Background: the HemiSPAIRE trial is being conducted to determine whether a modified
muscle sparing technique (SPAIRE - “Save Piriformis and Internus, Repairing Externus”) in
hip hemiarthroplasty brings clinical benefits compared to the standard lateral technique in
adults aged 60 years or older, with a displaced intracapsular hip fracture. This article describes
the detailed statistical analysis plan for the trial.
Methods and design: HemiSPAIRE is a definitive, pragmatic, superiority, multicentre,
randomised controlled trial (with internal pilot) with two parallel groups. Participants, ward
staff and all research staff involved in post-operative assessments are blinded to allocation.
This article describes in detail (1) the primary and secondary outcomes, (2) the statistical
analysis principles, including: a survivor average causal effect (SACE) method chosen
specifically to address the issue of potential bias from differential survival between trial arms,
which was seen from data review by the Trial Steering Committee; the participants that will be
included in each analysis; the covariates that will be included in each analysis, and how the
results will be presented, (3) planned main analysis of the primary outcome; (4) planned
analyses of the secondary outcomes, (5) planned additional analyses of the primary and
secondary outcomes.

BACKGROUND: Sleep disturbance (SD) is common among people living with dementia (PLwD) or mild cognitive impairment (MCI). It has a significant impact on the wellbeing of PLwD and caregivers, and makes care at home more difficult. Within primary care, assessment and management of SD for this population is complex and challenging. AIM: to identify what works, how, and for whom, in the assessment and management of SD for PLwD or MCI in primary care. METHOD: We conducted a realist review to develop explanations of causal relationships, using context-mechanism-outcome configurations (CMOCs). An initial programme theory was iteratively tested and refined, using data from relevant literature and stakeholder feedback. The study followed RAMESES reporting quality. RESULTS: in total, 71 papers from OECD countries were included for analysis, generating 19 CMOCs. Low awareness of SD and assessment methods resulted in underdiagnosis in primary care. Assessment and treatment of PLwD/MCI were, respectively, more challenging when people were unable to accurately report concerns or implement interventions independently. Sedative medication was commonly used to manage SD, often driven by low confidence in nonpharmacological strategies. Long-term medication use was common despite guidelines indicating limited efficacy, which was driven by perceived pressures to prescribe or concerns of relapse. In nursing homes, environments and routines could exacerbate SD. CONCLUSION: Increasing awareness, knowledge, and confidence in diagnosis and assessment of SD is needed. Primary care-specific assessment tools may help. Long-term medication use is the default option in absence of pragmatic and effective non-pharmacological interventions that can be easily incorporated into routine general practice.

AbstractBackgroundGlobal initiatives that promote public health responses to dementia have resulted in numerous countries developing new national policies. Current policy guidelines in England, for example, recommend that people diagnosed with mild‐to‐moderate dementia receive information and psychosocial interventions to improve their ability to ‘live well’. However, it remains unclear to what extent these recommendations are being achieved.MethodsSelf‐reported information from 1537 people living with dementia and informant‐reported information from 1277 carers of people living with dementia was used to quantify receipt of community‐based dementia support services, including health and social care services provided by statutory or voluntary‐sector organisations, in Britain from 2014 to 2016. Demographic factors associated with differences in receipt of support services were also investigated to identify particularly vulnerable groups of people living with dementia.ResultsBoth self‐ and informant reports suggested that approximately 50% of people living with dementia received support services for dementia. Receipt of support services was lower among people living with dementia who are older, female, and have fewer educational qualifications. Receipt of support services also differed according to diagnosis and carer status, but was unrelated to marital status.ConclusionsLimited receipt of dementia support services among people living with dementia in Britain provides a baseline to assess the efficacy of current policy guidelines regarding provision of information and support. Targeted efforts to facilitate receipt of support services among the particularly vulnerable groups identified in the current study could improve the efficacy of dementia support services both in Britain and internationally, and should inform policy development.

