two strategies may also vary with environmental conditions. We measured the effect of competition on the plasticity of foraging strategies when domestic chicks, Gallus gallus domesticus, foraged for familiar and novel coloured crumbs with or without competitor chicks. In addition we investigated the effect of prey detectability on the response of foragers to a competitor, by making the familiar food cryptic or conspicuous. AC birds responded to competition by accepting the novel prey more quickly than when foraging alone, regardless of how hard familiar food was to find. In contrast, DC birds failed to reduce their wariness in response to competition when the competitor’s food choice was obscured. The foraging strategies of the birds were thus found to be plastic in their expression, but this plasticity differed between inherently AC and DC individuals. The implications of these results are discussed in relation to the foraging strategies of wild and domestic birds. Crown Copyright Ó 2013 Published on behalf of The Association for the Study of Animal Behaviour by Elsevier Ltd. All rights reserved.
Foragers must decide whether to spend their time and energy searching for known profitable prey, or whether to risk dietary expansion to include novel prey of unknown palatability and profitability. Most animals (Barnett, 1958; Brigham & Sibley, 1999) show some degree of ‘dietary wariness’, causing them to hesitate before approaching and consuming unfamiliar food. Dietary wariness comprises two distinct behavioural processes, neophobia and dietary conservatism (DC). These operate over different timescales and differ in important respects. The initial transient fear of novelty, termed neophobia, was first described in rats encountering new objects (Barnett, 1958). Neophobia has since been observed in many animal groups (including birds, fish, mammals; reviews in Brigham & Sibley, 1999; Kelly & Marples, 2004; Mappes, Marples, & Endler, 2005; Marples, Kelly, & Thomas, 2005), in response not only to objects, but also to novel foods. This hesitation in approach is typically brief, lasting only a few minutes in most animals (Marples * Correspondence: N. M. Marples, School of Natural Sciences, Trinity College Dublin, Dublin 2, Ireland. E-mail address: [email protected] (N. M. Marples).
& Kelly, 1999), and is followed by investigation of the novel food or object (Coppinger, 1969). DC is a secondary, more lasting refusal to accept novel food into their diets by some (but not all) members of a population (Marples, Roper, & Harper, 1998; Thomas, Bartlett, Marples, Kelly, & Cuthill, 2004; Thomas, Marples, Cuthill, Takahashi, & Gibson, 2003). It is thus the time between first contact with the novel food (the end of neophobia) and the consumption of the novel food whenever it is encountered. The combined durations of neophobia plus DC can together be described as dietary wariness, incorporating the entire process of novel food acceptance. DC has been demonstrated both in birds (eight species; Marples et al., 2005) and in fish (five species; Thomas et al., 2010; Richards, Thomas, Marples, Snellgrove, & Cable, 2011; Richards et al. 2014). Marples and Kelly (1999) argued that neophobia and DC are two distinct processes, not only because of the large differences in duration of the avoidance, but also because neophobia is much easier to deactivate with experience of novel prey (Marples, Quinlan, Thomas, & Kelly, 2007). In addition, DC is a much more complex behaviour, comprising several steps (Marples & Kelly,
0003-3472/$38.00 Crown Copyright Ó 2013 Published on behalf of The Association for the Study of Animal Behaviour by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.anbehav.2013.12.025
64
K. McMahon et al. / Animal Behaviour 89 (2014) 63e69
1999), and is not correlated with neophobia (K. McMahon, unpublished data). All populations tested in either laboratory or field experiments have shown that only some individuals express DC (Marples et al., 1998), while other members of the same foraging population are quick to sample and incorporate new food items into their diets, and are known as adventurous consumers (AC; Marples et al., 2007; Thomas et al., 2010). This stable polymorphism of two foraging strategies within the population has been shown to be a genetically mediated trait in Japanese quail, Coturnix coturnix japonicus (see Marples & Brakefield, 1995). Because neophobia lasts only a few minutes in domestic chicks, Gallus gallus domesticus, the behavioural strategy of each individual can be identified using a relatively brief DC test (Marples & Kelly, 1999). While all individuals initially exhibit neophobia, DC individuals refuse to eat novel prey even after they have approached and handled the item (i.e. overcome their neophobia) or observed conspecifics eating it. Rather than a fear of approaching or touching the novel prey, DC is expressed as a persistent unwillingness of these individuals to broaden their diets to include the novel prey (Marples et al., 1998). It is thus possible to measure the two aspects of dietary wariness separately, with neophobia being the length of time until an individual first touches or handles a novel prey item, and DC being the remaining time before the novel prey is eaten consistently (i.e. eaten each time it is encountered; Kelly & Marples, 2004). The expansion of an individual forager’s diet therefore requires both its neophobia and its DC to be overcome (Marples et al., 2005, 2007). The fact that DC has been documented in two very divergent taxa (birds and fish), suggests that it may be a widespread foraging strategy (Marples et al., 2005; Marples & Kelly, 1999; Thomas et al., 2010). For any species with a