Research - Nelson - Biological Sciences - University of Canterbury - New Zealand

Dr Ximena Nelson - Research

Communication and cognition in kea.

Animal communication plays a fundamental role in the study of animal cognition, yet, despite evidence that the kea has cognitive abilities rivalling that of primates, this relationship remains unexplored what is arguably the world’s most charismatic bird.
Animal signals were traditionally seen as conveying information about an animal’s internal state. Now it is accepted that many animals communicate about events and stimuli external to themselves. These ‘functionally referential signals’ should be structurally discrete and be highly stimulus specific. A clear match between the object or context associated with that signal, the referent (e.g. predator type), and signal structure, is an indication of stimulus specificity. Some birds have functionally referential alarm calls, and observations suggest that kea also make these in response to aerial predators. In order to determine that a signal is functionally referential it must be shown that the discrete signal elicits in receivers a response as if the referent was present, but in its absence.

Photo: Natasha Fijn.

In this project we will characterise the acoustic structure and biological function of kea vocalisations, and in so doing, test the relationship between communication and cognitive ability using a world-renowned, unique New Zealand alpine parrot as an ideal subject. Our objective is to bring state-of-the-art techniques and innovative methods to bear for investigating how cognitive flexibility, sociality and communication systems, and the vagaries of a unique habitat interrelate in the kea. If habitat structure prevents discrimination between two sounds at a distance, such as, for example, dense vegetation attenuating signals, this will directly influence signal structure by selecting for relatively simple communication. Because kea venture from alpine beech forest onto snowy peaks above the tree-line, an open habitat, this study will inform us about the role of habitat structure in the use of acoustic signals by a bird with high cognitive abilities. Call function will be ascertained by playing recorded calls to conspecifics, as well as matched white noise controls, in order to determine the responses of the receiver/s.

Some relevant references:vision

  • Diamond J & Bond AB 1991. Social behaviour and the ontogeny of foraging in the kea (Nestor notabilis). Ethology 88: 128-144.
  • Gajdon GK, Fijn N & Huber L 2004. Testing social learning in a wild mountain parrot, the kea (Nestor notabilis). Learn. Behav. 32: 62-71.
  • Werdenich D & Huber L 2006. A case of quick problem solving in birds: string pulling in keas, Nestor notabilis. Anim. Behav. 71: 855–863.
  • Diamond J & Bond AB 1999. Kea, bird of paradox, the evolution and behavior of a New Zealand parrot. University of California Press, Berkeley, CA, USA.
  • Gajdon GK, Fijn N & Huber L 2006. Limited spread of innovation in a wild mountain parrot, the kea (Nestor notabilis). Anim. Cogn. 9: 173-181.
  • Evans CS & Evans L 2007. Representational signalling in birds. Biol. Lett. 3: 8-11.

Neural basis of information processing. Neuroethology of vision. close-up

In this collaborative project involving scientists from NZ and Australia, we are using a combination of behavioural, electrophysiological, psychophysical and modelling tools to determine how it is that jumping spiders (Plexippus unicolor), whose eyes are only a few hundred microns wide, at their largest, are able to see with a resolution not dissimilar to our own. Given that they can, how can they process this tremendous amount of information with a brain that makes the head of a pin look large? How are they using this “data” to make decisions?

We are beginning to unravel these fascinating questions- and generating new ones all the time. For more information see our spider neuroethology site:

Jumping spider in front of LED presenting light stimuli while photoreceptor responses are being recorded with a glass microelectrode.

Close up of spider, tip of the electrode and the LED.

Mimicry and deceptive signals.

jumping spiderThe use of deceptive signals is not uncommon in animals, although we tend to hear much more about honest signalling. The truth is that animals will use deceptive signals if they can get away with it- and often they can. In an interesting case of deceptive signalling we explored recently, sit-and-wait predators, such as the Death adder, may lure prey, such as lizards, toward them by using signals that may make the predators seem like prey. In the case of Death adders, they use a signal called “caudal luring”, in which the distal part of the tail is flicked in a manner that appears to trigger a predatory response in lizards, thus attracted the hapless lizard into the jaws of the adder.