INTRODUCTION: the increasingly ageing population is associated with greater numbers of people living with dementia (PLwD) and mild cognitive impairment (MCI). There are an estimated 55 million PLwD and approximately 6% of people over 60 years of age are living with MCI, with the figure rising to 25% for those aged between 80 and 84 years. Sleep disturbances are common for this population, but there is currently no standardised approach within UK primary care to manage this. Coined as a 'wicked design problem', sleep disturbances in this population are complex, with interventions supporting best management in context. METHODS AND ANALYSIS: the aim of this realist review is to deepen our understanding of what is considered 'sleep disturbance' in PLwD or MCI within primary care. Specifically, we endeavour to better understand how sleep disturbance is assessed, diagnosed and managed. To co-produce this protocol and review, we have recruited a stakeholder group comprising individuals with lived experience of dementia or MCI, primary healthcare staff and sleep experts. This review will be conducted in line with Pawson's five stages including the development of our initial programme theory, literature searches and the refinement of theory. The Realist and Meta-narrative Evidence Syntheses: Evolving Standards (RAMESES) quality and reporting standards will also be followed. The realist review will be an iterative process and our initial realist programme theory will be tested and refined in response to our data searches and stakeholder discussions. ETHICS AND DISSEMINATION: Ethical approval is not required for this review. We will follow the RAMESES standards to ensure we produce a complete and transparent report. Our final programme theory will help us to devise a tailored sleep management tool for primary healthcare professionals, PLwD and their carers. Our dissemination strategy will include lay summaries via email and our research website, peer-reviewed publications and social media posts. PROSPERO REGISTRATION NUMBER: CRD42022304679.

AbstractCognitive biases for encoding spatial information (orientation strategies) in relation to self (egocentric) or landmarks (allocentric) differ between species or populations according to the habitats they occupy. Whether biases in orientation strategy determine early habitat selection or if individuals adapt their biases following experience is unknown. We determined orientation strategies of pheasants, Phasianus colchicus, using a dual‐strategy maze with an allocentric probe trial, before releasing them (n = 20) into a novel landscape, where we monitored their movement and habitat selection. In general, pheasants selected for woodland over non‐woodland habitat, but allocentric‐biased individuals exhibited weaker avoidance of non‐woodland habitat, where we expected allocentric navigation to be more effective. Sex did not influence selection but was associated with speed and directional persistence in non‐woodland habitat. Our results suggest that an individual's habitat selection is associated with inherent cognitive bias in early life, but it is not yet clear what advantages this may offer.

. Memories about the spatial environment, such as the locations of foraging patches, are expected to affect how individuals move around the landscape. However, individuals differ in the ability to remember spatial locations (spatial cognitive ability) and evidence is growing that these inter-individual differences influence a range of fitness proxies. Yet empirical evaluations directly linking inter-individual variation in spatial cognitive ability and the development and structure of movement paths are lacking. We assessed the performance of young pheasants (
. Phasianus colchicus
. ) on a spatial cognition task before releasing them into a novel, rural landscape and tracking their movements. We quantified changes in the straightness and speed of their transitory paths over one month. Birds with better performances on the task initially made slower transitory paths than poor performers but by the end of the month, there was no difference in speed. In general, birds increased the straightness of their path over time, indicating improved efficiency independent of speed, but this was not related to performance on the cognitive task. We suggest that initial slow movements may facilitate more detailed information gathering by better performers and indicates a potential link between an individual's spatial cognitive ability and their movement behaviour.
.

Social environments influence important ecological processes and can determine how selection acts on traits. Cognitive abilities can shape these social environments and in turn, affect individuals' fitness. To understand how cognitive abilities evolve, we need to understand the complex interplay between an individual's cognitive abilities, the social environment that they inhabit and the fitness consequences of these relationships. We measured the associative learning ability of pheasant chicks, Phasianus colchicus, then released them into the wild where we quantified their social position by observing their associations at feeding stations and monitored the number of days survived. We observed disassortative mixing by learning performance at the population level, and poor learners had more associates than good learners. Learning was beneficial for survival when focal individuals had fewer than four associates, but survival probability across learning abilities equalized for individuals with more than four associates. While the mechanisms underlying these relationships remain to be determined, the patterns of association exhibited by pheasants at feeders can be predicted by individual variation in cognitive performances and we suspect these patterns are related to differences in information use. Critically, these resulting patterns of association have fitness consequences for individuals that cannot be explained directly by their cognitive ability, but which could mediate selection on cognition.

