NEWS & VIEWS RESEARCH

There is one caveat, however: clear skies are fully quantified. There is still a rich body 116, A11221 (2011). required to see and measure the pulsating- of research to be carried out regarding the 3. Jaynes, A. N. et al. J. Geophys. Res. Space Phys. 118, 4884–4894 (2013). aurora signals, so Earth’s terrestrial weather mysterious pulsating auroras. ■ 4. Kasahara, S. et al. Nature 554, 337–340 (2018). needs to cooperate. Furthermore, the chorus 5. Pulkkinen, T. I. & Tsyganenko, N. A. J. Geophys. Res. waves contain components of different fre- Allison N. Jaynes is in the Department of 101, 27431–27442 (1996). 6. Coroniti, F. V. & Kennel, C. F. J. Geophys. Res. 75, quency that interact with magnetospheric Physics and Astronomy, University of Iowa, 1279–1289 (1970). electrons in different ways depending on the Iowa City, Iowa 52242, USA. 7. Davidson, G. T. J. Geophys. Res. 84, 6517–6523 energy of the particles. This affects which e-mail: [email protected] (1979). particles end up travelling down to Earth’s 8. Miyoshi, Y. et al. J. Phys. Conf. Ser. 869, 012095 1. McEwen, D. J., Yee, E., Whalen, B. A. & Yau, A. W. (2017). atmosphere. These details are directly related Can. J. Phys. 59, 1106–1115 (1981). 9. Mende, S. et al. Space Sci. Rev. 141, 357–387 to geomagnetic activity and have not yet been 2. Nishimura, Y. et al. J. Geophys. Res. Space Phys. (2008).

ANIMAL BEHAVIOUR it undermines reliable testing of a hypothesis9. Ashton and colleagues turned instead to intraspecies comparisons. They studied 56 magpies, which were ringed to enable Brainpower boost for identification, from 14 different territorial groups that ranged in size from 3 to 12 birds (Fig. 1). Rather than measuring brain size, the birds in large groups authors conducted cognitive-performance tests in which the birds encountered wooden Whether intelligence is selected for in species that have a complex social life is or plastic devices that tested problem-solving debated and hard to test. Cognitive performance and associated reproductive skills; successful birds received a mozzarella- success are now linked to group size in wild magpies. See Letter p.364 cheese treat. Four different devices each tested a specific skill, including spatial memory and the ability to learn new associations between ANDREW WHITEN might resolve the dilemma of how to test stimuli and rewards. Humphrey’s ideas. Both measurements were The authors report that group size is linked bservations of primates’ everyday available for a range of primates — and, as to cognitive performance. Birds living in a lives led the psychologist Nicholas predicted, a positive relationship was found larger group displayed better performance at Humphrey to make a revolutionary between these factors. Multiple teams rep- the population level on each of the tests than Oproposal1 in 1976 to explain primate intelli­ licated the finding, using a range of related did birds living in smaller groups. At the indi- gence. Before then, it had been commonly variables — for example, measuring the rela- vidual level, exceptions to this trend could assumed that these animals’ cleverness was an tive sizes of the neocortex region rather than be found — some birds from smaller groups adaptation to their physical niches, reflected overall brain size — for primates7 and other outperformed birds from larger groups, and in their need for sophisticated skills in realms taxa8. Yet, as Ashton and colleagues acknowl- vice versa. such as foraging, navigation or avoiding pred- edge, analyses of large databases often provide The authors recorded the reproductive ators. Humphrey suggested instead that the conflicting results. When many variables differ success of individuals by counting the average complex social dynamics experienced when between species, cross-species comparisons can number of hatched clutches of eggs per year, such animals live in a group become the main lack robustness, because compensating for the and found that birds that performed well on selective force driving the evolution of primate differences can make a study so unwieldy that the tests had greater reproductive success than intelligence. On page 364, Ashton et al.2 offer support for Humphrey’s social-intelligence hypothesis, in a study of wild Australian mag- pies (Cracticus tibicen dorsalis, also known as BEN ASHTON Gymnorhina tibicen dorsalis). The causal relationships between social complexity, intelligence and reproductive success proposed by Humphrey inspired a generation of primatologists, who uncov- ered unexpected sophistication in monkeys’ and apes’ social knowledge and political manoeuvrings3–5. However, these discoveries arguably made it difficult to test Humphrey’s hypothesis directly, because intelligence — both in social interactions and in non-social realms, such as foraging or tool use — and social complexity were revealed to be com- posed of many components4,5. Primate social complexity, like intelligence itself, was found Figure 1 | Australian magpies in Guildford, Western Australia. Wild Australian magpies to be extraordinarily complex. (Cracticus tibicen dorsalis, also known as Gymnorhina tibicen dorsalis) are territorial and live in groups. In 1995, primatologist Robin Dunbar 2 6 Ashton et al. analysed the relationship between the birds’ group size and cognitive performance to test suggested that focusing on the typical group the long-debated idea1 that life in complex social groups can select for intelligence. At the population size of a species as a proxy for social complex- level, larger groups of birds performed better than smaller groups in cognitive-performance tests, and ity, and on its brain size instead of intelligence, cognitive performance was also linked to reproductive success.