One of the more common types of deceptive signal is Batesian mimicry, where palatable individuals of a species avoid predation by using their resemblance to an unpalatable or dangerous model species for deceiving potential predators. The most numerous Batesian mimics are Myrmarachne, of which we know have an in-depth understanding.

What with remarkable eyes, great eyesight, and ability to identify rivals, mates and prey from as far as 30 body lengths away, you’d think being a jumping spider (family Salticidae) was kudos enough, but the 200 or more species in the largest salticid genus (Myrmarachne) add dramatically more to the salticid portfolio by being remarkably similar to ants in appearance. One of the interesting things we have been discovering is that, for Myrmarachne, looking like ants isn’t all fun and games. Although resembling ants seems to function in general for Myrmarachne as Batesian mimicry, this gets complicated because, in order to be a convincing mimic, some Myrmarachne species go to considerable lengths to live in close vicinity of their models. That’s a complication because the company of ants can be dangerous. We have been looking at how Myrmarachne handles this difficulty. The story goes roughly like this:

Batesian mimics are palatable individuals that avoid predation by using their resemblance to unpalatable or dangerous models for deceiving potential predators. Would-be predators of salticids often seem to have an innate aversion to ants and we have shown experimentally that this aversion is sometimes generalized to an aversion for the ant-like salticid. Salticids are a great system for experimentally investigating mimicry, with salticids providing both the mimic (Myrmarachne) and the potential predator of the mimic. This is because the Salticidae also includes a sizeable number of species that specialise at eating ants and there are yet other salticids that specialise at eating salticids. Another advantage of working with salticids is how their extraordinary eyesight facilitates experimentation.

We know that, for many salticids, ants can be very dangerous in nature but salticid eyesight seems to be up to the challenge. Our experiments have demonstrated that many species of ordinary salticids (species that are neither ant-like nor ant-eating) readily identify by ants sight and then avoid their proximity. Ant-like and ant-eating salticids need to get close to ants, and they apparently have abilities to survive these close encounters better than other salticids. How closely different species of Myrmarachne resemble particular ant species varies considerably, and we have found that accurate mimics of a particular model species survive better than the more generalized ant mimics (spiders that resemble ants but have no specific model, see photo). This suggests that accurate mimics have evolved special adaptations that let them survive in the vicinity of their models and this has been one of the topics we decided to investigate further.

Myrmarachne assimilis is an accurate mimic of an especially aggressive model, the weaver ant Oecophylla smaragdina. Nonetheless, M. assimilis lives in proximity with this ant, and is very unusual among salticids because it aggregates. Furthermore, the advantage driving M. assimilis’ predisposition to aggregate appears not to be what we would expect from other social spiders. That is, aggregating does not seem to have anything to do enhancing foraging, nor with kin groupings. Instead, with M. assimilis, aggregations seem to come about primarily through adult females seeking each other out and forming crèches. Our work suggests that these crèches function to protect the females’ offspring from predators, with the most notable predator being the salticid’s own model, O. smaragdina.

Our experiments have established that Myrmarachne’s mimicry deceives highly visual predators, including mantises, ordinary salticids and, perhaps most interesting of all, salticids that specialise at eating ants. Although, for Myrmarachne, deceiving predators that eat ants sounds like a bad idea, Myrmarachne has evidently found ways to ameliorate mimicry that goes wrong. When an individual of Myrmarachne receives the unwanted attentions of ant-eating salticids, it actively defends itself by revealing its true identity. The mimic switches off its ant-like demeanour and acts like a conventional salticid by raising its legs and displaying at the ant-eating salticid in much the same way as salticids generally threaten each other. Our experiments demonstrated that it is specifically the display posture adopted by the ant mimic that deters the ant-eating predator.

This work was interesting as an illustration of how mimicry may be advantageous when it deceives ant-averse potential predators, but disadvantageous in encounters with ant-eating specialists, this being something that may apply more widely than just in the Salticidae. Another interesting implication is how it demonstrates that Myrmarachne’s resemblance to ants holds not only for our human eyes but also in the eyes of salticids. The story gets especially interesting when we consider encounters between Myrmarachne and Portia fimbriata, a salticid that has specific behaviour for capturing other salticids and prefers salticids as prey.