limited food supply, DC individuals in the population may be at a foraging disadvantage if they continue to avoid novel foods, but reducing their conservatism would also expose them to greater risks. A conservative approach to dietary sampling in individuals within a generalist species would help to minimize negative or unprofitable experiences, such as consumption of toxic prey (Lee, Marples, & Speed, 2009). Therefore, despite the reduction in potential food options, conservative feeders that foraged only on familiar prey of known value would avoid the risk of injury, illness or death through dietary indiscretion. These DC individuals might maintain their fitness levels alongside AC individuals, as the AC foragers may fall victim to prey defences at a greater rate, negating some of the fitness advantages conferred by a larger menu (Thomas et al., 2010). In addition, it is possible that DC foragers become more skilled at finding and handling their few favoured prey types, while AC foragers, with their wider diet, are likely to be less skilled at detection and handling of their many types of prey (Sherry & McDade, 1982). Investigating such choices in foraging strategies is crucial to understand the ecology and evolution both of foraging species and of their prey. Despite the relevance of understanding the foraging strategies of individuals, limited attention has been given to the plasticity of expression of these strategies under differing foraging conditions. It might be expected that individual experience and ecological circumstances may modify the expression of the underlying genetic propensity of an individual to be AC or DC. Indeed, there is evidence that the expression of AC or DC in individuals can be modified by experience. For example, Marples et al. (2007) showed that the expression of DC can be reduced in domestic chicks through extensive experience with benign novel foods, although full reversion to DC will occur after only one experience with an unpalatable food. Further evidence (Barnett, 2007; Sherratt, 2002; Skelhorn & Rowe, 2006a, 2006b) suggests that increased hunger levels may temporarily decrease DC expression towards aposematic prey. To understand the degree of plasticity in DC expression, we aimed in this study to evaluate the effects of conspecific
competition as an external factor affecting the costs and benefits of different foraging strategies. Birds foraging in a flock are likely to compete with other flock members for food, through both interference and exploitation competition (Sih, 1993) which together reduce the amount of time that an individual can spend making foraging decisions. Optimal foraging theory would predict individuals experiencing high levels of competition would broaden their diets to include less valuable or possibly less familiar foods, so as to avoid starvation (Stephens & Krebs, 1986). Additionally, social learning may further encourage broadening of the diet to incorporate unfamiliar prey on which conspecifics are observed to feed (Fryday & Greig-Smith, 1994). The combination of competition and social learning may therefore favour a relaxation of DC expression, but it is also possible that birds may ignore conspecific foraging behaviours and remain inflexible in their level of DC (cf. Richards et al., 2011). We used generalist foragers, domestic chicks, to test the hypothesis that DC expression is reduced when a forager is in competition with a conspecific that readily eats novel coloured food. A further experiment then evaluated DC expression in the same system, but in relation to the relative conspicuousness of novel and familiar foods. We hypothesized that when familiar food is relatively cryptic, increased search and attention costs will encourage foragers to reduce their expression of DC, and sample the more conspicuous novel prey (Jones, Krebs, & Whittingham, 2006). This study is the first to explore the degree of plasticity of AC and DC foraging strategies by birds in a complex foraging environment. METHODS Outline of Experiments To classify individual chicks as DC or AC, before each experiment we first tested each chick for its baseline level of DC using a ‘DC test’. We carried out three experiments to investigate: (1) the effects of testing for DC on the subsequent behaviour of the chicks; (2) the effects of a competitor on the expression of DC when the novel and familiar foods were equally conspicuous; and (3) the effects of a competitor on the expression of DC when the novel prey was more conspicuousness than the familiar prey. Chick Husbandry We carried out the three experiments using three separate batches of male ‘Cobb 500’ strain chicks, which were obtained at 1 day old from a commercial hatchery (Annyalla Chicks Ltd., Castlebury, Co. Monaghan, Ireland). The experiments were approved by the ethics committee of Trinity College Dublin, and the chicks were held under licence from the Department of Health and Children, Ireland number B100/2756 held by N. Marples. Batch 1 (used for experiment 1) contained 29 chicks, batch 2 (experiment 2) contained 55 chicks and batch 3 (experiment 3) contained 63 chicks. For each batch, individuals were housed together in a wooden holding pen (200 cm long 60 cm wide) with wood shavings covering the floor. The chicks were maintained on a 12:12 h light:dark regime, under standard fluorescent lights in addition to natural lighting. Ambient temperature was 24 4 C, but chicks could find warmer areas by standing under two infrared heat lamps suspended above their holding pen. From their arrival at the laboratory, the chicks were fed chick starter crumbs (William Connolly & Sons, Red Mills, Goresbridge, Co. Kilkenny, Ireland), dyed green (O’Brien’s liquid green 90; 1 ml dye: 30 ml water), which resulted in their subsequent treatment of this colour as familiar. Food and water were available ad libitum for the duration of the study, except
