2
Intelligence is not in itself an easily measurable quality because it consists of so many different manifestations. We know at least that birds have some substantial advantages over mammals. They mature earlier and produce more offspring than most mammals over a lifetime and generally do so more efficiently than mammals. They have superior energy efficiency and adapt faster to change than mammals and readily become colonisers of new terrain because of their locomotion (either fast on the ground or as able flyers). Indeed, Isler and van Schaik (2009a) asked recently, why are there so few smart animals but so many smart birds? Murray and Vickers-Rich (2004) had already suggested some years earlier that birds are the ‘rapid response unit’ of vertebrate evolution.
However, such adaptive advantages by themselves do not indicate the existence of ‘intelligence’. The horseshoe crab has survived for 400 million years without much change in design, and its survival, while remarkable, would not make one think of this being the result of its intelligence. This book searches for clues of cognitive complexity in the behaviour seen, described and documented about native Australian birds and will do so bearing in mind many of the pitfalls that have been debated in the last decades, some of which are outlined below.
The problem with brain size as a measure of intelligence
To establish whether an organism is ‘intelligent’ as a crude measurement, it was thought that brain size mattered most. Brain size increase has been touted as the major evolutionary achievement in vertebrates, including birds. When compared with mammals, birds have generally very small brains and have not developed the convoluted parts on the brain, called the neocortex, which all mammals have (Fig. 2.1).
Hence, birds were believed to be incapable of thinking and even the idea of testing complex cognitive ability in birds seemed pointless, if not absurd. Birds, more so than mammals, thus seemed to justify the theoretical model that the French philosopher René Descartes had proposed centuries earlier. It is this: animals act purely on instinct and deal with daily life purely in response to immediate and visible or perceivable stimuli. They have no feelings (either physical or mental), no thoughts, no minds of their own and, importantly, because they are automatons, we can do with them as we please. And this belief has stayed with us and with science well into the middle of the 20th century. Kleiner (2002), reviewing a book by Steven Wise Drawing the Line, begins by saying that his high school biology classes still taught that animals are mere instinct driven automatons, and that notions of intelligence, reasoning and even emotions were a mark of poor science and of anthropomorphising. It is thus in living memory that Descartes’ views have found a continued reflection in everyday attitudes. Evidence presented in this book will show that these assumptions are completely incorrect.
Fig. 2.1. Images of brains for comparison: 1. Zebra finch; 2. Australian magpie; 3. Human; 4. Comparison (dorsal view) between magpie brain (left) and that of young chicken (right). Scale bars below brain images 1–3 indicate size of zebra finch brain relative to magpie brain and to human brain. In 2004, an Avian Consortium changed all the names of regions of the brain to make them comparable to the human brain and to be able to start comparing links and functions. They argued that dissimilarities had been stressed too much and it was now important to locate regions of the brain that may have similar functions across species. Indeed, some similarities and dissimilarities are obvious even just by showing the outside of the brains. A = cerebellum; B = spinal cord; C = cerebrum (site of forebrain). The convoluted area in the human brain is the neocortex, a specific mammalian development most pronounced in humans and not found in any other classes of animals. Measurements in mm. 4. (lower panel): Note the smoothness of the surface of the two halves of the brain and the pronounced dividing line between the two hemispheres in the middle; the halves are however joined (not visible here). In the 19th century it was still believed that brain size alone determined intelligence.
The problem with reliance on body/brain ratios
Although measuring brain size was unceremoniously dropped as a valid measure of intelligence, measurements of the brain continued but now changed from brain size to a relational measure of brain to body weight, rather than absolute size. This ratio was used because large bodies require larger brains simply to move about. The idea of relating brain size to body size as a relational value has continued to this day and permeates the literature on cognitive complexity in vertebrates. Scientists have gone to work on this aspect too and have found that some birds (Cnotka et al. 2008) have brain to body weight ratios overlapping with those of mammals and some of these even overlap with primates (Jerison 1973; Rogers 1997). On that scale, birds do not do badly but one wonders to what extent weight overall, even if related to size, is a valid measure.
