Social buffering in a bird

Animal Behaviour 105 (2015) 11e19

Contents lists available at ScienceDirect

Animal Behaviour journal homepage: www.elsevier.com/locate/anbehav

Social buffering in a bird Joanne Edgar a, *, Suzanne Held a, Elizabeth Paul a, Isabelle Pettersson a, Robbie I'Anson Price b, Christine Nicol a a b

Department of Clinical Veterinary Science, University of Bristol, Langford, U.K. Department of Ecology and Evolution, Universit e de Lausanne, Lausanne, Switzerland

a r t i c l e i n f o Article history: Received 4 December 2014 Initial acceptance 15 January 2015 Final acceptance 25 March 2015 Available online 16 May 2015 MS. number: 14-00999R Keywords: bird chicken empathy maternal social buffering socially mediated arousal

The presence of a conspecific can ameliorate an individual's stress response. This social buffering is known to be widespread in social mammals but the capacity of birds to act as social buffers has not yet been determined. We previously demonstrated that domestic hens, Gallus gallus domesticus, show socially mediated arousal when watching their chicks receiving an aversive air puff. Furthermore, the hens' expectation of the situation strongly influenced the chicks' behaviour. Here we examined whether hens act as a social buffer; reducing their chicks' stress response to an aversive stimulus. Pairs of chicks were exposed to an air puff treatment and a control, each with and without their mothers. Chicks showed a suite of responses to the air puff (including increased standing, reduced eye temperature, preening and ground pecking). Maternal absence exacerbated the chicks' preening and ground-pecking responses to this stressor. Individual hens varied in their effectiveness as a social buffer and this was associated with their socially mediated arousal when (matched pairs of) their chicks received an air puff. Specifically, the hens' heart rate increase was strongly negatively correlated with the degree to which chick preening and ground pecking increased with maternal presence. This is the first demonstration that avian mothers are able to reduce their chicks' stress responses to an aversive stimulus. © 2015 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved.

The escalation or de-escalation of arousal and distress within social groups of animals will depend upon a number of potentially opposing social factors, including individual animals' capacities for emotional empathy and the extent to which social buffering occurs. Socially mediated arousal, the behavioural and/or physiological reactions of animals to the responses of conspecifics (demonstrators) showing signs of stress, is one of the component features of emotional empathy (for a review see Edgar, Nicol, Clark, & Paul, 2012). Social buffering is the process by which the presence of a conspecific can ameliorate an individual's response to a stressor. As well as being of fundamental interest, both social phenomena are of high relevance to the welfare of managed animals during exposure to conspecifics' responses to routine procedures, handling and slaughter. While there is opportunity for individuals to buffer a demonstrator's stress response, this may depend on whether, and to what extent, the individual is affected by the demonstrator's stress-related behaviour in the first place. Despite this potential for interaction between the two social phenomena, previous research

* Correspondence: J. Edgar, Department of Clinical Veterinary Science, University of Bristol, Langford House, Langford, Avon BS40 5DU, U.K. E-mail address: [email protected] (J. Edgar).

on empathy and socially mediated arousal has been conducted entirely in isolation from work on social buffering. Consideration of both social phenomena will shed important light on the spread of stress across groups of animals and is the focus of this paper. It is important to note that the reduction in stress responses in the presence of a conspecific (social buffering) is distinct from the return to baseline that is seen when social animals are brought together following isolation, or when bonded animals (mating pair or mother/young) are reunited following separation. To date, the literature on animal social buffering can generally be split into studies that focus on stress alleviation through conspecific presence during (e.g. Terranova, Cirulli, & Laviola, 1999; Varlinskaya, 2013) the stressor and those that focus on stress alleviation through social housing prior to and/or following (e.g. Hodges, Green, Simone, & McCormick, 2014; McCormick, Merrick, Secen, & Helmreich, 2007) the stressor (Kiyokawa, Takeuchi, & Mori, 2007). The vast majority of social-buffering studies have focused on mammals and many parameters have been used to measure social buffering. These include traditional stress indicators such as activation of the hypothalamic-pituitary-adrenal axis (e.g. increased heart rate, cortisol; Hodges et al., 2014) and behavioural responses (e.g. escape attempts, Gonzalez et al., 2013), through to neural responses (e.g. Fos gene expression; Kiyokawa et al., 2007;

http://dx.doi.org/10.1016/j.anbehav.2015.04.007 0003-3472/© 2015 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved.

12

J. Edgar et al. / Animal Behaviour 105 (2015) 11e19

Kiyokawa, Hiroshima, Takeuchi, Mod, 2014; Kiyokawa, Honda, Takeuchi, & Mori, 2014), motivation to seek social contact (Geverink et al., 1998; Ishiwata, Kilgour, Uetake, Eguchi, & Tanaka, 2007; Marin, Freytes, Guzman, & Jones, 2001) and recently even changes in ethanol intake (Hostetler & Ryabinin, 2014). The types of stressors used in social-buffering studies range from anxiety-inducing paradigms such as the elevated plus maze (Grippo et al., 2012), simulated predation (Mommer & Bell, 2013), social defeat (McQuaid, Audet, Jacobson-Pick, & Anisman, 2013), resident intruder tests (Grippo et al., 2012) and conditioned stimuli associated with electric shocks (Kiyokawa et al., 2007; Kiyokawa, Hiroshima, et al., 2014; Kiyokawa, Hiroshima, et al., 2014). Overall, the studies cited above point towards social buffering being a widespread phenomenon in social mammalian species. However, many studies purporting to measure social-buffering effects have used social isolation itself as the stressor (e.g. Hodges et al., 2014; Hostetler & Ryabinin, 2014; Kanitz, Hameister, Tuchscherer, Tuchscherer, & Puppe, 2014; Tuchscherer, Kanitz, Puppe, Hameister, & Tuchscherer, 2014). In this latter case changes in behaviour or physiology are likely to reflect a general return to baseline following separation, rather than a social-buffering effect. Surprisingly little is known about whether birds can act as social buffers for conspecifics. Studies with starlings, Sturnus vulgaris, and zebra finches, Taeniopygia guttata, have revealed that social isolation itself is stressful, but found no evidence that a familiar flock member (starlings and zebra finches) or a bonded mate (zebra finches) had any buffering effect on the stress response. Apfelbeck and Raess (2008) showed that, although visual separation of a starling from its group members had strong effects on the bird's behaviour and physiology, the presence of familiar group members did not attenuate the strong neophobic response exhibited during exposure to a novel object. Similarly, when zebra finches were exposed to 10 min of isolation in a novel environment, they showed an elevated corticosterone response, indicating a stress response to isolation. However, when the authors compared the response of isolated pair-bonded individuals with isolated nonpair-bonded birds, surprisingly they found no difference; the presence of their female partners did not attenuate the stress response of bonded male birds, providing no evidence for any social-buffering effect from this type of stressor from the bonded mate (Banerjee & Adkins-Regan, 2011). However, it would be premature to conclude that birds do not have the capacity to act as social buffers for one another. Domestic chickens, Gallus gallus domesticus, are an ideal model avian species to study social buffering; our own studies have shed light on their behavioural and physiological responses to stress, in particular within the mother/offspring bond (Edgar, Lowe, Paul, & Nicol, 2011; Edgar, Paul, & Nicol, 2013). Although the term ‘social buffering’ has not been directly used in previous chicken studies, there is some literature touching upon the concept of stress reduction through conspecific presence. Brooded domestic chicks show signs of stress, including a high rate of distress calling, when they are separated from their mother (Bermant, 1963; Collias, 1952; Hughes, Hughes, & Covalt-Dunning, 1982). Chicks also show stress responses when separated from same-age conspecifics and are highly motivated to move towards a conspecific (Jones & Williams, 1992; Suarez & Gallup, 1983; Vallortigara, Cailotto, & Zanforlin, 1990). Importantly, studies with broiler chickens have suggested that stress is influential in this ‘social reinstatement’ response; for example, mechanical restraint increases motivation for chicks to restore contact with conspecifics, especially familiar ones (Guzman & Marin, 2008; Marin et al., 2001). However, a lack of evidence of stress alleviation (provided through the measurement of behavioural and physiological indicators) means that it is unknown whether the presence of conspecifics buffers the stress response.