K. McMahon et al. / Animal Behaviour 89 (2014) 63e69
for the hour prior to the trials when food (but not water) was removed, to encourage active foraging during the tests. Each chick was marked with a unique colour combination on the head to distinguish between individuals. As in our previous studies, the chicks did not respond to their own marks or those of their conspecifics (Marples et al., 2007). At the end of the experiment, as the chicks were male and could not be rehoused on a farm or reused by any other researchers, in accordance with Irish welfare legislation they were euthanized. This was carried out by a qualified technician by means of intraperitoneal injection of 0.5 ml Euthatal in accordance with the licence B100/2756. Experimental Foods The chick starter crumbs (originally light brown) used in the study were dyed unique colours to be visually distinct and easily noticeable to the chicks (Marples et al., 2007). Crumbs were mixed with the dye solution and stirred for 3 min until all the crumbs had taken up the dye; they were then left to air dry completely before being used. Green was used as the familiar colour throughout the study, and green crumbs were provided ad libitum throughout, except where specified in the experimental methods below. Blue, yellow, orange and red coloured crumbs served as novel prey items during the experiments (coloured using O’Brien’s liquid blue (FCF/ 0.5KGS), O’Brien’s lemon yellow (T/0.5K), O’Brien’s liquid orange (SY/0.5K), O’Brien’s Christmas red; in each case diluting 1 ml of dye in 30 ml of water). Pretraining When the chicks were 2 days old, familiarization with foraging in the test arena began. The arena was a wooden square enclosure measuring 84 cm 84 cm, with 21 cm high walls. The wooden floor was covered with 13 13 rows of evenly spaced circular depressions spaced 3 cm apart, into each of which a single chick crumb could be placed. Since solitary chicks tend to become agitated and refuse to forage, two ‘buddy chicks’ were present in the arena to prevent this. The buddy chicks were allowed to eat to satiation before the trial, then kept in a ‘buddy area’ to prevent them from interfering with the foraging of the test chick (Marples et al., 2007). This buddy area square (A1 in Fig. 1) was fenced off from the rest of the arena with chicken wire to allow for visual and auditory contact between the test and buddy chicks. The buddy chicks were drawn from a pool of additional chicks not used as subjects in the experiments. New buddy chicks were randomly selected for each trial. Each test chick was familiarized with foraging in the arena by being offered familiar green crumbs during five successive
A
B
C
D
1
2
84 cm
3
4 Figure 1. Diagram of the floor of the foraging arena showing dimensions. Square A1 was the ‘buddy area’ and was unavailable to foraging chicks.
65
‘pretraining’ sessions over one and a half days. In the first pretraining session, the test chicks were allowed to forage in the training arena in a group of five chicks. Each subsequent session contained progressively fewer foraging companions, so that the final session involved a lone chick foraging in the arena but with the two buddy chicks added to the buddy area. All of the experimental chicks were watched to confirm that they consumed the familiar food in all of the pretraining sessions. Measuring DC Expression (DC Test) After the pretraining sessions, each of the chicks in batches 2 and 3 were tested individually for their levels of neophobia (latency to peck at the novel food) and DC (latency to eat three pieces of novel food once neophobia had been overcome). Our previous studies have shown that the latency to eat three food items signifies the acceptance of that particular novel food item into the diet, and thus an overcoming of DC (Marples & Kelly, 1999). In these ‘DC tests’, chicks were given a choice between a 50:50 ratio of familiar (green) and novel (orange) food. Each chick was allowed to forage individually for a 10 min period. Latency to eat three orange crumbs gave their ‘DC score’ with higher scores indicating more DC birds, as they had taken longer to eat the novel food. Chicks that did not eat any of the novel food within the 10 min were given the highest possible score (600 s) and classified as DC. Chicks that ate the novel food within this period (i.e. a latency of less than 600 s) were classified as AC individuals. In both batches of chicks the four most AC individuals were chosen as the competitors for the experiment and were housed together (but separately from all other chicks). All groups of chicks were held in identical pens in the same room under the same conditions. Pretraining of Competitor Chicks All competitor chicks were trained to eat both familiar and novel coloured crumbs, both as their maintenance diet in their housing pen and in the foraging arena prior to subsequent trials. Once all of the competitor chicks were readily eating both colours of crumbs, the experimental trials began. Experiment 1: Effects of Testing for DC on Behaviour Marples et al. (2007) reported that dietary wariness could be deactivated through experience of novel foods. Thus, by taking part in the initial DC test described above (which was necessary to establish a DC score, and therefore identify each test individual’s underlying foraging strategy), it was possible that chicks might have reduced their dietary wariness as a result of the exposure to novel food. If so, this would have affected their responses to novelty in subsequent