Relative brain size in birds has been correlated with greater flexibility of behaviour (Lefebvre 2011), shown in abilities to colonise new areas, including urban environments, and to adjust the time of breeding to changing climate. Until recently, studies of brain size relative to body size and behaviour had been conducted only on birds of the northern hemisphere but a recent study has looked at these relationships in Australian species. Franklin et al. (2014) analysed data for over 500 species of Australian birds and found relationships almost identical to those found for avian species elsewhere. The researchers also found relative brain size was large in parrots, cockatoos and owls and small in emus, ducks, megapodes (Fig. 2.2) and quails, as well as grebes, swifts and swallows. This may not mean that, for example, swifts and swallows are less intelligent than other birds, but rather that their brain size is constrained by the need to be agile flyers for aerial capture of insects. Nevertheless, parrots and cockatoos have larger relative brain size compared with all other species and, among the psittacine group, budgerigars and palm cockatoos stand out as having exceptionally large brain to body weight ratios while that of the eclectus parrot is exceptionally small, although, as the researchers point out, these results could be biased as a result of deriving from small sample sizes (Franklin et al. 2014).
Fig. 2.2. The brush turkey, a megapode, provides no parental care for the hatched offspring. The young are on their own from the day they struggle free from the mound and they usually have little tolerance for each other. The bird shown here is a juvenile.
While it has been shown and a case can be made for direct correlations between, say, the size of the brain region called the HVC (it used to be called the ‘high vocal centre’) in the forebrain of songbirds and complexity in song (Nottebohm and Nottebohm 1976), and between the size of the hippocampus and spatial ability (Clayton and Krebs 1994), there has been no evidence to date that has convincingly demonstrated that whole brain weight in relation to body size can be easily linked to ‘intelligence’ in general (Healy and Rowe 2007). Indeed, even the correlation between food-storing species and hippocampal volume has not been consistently confirmed (Brodin and Lundborg 2003). Our fascination with measuring the brain has such a long history that it is perhaps difficult to let go.
One valid criticism is that the brain is treated as a unitary organ and, as a whole, this is a very crude measure of its capacity but of course it has many parts. A likely better measure might be number of neurons (nerve cells) versus glial cells, number of dendrites per neuron, or the number of synaptic connections per neuron, all of which are known to vary in different parts of the brain and across mammalian species (Braitenberg and Schuz 1998), or some other measure reflecting neural mechanisms. Sawaguchi (1992) found in mammals that the neocortex is larger in species that feed primarily on fruit than in species that feed predominantly on leaves. This may depend on the need to remember where, when and what fruit is ripening, as will be taken up in the next chapter. However, the neocortex is made up of many parts with different functions and, if one part is larger in one species than in another, does it necessarily follow that the organism as a whole is more cognitively complex? The answer is ‘no’, even though there may be correlations of specific behaviours with specific parts of the brain.
Moreover, it is a misconception to think that a particular behaviour may involve just one specific part of the brain. Quite often, a multitude of loci are involved – rendering neat comparisons at times difficult or even meaningless. Different methods of scaling behaviour produce different results when they are correlated with the size of the brain or a part of the brain (Deaner et al. 2000). Brain-imaging studies with live birds have now revealed distinct neuronal circuitry involved in a crow’s assessments of different types of danger (Marzluff et al. 2012; Cross et al. 2013).
Butler and Hodos (2005) also took on the debate of brain versus body weight and they did so in great detail in their book on comparative vertebrate neuroanatomy. First of all, they pointed out that if the brain to body weight ratio is to be taken seriously, it cannot be cited selectively for some species as may suit the writer. If there are correlations that are alleged to be meaningful, they need to be applicable to all vertebrate species and the entire phylogenetic scale has to be examined. And when this is actually done (i.e. when not just mammals and birds but also fishes, amphibians and reptiles are included in the tally), a few surprises occur. Ancient ray-finned fishes, thought of as lowly in brainpower, are comparable to reptiles, and cartilaginous fishes are even comparable to mammals and birds of equivalent body size.
They summarised their objections by a very simple, but effective, crosscheck. Repeating the assumption: if two species of the same body weight have different brain weights, then the one with the heavier brain is presumed to be more intelligent. Now turning the logic around (which should also work but, in this case, does not): if two species have the same brain weight but different body weights, then the one with the lesser body weight would be presumed to be the more intelligent. Obviously, this would make no sense at all.