Additionally, the relevance of the conspecific's role, in terms of both its behaviour and its relationship to the subject, has not yet been determined, but would provide important insight into the extent to which social buffering occurs in birds. Studies on social buffering in guinea pigs, Cavia porcellus, have been pivotal in demonstrating the importance of the relationship between individuals on mammalian social-buffering effects. These studies have found that the mother is a more effective social buffer for her pups than other adult females (Graves & Hennessy, 2000; Hennessy, O'Leary, Hawke, & Wilson, 2002; Hennessy & Ritchey, 1987; Ritchey & Hennessy, 1987; Sachser, Durschlag, & Hirzel, 1998) and that a mate-bonded individual is more effective than other familiar conspecifics for both adult males and females (Graves & Hennessy, 2000; Sachser et al., 1998). However, the mechanisms that underpin the effectiveness of mothers or mates as social buffers have not been explored in mammals or in birds. We previously demonstrated that while watching their chicks subjected to an air puff, domestic mother hens show increased alertness, maternal vocalizations and heart rate, and reduced preening and eye temperature (which we together term ‘socially mediated arousal’; Edgar et al., 2011), phenomena that are not apparent when adult hens watch other familiar adults receive an air puff (Edgar, Paul, Harris, Penturn, & Nicol, 2012). The reduction in eye temperature was likely to be indicative of stress-induced hyperthermia, an increase in core body temperature associated with a decrease in peripheral temperature (Bouwknecht, Olivier, & Paylor, 2007; Busnado et al., 2010; Edgar, Nicol, Pugh, & Paul, 2013). Hens' arousal was influenced, not solely by chick distress cues, but by their own knowledge; mother hens responded behaviourally to expected, as well as actual, threat to their chicks (Edgar, Paul, et al., 2012). In contrast, chicks were strongly influenced by the hens' behaviour; chicks spent more time distress calling and less time preening when their mothers expected a potential threat (Edgar, Paul, et al., 2013). The fact that the hens' expectation of the situation affected the chicks' behaviour suggests that mother hens may influence the chicks' stress responses, raising their stress responses when threat is imminent but also perhaps reducing them when threat has passed and/or protection and care are available. We therefore aimed to determine whether hens act as social buffers for their chicks. We also aimed to assess the extent of natural variation in mother hens' socially mediated arousal and to examine how this relates to the hens' capacity to act as a social buffer during the same stressor. To do this, in Phase 1 we determined each mother hen's behavioural and physiological responses to their chicks receiving an air puff. Additionally, this phase provided us with information on the chicks' behavioural and physiological response in the presence of their mother, never before recorded. In Phase 2, with naïve chicks, we determined whether the presence or absence of the mother hen influenced the chicks' behavioural and physiological responses during the air puff administration. Using data from Phases 1 and 2, we finally explored whether the mothers' arousal (Phase 1) was correlated with their capacity to act as a social buffer (Phase 2). METHODS Ethical Note This project was carried out following ethical approval by the University of Bristol (University Investigation Number: UB/07/002) and in accordance with the ASAB Guidelines for the Treatment of Animals in Behavioural Research and Teaching. At the end of the study all animals were rehomed to responsible smallholders.