experiments. However, Marples et al. (2007) showed that DC was only reduced by long-term continuous exposure to the novel food in a social setting. Short-term exposures such as those experienced in the DC tests should not have been enough to reduce DC. Nevertheless, to ensure that the DC tests were not confounding our results, we tested whether this brief exposure to novelty was sufficient to reduce DC. We used the first batch of chicks for this purpose. One group (16 chicks) received a DC test in which they were presented with three pieces of familiar green food and three pieces of novel red food. They were exposed to the novel food for three trials, each of 3 min, with at least 2 h between each trial. The other 13 chicks were used as controls, receiving the same trials, but only familiar green food. This protocol differed from the one used in experiments 2 and 3 because it represented a more robust test of the effect of exposure to novelty. Chicks tend to lose interest in tasks longer than 3 min, so while keeping the length of exposure
K. McMahon et al. / Animal Behaviour 89 (2014) 63e69
similar to experiments 2 and 3 (below) we divided the DC test into three trials. The day after the first trial, all 29 chicks were tested for DC in exactly the same manner, but this time all chicks were given a choice between familiar green food and novel blue food. The total latency to eat three pieces of novel food across the three trials was recorded for each test, allowing us to detect any effect of the first DC test on the chicks’ latency to eat a new novel colour of food in the second test. Experiment 2: Conspicuous Foods Here, we assessed the effect of the presence of a competitor on an individual’s expression of dietary wariness. The second batch of chicks was first tested for DC (as described above for the first experiment) before progressing into trials with and without a competing forager. The foraging arena was prepared with two familiar green crumbs and two novel blue crumbs within each of the 15 test squares of the arena. Single crumbs were placed into randomly chosen depressions in the test square (Fig. 1). This gave a total of 60 available crumbs (30 familiar coloured and 30 novel coloured) per trial. Each trial lasted 8 min, as chicks tended to abandon foraging and focus on the buddy chicks or go to sleep after that time. We recorded the latency to peck a novel crumb, the number and time of subsequent pecks and the latency to eat a novel crumb. If no novel crumbs were eaten, the chick was noted as having taken 480 s, the full duration of the trial. Ten AC chicks and 10 DC chicks served as control groups and were each tested foraging alone in the arena (i.e. without competition from another chick, but with buddies present). The two experimental groups, 10 AC and 10 DC chicks, were observed foraging in competition with one chick from the pool of four competitor chicks. The competitor chicks were used in rotation, so as to maintain sufficient hunger levels to ensure that they foraged during the trials. In each competition trial, both the test chick and the competitor chick were placed in the arena at the same time, at opposite ends of the front row (A4 and D4) and allowed to forage together for 8 min. As with the control group, latencies to eat familiar green crumbs, to peck novel blue crumbs and to eat novel blue crumbs were recorded. These trials took place over 2 days. Experiment 3: Cryptic Familiar versus Conspicuous Novel Food The third experiment was carried out using the third batch of chicks, which were divided into two treatment groups exactly as in experiment 2. The familiar green crumbs were made cryptic by covering the foraging arena floor with a thin layer of wood shavings which had been dyed green with the same food colour as the familiar food. These ‘distractors’ served to obscure the green crumbs and so impair foraging. Prior to each trial the arena was set up with two familiar green crumbs and two novel blue crumbs in each square sector, as in the previous experiment. Trials lasted 8 min and latencies to eat green crumbs, to peck blue crumbs and to eat blue crumbs were recorded, with a maximum possible latency of 480 s. The same variables were recorded during these trials as in experiment 2. Data Analysis The time taken for each chick to overcome neophobia (latency to peck at a novel food item) and dietary wariness (latency to eat three novel food items, including both neophobia and DC components) was measured for AC and DC chicks in each treatment group across experiments 2 and 3 (no competition þ no distractors, competition þ no distractors, no competition þ distractors, competition þ distractors). Analyses of data were carried out using the statistical software R (version 2.13.2, R Development Core Team, 2009). Generalized
linear models (GLMs) were used to examine the role of different fixed factors in influencing the different measures of dietary wariness (neophobia and DC) recorded in the experiment. The factors considered were: (1) whether the focal chick had been classified as an AC or DC individual, (2) competition (i.e. presence of a competitor chick) or no competition, (3) distractors (i.e. green coloured wood shavings rendering the familiar coloured prey less conspicuous) or no distractors. Model selection and validation procedures followed Crawley (2007) and Zuur, Ieno, Walker, Saveliev, and Smith (2009). Neophobia latency data were analysed using Gaussian GLMs (with identity or log-link functions), whereas dietary wariness latency data were analysed using Poisson GLMs (with log-link functions). The data from experiment 1 were analysed using gamma GLMs (with inverse-link functions). The results of experiments 