There also remained a problem to be solved in the literature on human evolution. It became clear that human brains, while showing substantial increases in fossil finds at various points in human prehistory (Falk 1987; Holloway et al. 2004), eventually did not grow substantially beyond a certain size and yet many crucial cognitive innovations happened thereafter. The emphasis in comparative neuroanatomical studies has been to suggest that, in addition to an increase in size, human brain evolution must have been characterised by selective enlargement and reorganisation of specific cortical areas (Semendeferi et al. 2001, 2011). Of course, the development of subtle subcortical structures or structural changes (Barger et al. 2007, 2012) can only be inferred. It does not leave a fossil record, nor can these changes be judged by brain weight. These issues will be raised again and expanded throughout the book.
The problem of the neocortex as an index of complex cognition
In mammals with large brains, the neocortex expanded relative to the rest of the brain and the expansion of size of the neocortex occurred mainly along its surface rather than thickness, thereby causing fissures or crevices. Over the period of evolution from the earliest mammals to higher primates, the surface area of the neocortex increased more than 1000 fold without comparable increase in thickness. In fact, the neocortex was thought of as an indispensable precondition for higher cognition and, in humans, for language and consciousness (Bayer and Altman 1991; Eccles 1989; Innocenti and Kaas 1995; Krubitzer 1995).
I said before that birds were ‘lacking’ a neocortex. That was said advisedly, reflecting the prejudice and lack of knowledge in the field at the time. In fact, birds are not lacking anything but, in evolutionary terms, have been a highly successful class of animals. The idea that an organism could develop biologically in different ways and yet show some of the same capacities as another structural entity or reach the same result via different means (called convergent evolution) was a big conceptual jump and a move away from anthropocentrism.
There were several events that finally made the idea of the mindless animal, especially mindless birds, unsustainable. Charles Darwin had made a number of important assumptions that animals feel and think and need to be able to negotiate a complex environment, and many scientists set out to disprove him but instead found evidence in support of his views (Kaplan and Rogers 2004b).
From the mid-60s, investigations into birdsong eventually led to a complete mapping of the songbird brain and then, by the first decade of the 21st century, major changes were rife. First, neuroscientists and neuroethologists set out to rename each area of the brain in birds to be more comparable with naming of regions in the mammalian brain (Reiner et al. 2004; Jarvis et al. 2005). It was the most comprehensive and coordinated change made to avian brain nomenclature. These changes were based on known functions that had been researched for a number of decades and revealed similar functions as in the human brain. It was then admitted that, although the avian forebrain (now called the pallium) did not share the layered structure of the mammalian neocortex, it had many of the same functions, in particular those known to involve higher cognition. In a landmark publication called Neuroscience of Birdsong (Zeigler and Marler 2008), the point is made strongly that there are analogies between the neural structures in the avian and mammalian brain (i.e. bird brains have structures that may be equivalent to the neocortex of the mammalian brain), and such claims have since been strengthened by further research (Marzluff and Angell 2012). The extraordinary departure from beliefs held about the supremacy of humans, indeed of the primate line, places the capacities of the bird brain, including birdsong, in an entirely new light and this will be discussed in the following chapters in some detail.
Neuroscience has thus been instrumental in laying the foundations for research in cognitive behaviour of birds by identifying core regions and dynamic processes in which the avian forebrain is involved; and psychology and ethology have provided the tools for conducting rigorous experiments that can advance our understanding of how far and to what extent birds can ‘think’.
Second, from the point of view of sheer design efficiency, it was finally accepted that birds had to have small brains to remain aerodynamic and capable of flight. Conceptually, especially since the expansion of computer capabilities, we can now also accept that memory (i.e. even substantial amounts of information), can be stored in microscopically small spaces, hence the relatively small size of bird brains no longer constituted a conceptual problem in human debates.
Some have said that, from a design point of view, humans are the only species that can afford to carry such large heads relative to body size because of bipedalism (summarised by Falk 1992). We can balance our heads on a vertical and flexible spine and can therefore support a heavy head – animals that walk on four legs cannot have a head that pulls them down in front and animals that fly or fish under water must keep their bodies aerodynamically balanced.