J. Edgar et al. / Animal Behaviour 105 (2015) 11e19

Animals and Housing Twelve broody hens (Indian game, N ¼ 5, or Indian game crossed with Australorp, N ¼ 7), aged 50e100 weeks, were obtained from a breeder and housed individually in a floor pen (1.5  1 m). The pen was bedded with wood shavings (5 cm) and contained a feeder with layers mash, a drinker and a cardboard nestbox. The hens were allowed to sit on 12 infertile eggs within the nestbox for 24 h before the eggs were swapped with 12 fertile eggs. The hens were then allowed to incubate these until hatching. Throughout this period, once per day, hens were gently lifted and moved out of the nestbox, to encourage them to feed and drink. The temperature in the room was 23  C and the lighting schedule was 16:8 h light:dark. Brood size ranged from 10 to 11 chicks. The fertile eggs contained slow-growing Hubbard broiler chicks, meaning that the hens and chicks were unrelated. For all of the hens, this was their first experience of hatching chicks. Habituation and Behavioural Observations Week 1: habituation to test box and harness Following hatching on day 1, from days 2 to 5, each hen and her brood were gradually habituated to handling and the test apparatus. The habituation schedule had been developed for previous studies (Edgar, 2012) and ensured that hens and chicks showed no measurable responses to any aspect of the test procedure, including human presence and heart rate monitoring. Habituation involved picking up the hen and chicks and placing them into the test box for increasing periods of time (5e20 min). The test apparatus was a 100  50 cm wooden structure divided into two sections, the hen box and the chick box, which were separated by a wire-mesh panel. The hen and her chicks were always placed into their corresponding box throughout the habituation and test periods. Wire panels on the front of the chick and hen boxes allowed observation and subsequent thermal imaging. During habituation and subsequent testing, a human was positioned 1 m in front of and facing the wire panel on the hen box. Their presence during habituation, along with a 5 min settling period before testing, ensured that subjects did not respond and were not attentive to their presence. On day 4, to prepare the hens for noninvasive heart rate monitoring, they began habituation to wearing a heart rate monitor. Before being placed in the test box, each hen was fitted with a harness made from elastane, which fitted around the back and tail, between the legs, and was secured behind the neck with hook and loop fastenings, allowing free limb movement and normal behaviour. Hens were closely supervised while they were wearing the elastane harness, to which no adverse reactions were observed. On day 5 of the habituation phase, self-adhesive electrode sensors (Ambu Blue sensor M-00-S, Ambu, Cambridgeshire, U.K.) were applied before fitting the harness; this involved placing each hen gently on her back, cleaning two small sections of skin overlying the pectoralis muscle either side of the sternum (using surgical spirit and cotton wool) and placing electrode sensors on the cleaned skin. Naturally occurring brood patches on the hens meant that no feather removal was required. The hens were monitored closely following electrode application and all resumed normal behaviours. At the end of day 5, four chicks from each brood were randomly assigned to two notional groups: Group 1 (N ¼ 2) and Group 2 (N ¼ 2). Group 1 chicks were assigned to be used in Phase 1 to determine the hens' response to chicks receiving an air puff and Group 2 chicks were assigned to be used in Phase 2 to determine the hens' effectiveness as a social buffer. This was because the physiological monitoring of hens in Phase 1 required them to wear harnesses containing heart rate loggers, meaning

13

there was potential for this to influence the chicks' behaviour and thus potentially hamper the hens' ability to act as social buffers for their chicks. Our previous study showed that domestic hens react consistently to their chicks receiving an air puff, with no difference in their behaviour or physiology over two replicates (Edgar et al., 2011) and so, in the current study, the hens' effectiveness as social buffers was tested separately in Phase 2. A separate group of the hens' chicks was used so that these were naïve to the air puff. Domestic hens readily accept foreign chicks and do not discriminate between chicks of a similar colour and size (Kent, 1987), so the fact that we tested the hens' responses to two different groups of their chicks was deemed unlikely to affect the hens' response. To identify the chicks belonging to the two groups within a brood, chicks were marked using two coloured stock markers (Blue and Yellow, Richey Sprayline, Richey Tagg Ltd, North Yorkshire, U.K.) on the back and/or tail feathers. The assignment of the chick group to the colour and marker location was counterbalanced between hens. Marking of the chicks 3 days prior to testing ensured that the marking procedure did not influence the hens' response. Hens and chicks were closely monitored following marking to ensure that no aggression occurred (which would indicate that marking had interfered with chick identification/recognition) and that normal maternal interactions continued. Week 2: habituation to heart rate monitor and camera On days 8e10 of the habituation period, hens were fitted with the harness before being placed in the test box, but additionally, a noninvasive telemetric logger (Lowe, Abeyesinghe, Demmers, Wathes, & McKeegan, 2007) was placed in the pocket of the harness and connected to the sensors on the hens' skin using two attached wires. After the fitting, each hen was placed into the test apparatus, with her chicks, for 20 min. During this time an experimenter took thermal images of the hen and chicks every minute, through the wire-mesh panels in the test box, to habituate them to the camera and its movement. Phase 1: Hen and Chick Response to Chick Distress On days 11 and 12, hens were tested to determine their response to one pair of their chicks receiving an air puff. This was to enable us subsequently to correlate the hens' responsiveness with their effectiveness as a social buffer. Chicks from Group 1 were used in this test. Each hen was assigned to one condition (Control or Air puff to chicks) per day of testing, such that hens experienced the two conditions in a counterbalanced order. The methods below were adapted from Edgar et al. (2011). Settling period Hens were fitted with the heart rate monitor then placed into the hen box. Group 1 chicks were then placed into the chick box and a 5 min settling period commenced. After this, hens were exposed to one of two conditions. (1) Control: hen and chicks were left undisturbed for 20 min. For consistency with the other condition during analysis, this condition was split into a notional 10 min pretreatment period and a 10 min treatment period, although no specific treatment was applied. (2) Air puff to chicks: after a 10 min pretreatment period, during which the hen and chicks were left undisturbed, a 10 min treatment period began in which an air puff from a canister of inert compressed air (Sprayduster, AF International, Ashby de la Zouch, U.K.) was sprayed into the chick box for 1 s every 30 s. Immediately after each test the harness and heart rate monitor were removed and the hen and chicks returned to their home pen.

14

J. Edgar et al. / Animal Behaviour 105 (2015) 11e19

Physiological and Behavioural Responses All physiological and behavioural parameters were monitored throughout the pretreatment and treatment periods for both conditions. Physiology Every minute a thermal image of the head of the hen and the head of each chick was taken using a thermal imaging camera (ThermaCam E4, FLIR). Maximum eye (hen and chicks) and comb (hen) temperature were extracted from the images using ThermaCAM Reporter 2000 Professional. An allowance of 10 s at either side of the 1 min mark ensured that it was possible to obtain a clear image of the side of each individual's head. Distance from the test box was 1 m and the thermal camera set to an emissivity of 0.96. The ambient temperature of the test room was maintained at 23  C. Heart rate data for the hens were obtained and analysed using the noninvasive telemetric logging system and software described above (Lowe et al., 2007). Behaviour Hen and chick behaviour and vocalizations were recorded continuously using a video camera positioned over the test apparatus. Videos were analysed using Observer v.5.0 (Noldus Information Technology, Wageningen, The Netherlands). All the behaviours and vocalizations listed by previous authors (Collias & Joos, 1953; Nicol, Caplen, Edgar, & Browne, 2009) were included, using the same descriptors. Phase 2: Is the Mother a Social Buffer for Her chicks? On days 15 and 16, chicks were tested (in the same test box as in Phase 1) to determine whether their response to the air puff differed in the presence and absence of their mother. If social buffering occurred within the mother/offspring bond then we would expect the chicks to perform fewer of the behaviours that were exacerbated during the air puff administration in Phase 1. To ensure that chicks were naïve to the air puff, Group 2 chicks were used in this test. Each hen's group of chicks was assigned to two conditions per day of testing, such that they experienced the four conditions in a counterbalanced order. (1) Control without mother: the chicks were left undisturbed without their mother for 20 min. For consistency with Phase 1, this condition was split into a notional 10 min pretreatment period and a 10 min treatment period, although no specific treatment was applied. (2) Control with mother: as per Condition 1, but both the chicks and their mother were left undisturbed for 20 min. (3) Air puff without mother: the chicks were without their mother for 20 min. This consisted of a 10 min pretreatment period, during which no specific treatment was applied. After this, a 10 min treatment period began in which an air puff from a canister of inert compressed air was sprayed into the chick box for 1 s every 30 s. (4) Air puff with mother: as Condition 3, but the mother was present for the 20 min test. For all four conditions, only data from the treatment period were used in the analysis. All physiology and behaviour were recorded in the chicks as per Phase 1. Statistical Analyses Data were analysed using SPSS Statistics 21 (IBM, Portsmouth, U.K.). Data for the chicks were calculated as a mean for both chicks. Hens' heart rates (bpm, beats/min) were calculated every 1 min. Heart rate (hens) and temperature data (hens and chicks) were