2 and 3 were compared in the analyses below, using a gamma GLM (with inverse-link functions) to investigate whether differences in latencies to eat three pieces of novel food existed between chicks from batches 2 and 3. All analyses used two-tailed hypothesis testing. RESULTS Experiment 1: Effects of Testing for DC on Behaviour The results of the first experiment revealed that there was no difference in DC expressed in the second trial of this experiment between the control group, which had not previously received a DC test, and the treatment group which had received a DC test in the first trial of the experiment (gamma GLM, inverse link: t1,28 ¼ 0.537, P ¼ 0.596; Fig. 2). Thus the exposure to novelty in the context of a DC test was not enough to reduce subsequent DC and therefore did not confound the results of the following experiments. DC Tests for Experiments 2 and 3 The initial DC test scores for chicks in batches 2 and 3 based on the latency to eat three novel coloured crumbs when foraging alone allowed us to define 42 chicks as AC, as they accepted the novel food within 10 min. The other 66 chicks were classified as DC as they refused to accept the novel food within the 10 min trial (Fig. 3). The difference in DC scores between chicks in the two batches was marginally nonsignificant (gamma GLM, inverse link: F1,116 ¼ 1.307, P ¼ 0.054). Factors Influencing Neophobia and Dietary Wariness As there were no differences in methodology between experiments 2 and 3, they were analysed together to explore simultaneously the roles of competition pressure and prey crypticity. The
300 DC Score (s)
66
200 100 0 DC Test
No DC Test Treatment group
Figure 2. Box plot of the dietary wariness of chicks for both treatment groups of experiment 1: chicks that received a DC test and those that did not. The box plots show the median and 25th and 75th percentiles; the whiskers indicate the values within 1.5 times the interquartile range and the circle is an outlier.
K. McMahon et al. / Animal Behaviour 89 (2014) 63e69
identity link, square root-transformed dependent variable: F1,85 ¼ 4.1274, P ¼ 0.045). Similarly, in the model explaining the time taken to eat three novel prey (dietary wariness), the three-way interaction term was also significant (Poisson GLM, log-link function, likelihood ratio test: LRT1,85 ¼ 1826.8, P < 0.001). In other words, the manner in which prey conspicuousness, and the competitive environment, influenced both neophobia and dietary wariness differed significantly between AC and DC focal birds. We therefore analysed the AC birds separately from the DC birds in our further analyses.
AC Chicks | DC Chicks
600 Latency (s)
500 400 300 200 100 0 0
20
40
60
80
100
120
Chick ID
Foraging Behaviour of AC Birds
Figure 3. Scatter plot depicting the initial DC test scores for all chicks, used in their assignment as AC or DC birds. DC test scores represent the latencies (s) for each chick to eat three novel coloured crumbs when foraging alone.
For the AC birds (Fig. 4a), the model explaining neophobia did not show a significant two-way interaction between the presence of a competitor and the presence of distractors (Gaussian GLM, log link, square root-transformed dependent variable: F1,34 ¼ 2.156, P ¼ 0.158). In other words, there was no significant difference in the effect of a competitor between experiment 2, when distractors were absent, and experiment 3, when distractors were present. When the two-way interaction was excluded from the model, there was a significant effect of competition (Gaussian GLM, log link, square root-transformed dependent variable: F1,35 ¼ 7.4322, P ¼ 0.009), with the presence of a competitor chick being associated with a reduction in neophobia, whereas there was no significant effect of the presence/absence of distractors (Gaussian GLM, log link, square root-transformed dependent variable: F1,35 ¼ 0.7325, P ¼ 0.398). In other words, the AC chicks reduced their neophobia in response to a competitor, but distractors did not affect this response. When the dietary wariness in the AC birds was analysed, a similar pattern as for neophobia was evident. However, in this case
results revealed distinct differences between AC and DC foragers in terms of the plasticity of expression of both the time to peck at the novel prey (i.e. their neophobia) and the time to eat three novel prey (i.e. their dietary wariness, incorporating both neophobia and DC). These two responses were measured in the presence or absence of competition pressure and under two levels of prey crypticity (with and without distractors resembling the familiar food). We first tested for the three-way interaction between (1) whether the focal chick had initially exhibited an AC or a DC foraging strategy, (2) the presence of a competitor chick and (3) prey conspicuousness (presence of distractors). In the model explaining the time taken to peck at the novel prey (neophobia), the three-way interaction term was significant (Gaussian GLM,
Neophobia 600
67
|
Dietary wariness
(a)
500 400 300 200
Latency (s)
100 0 NC−ND C−ND NC−D 600
C−D
NC−ND C−ND NC−D
C−D
C−D NC−ND C−ND NC−D Treatment groups
C−D
(b)
500 400 300 200 100 0 NC−ND C−ND NC−D
Figure 4. Box plot of the median latency for (a) AC and (b) DC designated chicks to end neophobia and dietary wariness, under the conditions of the four treatments, as indicated below each box plot. NC ¼ no competitor present; C ¼ competitor present; ND ¼ no distractors present; D ¼ distractors present, making the familiar food less conspicuous than the novel food. The box plots show the median and 25th and 75th percentiles; the whiskers indicate the values within 1.5 times the interquartile range and the circles are outliers.
