Vol. 14, No. 5, May 1999 News and Comment Animal communication: what is the signal to noise ratio? Staffan Andersson a s.andersson@zool.gu.se and Peter K. McGregor b pkmcgregor@zi.ku.dk Trends in Ecology & Evolution 1999, 14:174-175 [a] Dept of Zoology, Gteborg University, Box 463, SE-405 30 Gteborg, Sweden[b] Dept of Animal Behaviour, Zoological Institute, University of Copenhagen, Tagensvej 16 DK-2200, Copenhagen N, Denmark

Article Outline


From chemical communication between cells to whale song traversing the oceans, the fascinating diversity of biological signals call for evolutionary explanations, and their investigation often requires sophisticated, multi-disciplinary techniques ¹. A workshop last September in Kongsvoll (Norway), titled 'Adaptive significance of signalling and signal design in animal communication', gave a nice snapshot of current trends and devel-opments in this active field. The meeting was hosted by The Royal Norwegian Society of Sciences and Letters ², and organized by Yngve Espmark, Trond Amundsen and Gunilla Rosenqvist at the University in Trondheim (Norway).

Recent advances in studies of visual signalling include spectrometric analyses of colour signals in relation to backgrounds, ambient light and receiver visual systems. John Endler (University of California, Santa Barbara, USA) showed how Australian bowerbirds (Ptilonorhyncidae) decorate their bowers with coloured ornaments that contrast maximally with their plumage at the time and place of display. However, as in Endler's earlier work ^3,4 , predators seem to impose important constraints. Perhaps to avoid attracting visually hunting predators, bower decorations are at the extremes of, but still bordering, the range of nearby background colours.

Although John Endler has gone terrestrial with his colour measurements, Justin Marshall (University of Queensland, Brisbane, Australia) remains submerged at the Great Barrier Reef, arguably the planet's most dazzling colour exhibition. A variety of light microhabitats seems one likely reason for the explosion of fish colour patterns, but many questions remain. As if the variation we humans can see was not enough, many reef fish colours have an ultraviolet (UV, 320–400 nm) and a non-UV version, which we cannot distinguish but which juvenile and many adult fish probably can. Beware, however, of taking reflectance as evidence for a signal function; because of UV-absorbing lens pigments, several strongly UV-coloured fish seem just as blind to these wavelengths as we are.

Moving back to birds, lens transmission has admittedly been measured in very few species, but other evidence suggests that UV perception is even more general in birds than it is in fish. Innes Cuthill (University of Bristol, UK) reported on a consistent occurrence of a UV (360 nm) photoreceptor in passerines, and summarized the accumulating evidence that UV is an integral part of avian colour communication.

A uniquely complete study of adaptive radiation of colour signals came from Leo Fleishman (Union College, Schenectady, NY, USA) and 17 species of Anolis lizards. The jerky head and dewlap movements of males contrast with the backdrop of gently moving vegetation and have been shown to grab the attention of females efficiently ⁵. New experiments show that detectability also depends on brightness and colour contrast with the background. Dewlap colouration varies greatly between species and includes black, white, red, purple and UV colours. With a model based on reflectance variation, background radiance in three habitats, and the spectral tuning of the four Anolis cone receptors, Fleishman showed that dewlap colours have arisen, at least in part, as adaptations for signal visibility in habitats with different light conditions.

It is important to remember that the terms 'signal' and 'information' are not synonymous, and that signals are not the only source of information in the environment. In the words of Oren Hasson (Ramat-Gan, Israel), true signals have 'evolved to decrease the cost of information-gathering'. Of course it is not always easy to make the distinction. Brian Wisenden (Moorhead State University, Moorhead, MT, USA) gave several examples of aquatic organisms that respond with predator avoidance behaviour when exposed to water containing the faeces of predators fed on conspecifics ⁶. If these predation-revealing chemicals have evolved to signal such information, then this must be one of the few examples of post-mortem communication. It is more likely, of course, that prey have evolved great sensitivity to any cues informing them that they are on a predator's menu.

Also balancing on the borderline between signals and cues, John Swaddle (University of Chicago, IL, USA) reviewed the mixed evidence that fluctuating bilateral asymmetry (FA) signals developmental stability and thus 'quality' to mates or rivals. Only 66 of 190 (35%) published attempts have supported FA signalling. Furthermore, even if FA is present, is it perceived in natural communication? Of eight experiments suggesting the latter, only three manipulated FA within its natural range. Neither did Swaddle find pecking preferences for symmetric patterns in Skinner-boxed Chicago starlings (Sturnus vulgaris). Some might argue that discriminative abilities have been lost in American founder populations. However, British starlings also seem to ignore FA cues ⁷.

The list of more conspicuous quality advertisements is growing, but how stable are such signalling systems over time and environmental fluctuations? Ben Sheldon (Uppsala University, Sweden) addressed the largely unexplored genotype × environment interactions in signal systems and presented data from collared flycatchers (Ficedula albicollis), where the correlations between a male ornament (the white forehead patch) and offspring fitness varied tremendously between seasons.

Turning to very obvious and stable signal functions, offspring begging taps into one of the most attentive and powerful receiver biases on earth – the parent: a food and protection machine bound to be exploited by anyone who can get away with it. Rebecca Kilner (University of Cambridge, UK) filled in more gaps on the amazing story of how the European cuckoo chick (Cuculus canorus) mimics the begging sound of an average brood of reed-warbler chicks (Acrocephalus scirpaceus) ⁸. The visual stimulus of a red gape also has an effect on parental feeding effort. As the cuckoo chick grows older and less similar in gape area (compared with the would-be reed-warbler brood it has evicted), it seems successfully to compensate for this loss of visual mimicry by a steeper increase in call rate. Given that the future reproduction of a reed warbler is of no concern to a cuckoo, why not call faster still and work the poor foster parents even harder?

One way to be sure that you are studying true signals is to model communication mathematically – at least you can then define what the signal is. Rufus Johnstone (University of Cambridge) incorporated signal detection theory with the strategic considerations of a game theory approach. As Oren Hasson also pointed out, signal detection theory probably has more to offer animal communication models than does information theory ⁹. Johnstone's model showed that even in perfectly cooperative systems there will be conflict between sender and receiver over how the costs of production and perception should be split. He also modelled a situation in which an eavesdropper gained information as well as the signaller and receiver. In such (probably very common) circumstances, the outcome of the model depends on details of how the costs and benefits were assigned between the three participants.

The study of signal context, signal form and receiver responses can reveal a lot about the information content and function of signals, as illustrated by several interesting talks on bird song. For example, Sandy Vehrencamp (University of California, San Diego) argued that the type of cost paid by senders to guarantee signal honesty is an important determinant of signal form and function, and classified the variable elements of bird song into handicap, index and conventional signals.

Mats Grahn (Karolinska Institute, Stockholm, Sweden) and Claus Wedekind (University of Bern, Switzerland) discussed signalling based on the major histocompatibility complex (MHC), an essential genetic component of immunocompetence. The current fashion of Swedish pheasants (Phasianus colchicus) seems to be spur length, which, according to Grahn, varies with MHC haplotype. Wedekind reported on the avoidance of MHC inbreeding, mediated by odour in humans and mice, and startling new results suggesting MHC outcrossing through selective fertilization ('egg choice') in mice (Mus musculus).

Continuing the 'Signalling and health' theme, Marlene Zuk (University of California, Riverside) discussed how principles of animal communication also apply to intercellular signalling, and how parasites exert powerful selection on communication within, as well as between, individuals. Receiver exploitation is basically the same phenomenon whether it is a cuckoo chick manipulating its foster parent to obtain food or a virus producing an interleukin mimic that inhibits the immune response of its host.

Did a clear signal emerge from the workshop? The answer was a resounding 'yes'. Regardless of signal modality, there is a 'grand unifying theory' of animal signalling. In fact, it has been around for a while – it is called evolution by natural selection.



Acknowledgements

We are grateful to Malte Andersson for comments on this article.

References

[1] Bradbury, J.W. and Vehrencamp, S. (1998) Principles of Animal Communication, Sinauer

[2] Espmark, Y., Amundsen, T. and Rosenqvist, G. (1999) Adaptive Significance of Signalling and Signal Design in Animal Communication, The Royal Norwegian Society for Sciences and Letters, Tapir Publishers, Trondheim, Norway

[3] Endler J.A. (1980) Natural selection on color patterns in Poecilia reticulata.
Evolution, 34:76-91. [Cited by]

[4] Endler J.A. and Théry M. (1996) Interacting effects of lek placement, display behavior, ambient light, and color patterns in three neotropical forest-dwelling birds.
Am. Nat., 148:421-452. [Cited by]

[5] Fleishman L.J. (1992) The influence of the sensory system and the environment on motion patterns in the visual-displays of anoline lizards and other vertebrates.
Am. Nat., 139:36-61. [Cited by]

[6] Wisenden B.D. and Sargent R.C. (1997) Antipredator behaviour and suppressed aggression by convict cichlids in response to injury-released chemical cues of conspecifics but not to those of an allopatric heterospecific.
Ethology, 103:283-291. [Cited by]

[7] Swaddle J.P. and Witter M.S. (1995) Chest plumage, dominance and fluctuating asymmetry in female starlings.
Proc. R. Soc. London Ser. B, 260:219-223. [Cited by]

[8] Davies N.B., Kilner R.M. and Noble D.G. (1998) Nestling cuckoos, Cuculus canorus, exploit hosts with begging calls that mimic a brood.
Proc. R. Soc. London Ser. B, 265:673-678. [Cited by]

[9] Wiley R.H. (1994) Errors, exaggeration, and deception in animal communication.
In: Real L (Eds.) Behavioral Mechanisms in Evolutionary Ecology. (pp. 157-189) : University of Chicago Press