AbstractTo understand the evolution of cognitive abilities, we need to understand both how selection acts upon them and their genetic (co)variance structure. Recent work suggests that there are fitness consequences for free-living individuals with particular cognitive abilities. However, our current understanding of the heritability of these abilities is restricted to domesticated species subjected to artificial selection. We investigated genetic variance for, and genetic correlations among four cognitive abilities: inhibitory control, visual and spatial discrimination, and spatial ability, measured on >450 pheasants, Phasianus colchicus, over four generations. Pheasants were reared in captivity but bred from adults that lived in the wild and hence, were subject to selection on survival. Pheasant chicks are precocial and were reared without parents, enabling us to standardize environmental and parental care effects. We constructed a pedigree based on 15 microsatellite loci and implemented animal models to estimate heritability. We found moderate heritabilities for discrimination learning and inhibitory control (h2 = 0.17–0.23) but heritability for spatial ability was low (h2 = 0.09). Genetic correlations among-traits were largely positive but characterized by high uncertainty and were not statistically significant. Principle component analysis of the genetic correlation matrix estimate revealed a leading component that explained 69% of the variation, broadly in line with expectations under a general intelligence model of cognition. However, this pattern was not apparent in the phenotypic correlation structure which was more consistent with a modular view of animal cognition. Our findings highlight that the expression of cognitive traits is influenced by environmental factors which masks the underlying genetic structure.

The differential specialization of each side of the brain facilitates the parallel processing of information and has been documented in a wide range of animals. Animals that are more lateralized as indicated by consistent preferential limb use are commonly reported to exhibit superior cognitive ability as well as other behavioural advantages. We assayed the lateralization of 135 young pheasants (Phasianus colchicus), indicated by their footedness in a spontaneous stepping task, and related this measure to individual performance in either 3 assays of visual or spatial learning and memory. We found no evidence that pronounced footedness enhances cognitive ability in any of the tasks. We also found no evidence that an intermediate footedness relates to better cognitive performance. This lack of relationship is surprising because previous work revealed that pheasants have a slight population bias towards right footedness, and when released into the wild, individuals with higher degrees of footedness were more likely to die. One explanation for why extreme lateralization is constrained was that it led to poorer cognitive performance, or that optimal cognitive performance was associated with some intermediate level of lateralization. This stabilizing selection could explain the pattern of moderate lateralization that is seen in most non-human species that have been studied. However, we found no evidence in this study to support this explanation.

The ability to inhibit prepotent actions towards rewards that are made inaccessible by transparent barriers has been considered to reflect capacities for inhibitory control (IC). Typically, subjects initially reach directly, and incorrectly, for the reward. With experience, subjects may inhibit this action and instead detour around barriers to access the reward. However, assays of IC are often measured across multiple trials, with the location of the reward remaining constant. Consequently, other cognitive processes, such as response learning (acquisition of a motor routine), may confound accurate assays of IC. We measured baseline IC capacities in pheasant chicks, Phasianus colchicus, using a transparent cylinder task. Birds were then divided into two training treatments, where they learned to access a reward placed behind a transparent barrier, but experienced differential reinforcement of a particular motor response. In the stationary-barrier treatment, the location of the barrier remained constant across trials. We, therefore, reinforced a fixed motor response, such as always go left, which birds could learn to aid their performance. Conversely, we alternated the location of the barrier across trials for birds in the moving-barrier treatment and hence provided less reinforcement of their response learning. All birds then experienced a second presentation of the transparent cylinder task to assess whether differences in the training treatments influenced their subsequent capacities for IC. Birds in the stationary-barrier treatment showed a greater improvement in their subsequent IC performance after training compared to birds in the moving-barrier treatment. We, therefore, suggest that response learning aids IC performance on detour tasks. Consequently, non-target cognitive processes associated with different neural substrates appear to underlie performances on detour tasks, which may confound accurate assays of IC. Our findings question the construct validity of a commonly used paradigm that is widely considered to assess capacities for IC in humans and other animals.