68
K. McMahon et al. / Animal Behaviour 89 (2014) 63e69
the two-way interaction between competition and prey crypticity was highly significant (Poisson GLM, log link: LRT1,34 ¼ 1453, P < 0.0001). In other words, when competition was present there was a reduction in wariness, and this reduction was more pronounced in the experiment with distractors present compared to the experiment with no distractors (Fig. 4a). Foraging Behaviour of DC Birds In the DC birds (Fig. 4b), the model explaining neophobia included a nonsignificant interaction between the presence of a competitor and the presence of distractors (Gaussian GLM, log link, square root-transformed dependent variable: F1,52 ¼ 1.0682, P ¼ 0.306). When the two-way interaction was excluded from the model, there was a significant independent effect of competition on neophobia (Gaussian GLM, log link, square root-transformed dependent variable: F1,52 ¼ 4.391, P ¼ 0.041). There was no significant independent effect of distractors (F1,52 ¼ 1.068, P ¼ 0.306). In other words, competition reduced neophobia, but prey conspicuousness did not. In contrast, the model explaining dietary wariness included a highly significant interaction between the presence of a competitor and the presence of distractors (Poisson GLM, log link: LRT1,51 ¼ 391.11, P < 0.0001). In other words, wariness was reduced in response to competition when there were no distractors present, but no such effect was found when familiar prey was more cryptic than novel prey (Fig. 4b). In summary, Fig. 4b shows that competition reduced the expression of both neophobia and dietary wariness in DC birds only in experiment 2, that is, when the chicks could clearly see what their companion was eating. In experiment 3 where the familiar crumbs were cryptic, the presence of a competitor had no significant effect on the latency to peck at, and eat, the novel crumbs. This contrasts with the AC chicks, which significantly reduced their latency in these circumstances. DISCUSSION The findings of this study highlight the contrast between AC and DC foraging strategies. While it has previously been shown that individuals have a genetic propensity towards being either AC or DC foragers (Marples & Brakefield, 1995), this study demonstrates for the first time that both competition and prey conspicuousness may modulate the expression of these foraging strategies. Furthermore, this modulation differed depending on whether the individual was inherently an AC or a DC forager. This demonstrates that the underlying foraging strategies of these birds are plastic to some degree, as birds that were strongly DC (in their DC test) acted as far less conservative in some experimental circumstances. When faced with a competitor that readily ate both types of food, the AC chicks increased their acceptance of novel food, but did so significantly only when distractors made the familiar food cryptic. In other words, when the AC birds had a rival competing for the food, and had to work harder to find the familiar food, they gave up any preference for it and ate the novel food (which was more conspicuous). In contrast, the DC birds accepted the novel food more quickly in the presence of a rival forager only when there were no distractors to disguise the prey. They maintained their avoidance of the novel prey when a rival was present, even though distractors made the familiar prey more cryptic. Thus, under competition, DC birds remained highly conservative in their foraging under circumstances when we would expect DC foraging to be more costly, because familiar prey was more difficult to find. This finding contrasts with previous work which has shown a reduction in dietary wariness in circumstances where a
conservative foraging strategy would be expected to be costly (Barnett, Bateson, & Rowe, 2007; Lee et al., 2009; Mappes et al., 2005; Richards et al., 2011). However, in our study, DC foragers may have benefitted from increased efficiency in foraging if they were able to form a search image for the cryptic familiar food (Tinbergen, 1960). AC foragers, in contrast, expanded their diet to include novel prey instead of specializing on the familiar food. The differences between AC and DC birds in their responses to the rival forager suggest that they may have perceived the rival in different ways. The AC birds responded to a competitor as predicted, by increasing their intake of the novel food when it was more conspicuous than the alternative food source. Competition pressure reduces time available for viewing prey (Gamberale-Stille, 2001; Guilford, 1986) decreasing attack latencies of competing chicks feeding on both palatable and aposematic prey (GamberaleStille, 2001). This is consistent with the observed decreases in both neophobia and DC expression in AC birds demonstrated in our study. In contrast, the DC birds reduced their wariness in the presence of a competitor only when they could see the food clearly (treatments without distractors), and remained fully averse to novel food when there were distractors present. This result could be interpreted as the DC birds considering the companion forager as a demonstrator of food choice in a social context, in that they appeared to copy the choice of novel food more readily when both foods were clearly visible. Such social learning based on observation can increase feeding rate (Fryday & Greig-Smith, 1994; Gentle, 1985; Wauters, Richard-Yris, & Talec, 2002) as chicks can learn to identify profitable foods by watching and