Can we use anthropomorphic language in animal behaviour research?

A few months ago during the coffee break at an animal behaviour conference I was talking to a colleague about her research when she told me that to suggest that animals could feel fear or be afraid was anthropomorphism, the mistake of assigning of uniquely human characteristics to other animals. This view is not at all uncommon among practicing scientists and the term anthropomorphism is often extended to include a whole range of behavioural traits and emotions such as impatience, joy, expectation, boredom, anger, happiness or sadness, and yet there is good evidence that these emotions are not unique to humans. For example, dogs have been shown to exhibit jealousy, elephants have empathy, and Capuchin monkeys get visibly angry when treated unfairly as this video shows.

The possibility than animals can think for themselves is also often questioned and yet we know that some animals such as chimps and dolphins have a sense of self. There are also examples of animal behaviours that surely require some degree intelligence and forward planning such as innovative tool use by chimps and deception by ravens.

Deep in thought. Do chimps think like we do?
Deep in thought. Do chimps think like we do?

One of the main arguments against using anthropomorphic language to describe animal behaviours is that there is no way to know how an animal is really feeling, we can only describe what it looks like it’s feeling but not what’s actually happening inside it’s head. But the same is also true of humans, yet no one would question the use of anthropomorphic language to describe human behaviours.

If a person says they are excited we don’t actually know that what they feel as excitement is the same as what you or I feel as excitement, to them it may be a very different thing. All we can do is observe how that person acts and behaves and decide for ourselves if that matches up with our interpretation of what excitement is. The same is true of any emotion. If I say I am or happy or bored how could you tell that what I feel as happiness or boredom is the same as what you or anyone else feels as those things? At some point we have to use our subjective judgement to decide how a person is feeling. Do their actions match our expectations for a happy person? Then we can say they are happy. Do they behave as if they are sad? Then we can say they are sad. If this applies to humans then surely it can also apply to animals, at least in some cases.

I am not suggesting that we should abandon all caution and start using anthropomorphic terms carelessly. What I am suggesting is that so long as we clearly define our terms we should be able to use words like ‘afraid’ or ‘excited’ to describe animal behaviours when those terms well match what we see. If I want to describe anger in animals I should be able to use the word anger so long as I clearly state what I mean by the term and the behaviour that I am describing well matches what most of us would recognise as anger.

Of course, there are cases where using anthropomorphic terms really isn’t appropriate. If I read a paper that described ‘angry’ aphids or ‘jealous’ earthworms I would be dubious that those animals really could feel those emotions. In other cases I would have much less of a problem. Can chimps get angry? I think so. Are rats afraid of predators? It seems likely.

For many scientists, including people who I work with, the fear of using anthropomorphic language seems deeply ingrained and I think this affects how we view the behaviour of animals. Not all animals are mindless automatons that blindly follow their pre-programmed instincts. Many animals, especially among the vertebrates, have complex behaviours and emotions which are best described using the same terms we use for those things in human animals. I think it is time we started describing animals behaviours exactly as we see them. We must define our terms and we must be clear but so long as we are there should be little problem to this approach.

For those that still doubt that animals have can have thoughts and emotions like ours I recommend this TED talk.

For another view I also strongly recommend this post by Jilly at her blog fluffysciences

What do you think? If you have an opinion please leave a comment below.




Great tits hunt for Pipistrelle bats
When under strong ecological pressure, or when a good opportunity arises, animals have often shown themselves to be surprisingly innovative in how they adapt to new pressures or take advantage of new resources. Many examples of this have been observed in the wild including the discovery of tool use by chimpanzees, problem solving in guppies and the development of a novel ‘body-slapping’ behaviour as a means of communication in grey seals. No behaviour has surprised me more however than the discovery that in Hungary a population of a small seed-eating song bird, the great tit (Parus major), has switched from its staple diet of seeds and insects and has learnt to search for, kill and eat hibernating bats (Pipistrellus pipistrellus).

At around five inches long great tits are small birds, but pipistrelle bats are even smaller at just an inch in size. During the winter these bats hibernate in cracks and crevices in dark caves or old buildings where they are safe and well hidden, but when they awaken they start making noises which draws the attention of nearby predators, including great tits.

A Great tit (Parus major). Not a typical carnivore.
A Great tit (Parus major). Not a typical carnivore.