Here, it was not obvious which way things would go because P. fimbriata is renowned for having perhaps the best eyesight of any salticid. Experiments using computer animation and virtual prey have shown that an important salticid-identification cue for P. fimbriata is what human taxonomists also use, the large front eyes of the salticid. Like all salticids, Myrmarachne has large forward-facing eyes (see photo). What we found is that P. fimbriata rejects as prey not only ants but also Myrmarachne. An intriguing question remains. Is it that P. fimbriata is simply deceived or do Myrmarachne’s large forward-facing eyes tell P. fimbriata that this is a salticid but with the otherwise repugnant (for P. fimbriata) ant-like appearance of Myrmarachne making P. fimbriata ‘lose’ its appetite?

There would seem to be another problem for Myrmarachne. Salticids have long been known for their elaborate vision-guided courtship displays. They are known to be remarkably good at identifying members of their own species and are able to distinguish between males and females using sight alone. But what about Myrmarachne? What is courtship like for an ant-mimicking salticid?The answer is of interest because, after all, Batesian mimicry for M. assimilis depends on making vision-based identification difficult. That’s good if the eyes are on a predator, but not so good when the eyes are on a potential mate.

Again we chose M. assimilis as experimental subjects. First we found that this ant mimic has displays dedicated explicitly for use during intraspecific communication. This was good because it meant we could use performance of these displays as a bioassay for determining whether M. assimilis distinguishes, by sight alone, between its ant-like conspecifics and its model, real ants. Our experiments suggest that M. assimilis, as a Batesian mimic of a particularly dangerous model, has the especially good perceptual ability needed for rapidly discriminating between conspecific and model.

Myrmarachne bakeri (a generalised
ant mimic).
Photo: Robert Jackson.

Now the plot thickens. Besides their ant mimicry, there is something else for which the genus Myrmarachne is renowned. It is as if the males of these salticids have been badly mangled by sexual selection. Myrmarachne males have enormously elongated chelicerae and, at first glance, these would seem to render them relatively unlike ants in appearance. Maybe this is, for Myrmarachne, a cost of sexual selection, but this cost seems to be ameliorated by the Myrmarachne males retaining Batesian mimicry by resembling something more than an ant. The large chelicerae resemble an ant carrying an object in its mandibles. We have called this ‘compound mimicry’, as the model for Myrmarachne males seems to be a compound model, an ant and something being carried by the ant. Resembling an ant that is carrying an object in its mandibles, however, appears to have an unwelcome effect for the Myrmarachne male, as it makes the male more attractive to ant-eating jumping spiders. It turns out that ant-eating salticids prefer ants carrying objects in their mandibles as prey, probably because it is less dangerous than an ant with free mandibles. Ant-eating jumping spiders also ‘prefer’ male over female Myrmarachne.

Some recent papers pertaining to ant-spider relationships:

  • Edmunds, M. (2006). Do Malaysian Myrmarachne associate with particular species of ant? Biol. J. Linn. Soc. 88, 645-653.
  • Jackson, R. R. & Li, D. (2001). Prey-capture techniques and prey preferences of Zenodorus durvillei, Z. metallescens and Z. orbiculata tropical ant-eating jumping spiders (Araneae: Salticidae) from Australia. N. Z. J. Zool. 28, 299-341.
  • Jackson, R. R., Nelson, X. J., Pollard, S. D., Edwards, G. B. & Barrion, A. T. (2004). Predation by ants on jumping spiders (Araneae: Salticidae) in the Philippines. N. Z. J. Zool. 31, 45-46.
  • Nelson, X. J. & Jackson, R. R. (In Press). Anti-predator crèches and aggregations of ant-mimicking jumping spiders (Araneae: Salticidae). Biol. J. Linn. Soc.
  • Nelson, X. J. & Jackson, R. R. (2007). Complex display behaviour during the intraspecific interactions of myrmecomorphic jumping spiders (Araneae, Salticidae). J. Nat. Hist. 41,1659-1678.
  • Nelson, X. J. & Jackson, R. R. (2007). Vision-based ability of an ant-mimicking jumping spider to discriminate between models, conspecific individuals and prey. Insect. Soc. 54, 1-4.
  • Nelson, X. J. & Jackson, R. R. (2006). Compound mimicry and trading predators by the males of sexually dimorphic Batesian mimics. Proc. R. Soc. B. 273, 367-372.
  • Nelson, X. J. & Jackson, R. R. (2006). Vision-based innate aversion to ants and ant-mimics. Behav. Ecol. 17, 676-681.
  • Nelson, X. J., Jackson, R. R. & Li, D. (2006). Conditional use of honest signalling by a Batesian mimic. Behav. Ecol. 17, 575-580.
  • Nelson, X. J., Li, D. & Jackson, R. R. (2006). Out of the frying pan and into the fire: a novel trade-off for Batesian mimics. Ethology 112, 270-277.
  • Nelson, X. J., Jackson, R. R., Edwards, G. B. & Barrion, A. T. (2005). Living with the enemy: jumping spiders that mimic weaver ants. J. Arachnol. 33, 813-819.
  • Nelson, X. J., Jackson, R. R., Li, D., Edwards, G. B. & Barrion, A. T. (2006). Innate aversion to ants (Hymenoptera: Formicidae) and ant mimics: experimental findings from mantises (Mantodea). Biol. J. Linn. Soc. 88, 23-32.

Predator and prey assessment.

Decision-making is a fundamental part of an animal’s behaviour. To be able to detect and avoid predators, while accurately detecting and obtaining prey and classifying conspecifics either as rivals or potential mates, often from some distance, requires both acute sensory systems able to cope with these demands, and the capacity to process sensory information rapidly in order to make a decision. Using live animal tests, playback studies and 3D animation techniques, I have explored the effects of apparent size, speed, motion characteristics, as well as shape, as some of the visual parameters used by animals when making these decisions. My work on visually-mediated decision-making has included jumping spiders, lizards and birds.

Through this work we know, for example, that an African species of jumping spider, Evarcha culicivora, can differentiate its favourite prey of mosquitoes in the genus Anopheles from other mosquitoes based solely on the posture in which they rest (see: Nelson, X. J., Jackson, R. R. (2006). A predator from East Africa that chooses malaria vectors as preferred prey. PloS ONE 1(1): e132. doi:10.1371/journal.pone.0000132.


Apparatus for virtual-prey testing. Spider (not to scale) at top of inclined metal ramp, oriented toward one of two side-by-side virtual mosquitoes. Virtual mosquitoes on small screen positioned in front of higher end of ramp. Images pass from projector lens (connected to computer; projector and computer not shown) through second lens (for reducing image size) on to screen. Inset: virtual mosquitoes in Anopheles resting posture (left) and in non-Anopheles resting posture (right).

Below are some examples of visual stimuli presented to lizards, using a combination of digital video playback and 3D animation.

visual stimuli

Visual stimuli presented to Jacky dragons to assess behavioural decision-making. (a) Silhouette of animated falcon viewed at an apparent size of 3o at lizard eye height against ‘grey’ background. (b) Frame from the conspecific display presentation. (c) ‘Cybercricket’ against grassy background. (d) ‘Cybersnake’ against grassy background.

"Personality” traits or behavioural syndromes.

FightAnimals are typically faced with the challenge of acquiring resources, such as mates or food, while minimizing associated costs, such as exposure to predators or rivals. Individuals of many species differ consistently in their behavioural reactions toward different stimuli, such as predators, rivals, and potential mates, reflecting their assessment of this trade-off. These typical reactions, described as ‘personalities’, or ‘behavioural syndromes’, appear to be heritable and therefore subject to selection.

Using a combination of high-definition digital video and 3D animation techniques, I have explored some of the constraints on the behaviour of both birds and lizards. The study of animal personalities is particularly difficult because often times, as we have shown, what appears to be a behavioural syndrome is in fact a by-product of something entirely different and external to the animal, such as social position in the hierarchy.

The rooster photographed below is aggressively responding to an HDV playback of a rival male, illustrating how effective video playback can be as a behavioural tool.