In a way, the avian brain itself ironically provides the most important counter-argument against using relative brain size as an indicator of cognitive capacity. It is possible that birds process information differently and even more efficiently than do mammals. Hence, relative brain size alone does not appear to get near analysing the source of the cognitive ability of birds.
The problem of reductionism
There are claims made recently (Jønsson et al. 2012) that corvids ‘are among the few non-hominid organisms on Earth to be considered intelligent and well-known examples exist of several crow species having evolved innovative strategies and even use of tools in their search for food’. Such statements are risky (see Figs 2.3–2.5) and generalise from few known facts to all the rest of the tens of thousands of species of vertebrates of which we know next to nothing.
Fig. 2.3. (Left) The blue jay belongs to the select group of corvids that have been studied and are very familiar in the literature. (Right) The Australian currawong has substantial abilities but it has never been tested systematically for cognitive abilities. So far, we only have circumstantial evidence. Currawongs belong to Artamidae.
Fig. 2.4. (Left) The Australian magpie and (right) the European raven are worlds apart yet the two species may have very similar cognitive abilities. (Raven photo: Peter Wallack/CC-BY-SA-3.0, http://creativecommons.org/licenses/by-sa/3.0/legalcode.)
Fig. 2.5. The parrots of the world have always delighted humans for their stunning plumage, their playful nature, their ability to mimic human speech and form friendships with humans. Importantly, they represent a group of birds with the largest brain to body weight ratio and they are very much on par with the corvids. Some of the parrots have been experimentally tested in cognitive tasks. Noticeably, all of the ones so far tested are found in the southern hemisphere. A = the African Grey Parrot (Africa); B = Galah (Australia); C = Budgerigar (Australia); D = Palm Cockatoo (Australia/Papua New Guinea); E = Kea (New Zealand); F = Blue-and-yellow Macaw (South America). These images are not to scale). Most of those mentioned have performed in tests and were often equal to or even better than chimpanzees. Their size is obviously not as important (because the budgerigar is the smallest of all psittacine species and the kea is medium size) but they all have in common that they are very long lived and this life history criterion has been related to cognition.
A bias in favour of a few species may be misleading for several reasons. They imply a scala naturae, saying only some animals are worthwhile while others are less worthy of our attention, sympathy and welfare (moral obligation). Second, elevating some, and not other, species may one day prove to be false, because the number of species of birds specifically tested for their intelligence is extremely small. Third, potential for intelligence does not equal its expression. Motivation, opportunity and needs have a good deal to do with expressions of behaviour and, conversely, interventions such as stress, starvation, oppression or chronic fear may suppress behaviour, including intelligence. Importantly, biological organisms are dynamic and interact with their environment (Kaplan and Rogers 2004b). The brain is particularly receptive to change and its structural components may alter (shrink or grow) within a lifetime of an individual. To use the hippocampus, a part of the brain particularly involved in spatial memory, as an example (because this has been tested): it was found that birds that store or cache their food have a relatively larger hippocampal region of the brain than do non-storing birds, but this is the case only if the storing birds have an opportunity to store food. If they are kept under conditions preventing food storing, their hippocampus does not enlarge (Krebs et al. 1996). That is, the brain can allocate more resources to one area if needed. Fourth, temperament and individual differences are perhaps also important to consider because variability is an important aspect of biology.
Explaining big-brained birds in the context of life histories
Leaving aside the critique of these particular brain measures, there is an important biological contradiction about brain size that has fascinated scientists. It is one that most birds appear to have solved and many mammals have not.
Briefly, this contradiction is played out in two ways: on the one hand, having a bigger brain is said to be of substantial advantage, or substantial advantage leads to bigger brains. Sol et al. (2007) categorically declared in the title of their paper that ‘big-brained birds survive better in nature’. Sol and colleagues tested their hypothesis by choosing just two variables in birds: namely linking brain size (relative to body size) to mortality rates. They found a negative correlation between the two variables: the larger the brain the lower was the mortality rate. Hence, they appeared to have established in their study a clear and identifiable benefit in having a large brain.