averaged over each 10 min period to produce one data point per 10 min period. The duration (percentage of time) of all performed behaviours and vocalizations were calculated for each 10 min period. All data were checked for normality using a KolmogoroveSmirnov test. All behavioural data were non-normally distributed and were transformed using the formula x ¼ square root(x þ 0.5). Data were checked again for normality and to ensure that they did not violate the assumption of sphericity (Mauchly's sphericity test). Phase 1 A repeated measures ANOVA was conducted with condition (Air puff and Control) and period (pretreatment and treatment period) as within-subjects factors. Post hoc tests (Bonferroni) were conducted in the event of a significant main or interaction effect from an ANOVA, during which pretreatment and treatment periods were compared for each condition. Phase 2 For normal data (eye temperature), a repeated measures ANOVA was conducted with stressor condition (Control or Air puff) and social condition (mother present or mother absent) as withinsubjects factors. Post hoc tests (Bonferroni) were conducted in the event of a significant interaction effect from an ANOVA, during which pretreatment and treatment periods were compared for each condition. For non-normal data (behavioural data), a Friedman test was conducted to determine whether there was a significant difference between the four conditions (Control without mother, Control with mother, Air puff without mother, Air puff with mother). If a significant difference was found, post hoc Wilcoxon signed-rank tests were conducted to determine where the difference was. Spearman correlations were used to determine the presence of correlations between mothers' response to their chicks receiving an air puff (i.e. hen Phase 1 response to Air puff minus Phase 1 response to Control) and the difference in chick behaviour as a result of the presence of their mother during the air puff (i.e. chick Phase 2 response to Air puff with mother minus Phase 2 response to Air puff without mother). Data Accessibility Hen and chick behaviour, thermal imaging and heart rate data are openly available on the University of Bristol Research Data Repository. RESULTS Phase 1a: Response of Hens As in our previous work (Edgar et al., 2011), we found a pronounced set of behavioural and physiological changes in hens when their chicks were subjected to an air puff. Hens had an increased heart rate (pretreatment period mean ¼ 275 ± 14 bpm; treatment period mean ¼ 337 ± 14 bpm; Wilks's lambda ¼ 2.64, F1,11 ¼ 30.625, P < 0.001, partial eta2 ¼ 0.736), time spent standing alert (pretreatment period median ¼ 19.9%, interquartile range, IQR ¼ 82.6; treatment period median ¼ 78.7%, IQR ¼ 48.8; Wilks's lambda ¼ 0.598, F1,11 ¼ 7.392, P ¼ 0.020, partial eta2 ¼ 0.402) and time spent vocalizing (pretreatment period median ¼ 1.2%, IQR ¼ 7; treatment period median ¼ 5.5%, IQR ¼ 38.6; Wilks's lambda ¼ 0.591, F1,11 ¼ 7.622, P ¼ 0.019, partial eta2 ¼ 0.409) and a decreased eye temperature (pretreatment period mean ¼ 33.8 ± 0.4  C; treatment period mean ¼ 32.4 ± 0.3  C; Wilks's

J. Edgar et al. / Animal Behaviour 105 (2015) 11e19

Phase 1b: Response of Chicks We observed a range of behavioural and physiological responses when the Group 1 chicks received the air puff in the presence of their mother. The chicks' eye temperature (Fig. 1a) dropped significantly during the Air puff treatment, but not during the same period in the Control condition (Wilks's lambda ¼ 0.198, F1,11 ¼ 44.683, P < 0.001, partial eta2 ¼ 0.802). In addition, the chicks changed their behaviour while exposed to the air puff, with an increase in time spent standing (Fig. 1b; Wilks's lambda ¼ 0.244, F1,11 ¼ 30.973, P < 0.001, partial eta2 ¼ 0.756) and walking (Wilks's lambda ¼ 0.245, F1,11 ¼ 30.815, P < 0.001, partial eta2 ¼ 0.755), and corresponding reduction in time spent sitting (Fig. 1c; Wilks's lambda ¼ 0.164, F1,11 ¼ 51.049, P < 0.001, partial eta-squared ¼ 0.836), ground pecking (Fig. 1d; Wilks's lambda ¼ 0.639, F1,11 ¼ 6.212, P ¼ 0.030, partial eta2 ¼ 0.361) and preening (pretreatment period median ¼ 9.5%,

35 (a)

Chick eye temperature (oC)

a

Period Pretreatment Treatment a

a

34

33

b

32

IQR ¼ 11.2; treatment period median ¼ 0, IQR ¼ 1.2; Wilks's lambda ¼ 0.328, F1,11 ¼ 20.442, P ¼ 0.001, partial eta2 ¼ 0.672). Chicks also spent more time walking into the mesh divider separating the chicks and the mother during the Air puff treatment period than in the Control treatment period (pretreatment period median ¼ 1%, IQR 0; treatment period median ¼ 28.0%, IQR 29.3; Wilks's lambda ¼ 0.307, F1,11 ¼ 22.615, P ¼ 0.001, partial eta2 ¼ 0.693). Phase 2a: Control Condition The presence of the hen had an effect on the behaviour and physiology of the Group 2 chicks during the Control condition. Chicks showed a general response to separation from their mother with reduced preening (Control with mother median ¼ 6.1%, IQR 11.8; Control without mother median ¼ 0%, IQR 0.8; Wilcoxon signed-rank test: W ¼ 78, z ¼ 3.059, P ¼ 0.002), ground pecking (Control with mother median ¼ 2.7%, IQR 9.1; Control without mother median ¼ 0%, IQR 0.1; Wilcoxon signed-rank test: W ¼ 61, z ¼ 2.490, P ¼ 0.013) and reduced eye temperature (Control with mother mean ¼ 33.7 ± 0.1  C; Control without mother mean ¼ 33.4 ± 0.2  C; Wilks's lambda ¼ 0.482, F1,11 ¼ 11.826, P ¼ 0.006, partial eta2 ¼ 0.518) during the Control without mother compared with the Control with mother.