©2018 Mac millan Publishers Li mited, part of Spri nger Nature. All ri ghts reserved. ©2018 Mac millan Publishers Li mited, part of Spri nger Nature. Al15l ri gFEBRUARYhts reserved. 2018 | VOL 554 | NATURE | 303

RESEARCH NEWS & VIEWS birds that performed poorly. Evidence for a as domain-general intelligence), or for e-mail: [email protected] connection between test performance and specialized forms of intelligence, such as the 1. Humphrey, N. K. in Growing Points in Ethology biological fitness could not have been assessed social skill of understanding what others might (eds Bateson, P. P. G. & Hinde, R. A.) 303–317 by the earlier approaches based on interspecies be thinking. This might sound like a tall order (Cambridge Univ. Press, 1976). comparisons. for a magpie. But there is evidence10 for such 2. Ashton, B. J., Ridley, A. R., Edwards, E. K. & Thornton, A. Nature 554, 364–367 (2018). For any given individual, the bird’s sophisticated behaviour in the crow family — 3. de Waal, F. B. M. Chimpanzee Politics: Power and Sex performance on each of the four skill tests a western scrub-jay (Aphelocoma californica) Among Apes (Cape, 1982). was correlated. A principal-components sta- might make a theft-prevention manoeuvre by 4. Byrne, R. W. & Whiten, A. Machiavellian Intelligence: Social Complexity and the Evolution of Intellect in tistical analysis, which identifies the number relocating hidden food elsewhere if it spots Monkeys, Apes and Humans (Oxford Univ. Press, of factors accounting for variance in an array that another bird observed where the food 1988). of scores, showed that a single unknown was hidden. Social cognition has numerous 5. Seyfarth, R. M. & Cheney, D. L. Anim. Behav. 103, 11,12 191–202 (2015). factor accounted for 65% of the variance in other manifestations . Yet, what Ashton 6. Dunbar, R. I. M. J. Hum. Evol. 28, 287–296 (1995). 2 test scores. The authors refer to this factor as et al. tested was essentially non-social cog- 7. Dunbar, R. I. M. & Shultz, S. Science 317, ‘general intelligence’, analogous to the ‘g factor’ nition. What now begs to be fleshed out is 1344–1347 (2007). 8. Perez-Barberia, F. J., Shultz, S. & Dunbar R. I. M. used to assess general intelligence in humans. the nature of both the social and non-social Evolution 61, 2811–2821 (2007). However, it is worth noting that the magpie intelligence skills that may have been at work 9. Powell, L. E., Isler, K. & Barton, R. A. Proc. R. Soc. B tests assess learning abilities rather than test- in the phenomena these authors observed. ■ 284, 20171765 (2017). 10. Clayton, N. S., Dally, J. M. & Emery, N. J. Phil. Trans. ing the capacity to invent creative solutions to R. Soc. B 362, 507–522 (2007). problems — a talent sometimes considered Andrew Whiten is in the Centre 11. Emery, N. J., Clayton, N. S. & Frith, C. D. Phil. Trans. to be a defining characteristic of animal or for Social Learning and Cognitive R. Soc. B 362, 485–488 (2007). 12. Whiten, A. & van de Waal, E. Neurosci. Biobehav. Rev. human intelligence. Evolution, School of Psychology and 82, 58–75 (2017). Humphrey proposed that intelligence Neuroscience, University of St Andrews, evolved in response to the pressures of social St Andrews KY16 9JP, UK. This article was published online on 7 February 2018. complexity. Ashton and colleagues, however, did not directly address this evolutionary hypothesis; instead, they investigated the rela- NEUROSCIENCE tionship between social-group size and the development of cognition in the birds’ early life, a linkage that was observed to emerge by the time the birds were 200 days old. Neverthe- Burst firing sets the less, the link between cognitive prowess and fitness identified by Ashton and colleagues has major implications for connecting the stage for depression social-intellect hypothesis to an underlying evolutionary mechanism, and it suggests that Salvos of neuronal activity in the brain’s lateral habenula, regulated by astrocyte