Because many Australian birds are long lived (see Appendix 1), one ought to be able to infer that Australian birds in general might have large brains, or at least larger brains compared with body weight than other birds with shorter lifespans. It is a difficult argument to make – the highly complex species of octopuses, even the giant octopus, have very short lives (only 4 years) and yet seem to be cognitively complex. Some researchers have examined the brain size in octopuses and have so far found that it is a puzzle. Their central brain is small but integrates large amounts of visual and tactile information from its arms. This has so far resulted in the conclusion that the octopus brain may be organised very differently than anything we know or perhaps even that the central brain may not be the only locus for processing information (Zullo et al. 2009). It remains to be seen whether invertebrates will change the debate about brain size/weight and cognition in future. Their abilities have so impressed researchers that many welfare regulations for primates and other vertebrates have now been extended to the octopus or are under debate that this ought to happen (Fisher 2012). Fischer, from the veterinarian department at the University of Sydney argued that:
Octopus engage in complex social interactions, learn rapidly and display a high level of exploratory behaviours … Their self-monitoring, ability to recognise individuals and, more importantly, their adverse response to degraded conditions, all point towards a sense of consciousness (…).
Identical arguments can be made for birds, but the possibility that they may be conscious and sentient has not been forcefully argued and, I suspect, rarely considered possible.
Why bigger brained birds live longer may be partially explained by a theory called the cognitive buffer, useful both in food searches and in case of a threat by a novel predator, for instance. Being able to categorise an object (Aust and Huber 2006) and fit it into a particular group and/or be aware of potential risk is a matter for the cognitive domain. Some of that, depending on complexity, can also be innate, of course (Sewards and Sewards 2002). A ‘buffer’ is created by being able to think about a novel item and by having accurately categorised it as a risk of a certain type, the clever animal will be able to buy time, avoid the danger and live another day. Of course, maximum lifespan is not related to intelligence in land animals (including humans), as one of the longest-lived land animals, the Galapagos turtle with a maximum lifespan of nearly 200 years, clearly shows. Such an argument would be entirely untenable. However, within a species or related group of species, the potential for a long life is increased if an organism is innovative and flexible enough to meet environmental challenges more rapidly and effectively than if only standard responses are employed in novel situations. Evidence of the ability to show such responsive and innovative behaviour is referred to as cognitive complexity. Here, birds tend to do better than mammals. As I shall try to show in this book, Australian birds evolved to be responsive, innovative and flexible as a dictate of survival in a particularly harsh and unreliable environment in Australia.
Downside of a large brain
While the benefits of having a large brain are very persuasive, this is not as straightforward as it may seem, and herein lies the contradiction. Scientists have long since discovered that having a large brain (in relation to body weight) is actually an extravagant business. The larger the brain the smaller the maximum number of offspring can be produced because energy demands of building a larger brain results in a slowing down of reproduction and development. Biologically, the brain is the most energy-demanding organ in the body. Larger brains take longer to grow and thus require more time and more resources. Fewer offspring can be generated and thus reproduction is not only slowed but also fewer are produced. Hence a large brain, in a sense, is a frivolity unless the species or organism can find ways of reining in these biological costs. Reining in such costs means finding a counter measure or measures against a creeping extinction: that is, a process in which replenishment of offspring begins to fall below death rate. This is called ‘the expensive brain hypothesis’ (Isler and van Schaik 2009a).
It seems that the best known and now statistically confirmed way to offset growing a large brain and maintain viable population size at the same time is to get help with raising offspring so that the chance of their survival is maximised. Mammals do rather poorly at this. In approximately only 5% of mammals do parents jointly raise offspring and even fewer can secure additional care for their offspring from helpers for such things as providing food. Mammalian youngsters are of course suckled and that makes the relationship with their mother very exclusive for as long as the offspring are entirely dependent on this form of provisioning. Dogs have somehow ended up in a compromise solution: while the female nurtures the pups, the only help the male or the entire group can provide is protection and, in some instances, feeding the lactating mother so that she continues to have enough energy to continually nurse her litter. Once the suckling stage is over, however, all adult members of the group jointly help to provide food by an ingenious solution of bringing back food in their stomachs and regurgitating it for the pups (see Rogers and Kaplan 2001). In the marmoset, a New World monkey, the father helps by carrying the young and returning the offspring only for feeding. In such a way, the lactating female can forage freely and save the energy of carrying the youngsters as well, allowing marmosets to raise twins, sometimes even triplets (Yamamoto 1993). Overall, though, bi-parental support in raising the offspring is negligible in mammals compared with birds, which have raised joint parental care to a highly efficient level. The vast majority of Australian landbirds raise their offspring together as a pair and most pairs stay together not just for the season but for the long haul.