60 (b) % Time chicks spent standing

lambda ¼ 0.360, F1,11 ¼ 19.573, P ¼ 0.001, partial eta2 ¼ 0.640) when observing the Group 1 chicks receiving an air puff. There were no other significant behavioural or comb temperature changes. There was no effect of maternal age or breed on the hens' response during phase 1.

15

Period Pretreatment Treatment

b

50 40 30 a a

20

a 10

% Time chicks spent sitting

100

80

C (c)

a

0

AP

a

Period Pretreatment Treatment a

60

40

20

0

b

C

AP

C

AP

6 (d) % Time chicks spent ground pecking

31

a

Period Pretreatment Treatment

5 a 4 a 3 2 b

1 0

C

AP

Figure 1. Chicks' (a) eye temperature, (b) standing, (c) sitting and (d) ground pecking in response to the pretreatment and treatment periods of the Control (C) and Air puff (AP) conditions (different letters signify significant differences).

16

J. Edgar et al. / Animal Behaviour 105 (2015) 11e19

Phase 2b: Air Puff Condition As well as the mother hens having a general effect on behaviour (during the Control condition), their presence influenced the behaviour of the chicks while they were subjected to the air puff. The mothers' presence resulted in chicks spending more time preening (Fig. 2a; Wilcoxon signed-rank test: W ¼ 41, z ¼ 2.192, P ¼ 0.028) and ground pecking (Fig. 2b; Wilcoxon signed-rank test: W ¼ 40, z ¼ 2.073, P ¼ 0.038) during administration of the air puff, than when the mother was absent during this treatment. The mothers' presence had no significant effect on any of the other behaviours during the tests, nor did the mothers' presence during the air puff administration have an effect on chick eye temperature. There was no effect of maternal age or breed on the chicks' eye temperature or behavioural response during phase 2. Correlations Between Hens' Response and Social-buffering Effects To determine whether the degree of socially mediated arousal expressed by mothers was linked with their effectiveness as a social

Treatment C AP

% Time chicks spent preening

12.5

b

120 (a)

10

7.5

5 c 2.5 a 0

Hen difference in heart rate due to AP to chick (bpm) ‘socially mediated arousal’

(a)

buffer, we correlated the parameters measured in the mother hens that differed when the chicks were receiving an air puff in Phase 1 (i.e. their responsiveness to the Air puff treatment period compared to the Control condition period) measured by changes in heart rate, eye temperature, standing, maternal vocalizations; together termed ‘socially mediated arousal’, with the parameters measured in chicks that differed in response to the presence of the mother when the chicks were receiving an air puff during Phase 2 (i.e. the degree to which the chicks showed fewer stress-related behaviours in response to the air puff in the presence of the mother compared to the air puff in the absence of the mother) measured by greater preening and ground pecking. Bonferroni corrections were applied, to take account of the eight correlations that were performed. This resulted in the required P value becoming 0.006. The hens' heart rate change in response to the chicks receiving an air puff was strongly negatively correlated with the chicks' increase in ground pecking (Fig. 3a; rS ¼ 0.927, N ¼ 11, P < 0.001) and preening (Fig. 3b; rS ¼ 0.789, N ¼ 11, P ¼ 0.004) as a result of the presence of their mother during the air puff. This heart rate change was not correlated with the chicks' change in ground pecking

90

60

30

0

-2

a

Without hen

0

2

4

6

8

10

Chick difference in preening due to presence of mother during AP (%) ‘social buffering’

With hen

% Time chicks spent ground pecking

Treatment C AP

5

b

4

3

2

c

Hen difference in heart rate due to AP to chick (bpm) ‘socially mediated arousal’

(b) 6

120 (b) 90

60

30

0

1 a 0

-2

a

Without hen

With hen

Figure 2. Chicks' (a) preening and (b) ground pecking in response to the Control (C) and Air puff (AP) conditions in the presence and absence of their mother (different letters signify significant differences).

4 0 2 Chick difference in ground pecking due to presence of mother during AP (%) ‘social buffering’

Figure 3. Correlations between the hens' ‘socially mediated arousal’ and their ‘socialbuffering’ effects on the chicks' (a) preening and (b) ground pecking (AP ¼ air puff).

J. Edgar et al. / Animal Behaviour 105 (2015) 11e19

(rS ¼ 0.042, N ¼ 11, P ¼ 0.897) and preening (rS ¼ 0.259, N ¼ 11, P ¼ 0.417) as a result of the presence of their mother during the Control condition. The hens' overall heart rate was not correlated with chick preening or ground pecking during either the Air puff or Control condition. There were no other significant correlations between the hen's and chicks' response. Before the correlation analyses were conducted, one outlier hen was identified and removed from the analysis. This outlier had a much larger increase in heart rate (>120 bpm) during air puff administration than the rest. Keeping this outlier in resulted in the same negative correlations, but with the correlation between the chick's difference in preening duration and the mother hen's heart rate difference becoming nonsignificant (due to Bonferroni correction) at P ¼ 0.01. DISCUSSION Having previously demonstrated that chicks respond behaviourally to their mother's expectation of an aversive stimulus (Edgar, Paul, et al., 2013), our aim here was to investigate whether hens are able to buffer their chicks' stress response. Second, we sought to determine the extent of individual variation in the hen's responsiveness to chick stress and how this affects her socialbuffering capacity. Our previous studies identified an air puff as an aversive stimulus for chicks (Edgar, 2012) but their behavioural and physiological responses had not previously been determined. We found a suite of chick behavioural and physiological responses to receiving an air puff, including an increase in time spent standing and walking, and a corresponding reduction in eye temperature, time spent sitting, ground pecking and preening. changes previously linked with a heightened stress response in chickens (Edgar et al., 2011; Edgar, Nicol, et al., 2013; Fortomaris, Arsenos, Tserveni-Gousi, & Yannakopoulos, 2007; Pohle & Cheng, 2009). During the Control (no air puff) condition the maternal presence had a strong effect on chick behaviour and physiology. This ‘social presence’ effect is distinct from social buffering (as it occurs without an independent stressor), and has been previously demonstrated between domestic chicks (Feltenstein, Lambdin, Webb, Acevedo, & Sufka, 2003; Jones & Williams, 1992; Kaufman & Hinde, 1961), adult starlings (Apfelbeck & Raess, 2008) and zebra finches (Banerjee & AdkinsRegan, 2011), as well as when brooded domestic chicks are separated from their mother (Bermant, 1963; Collias, 1952; Hughes et al., 1982). Additionally, and importantly, however, the mother hen had a specific effect on the chicks' behaviour during periods when the stressor was presented. Namely, chicks spent more time preening and ground pecking during the air puff administration if their mother was present. Taken together, these results show that mother hens act as social buffers for their chicks. Mother hens varied in both their effectiveness as a social buffer and the degree to which they showed socially mediated arousal, with a strong negative correlation between the two social phenomena. In particular, the extent to which a hen's heart rate increased when observing her chicks receiving an air puff was strongly negatively correlated with the degree of increase in preening and ground pecking shown by chicks when the maternal presence condition was compared with maternal absence. This heart rate change was not correlated with changes in chick preening or ground pecking due to the mothers' presence during the Control condition, nor were the hens' and chicks' behaviours synchronized in either condition, implying that social facilitation was not the causal mechanism for the correlation. Instead, the strong negative correlation between mother arousal and change in chick behaviour suggests that hens are able to modify their chicks' stress response and that socially aroused avian mothers are less