Inhibitory control (IC) is the ability to intentionally restrain initial, ineffective responses to a stimulus and instead exhibit an alternative behaviour that is not pre-potent but which effectively attains a reward. Individuals (both humans and non-human animals) differ in their IC, perhaps as a result of the different environmental conditions they have experienced. We experimentally manipulated environmental predictability, specifically how reliable information linking a cue to a reward was, over a very short time period and tested how this affected an individual’s IC. We gave 119 pheasants (Phasianus colchicus) the opportunity to learn to associate a visual cue with a food reward in a binary choice task. We then perturbed this association for half the birds, whereas control birds continued to be rewarded when making the correct choice. We immediately measured all birds’ on a detour IC task and again 3 days later. Perturbed birds immediately performed worse than control birds, making more unrewarded pecks at the apparatus than control birds, although this effect was less for individuals that had more accurately learned the initial association. The effect of the perturbation was not seen 3 days later, suggesting that individual IC performance is highly plastic and susceptible to recent changes in environmental predictability. Specifically, individuals may perform poorly in activities requiring IC immediately after information in their environment is perturbed, with the perturbation inducing emotional arousal. Our finding that recent environmental changes can affect IC performance, depending on how well an animal has learned about that environment, means that interpreting individual differences in IC must account for both prior experience and relevant individual learning abilities.

Sexual segregation is common and can occur when sexes occupy different habitats, and/or when sexes aggregate assortatively within the same habitats. However, it is rarely studied in birds, with most previous work concentrating on differential settlement by the sexes in discrete habitats, often separated by large distances. Little attention has been paid to patterns of segregation within the same site. We reared 200 Common Pheasants Phasianus colchicus and released them onto a relatively small site of 250 ha and recorded their patterns of association and differential use of artificial feeders in space and time. Particular feeders were preferred by one sex, although we found no features of the local habitat which explained such preferences. Furthermore, we found sex differences in the use of feeders throughout the day, with females preferentially visiting them in the morning and the proportion of females visiting feeders increasing as the year progressed. Social network analyses found that in the first month after release into the wild, females did not associate strongly with other females, which was surprising as, prior to release, females have been shown to associate with other females in both semi-natural conditions and when tested in isolation. However, sexual segregation was clearly seen after 1 month of being released and became more pronounced as the year progressed. Females associated with other females from November to February, whereas males avoided other males over this same period. Sexes became less likely to associate with one another in 5 of the 6 months monitored. Such avoidance observed in males suggests that they start to form territories much sooner than previously thought. Pheasants exhibit clear patterns of fine-scale sexual segregation based on space and time, which was observed in their social preferences at feeding sites. Such detailed fine-scale segregation is rarely observed in birds.

The ability to control impulsive actions is an important executive function that is central to the self-regulation of behaviours and, in humans, can have important implications for mental and physical health. One key factor that promotes individual differences in inhibitory control (IC) is the predictability of environmental information experienced during development (i.e. reliability of resources and social trust). However, environmental predictability can also influence motivational and other cognitive abilities, which may therefore confound interpretations of the mechanisms underlying IC. We investigated the role of environmental predictability, food motivation and cognition on IC. We reared pheasant chicks, Phasianus colchicus, under standardised conditions, in which birds experienced environments that differed in their spatial predictability. We systematically manipulated spatial predictability during their first 8 weeks of life, by either moving partitions daily to random locations (unpredictable environment) or leaving them in fixed locations (predictable environment). We assessed motivation by presenting pheasants with two different foraging tasks that measured their dietary breadth and persistence to acquire inaccessible food rewards, as well as recording their latencies to acquire a freely available baseline worm positioned adjacent to each test apparatus, their body condition (mass/tarsus3) and sex. We assessed cognitive performance by presenting each bird with an 80-trial binary colour discrimination task. IC was assessed using a transparent detour apparatus, which required subjects to inhibit prepotent attempts to directly acquire a visible reward through the barrier and instead detour around a barrier. We found greater capacities for IC in pheasants that were reared in spatially unpredictable environments compared to those reared in predictable environments. While IC was unrelated to individual differences in cognitive performance on the colour discrimination task or motivational measures, we found that environmental predictability had differential effects on sex. Males reared in an unpredictable environment, and all females regardless of their rearing environment, were less persistent than males reared in a predictable environment. Our findings, therefore, suggest that an individual's developmental experience can influence their performance on IC tasks.

Transparent Cylinder and Barrier tasks are used to purportedly assess inhibitory control in a variety of animals. However, we suspect that performances on these detour tasks are influenced by non-cognitive traits, which may result in inaccurate assays of inhibitory control. We therefore reared pheasants under standardized conditions and presented each bird with two sets of similar tasks commonly used to measure inhibitory control. We recorded the number of times subjects incorrectly attempted to access a reward through transparent barriers, and their latencies to solve each task. Such measures are commonly used to infer the differential expression of inhibitory control. We found little evidence that their performances were consistent across the two different Putative Inhibitory Control Tasks (PICTs). Improvements in performance across trials showed that pheasants learned the affordances of each specific task. Critically, prior experience of transparent tasks, either Barrier or Cylinder, also improved subsequent inhibitory control performance on a novel task, suggesting that they also learned the general properties of transparent obstacles. Individual measures of persistence, assayed in a third task, were positively related to their frequency of incorrect attempts to solve the transparent inhibitory control tasks. Neophobia, Sex and Body Condition had no influence on individual performance. Contrary to previous studies of primates, pheasants with poor performance on PICTs had a wider dietary breadth assayed using a free-choice task. Our results demonstrate that in systems or taxa where prior experience and differences in development cannot be accounted for, individual differences in performance on commonly used detour-dependent PICTS may reveal more about an individual’s prior experience of transparent objects, or their motivation to acquire food, than providing a reliable measure of their inhibitory control.

Dominant individuals differ from subordinates in their performances on cognitive tasks across a suite of taxa. Previous studies often only consider dyadic relationships, rather than the more ecologically relevant social hierarchies or networks, hence failing to account for how dyadic relationships may be adjusted within larger social groups. We used a novel statistical method: randomized Elo-ratings, to infer the social hierarchy of 18 male pheasants, Phasianus colchicus, while in a captive, mixed-sex group with a linear hierarchy. We assayed individual learning performance of these males on a binary spatial discrimination task to investigate whether inter-individual variation in performance is associated with group social rank. Task performance improved with increasing trial number and was positively related to social rank, with higher ranking males showing greater levels of success. Motivation to participate in the task was not related to social rank or task performance, thus indicating that these rank-related differences are not a consequence of differences in motivation to complete the task. Our results provide important information about how variation in cognitive performance relates to an individual’s social rank within a group. Whether the social environment causes differences in learning performance or instead, inherent differences in learning ability predetermine rank remains to be tested.

To understand how natural selection may act on cognitive processes, it is necessary to reliably determine interindividual variation in cognitive abilities. However, an individual's performance in a cognitive test may be influenced by the social environment. The social environment explains variation between species in cognitive performances, with species that live in larger groups purportedly demonstrating more advanced cognitive abilities. It also explains variation in cognitive performances within species, with larger groups more likely to solve novel problems than smaller groups. Surprisingly, an effect of group size on individual variation in cognitive performance has rarely been investigated and much of our knowledge stems from impaired performance of individuals reared in isolation. Using a within-subjects design we assayed individual learning performance of adult female pheasants, Phasianus colchicus, while housed in groups of three and five. Individuals experienced the group sizes in a different order, but were presented with two spatial discrimination tasks, each with a distinct cue set, in a fixed order. We found that across both tasks individuals housed in the large groups had higher levels of success than individuals housed in the small groups. Individuals had higher levels of success on their second than their first task, irrespective of group size. We suggest that the expression of individual learning performance is responsive to the current social environment but the mechanisms underpinning this relationship require further investigation. Our study demonstrates that it is important to account for an individual's social environment when attempting to characterize cognitive capacities. It also demonstrates the flexibility of an individual's cognitive performance depending on the social context.

Brain lateralization is considered adaptive because it leads to behavioral biases and specializations that bring fitness benefits. Across species, strongly lateralized individuals perform better in specific behaviors likely to improve survival. What constrains continued exaggerated lateralization? We measured survival of pheasants, finding that individuals with stronger bias in their footedness had shorter life expectancies compared to individuals with weak biases. Consequently, weak, or no footedness provided the highest fitness benefits. If, as suggested, footedness is indicative of more general brain lateralization, this could explain why continued brain lateralization is constrained even though it may improve performance in specific behaviors.

Cognitive abilities probably evolve through natural selection if they provide individuals with fitness benefits. A growing number of studies demonstrate a positive relationship between performance in psychometric tasks and (proxy) measures of fitness. We assayed the performance of 154 common pheasant (Phasianus colchicus) chicks on tests of acquisition and reversal learning, using a different set of chicks and different set of cue types (spatial location and colour) in each of two years and then followed their fates after release into the wild. Across all birds, individuals that were slow to reverse previously learned associations were more likely to survive to four months old. For heavy birds, individuals that rapidly acquired an association had improved survival to four months, whereas for light birds, slow acquirers were more likely to be alive. Slow reversers also exhibited less exploratory behaviour in assays when five weeks old. Fast acquirers visited more artificial feeders after release. In contrast to most other studies, we showed that apparently 'poor' cognitive performance (slow reversal speed suggesting low behavioural flexibility) correlates with fitness benefits in at least some circumstances. This correlation suggests a novel mechanism by which continued exaggeration of cognitive abilities may be constrained.This article is part of the theme issue 'Causes and consequences of individual differences in cognitive abilities'.

Individual differences in performances on cognitive tasks have been found to differ according to social rank across multiple species. However, it is not clear whether an individual's cognitive performance is flexible and the result of their current social rank, modulated by social interactions (social state dependent hypothesis), or if it is determined prior to the formation of the social hierarchy and indeed influences an individual's rank (prior attributes hypothesis). We separated these two hypotheses by measuring learning performance of male pheasants, Phasianus colchicus, on a spatial discrimination task as chicks and again as adults. We inferred adult male social rank from observing agonistic interactions while housed in captive multi-male multi-female groups. Learning performance of adult males was assayed after social rank had been standardised; by housing single males with two or four females. We predicted that if cognitive abilities determine social rank formation we would observe: consistency between chick and adult performances on the cognitive task and chick performance would predict adult social rank. We found that learning performances were consistent from chicks to adults for task accuracy, but not for speed of learning and chick learning performances were not related to adult social rank. Therefore, we could not support the prior attributes hypothesis of cognitive abilities aiding social rank formation. Instead, we found that individual differences in learning performances of adults were predicted by the number of females a male was housed with; males housed with four females had higher levels of learning performance than males housed with two females; and their most recent recording of captive social rank, even though learning performance was assayed while males were in a standardized, non-competitive environment. This does not support the hypothesis that direct social pressures are causing the inter-individual variation in learning performances that we observe. Instead, our results suggest that there may be carry-over effects of aggressive social interactions on learning performance. Consequently, whether early life spatial learning performances influence social rank is unclear but these performances are modulated by the current social environment and a male's most recent social rank.

The name of Christine E. Beardsworth was incorrectly spelled in the original version of this article. The original article was corrected.

The playful (i.e. not overtly functional) combination of objects is considered a potential ontogenetic and phylogenetic precursor of technical problem solving abilities, as it may lead to affordance learning and honing of mechanical skills. We compared such activities in 6 avian species: 3 psittaciforms (black-headed caiques, red-shouldered macaws, and Goffin cockatoos) and 3 corvids (New Caledonian crows, ravens, and jackdaws). Differences in the type and frequency of object combinations were consistent with species' ecology. Object caching was found predominately in common ravens, which frequently cache food. The most intrinsically structured object combinations were found in New Caledonian crows and Goffin cockatoos, which both stand out for their problem solving abilities in physical tasks. Object insertions prevailed in New Caledonian crows that naturally extract food using tools. Our results support the idea that playful manipulations of inedible objects are linked to physical cognition and problem-solving abilities.

Renewed interest in the field of comparative cognition over the past 30 years has led to a renaissance in our thinking of how cognition evolved. Here, we review historical and comparative approaches to the study of psychological evolution, focusing on cognitive differences based on evolutionary divergence, but also cognitive similarities based on evolutionary convergence. Both approaches have contributed to major theories of cognitive evolution in humans and non-human animals. As a result, not only have we furthered our understanding of the evolution of the human mind and its unique attributes, but we have also identified complex cognitive capacities in a few large-brained species, evolved from solving social and ecological challenges requiring a flexible mind. © 2011 John Wiley and Sons, Ltd.

We quantified cache spacing patterns, cache retrieval rates and rates of cache theft in New Zealand robins, Petroica australis. In a field experiment we presented wild birds with a superabundant supply of mealworms. Trials were conducted on competitively dominant males and subordinate females, both when birds were alone and when they were accompanied by their mate. We hypothesized that (1) dominant males would aggregate caches to facilitate their defence, whereas subordinate females would scatter their caches more widely to avoid their discovery by males, (2) sexual differences in cache spacing would be context dependent, or occur only when birds were in pairs, and (3) patterns in cache spacing would facilitate cache retrieval and reduce cache theft. Females cached food further from food sources than males when they occurred in pairs. However, both sexes cached at similar distances when they were alone. Regardless of social context, females created more cache sites than males, and both sexes created more cache sites when they were together. Cache theft was frequently observed. Female-made caches were retrieved at similar rates by both males and females whereas male-made caches were more likely to be stolen by females than retrieved by the males that made them. Therefore, the results showed no support for the hypothesis that cache spacing patterns reduce cache theft. We suggest that the apparently conflicting relations between cache spacing, recovery and theft are determined by a trade-off between winter survival tactics and mate provisioning. © 2007 the Association for the Study of Animal Behaviour.

We presented monogamous pairs of New Zealand robins Petroica australis with a superabundant supply of artificial prey and then observed birds re-cache previously hoarded prey. Males and females frequently relocated and redistributed previously hoarded prey. Both sexes also re-cached prey that was originally hoarded by their mate. However, re-caching rates of retrieved and pilfered prey differed between sexes. Female-made caches were re-cached at the same rate by both females and pilfering males, while male-made caches were re-cached more frequently by pilfering females. The re-caching of previously hoarded prey is consistent with the hypothesis that re-caching reduces theft. However, re-caching pilfered prey suggests that it might also play a role in cache theft strategies. Higher rates of re-caching by pilfering females cannot be explained by current hypotheses used to account avian re-caching behaviours. We suggest two new hypotheses, sexual dimorphism in spatial memory and indirect mate provisioning, which may help to explain the evolution of re-caching in New Zealand robins. © Journal of Avian Biology.

We conducted acoustic and behavioural observations on wild New Zealand North Island kaka (Nestor meridionalis septentrionalis) to assess the behavioural context of their most common calls. We distinguished several call types by ear in the field and then quantitatively evaluated our call type classifications using spectrographic analyses. Next, we established the behavioural context of each call type during 500 h of field observations. We observed five distinctive call types that were clearly segregated in subsequent spectrographic analyses. Behavioural observations showed that each call type was generally associated with particular behaviours used by birds separated by different distances. Some call types were used by distantly‐separated solitary birds that were foraging or preening, while others were used mostly during copulation. Overall results indicate that kaka have a range of distinctive call‐types for communication under different spatial and social circumstances. © the Royal Society of New Zealand 2007.

We experimentally evaluated the food hoarding behaviour of North Island robins (Petroica australis longipes) at Karori Wildlife Sanctuary, Wellington. Mealworms were offered to free-ranging pairs of male and female robins to evaluate whether their food hoarding behaviour was similar to previous observations of South Island robins. We also tested theoretical predictions derived in the Northern Hemisphere, which argue that competitively subordinate birds should hoard more food than dominant birds. Results showed that the food hoarding behaviour of North Island robins was similar to South Island robins, except that North Island robins repeatedly used the same cache sites, which is rare in South Island robins. Data did not support the prediction that competitively subordinate birds hoard more food than dominant birds. Males acquired most of the mealworms offered to birds during trials, and won nearly all aggressive interactions observed between sexes. Therefore, males appeared to be competitively dominant to females in winter. However, males stored over five times as many mealworms as females, which is opposite to theoretical predictions. We interpret the reluctance of females to cache food as a strategy to avoid food loss to competitively dominant males. © the Ornithological Society of New Zealand, Inc. 2005.