Apart from the pair, it may be necessary to draw on additional help possibly from offspring of last breeding seasons. As discussed in Chapter 1, siblings can be called upon to help at feeding even when they may still have problems in learning how to provision themselves. This may be the case in white-winged choughs (Heinsohn and Cockburn 1994). Overall, as already said, Australian birds show an unusually high number of cooperative behaviours and the link that has been perceived between the necessity for such cooperation for biological, rather than ecological, reasons has added an extra dimension to the question why some avian species may have larger brains than others.
In summary, a large brain may be better able to meet change and unpredictable events but, biologically, a large brain may well be too expensive to run and keep. However, on the basis of Cockburn’s summary tables of evidence of cooperative behaviour, we now have tangible evidence of a mechanism that may resolve the problem. Slower rate of development and slower reproduction (and reduced clutch size) reduces the problem while at the same time successful raising of any young may be bolstered by cooperative behaviour. If this hypothesis holds, then a long-held puzzle of helping behaviour of offspring not directly related seen in many Australian avian species might also be resolved. At the very least, the ‘expensive brain hypothesis’ clearly implies that cooperation may be a precondition for maintaining a large brain and what goes with it, namely slower development and slower reproduction and so might another important variable of life history: longevity.
Scholars of human brain evolution have rightly suggested that rates of brain growth and length of time needed for maturation are not always a good indication of brain size. Brain growth rates differ between species and slow growth rates do not necessarily indicate more brainpower and larger brain size. For instance, some small primates have small brains with very slow growth rates before reaching adult size. In some cases such delayed maturation may be a sign of starvation or at least of insufficient provisioning and has thus to do with resource allocation during development (Leigh 2012). These are insights that are important to bear in mind. For instance, is the extraordinarily slow process of maturation in white-winged choughs (over 4 years, Heinsohn 1991b), an indication of a large brain needing this time for maturation or is it a sign of resource deprivation and low intake of nutrients so that it requires extra time for the brain to mature (Heinsohn 1992; Heinsohn et al. 2000)? Unless comparative neuro-anatomical studies are undertaken at the species level, this question cannot be answered. Both hypotheses are distinct possibilities and need to be tested. We can only infer from observed behaviour and hypothesise that a certain accumulation of cognitive abilities in a species may tell us something about their overall ‘brainpower’.
Cognitive complexity
The question is how cognitive complexity can be measured. As a single entity, it cannot be measured. Cognitive abilities come in a number of discrete ‘packages’, in the sense that one animal may develop one skill and another a different skill or another shows two disparate traits. They need not be related and they may be processed in different parts of the brain. In other words, problem solving, social learning or imitation are now recognised as very different capacities that can be examined and, eventually, these separate capacities might be traced back in evolution.
We have long given up on the idea that evolution is a direct path from low to high complexity. Indeed, we now suspect that in mammals, birds, reptiles and even invertebrates (Pahl et al. 2013), entirely unrelated clades and species in different orders may have developed advanced cognitive abilities and skills. Thinking of hunting spiders that have flexible hunting techniques, the honeybee that is capable of forming memories and learning to navigate in complex mazes or the octopus that solves problems, one can see that under certain circumstances extraordinary cognitive abilities develop (Morrel 2013).
It seems, however, that complex cognitive abilities are more readily found among altricial rather than precocial species. Altricial species tend to be helpless, naked and blind at hatching and such hatchlings remain entirely dependent on adults to feed and protect them. Growth and development (of body generally and of the brain) are thus incomplete at birth/hatching. This has the disadvantage of being very vulnerable but the advantage that development can continue over a longer period of time. Precocial species are ready to feed themselves and walk about immediately or very soon after hatching (Figs 2.6–2.7).