17

effective social buffers for their chicks. The mother hen may function as an ‘interpreter’ for her chicks, which are naïve to many aspects of the environment. In the current study, having never experienced the air puff before, the chicks might have been guided by their mother's response. Possibly, high maternal arousal signalled that the air puff was a fitness-threatening stimulus, and this might have overshadowed the predicted ‘calming’ socialbuffering effect. Further studies are needed to assess whether this correlation applies only to socially mediated arousal or whether mothers that are generally more stress-reactive are less effective social buffers. In the current study, differences in chick behavioural response to the stressor were minimized by using chicks of the same age and breed, subject to identical habituation and test procedures. However, an interesting question is the degree to which chick differences could act to mediate the hens' socially mediated arousal. Future studies using one-way glass or video playback could manipulate differences in both hen and chick stress response and could help to disentangle this potential two-way arousal. The existence of variation in the mothers' socially mediated arousal highlights the need for further work to examine whether these differences persist over successive tests and broods. All hens in the current study had the same experience of motherhood. Additionally, our analyses showed that the differences were not accounted for by age or breed, leaving the possibility that they may be representative of ‘types’ of mothers. The idea that individual mothers differ with regard to their maternal behaviour is well documented in mammals. These individual differences in maternal care, termed maternal styles, are consistent across generations of offspring and influence their behavioural phenotype (Fairbanks & McGuire, 1988). Recently, maternal styles have been identified for the first time in another precocial bird, the Japanese quail, Coturnix japonica (Pittet et al., 2014). In a cross-fostering experiment, naturally occurring individual differences in the characteristics of maternal care by mother quails were split into two principal independent dimensions: mothers' aggressiveness and their propensity to warm or reject their chicks. The authors found correlations between the mothers' temperament, maternal styles and the chicks' development, with the mothers' maternal rejection scores being positively correlated with the chicks' social motivation. The extent to which the interindividual variation observed in the current study persists over successive tests and broods and is representative of a ‘maternal style’ are key questions. Determining these links, using mothers of the same breed, age and maternal experience, would shed light on the importance of mothering for the domestic chicken and is an important line of further enquiry. Measuring the responses of both the subject (hen) and the demonstrators (chicks) can allow us to make inferences about the possible mechanism of the social-buffering effects. There are two mechanisms through which animals could act as social buffers: first by their simple presence (passive social buffering) and second through changes in their behaviour (active social buffering), including prosocial behaviour. To our knowledge, social-buffering studies have yet to attempt to distinguish between these mechanisms. In the current study, the presence of a correlation between mothers' arousal and their effectiveness as a social buffer suggests that it is not simply the presence of the hen that is important for the chicks, but some other aspect of her presence as well. Although we found that the mothers' heart rate was correlated with their effectiveness as a social buffer, none of the observed behaviours were. It is therefore not known exactly which aspect of the hens' presence and/or behaviour facilitated the chicks' social buffering. The chicks were unlikely to directly detect their mothers' increase in heart rate, although this cannot be ruled out. Possibly the chicks detected olfactory or temperature changes, a general shift in their mother's behaviour or subtle behavioural or postural changes that