The earliest suggestion that great tits might hunt for bats goes back to at least 1947 when a Swedish biologist named Olaf Ryberg observed dead bats in Sweden with “injury, caused e.g. by titmice (possibly also bigger birds)“. It was to be almost half a century before the subject was raised again when in 1996 a great tit was seen feeding on a dead bat in a cave in Poland. Three years later at the same site in Poland three more bats were found, one dead and two alive, with injuries which looked like they were caused by tit beaks. Despite these observations it was still not clear that in any of these cases great tits were actually hunting for bats actively and it remained a possibility that they were simply scavenging on bats which had already died. A chance observation of a great tit capturing a live pipistrelle in a cave in Hungary in 1996 provided the only evidence at this point that great tits ever actively preyed on live bats.

That first observation was made by Péter Estók from Germany’s Max Planck Institute for Ornithology and intrigued by what he had seen he and his research team returned to the cave in Hungary on three separate occasions from 2004 to 2009. Using experiments and old-fashioned observation  they aimed to discover whether feeding on bats by great tits was simply opportunistic, or whether great tits had learnt to deliberately and systematically hunt for and feed on pipistrelles.

A common pipistrelle bat (Pipistrellus pipistrellus). Image from
A common pipistrelle bat (Pipistrellus pipistrellus). Image from

The research team quickly found their answer. During the first winter of observations they witnessed great tits capture and consume live bats seventeen times in just ten days. Yet despite this it was still not known why this behaviour had developed in the first place.

One possibility was that great tits used bats as a last-ditch food source when their regular food was in short supply. To test this possibility the researchers left a mixture of sunflower seeds and bacon in feeders around the cave entrance to provide an easy and irresistible meal for any passing great tits. Sure enough, when plentiful food was provided they found that hunting for bats by great tits stopped almost completely with only one case observed over a ten-day period. This provided good evidence that feeding on bats was driven by an urgent need for food and did not represent a more general shift in diet.

Now just one question remained to be answered. How do great tits find the bats in the first place? It was thought that they might be able to home in on the bat’s calls so to test this possibility Estók recorded the bats and played their calls back to great tits from a speaker. Around 80% of the birds reacted strongly to the sounds often turning their heads towards the speaker and approaching to investigate. This was particularly interesting because in one study bat calls were shown to act as a deterrent to mammalian predators, possibly by signalling that the bats are awake and cannot be caught. For great tits however it seems that bat calls are far from a deterrent, possibly because they can easily outmanoeuvre a bat in flight.

Eight years passed between the first observation of a great tit preying on a live bat and the start of Estók’s study. Given that the typical lifespan of great tits is three years the birds observed in 2004 couldn’t possibly have been the same birds that were seen in 1996. This raises the fascinating possibility that the bat killing behaviour is passed from one generation to the next by some form of cultural transmission. Whether this is or is not the case is not yet known and so it seems there is still much to learn about the unassuming great tit.



For the study of great tits hunting bats

Estók P, Zsebok S, & Siemers BM (2010). Great tits search for, capture, kill and eat hibernating bats. Biology letters, 6 (1), 59-62 PMID: 19740892

Bat calls as a deterrent to mammalian predators

Martin, K., & Fenton, M. (1978). A possible defensive function for calls given by bats (Myotis lucifugus) arousing from torpor
Canadian Journal of Zoology, 56 (6), 1430-1432 DOI: 10.1139/z78-196

Innovative behaviour in other animals

Body slapping seals
Bishop, A., Lidstone-Scott, R., Pomeroy, P., & Twiss, S. (2013). Body slap: An innovative aggressive display by breeding male gray seals (Halichoerus grypusMarine Mammal Science DOI: 10.1111/mms.12059

Problem solving guppies
Laland KN, & Reader SM (1999). Foraging innovation in the guppy. Animal behaviour, 57 (2), 331-340 PMID: 10049472

Tool use in chimpanzees
Goodall, J. (1964). Tool-Using and Aimed Throwing in a Community of Free-Living Chimpanzees Nature, 201 (4926), 1264-1266 DOI: 10.1038/2011264a0