Hence, concerning behavioural complexity, it is now known to be of less importance that birds are only distantly related to humans while primates are closely related to us. Clearly, birds and humans have developed the ability for vocal learning, but primates have not, not even those very close to humans in evolution (orang-utans, chimpanzees). Traits can appear and disappear throughout evolution. One can demonstrate this well with the example of colour vision. Colour vision is not unique to humans and it has not evolved slowly or only once throughout the animal kingdom. A trait as complex as colour vision is believed to have evolved at least 40 times (von Salvini-Plawen and Mayr 1977; Cronly-Dillon and Gregory 1990; Bowmaker 1998). The mantis shrimp has better and more complex eyes than humans or probably any other organism (Marshall and Oberwinkler 1999). Recently, by examining mineralised eye tissue, it has been found that a 300 million-year-old fossil fish (Acanthodes bridgei) was likely to have had colour vision (Tanaka et al. 2014). And flight has been lost and gained at least three times during evolution (Moore 1996). Equally, abilities in complexity of thinking might be qualities that have been lost and gained many times over, depending on ecological niche, environmental pressures and other variables.
Fig. 2.6. Precocial species. (A) One-day-old domestic chick (CSIRO ScienceImage Archive); (B) Australasian grebe, with offspring (photo by Keith Lightbody, Wikipedia commons). Note that the orders Galliformes (chicken and allies) and Anseriformes (ducks and allies) are well known to produce precocial offspring (i.e. they are fully feathered, can hear and see from the moment of hatching). Chicks may be able to feed on day 1 after hatching, but they usually feed themselves by day 3 (the egg yolk sustains them to this point). Chicks and ducks imprint on the hen so they continue to enjoy the protection of an adult until they are fully grown. Ratites, such as emus and cassowaries, and megapodes to which the malleefowl and brush turkeys belong, also produce precocial offspring. In megapodes, it is the most radical form of independence. The newly hatched offspring never see their parents and get no guidance or support from adults. Yet even in species with strong innate predispositions, complex cognitive behaviour such as referential signalling and a range of important discrimination tasks have been discovered, at least in the chicken.
While there may be a clear link between a larger brain and overall survival rate, that alone does not tell us what it is that leads to these outcomes. Marino (2005) suggested that summary terms may need to be found that could be regarded as inclusive of all the facets. A quality such as ‘unpredictability’ might well serve as an example of a higher order ecological demand that, alone, may be sufficient to explain the evolution of larger brains because it may demand new and flexible cognitive strategies to survive. I particularly like his suggestion because it can be readily applied to Australian conditions, as already briefly touched upon in Chapter 1.
However, one aspect of all this debate about brains, adaptation and survival has not been shaken and that is that birds generally are believed to do better than mammals. In the scientific evaluation of birds versus mammals, birds are now considered an ideal class of animals in which to study the advantages of having a large brain relative to body weight.
Fig. 2.7. Altricial species. Altricial hatching seemed a poor evolutionary invention and none too attractive. Offspring hatch naked and blind and are entirely dependent on adult care for weeks, even months, before being able to fly or feed independently. (A) Two kookaburra siblings from the same clutch, aged 6 and 10 days, respectively (asynchronous breeding). Both have their eyes open and the older one has developed pin feathers on abdomen, back and on the head. The other sibling is still naked, with the first pin feathers showing. Neither the 2-day-old rosella nor the 10-day-old kookaburra is yet able to stand up but both can raise their heads just briefly enough to be fed. (B) An eastern rosella, 2-days-old, not quite the size of a thumb. The egg tooth on the tip of the upper mandible is still clearly visible. The eyes are firmly shut and a very thin down covers part of the body – thermoregulation is inadequate and brooding by an adult essential. The bird was raised to become a healthy adult and rehabilitated and returned to the wild. (C) A common starling hatchling (an introduced species in Australia). The development documented here is over a period of just 8 days from the tiny body, the head no larger than the top of a little finger, blind and naked (i–ii) to alert with eyes open (iii), held between thumb and index finger to take this photo and finally fully feathered (iv).The seeming contradiction is that the immaturity and helplessness made the birds vulnerable to predation but, at the same time, had a built in plan for a longer developmental plan that permitted the slowest developing part of the organism, the brain, also to have time to grow. All three species show play behaviour and the parrots and starlings are excellent mimics.