18

J. Edgar et al. / Animal Behaviour 105 (2015) 11e19

were not detected in our behavioural analysis. Stress can affect behavioural structure, including patterns or temporal regularity of behaviours (Asher et al., 2009; Rutherford, Haskell, Glasbey, Jones, & Lawrence, 2003). Further studies using technology that would enable recording of more subtle behavioural and postural changes (e.g. facial expressions, pupil dilation) could also shed light on this mechanism. Conclusion We demonstrated, for the first time, that avian mothers are able to buffer their chicks' stress response. Furthermore, mothers showing more socially mediated arousal were less effective social buffers for their chicks. This study raises important fundamental questions on the extent to which the variation in response to chick stress is representative of a ‘maternal style’, correlates with general (nonmaternal) stress reactivity and influences longer-term chick stress reactivity and development. Acknowledgments This project was funded by the BBSRC (BB/J021679/1). We thank the two anonymous referees for their comments which helped to improve the manuscript. References Apfelbeck, B., & Raess, M. (2008). Behavioural and hormonal effects of social isolation and neophobia in a gregarious bird species, the European starling (Sturnus vulgaris). Hormones and Behaviour, 54(3), 435e441. http://dx.doi.org/ 10.1016/j.yhbeh.2008.04.003. Asher, L., Collins, L. M., Ortiz-Pelaez, A., Drewe, J. A., Nicol, C. J., & Pfeiffer, D. U. (2009). Recent advances in the analysis of behavioural organization and interpretation as indicators of animal welfare. Journal of the Royal Society Interface, 6(41), 1103e1119. http://dx.doi.org/10.1098/rsif.2009.0221. Banerjee, S. B., & Adkins-Regan, E. (2011). Effect of isolation and conspecific presence in a novel environment on corticosterone concentrations in a social avian species, the zebra finch (Taeniopygia guttata). Hormones and Behaviour, 60(3), 233e238. http://dx.doi.org/10.1016/j.yhbeh.2011.05.011. Bermant, G. (1963). Intensity and rate of distress calling in chicks as a function of social contact. Animal Behaviour, 11, 514e517. Bouwknecht, J. A., Olivier, B., & Paylor, R. E. (2007). The stress-induced hyperthermia paradigm as a physiological animal model for anxiety: a review of pharmacological and genetic studies in the mouse. Neuroscience and Biobehavioural Reviews, 31, 41e59. Busnardo, C., Tavares, R. F., Resstel, L. B. M., Elias, L. L. K., & Correa, F. M. A. (2010). Paraventricular nucleus modulates autonomic and neuroendocrine responses to acute restraint stress in rats. Autonomic Neuroscience, 158, 51e57. Collias, N. E. (1952). The development of social behavior in birds. Auk, 69, 127e159. Collias, N., & Joos, M. (1953). The spectrographic analysis of sound signals of the domestic fowl. Behaviour, 5(3), 175e188. http://dx.doi.org/10.1163/ 156853953x00104. Edgar, J. L. (2012). Are chickens affected by the distress of conspecifics? (Unpublished Ph.D. thesis). Bristol, U.K.: University of Bristol. Edgar, J. L., Lowe, J. C., Paul, E. S., & Nicol, C. J. (2011). Avian maternal response to chick distress. Proceedings of the Royal Society B: Biological Sciences, 278(1721), 3129e3134. http://dx.doi.org/10.1098/rspb.2010.2701. Edgar, J. L., Nicol, C. J., Clark, C. C. A., & Paul, E. S. (2012). Measuring empathic responses in animals. Applied Animal Behaviour Science, 138(3e4), 182e193. http://dx.doi.org/10.1016/j.applanim.2012.02.006. Edgar, J. L., Nicol, C. J., Pugh, C. A., & Paul, E. S. (2013). Surface temperature changes in response to handling in domestic chickens. Physiology and Behaviour, 119, 195e200. http://dx.doi.org/10.1016/j.physbeh.2013.06.020. Edgar, J. L., Paul, E. S., Harris, L., Penturn, S., & Nicol, C. J. (2012). No evidence for emotional empathy in chickens observing familiar adult conspecifics. PLoS One, 7(2), 0031542. http://dx.doi.org/10.1371/journal.pone.0031542. Edgar, J. L., Paul, E. S., & Nicol, C. J. (2013). Protective mother hens: cognitive influences on the avian maternal response. Animal Behaviour, 86(2), 223e229. http://dx.doi.org/10.1016/j.anbehav.2013.05.004. Fairbanks, L. A., & McGuire, M. T. (1988). Long-term effects of early mothering behavior on responsiveness to the environment in vervet monkeys. Developmental Psychobiology, 21(7), 711e724. http://dx.doi.org/10.1002/dev.420210708. Feltenstein, M. W., Lambdin, C., Webb, H. E., Acevedo, E. O., & Sufka, K. J. (2003). Corticosterone responses in the chick social separationestress paradigm. Physiology and Behaviour, 78(3), 489e493.

Fortomaris, P., Arsenos, G., Tserveni-Gousi, A., & Yannakopoulos, A. (2007). Performance and behaviour of broiler chickens as affected by the housing system. Archiv für Geflugelkunde, 71(3), 97e104. Geverink, N. A., Buhnemann, A., Van de Burgwal, J. A., Lambooij, E., Blokhuis, H. J., & Wiegant, V. M. (1998). Responses of slaughter pigs to transport and lairage sounds. Physiology and Behaviour, 63(4), 667e673. http://dx.doi.org/10.1016/ s0031-9384(97)00513-1. Gonzalez, M., Averos, X., Beltran de Heredia, I., Ruiz, R., Arranz, J., & Estevez, I. (2013). The effect of social buffering on fear responses in sheep (Ovis aries). Applied Animal Behaviour Science, 149(1e4), 13e20. http://dx.doi.org/10.1016/ j.applanim.2013.09.011. Graves, F. C., & Hennessy, M. B. (2000). Comparison of the effects of the mother and an unfamiliar adult female on cortisol and behavioral responses of pre- and postweaning guinea pigs. Developmental Psychobiology, 36(2), 91e100. http:// dx.doi.org/10.1002/(sici)1098-2302(200003)36:2<91::aid-dev1>3.3.co;2-t. Grippo, A. J., Pournajafi-Nazarloo, H., Sanzenbacher, L., Trahanas, D. M., McNeal, N., Clarke, D. A., et al. (2012). Peripheral oxytocin administration buffers autonomic but not behavioral responses to environmental stressors in isolated prairie voles. Stress, 15(2), 149e161. http://dx.doi.org/10.3109/10253890.2011.605486. Guzman, D. A., & Marin, R. H. (2008). Social reinstatement responses of meat-type chickens to familiar and unfamiliar conspecifics after exposure to an acute stressor. Applied Animal Behaviour Science, 110(3e4), 282e293. http:// dx.doi.org/10.1016/j.applanim.2007.04.017. Hennessy, M. B., O'Leary, S. K., Hawke, J. L., & Wilson, S. E. (2002). Social influences on cortisol and behavioral responses of preweaning, periadolescent, and adult guinea pigs. Physiology and Behaviour, 76(2), 305e314. http://dx.doi.org/ 10.1016/s0031-9384(02)00712-6. Hennessy, M. B., & Ritchey, R. L. (1987). Hormonal and behavioral attachment responses in infant guinea-pigs. Developmental Psychobiology, 20(6), 613e625. http://dx.doi.org/10.1002/dev.420200607. Hodges, T. E., Green, M. R., Simone, J. J., & McCormick, C. M. (2014). Effects of social context on endocrine function and Zif268 expression in response to an acute stressor in adolescent and adult rats. International Journal of Developmental Neuroscience, 35, 25e34. http://dx.doi.org/10.1016/j.ijdevneu.2014.03.001. Hostetler, C. M., & Ryabinin, A. E. (2014). Social partners prevent alcohol relapse behavior in prairie voles. Psychoneuroendocrinology, 39, 152e157. http:// dx.doi.org/10.1016/j.psyneuen.2013.10.006. Hughes, M. K., Hughes, A. L., & Covalt-Dunning, D. (1982). Frequency and intensity of chick distress calls: effects on maternal foodcalling in chickens. Behavioural Processes, 7, 335e340. Ishiwata, T., Kilgour, R. J., Uetake, K., Eguchi, Y., & Tanaka, T. (2007). Choice of attractive conditions by beef cattle in a Y-maze just after release from restraint. Journal of Animal Science, 85(4), 1080e1085. http://dx.doi.org/10.2527/jas.2006405. Jones, R. B., & Williams, J. B. (1992). Responses of pair-housed male and female domestic chicks to the removal of a companion. Applied Animal Behaviour Science, 32(4), 375e380. http://dx.doi.org/10.1016/s0168-1591(05)80029-2. Kanitz, E., Hameister, T., Tuchscherer, M., Tuchscherer, A., & Puppe, B. (2014). Social support attenuates the adverse consequences of social deprivation stress in domestic piglets. Hormones and Behaviour, 65(3), 203e210. http://dx.doi.org/ 10.1016/j.yhbeh.2014.01.007. Kaufman, I. C., & Hinde, R. A. (1961). Factors influencing distress calling in chicks, with special reference to temperature changes and social isolation. Animal Behaviour, 9(3e4), 197e204. http://dx.doi.org/10.1016/0003-3472(61)90009-4. Kent, J. P. (1987). Experiments on the relationship between the hen and chick (Gallus gallus): the role of the auditory mode in recognition and the effects of maternal separation. Behaviour, 102, 1e14. Kiyokawa, Y., Hiroshima, S., Takeuchi, Y., & Mod, Y. (2014). Social buffering reduces male rats' behavioral and corticosterone responses to a conditioned stimulus. Hormones and Behaviour, 65(2), 114e118. http://dx.doi.org/10.1016/ j.yhbeh.2013.12.005. Kiyokawa, Y., Honda, A., Takeuchi, Y., & Mori, Y. (2014). A familiar conspecific is more effective than an unfamiliar conspecific for social buffering of conditioned fear responses in male rats. Behavioural Brain Research, 267, 189e193. http:// dx.doi.org/10.1016/j.bbr.2014.03.043. Kiyokawa, Y., Takeuchi, Y., & Mori, Y. (2007). Two types of social buffering differentially mitigate conditioned fear responses. European Journal of Neuroscience, 26(12), 3606e3613. http://dx.doi.org/10.1111/j.1460-9568.2007.05969x. Lowe, J. C., Abeyesinghe, S. M., Demmers, T. G. M., Wathes, C. M., & McKeegan, D. E. F. (2007). A novel telemetric logging system for recording physiological signals in unrestrained animals. Computers and Electronics in Agriculture, 57(1), 74e79. http://dx.doi.org/10.1016/j.compag.2007.02.003. Marin, R. H., Freytes, P., Guzman, D., & Jones, R. B. (2001). Effects of an acute stressor on fear and on the social reinstatement responses of domestic chicks to cagemates and strangers. Applied Animal Behaviour Science, 71(1), 57e66. http:// dx.doi.org/10.1016/s0168-1591(00)00167-2. McCormick, C. M., Merrick, A., Secen, J., & Helmreich, D. L. (2007). Social instability in adolescence alters the central and peripheral hypothalamic-pituitary-adrenal responses to a repeated homotypic stressor in male and female rats. Journal of Neuroendocrinology, 19(2), 116e126. http://dx.doi.org/10.1111/j.13652826.2006.01515x. McQuaid, R. J., Audet, M. C., Jacobson-Pick, S., & Anisman, H. (2013). The differential impact of social defeat on mice living in isolation or groups in an enriched environment: plasma corticosterone and monoamine variations. International