Awesome orcas

It has been known for a long time that whales and dolphins are incredibly intelligent animals but it’s not often we see that intelligence so impressively displayed as when orcas (often called killer whales) hunt. Orcas can actually be divided into several different ‘types’ which are found in different areas of the world and often specialise in hunting different prey. Some, such as those around Norway and Greenland, are particularly adept at hunting herring and follow the fishes migration path. Others, such as those in the north-east Pacific are skilled salmon hunters, and some have even learnt to strip tuna fish from fisherman’s lines. There is one group however that outclasses them all, the orcas of the Antarctic peninsula have become specialised at taking seals from floating ice and the way they do it is simply breathtaking. Ingrid Visser and her colleagues were lucky enough to observe the attack in 2006 and described it like this

…one killer whale remained in position with its rostrum against the ice floe while four killer whales moved away from the ice floe with the seal on it. These four killer whales reappeared simultaneously, approximately 20 seconds later in line-abreast with all submerged just under the surface. All four were coordinated-swimming, with their left sides orientated towards the surface. A trail of bubbles emanated from each of the animals blowholes as they accelerated and passed directly under the ice floe, two on each side of the stationary killer whale. This generated a large wave, which tipped the ice floe initially towards the wave, then as the wave poured over and crested under the ice, it pivoted and tilted the ice in the other direction where the attacking whales were now waiting. The breaking wave washed the seal into the water…

Coordinated orcas about to launch a "wave washing" attack.
Coordinated orcas about to launch a “wave washing” attack.

This same hunting technique, sometimes termed “wave-washing”, was later filmed by the BBC for the series Frozen Planet (highly recommended if you haven’t yet seen it).

As someone who has worked with seals a lot over the last few years I have mixed feelings about this. On the one hand the seal is clearly distressed and is tormented for a very long time before it is finally killed, but on the other I can’t fail to be impressed by the skill and intelligence of the orcas that is required to pull off an attack like this. For this hunting strategy to be succesful there must be forward planning, and a high level of communication and coordination between individual orcas. These characteristics are not often associated with animals.

What is really interesting is that in the case described by Visser and her colleagues the seal was caught after around 15 minutes but then released and allowed back onto the ice. It then had to endure a second wave-washing attack before being finally killed almost 15 minutes later. Why did the orcas not kill and eat the seal immediately? The answer is not known, it could simply be play behaviour or, more interestingly, it may be that the adults are training their young to hunt. We clearly have a lot more to learn from these amazing animals and I expect there will be many more discoveries in the future.


For a detailed description of this behaviour see:

Visser I.N., Smith T.G., Bullock I.D., Green G.D., Carlsson O.G.L. & Imberti S. (2008). Antarctic peninsula killer whales (Orcinus orca) hunt seals and a penguin on floating ice, Marine Mammal Science, 24 (1) 225-234. DOI:

My poster for Behaviour 2013

I’m going to be spending the next few days in Newcastle at Behaviour 2013, a conference billed as the largest gathering of researchers working in the field of animal behaviour this year. Today we had talks on assassin bugs that actually attack and eat spiders on their webs, to bats that home in on the call of their unfortunate prey, the túngara frog. Tomorrow it all starts again bright and early with talks on everything from avian cognition to social learning and parental care. It’s so rare to be in a place where everyone shares my enthusiasm for animal behaviour, it’s great!

I’m not just going to listen to talks though, oh no,  I’m also presenting a poster showing some of what I’ve been working on for the last year. For those of you that won’t be there I thought I’d put the poster up here, I’ve received some good feedback on this so far but if you have any comments I’d love to hear them.

Click here for a high-res pdf version.

ASAB Behaviour 2013 poster


The enigmatic fossa and the evolution of sociality

Madagascar is a land filled with wonderful creatures, from charismatic lemurs to chameleons so small they can stand on the tip of a pencil. It is also home to the fossa, Cryptoprocta ferox, a carnivorous cat-like mammal endemic to the island and thought to be closely related to mongooses (although this is contested). Unusually for a mammal, male fossas sometimes form long-term social groups of two, or occasionally three, individuals that work as a team when hunting allowing them to take down big prey such as some of the larger lemurs. Sociality clearly provides animals with major benefits, not only when hunting but also in terms of decreased predation risk and an improved ability to defend resources such as territory, food supplies or mates. However, there are also significant costs such as increased competition for food when supplies are low or for mates when mating opportunities are limited.  In most male mammals it seems that the costs of sociality outweigh the benefits. Although it is known from a few other species such as cheetahs and kinkajous, for male mammals to form long-lasting social bonds is extremely unusual, especially when the females are solitary, as is the case with the fossa.