In most papers on human evolution, innovation features as an important concomitant of complex cognition or shows that a great deal is attributed to innovation. Indeed, most stages of early human history are named after the most crucial innovation of the age, such as Stone Age and Bronze Age and these periods are subdivided into early, middle and late, usually based on the perceived increase of complexity of tools and artifacts. Such categorisations suggest that innovation drives the evolution of large brains, aided by a number of rather important contributing factors. The same criteria are translatable to birds and animals, making the topic of innovation an important one (see Fig. 2.8). Reader and Laland made a compelling argument. Using primates, they linked hundreds of documented cases of instances of innovation, social learning and tool use, and established statistically that all these variables were positively correlated with a species’ relative and absolute ‘executive’ brain volumes (Reader and Laland 2002). Or is it the other way around? As Dunbar had argued, large brains first and foremost had to do with social bonds and bonded relationships; this came to be known as the social brain hypothesis, as shall be discussed later (Dunbar 1998, 2009).
Innovation is usually not discussed in general terms but in terms of ‘modules’. This refers to aspects of behaviour that might identify evidence of cognition in birds. General categories or ‘modules’ that have been promoted and used widely in comparative psychology and primatology, include problem-solving ability, tool use, extractive foraging, foraging innovations generally, social intelligence and even resource mapping. Life history data are now also considered important, such as longevity, length of learning periods, and play behaviour, if one wishes to follow Ricklefs’ suggestion that cognitive research has to be viewed in a life-history context (Ricklefs 2004). His points have a good deal of merit. The question is surely also how youngsters grow up. Questions might be: what is the quality and durability of the parental bond, what is the level of cooperation in teaching and protecting offspring; how much room is there for exploration and discovery; and how much time do youngsters have to learn important life skills? I believe that Ricklefs’ suggestions are not only valuable but have opened new doors for investigating cognition in animals generally. The emphasis on individual or group performance is certainly important. However, in order to gain an insight as to how such individual performance is made possible, the link between environment and individual organisms is fundamentally important. In Australian birds, ‘growing up’ may happen over protracted periods and this is certainly significant.
Fig. 2.8. Residual brain size and innovation. There is a broad correlation between residual brain size and innovation rate (A) and tool use (B) in the same group of species. Feeding innovations (C) also correlate with brain size. The species represented in this graph covers most birds discussed in this book (all song birds, birds of prey and psittacine species). Corvida and Passerida together make up all true songbirds and refer to the broad subdivision of songbirds into two parvorders, as said before and as developed by Sibley and Ahlquist (1990) and thereafter used for many calculations of broad trends and similarities. However, ongoing taxonomical assessments have since changed the placement of a number of species. This figure is a simplified composite diagram of some elements of figures from two publications by Lefebvre and colleagues (A/B: Lefebvre et al. 2001; C: Lefebvre et al. 2004).
Reader and Laland (2003) defined innovation as ‘a process that results in new or modified learned behaviour and that introduces novel behavioural variants into a population’s repertoire’ (p. 14). This raised many intriguing questions why innovation happened, which species did so and whether any of this was related to variations in brain size as a ratio of body weight to brain weight between species and even between related species. What purpose was there for some cockatoos to have developed larger brains than some other parrots or for the variation in brain size between corvids and why were those of corvids and parrots so much larger than those of brush turkeys? What are the costs and benefits and do the differences reflect different cognitive abilities (Lefebvre et al. 2004)? There remain many unanswered questions. Still, the last 20 years of intense research into bird brains, into their learning and cognitive behaviour, has given us new tools for testing their complexity.
A decade ago, Lefebvre and colleagues set out to build a statistical framework for such variables as innovation and brain size (Fig. 2.8). They found positive correlations between innovation and tool use and feeding innovations in bird species across the world, including Australian species. Although these correlations are constructed with very broad brushstrokes, Lefebvre and colleagues have shown that innovation is crucially related to brain size and there are specific ways one can examine living birds even in their natural environment. Indeed, in the next six chapters, the topics that have been so dominant in research are to be to be dealt with in detail, but applied to native Australian birds.