selection is acting on a relationship between cells, drive depression-like behaviours in rodents. The finding might help us to sociality and cognition. understand one antidepressant and to develop more. See Articles p.317 & p.323 Group size is, of course, a crude index of social complexity, just as brain size is a crude indicator for the complexities of cognition WILLIAM M. HOWE & PAUL J. KENNY electrical potential across the cell membrane being selected for. Indeed, group size itself can- (in a resting state, the inside of the cell is nega- not be the key causal factor — the immensity pposing forces shape our everyday tively charged relative to the extracellular of a wildebeest herd, for example, is unlikely lives — for instance, stimuli can space around it). Hyperpolarization, in which to select strongly for intellect. And even three encourage us to move or stop, and the cell interior becomes more negative than individuals can suffice to create a high level Oevents can make us happy or sad. Accordingly, normal, is typically associated with neuronal of social complexity, as demonstrated in the our brains are designed with ‘yin–yang’ systems inhibition. humorous account provided by Jerome K. that guide our actions and influence our feel- By studying brain slices ex vivo, Yang and co- Jerome’s 1889 book Three Men in a Boat, or ings. Neurons in the brain’s mesolimbic system workers showed that LHb neurons were more documented in an analysis3 of three adult promote reward-seeking behaviour and help likely to fire in a pattern of rapid bursts in the chimpanzees that repeatedly shifted alliances to process information about actions that ‘depressed’ rats than in control animals. They in a way that allowed each to be supreme for result in pleasurable outcomes1–3. By contrast, also observed that, when the LHb neurons a while in their ‘game of thrones’. Ashton and neurons in the lateral habenula (LHb) encode were hyperpolarized, this increased the likeli- colleagues can only speculate on how mag- information related to noxious outcomes and hood that these cells would fire in bursts rather pies’ cognitive powers relate to social-group suppress reward-seeking4–6. Unbalancing these than steady volleys. The researchers went on to size and number of offspring, and suggest that opposing systems might therefore affect our show that they could increase depression-like some kind of political skill, such as the ability behaviour. Indeed, emerging evidence7 sug- behaviours in rats using a genetic manipula- to successfully handle or avoid conflicts, might gests that LHb hyperactivity contributes to tion to drive hyperpolarization, and so burst be involved. mood disorders such as major depression. firing, in LHb neurons. Which types of intelligence deserve closer Two papers8,9 in Nature now shed light on the Next, the group investigated the signals attention in future studies of the social- mechanisms that underlie LHb hyperactivity, that regulate this burst firing. In other intelligence hypothesis? Some substantially and on how the antidepressant drug ketamine brain regions10, burst firing is controlled by different ideas exist regarding which form modulates this state. N-methyl-d-aspartate receptors (NMDARs) of intelligence might be selected for4. One In the first paper, Yang and colleagues8 — membrane-spanning channel proteins such contrast is between whether social (page 317) assessed the firing activity of LHb whose activation leads to an influx of positively complexity selects for overall intelligence neurons in two rat models of depression. charged calcium ions into neurons, resulting across many different contexts (also known Neuronal firing involves depolarization of the in depolarization and neuronal firing. Yang

304 | NATURE | VOL 554 | 15 FEBRUARY©2018 M 2018ac millan Publishers Li mited, part of Spri nger Nature. All ri ghts reserved. ©2018 Mac millan Publishers Li mited, part of Spri nger Nature. All ri ghts reserved.