J. Edgar et al. / Animal Behaviour 105 (2015) 11e19 Journal of Neuropsychopharmacology, 16(2), 351e363. http://dx.doi.org/10.1017/ s1461145712000120. Mommer, B. C., & Bell, A. M. (2013). A test of maternal programming of offspring stress response to predation risk in threespine sticklebacks. Physiology and Behaviour, 122, 222e227. http://dx.doi.org/10.1016/j.physbeh. 2013.04.004. Nicol, C. J., Caplen, G., Edgar, J., & Browne, W. J. (2009). Associations between welfare indicators and environmental choice in laying hens. Animal Behaviour, 78(2), 413e424. http://dx.doi.org/10.1016/j.anbehav.2009.05.016. Pittet, F., Houdelier, C., de Margerie, E., Le Bot, O., Richard-Yris, M.-A., & Lumineau, S. (2014). Maternal styles in a precocial bird. Animal Behaviour, 87, 31e37. http:// dx.doi.org/10.1016/j.anbehav.2013.10.025. Pohle, K., & Cheng, H. W. (2009). Furnished cage system and hen well-being: comparative effects of furnished cages and battery cages on behavioral exhibitions in White Leghorn chickens. Poultry Science, 88(8), 1559e1564. http:// dx.doi.org/10.3382/ps.2009-00045. Ritchey, R. L., & Hennessy, M. B. (1987). Cortisol and behavioral-responses to separation in mother and infant guinea-pigs. Behavioural and Neural Biology, 48(1), 1e12. http://dx.doi.org/10.1016/s0163-1047(87)90514-0. Rutherford, K. M. D., Haskell, M. J., Glasbey, C., Jones, R. B., & Lawrence, A. B. (2003). Detrended fluctuation analysis of behavioural responses to mild acute stressors

19

in domestic hens. Applied Animal Behaviour Science, 83(2), 125e139. http:// dx.doi.org/10.1016/s0168-1591(03)00115-1. Sachser, N., Durschlag, M., & Hirzel, D. (1998). Social relationships and the management of stress. Psychoneuroendocrinology, 23(8), 891e904. http://dx.doi.org/ 10.1016/s0306-4530(98)00059-6. Suarez, & Gallup. (1983). Social reinstatement and open field testing in chickens. Animal Learning and Behaviour, 11, 119e126. Terranova, M. L., Cirulli, F., & Laviola, G. (1999). Behavioral and hormonal effects of partner familiarity in periadolescent rat pairs upon novelty exposure. Psychoneuroendocrinology, 24(6), 639e656. Tuchscherer, M., Kanitz, E., Puppe, B., Hameister, T., & Tuchscherer, A. (2014). Social support modulates splenocyte glucocorticoid sensitivity in piglets exposed to social deprivation stress. Physiology and Behaviour, 131, 25e32. http:// dx.doi.org/10.1016/j.physbeh.2014.04.010. Vallortigara, G., Cailotto, M., & Zanforlin, M. (1990). Sex-differences in social reinstatement motivation of the domestic chick (Gallus gallus) revealed by runway tests with social and nonsocial reinforcement. Journal of Comparative Psychology, 104(4), 361e367. Varlinskaya. (2013). Social context induces two unique patterns of c-Fos expression in adolescent and adult rats. Developmental Psychobiology, 55(7), 684e697. http://dx.doi.org/10.1002/dev.21064.