As a general rule female social groups occur in response to the distribution of food in the environment, while male social groups form in response to the distribution of females. When food sources are clumped females tend to aggregate at feeding areas while males form groups around them. In contrast, when food is evenly distributed both sexes are expected to lead mostly solitary lifestyles coming together only to reproduce. The fossa then is a bit of an oddball. It is the largest carnivore in Madagascar preying on small to medium-sized animals, its food supply is evenly distributed and, as theory predicts, the females live solitary lives. Why then do some males go against theory and form social groups? It may be that more sociable males benefit not only from cooperative hunting but also from increased mating success.

The fossa’s mating system is unique among mammals. During the breeding season large numbers of males gather at traditional mating trees which are occupied by small groups of up to three females. This leads to fierce contest competition among the males for access to mates and reportedly intense sperm competition. Those males that are able to monopolise access to females and gain the longest copulations are likely to reap the benefits of increased reproductive success. It may be that those males that form social groups share access to females and so gain a reproductive advantage over their solitary counterparts. However, intense sperm competition and typically small litter sizes may also impose high fitness costs to individuals belonging to groups of unrelated males thereby constraining the evolution of sociality.

This image from Arkive shows the fossa mating tree.
This image from Arkive shows the fossa mating tree.

The evolution of group-living from solitary ancestors has long been a hot topic in behavioural ecology. The unusual social behaviour of male fossas makes them an ideal study species for testing predictions about why sociality arose in the first place. Answering the question of why some male fossas go it alone while others form associations is the subject of recently published research by Mia-Lana Lührs and her colleagues.

Lührs and her fellow researchers were able to track and gather data on 22 wild male fossas between 2007 and 2010. They began by capturing individuals using live-traps and taking DNA, tissue and hair samples together with data on body mass and length, testis volume, and canine width. Some of these males were also fitted with GPS-accelerometer collars before release.

Since social groups of males were expected to move and hunt in synchrony, accelerometers were used to record when, and how vigorously, individuals were moving. This allowed the researchers to determine when their tracked males were moving and hunting and if they were hunting as a group or alone. In order to determine male mating success seven females were observed during the breeding seasons between 2007 and 2010 and the number and duration of copulations by known males recorded.

The results of this study are fascinating. Male fossas were found to exhibit two distinct phenotypes which differ according to social organisation. Those males that formed associations were found to be 38% heavier and 13% larger than solitary males, most likely due to differences in diet and an increased foraging efficiency. Dietary analyses revealed that the diet of social males contains larger and more agile bodied prey than solitary males and this has been linked to cooperative hunting. Data from associated males fitted with accelerometer tags shows that their movements, including movements typically associated with hunting, were highly synchronised, supporting the view that associated males hunt cooperatively. These results suggest that the benefits of hunting in a group may more than compensate for the costs of increased competition for scarce food resources. The physical superiority of social males may also grant them reproductive advantages as they are able to out-compete solitary males for access to females.

This boxplot shows the proportions of major mammal species in the diet of solitary (grey bars) and social (black bars) males. Prey species are sorted by size from left to right, largest to smallest. You can clearly see that the diets of the two phenotypes differ with social males consuming far more large bodied prey than solitary individuals.
These boxplots shows the proportions of major mammal species in the diet of solitary (grey bars) and social (black bars) males. Prey species are sorted by size from left to right, largest to smallest. You can clearly see that the diets of the two phenotypes differ with social males consuming far more large bodied prey than solitary individuals.

Since males appear to benefit from sociality it is worth considering why females of this species do not also form associations. Lührs and her colleagues propose that the energetic costs of competition for food when communally raising offspring are too high, for female fossas it just does not pay to be social. This may also explain why males usually only associate in pairs and, when groups of three have been observed, one member is usually in very poor condition. There is a fine balance between the benefits of cooperation and the costs of increased competition, two’s company but three’s a crowd.

The increased competition that comes with sociality not only applies to food but also to mating opportunities. This raises the question of how associated males share these limited opportunities to reproduce. A likely answer is that the costs of competition are accounted for by inclusive fitness benefits. This does appear to be the case. In four out of six associations the males were found to be likely litter mates, presumably brothers. Furthermore, the only unrelated male associates in this study were found to share mating opportunities. This suggests that forming associations provides males with significant benefits whether they are related or not.