copying others (Wauters et al., 2002). It is possible that seeing the competitor chick eat the novel food without harm encouraged the DC foraging chick to sample and accept the novel food as palatable and profitable, a behavioural modification previously seen in house sparrows, Passer domesticus (see Fryday & Greig-Smith, 1994). The differences between AC and DC foragers in their response to a competitor supports the hypothesis that these are two distinct processes (Marples & Kelly, 1999) even though the precise mechanism through which they act is not yet known. In addition, the existence of AC and DC foraging strategies as a dichotomy, rather than as a continuum of latencies to attack novel prey, is a matter of some debate which has yet to be resolved. The genetic basis and heritability of the trait (Marples & Brakefield, 1995) support the view that these strategies are distinct. Individuals show consistent behaviour towards novel foods across a range of foraging contexts. Differences in behaviours other than latency to accept novel foods that correlate with the two foraging strategies, such as those found between AC and DC individuals in the present experiment, add support to the view that they are distinct strategies. In this study, the expression of AC and DC foraging strategies was modified by both competition and food conspicuousness in different ways. This study is the first to demonstrate plasticity in the expression of DC in response to competition. These insights into the plasticity of DC in both AC and DC individuals have several important applications. For instance, the domestic poultry industry may benefit by allowing birds to observe each other foraging whenever food types need to be changed, reducing the drop in growth rate that is typical of dietary changes (Gentle, 1985; Jones, 1986; Marples et al., 2007). Furthermore, reintroduction programmes for endangered socially foraging species may be more successful if natural levels of DC expression are enabled to be expressed by offering social groups of individuals a variety of foods prior to release. Ecosystem changes may result in an alteration in prey available to predators (Møller, Fiedler, & Berthold, 2010; Schaefer, Jetz, & Böhning-Gaese, 2008; Walther et al., 2002). As a result of any of these changes in prey availability, increased competition between foragers could also
K. McMahon et al. / Animal Behaviour 89 (2014) 63e69
change the relative costs and benefits of the two strategies, and hence change the balance between AC and DC foragers in the population. Understanding how AC and DC foragers respond to such changes will help to predict shifts in ecosystem function, and to assist in prioritizing conservation actions. Our results demonstrate that birds using the AC and DC foraging strategies differ markedly in their response to the combination of competition and prey crypticity. Determining both the drivers and effects of DC expression in foragers will help to uncover evolutionary relationships between predators and prey, and to discover which ecological conditions maintain the observed polymorphism in foraging strategies. Acknowledgments We thank Peter Stafford and Alison Boyce for help with bird husbandry throughout this project. We are also grateful to the journal editors and anonymous referees for improvements to the manuscript. This work was part funded by the IRCSET Embark Initiative. References Barnett, C. A., Bateson, M., & Rowe, C. (2007). State-dependent decision making: educated predators strategically trade off the costs and benefits of consuming aposematic prey. Behavioral Ecology, 18, 645e651. Barnett, S. A. (1958). Experiments on ‘neophobia’ in wild and laboratory rats. British Journal of Psychology, 49, 195e201. Brigham, A. J., & Sibley, R. M. (1999). A review of the phenomenon of neophobia. In P. D. Cowan, & C. J. Feare (Eds.), Advances in vertebrate pest management (pp. 67e 84). Furth, Germany: Filander Verlag. Coppinger, R. P. (1969). The effect of experience and novelty on avian feeding behavior with reference to the evolution of warning coloration in butterflies. Part I: reactions of wild-caught adult blue jays to novel insects. Behaviour, 35, 45e60. Crawley, M. J. (2007). The R book. Chichester, U.K.: J. Wiley. Fryday, S. L., & Greig-Smith, P. W. (1994). The effects of social learning on the food choice of the house sparrow (Passer domesticus). Behaviour, 128, 281e300. Gamberale-Stille, G. (2001). Benefit by contrast: an experiment with live aposematic prey. Behavioral Ecology, 12, 768e772. Gentle, M. J. (1985). Sensory involvement in the control of food intake in poultry. Proceedings of the Nutrition Society, 44, 313e321. Guilford, T. (1986). How do ‘warning colours’ work? Conspicuousness may reduce recognition errors in experienced predators. Animal Behaviour, 34, 286e288. Jones, K. A., Krebs, J. R., & Whittingham, M. J. (2006). Interaction between seed crypsis and habitat structure influence patch choice in a granivorous bird, the chaffinch Fringilla coelebs. Journal of Avian Biology, 37, 413e418. Jones, R. B. (1986). Responses of domestic chicks to novel food as a function of sex, strain and previous experience. Behavioural Processes, 12, 261e271. Kelly, D. J., & Marples, N. M. (2004). The effects of novel odour and colour cues on food acceptance by the zebra finch. Animal Behaviour, 68, 1049e1054. Lee, T. J., Marples, N. M., & Speed, M. P. (2009). Can dietary conservatism explain the primary evolution of aposematism? Animal Behaviour, 79, 63e74.
69
Mappes, J., Marples, N. M., & Endler, J. A. (2005). The complex business of survival by aposematism. Trends in Ecology & Evolution, 20, 598e603. Marples, N. M., & Brakefield, P. M. (1995). Genetic variation for the rate of recruitment of novel insect prey into the diet of a bird. Biological Journal of the Linnean Society, 55, 17e27. Marples, N. M., & Kelly, D. J. (1999). Neophobia and dietary conservatism: two distinct processes? Evolutionary Ecology, 13, 641e653. Marples, N. M., Kelly, D. J., & Thomas, R. J. (2005). Perspective: the evolution of warning colouration is not paradoxical. Evolution, 59, 933e940. Marples, N. M., Roper, T. J., & Harper, D. G. C. (1998). Responses of wild birds to novel prey: evidence of dietary conservatism. Oikos, 83, 161e165. Marples, N. M., Quinlan, M., Thomas, R. J., & Kelly, D. J. (2007). Deactivation of dietary wariness through experience of novel food. Behavioral Ecology, 18, 803e810. Møller, A. P., Fiedler, W., & Berthold, P. (2010). Effects of climate change on birds. Oxford, U.K.: Oxford University Press. R Development Core Team. (2009). R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. Richards, E. L., Thomas, R. J., Marples, N. M., Snellgrove, D. L., & Cable, J. (2011). The expression of dietary conservatism in solitary and shoaling 3-spined sticklebacks Gasterosteus aculeatus. Behavioral Ecology, 22, 738e744. Richards, E. L., Alexander, L. G., Snellgrove, D., Thomas, R. J., Marples, N. M., & Cable, J. (2014). Variation in the expression of dietary conservatism within and between fish species. Animal Behaviour, 88, 49e56. Schaefer, T. C., Jetz, W., & Böhning-Gaese, K. (2008). Impact of climate change on migratory birds: community reassembly vs. adaptation. Global Ecology and Biogeography, 17, 38e49. Sherratt, T. N. (2002). The coevolution of warning signals. Proceedings of the Royal Society B, 269, 741e746. Sherry, T. W., & McDade, L. A. (1982). Prey selection and handling in two neotropical hover-gleaning birds. Ecology, 63, 1016e1028. Sih, A. (1993). Effects of ecological interactions on forager diets: competition, predation risk, parasitism and prey behaviour. In R. N. Hughes (Ed.), Diet selection: An interdisciplinary approach to foraging behaviour (pp. 182e208). Oxford, U.K.: Blackwell Scientific Publications. Skelhorn, J., & Rowe, C. (2006a). Taste-rejection by predators and the evolution of unpalatability in prey. Behavioral Ecology and Sociobiology, 60, 550e555. Skelhorn, J., & Rowe, C. (2006b). Avian predators taste-reject aposematic prey on the basis of their chemical defence. Biology Letters, 2, 348e350. Stephens, D. W., & Krebs, J. R. (1986). Foraging theory. Chichester, U.K.: Princeton University Press. Thomas, R. J., Bartlett, L., Marples, N. M., Kelly, D. J., & Cuthill, I. C. (2004). Prey selection by wild birds can allow novel and conspicuous colour morphs to spread in prey populations. Oikos, 106, 285e294. Thomas, R. J., King, T. A., Forshaw, H. E., Marples, N. M., Speed, M. P., & Cable, J. (2010). The response of fish to novel prey: evidence that dietary conservatism is not restricted to birds. Behavioral Ecology, 21, 669e675. Thomas, R. J., Marples, N. M., Cuthill, I. C., Takahashi, M., & Gibson, E. (2003). Dietary conservatism may facilitate the initial evolution of aposematism. Oikos, 101, 458e466. Tinbergen, L. (1960). The natural control of insects in pine woods: I. Factors influencing the intensity of predation by songbirds. Archives Néerlandaises de Zoologie, 13, 265e343. Walther, G. R., Post, E., Convey, P., Menzel, A., Parmesan, C., Beebee, T. J. C., et al. (2002). Ecological responses to recent climate change. Nature, 416, 389e395. Wauters, A. M., Richard-Yris, M. A., & Talec, N. (2002). Maternal influences on feeding and general activity in domestic chicks. Ethology, 108, 529e540. Zuur, A. F., Ieno, E. N., Walker, N., Saveliev, A. A., & Smith, G. M. (2009). Mixed effects models & extensions in ecology with R. Secaucus, NJ: Springer.