This raises one final question, if forming associated pairs is so beneficial to males then why are there still any solitary males at all? Why hasn’t natural selection removed solitary behaviour from the population? One possibility is that the solitary males that were observed did not have any litter mates and so are constrained in their opportunities to associate with a familiar male. Put simply, they are solitary because they had no choice. No male littermates were found for any of the solitary males observed in this study, supporting this conclusion. Another possibility is that solitary males represent an alternative life strategy maintained by frequency dependent selection. Lührs and her colleagues suggest that the small size of solitary males may allow them to extract small lemurs from tree-holes or rodents from burrows. This seems unlikely since, when it comes to competition for mates, solitary males are at a significant disadvantage. Interestingly, Lührs and her colleagues note that in some species inferior males invest relatively more in sperm competition by producing more sperm. In fossas however this is not the case and this suggests that, for fossas, going solo is not an alternative strategy. At present the view that solitary males occur due to a lack of littermates with which to associate seems the most likely.


claimtoken-51352dc0d52f9Luhrs M.L., Dammhahn M. & Kappeler P. (2012). Strength in numbers: males in a carnivore grow bigger when they associate and hunt cooperatively, Behavioral Ecology, 24 (1) 21-28. DOI:

Highlights from the field season 2012

I’ve spent most of the last two months sitting in a small wooden box in the middle of a grey seal colony on the east coast of England. It wasn’t just for fun though, this was part of my research into the links between ‘personality’ and mate choice in female grey seals (you can read about that here). The data for my research comes from behavioural observations of grey seals in the wild. After 320+ hours of observation in the field I think I’ve seen almost everything that happens on a grey seal colony, these are some of the highlights…

As you might expect the weather in November and December can be very cold, we experienced everything from rain, hail and snow to howling gales to bright sunshine. The weather might not always have been welcome but it did make for some nice photographs.

Sunrise over the beach
Sunrise at 7am

We started each day at 6.00am, the mornings were hard but when the weather was good we were rewarded with sunrises like this.

The rainbow after the storm
The rainbow after the storm

The weather on the day this photo was taken was really bad, strong winds and heavy rain from morning till night. This rainbow appeared during a brief window of sunshine at about midday and made an otherwise bad day in the field worthwhile.

Fights were a regular occurrence on the colony, females would fight off the unwanted advances of the ever-present males or fend off other females if they got too close.

Two female grey seals fighting
Two female grey seals fighting

Some of the most memorable fights were between two males competing for access to females.

This photo was taken just before the fight began. The open mouth is a threat.
This photo was taken just before the fight began. Grey seals open their mouths and bare their teeth as a threat behaviour.

This video filmed at the start of the season shows just how aggressive males can be, and this fight was relatively tame!

There were of course lots of opportunities for photographing cute pups, these are some of my favourite photos.

Posing for the camera
Posing for the camera
Chilling out
Chilling out
Grey seal pup in the grass
Grey seal pup in the grass

Grey seal births can happen very quickly and so they are easy to miss. We were very lucky then to see this one up close, the female came right up to the hide before giving birth in front of the video camera!

Those are favourite moments from this season. I’ll be writing up more detailed posts about my research over the next year, in the meantime my lab group has a blog at that you might be interested in.

Gouldian finches’ head colour reflects their personality

Gouldian finches, Erythrura gouldiae, are an extraordinarily colourful species of passerine bird endemic to subtropical woodlands of northern Australia. Both sexes are brightly coloured with red, green, black, yellow, red and purple markings, it is for this reason that they are also sometimes known as rainbow finches.

Just look at those colours! Note the different yellow, red and black head colouration.

In the wild the birds exhibit two main head colour morphs, black and red. There is also a rare yellow colour morph as shown in the image above. Interestingly, studies of captive birds have shown that males with red heads are on average more aggressive than those with black heads and that females have a preference for red-headed over black-headed individuals. Red headed males were also found to have higher levels of testosterone and corticosterone than black headed males when faced with socially challenging situations.

What this suggests is that behavioural characteristics, such as aggression and other traits, may be correlated with particular head colour morphs meaning that head colour is indicative of different personality types. This idea has been tested in a new paper by Leah Williams and her colleagues.

In order to determine if head colour really does indicate personality traits in Gouldian finches Williams and her colleagues tested a number of predictions. First they looked at pairs of black-headed birds which were expected to show less aggression towards each other than pairs of red-headed birds, this makes sense since red-headed birds had previously been found to exhibit higher levels of aggression.

The second prediction was that red-headed birds should be bolder, more explorative and take more risks than black-headed birds. This hypothesis is based on previous studies of other species that have shown a correlation between aggression and these behavioural characteristics. However, there is another possibility, red-headed birds could take fewer risks for two reasons; first, they may be more conspicuous to predators due to their bright colouration and second, it may pay black headed birds to take more risks and be more explorative so they find food resources before the dominant red-headed birds do.

In order to test the first prediction paired birds of matching head colour were moved into an experimental cage without food. After one hour of food deprivation a feeder was placed into the corner of the cage where there was only enough room for one bird to feed at a time. aggressive interactions such as threat displays and displacements were then counted over a 30 minute period.

The results as shown in the figure below were striking. Red-headed birds were significantly and consistently more aggressive than black-headed birds.

Figure shows the mean (+SE) number of aggressive interactions by individuals in relation to their head colour.

To test the birds willingness to take risks they were deprived of food for one hour before their feeder was replaced. After the birds had calmly begun to feed a silhouette of an avian predator was moved up and down in front of the cage to scare the birds from the feeder. The time it took for them to return to the feeder was taken as a measure of their willingness to take risks, birds that returned quickly were considered to be greater risk takers than those that were more cautious.

This time the results were surprising. Red-headed birds were considerably more cautious than those with black heads at returning to the feeder after a “predator” had been introduced. As the figure below shows they took on average 4x longer to begin feeding again than the less aggressive black-headed birds.

Figure shows the mean (+SE) time taken for birds to return to their feeder after a “predator” was introduced.

Finally, the authors investigated the birds interest in novel objects or “object neophilia” which is defined in the paper as “exploration in which investigation is elicited by an object’s novelty“. To do this a bunch of threads was placed on a perch within the cage, the time taken for the birds to approach the threads within one body length and to touch them were recorded over a one hour period. In line with the results from the risk taking experiment it was found that the aggressive red-headed birds showed less interest in novel objects than did black-headed birds. The difference is not so striking as the previous experiments but was statistically significant nonetheless.

Figure shows the mean (+SE) time taken for birds to approach a novel object relative to their head colour.

These experiments were repeated after a two month interval and showed that different birds differed in their responses but the responses of individual birds were consistent over time. Head colour was found to predict the behavioural responses of the birds. Red-headed birds were more aggressive than black-headed birds but took fewer risks and were not explorative.

What is surprising about these results is that aggression does not correlate with risk taking behaviour, however, the authors do provide a convincing explanation, suggesting that…

…red coloration has been found to be conspicuous against natural backgrounds, and more conspicuous birds have been found to suffer higher predation rates. Thus, selection could favour more conspicuous red-headed birds taking fewer risks.

Interestingly boldness and risk taking behaviours were found to be strongly correlated, regardless of head colour they always occurred together forming a “behavioural syndrome”. This implies that there is selection in favour of specific combinations of traits and of head colour in relation to those traits. Selection favours aggression in red-headed birds and the boldness/risk taking behavioural syndrome in black-headed birds. This makes sense when you consider the high risk of predation faced by red-headed birds if they take too many risks and the need for black-headed birds to find food away from the dominant red heads which occupy the safest foraging locations.

Williams and her colleagues suggest that if red-headed birds are aggressive, and black-headed birds take more risks, this could lead to differences in foraging tactics. For example, black headed birds could increase their foraging opportunities by feeding at more risky sites away from interference by the dominant red-headed birds which feed in safer locations. The lower conspicuousness of their black heads means they are at less risk of predation at exposed sites that red-headed birds would be.

The results of this fascinating study strongly support the hypothesis that head colour does indeed signal personality in Gouldian finches. I would love to see some more research in this area. The authors themselves suggest that more research is needed to find out what roles head colours play in social situations. It would also be interesting to find out how widespread this phenomenon is, given that birds frequently use plumage colouration as signals it seems likely to me that colour may indicate personality in other avian species.


Williams L.J., King A.J. & Mettke-Hofmann C. (2012). Colourful characters: head colour reflects personality in a social bird, the Gouldian finch, Erythrura gouldiae, Animal Behaviour, 84 (1) 159-165. DOI: