The Nowicki Lab at Duke University



Assessment & Reliability

Signal reliability presents a puzzle because the evolutionary interests of signalers and receivers generally differ. For example, when a female assesses a male's quality based on a signal characteristic, we might expect the male to exaggerate his quality; that is, to signal deceptively. But selection acts on receivers as well as on signalers, so receivers ought to respond to signals only if doing so is on average to their advantage as well. Taken to the limit, this logic suggests that signaling systems are unlikely to be evolutionarily stable unless there's some resolution to this conundrum, i.e., a mechanism to enforce honesty. A long-standing focus of our research, in collaboration with Bill Searcy at the University of Miami, has been to understand how signal reliability is maintained.

One facet of this work has been to test the developmental stress hypothesis, an idea we proposed that suggests a means by which learned features of birdsong—such as copy quality, local dialect structure, or song complexity—can serve as reliable indicators of male quality in the context of mate choice. The link depends on the fact that brain structures underlying song learning develop early in life, during a time when songbirds are likely subject to nutritional and other stressors. Individuals faring well in the face of stress are able to invest more resources to brain development and should be correspondingly better at song learning, so reliability is maintained by the developmental costs of building the brain.

Our lab also has explored signal reliability in the context of aggression. Receivers in aggressive situations benefit from knowing about at least two aspects of a signaler: its fighting ability and its aggressive intentions. Signals of fighting ability are often index signals, constrained to be honest because of a causal connection between physical attributes of the sender and characteristics of the signal. Signals of aggressive intent, however, are often "free strategic choice" signals, which any signaler is able to produce. Even with receiver-dependent costs imposed, such signals may be expected to be less reliable on average. Our empirical work on aggressive signaling in both song sparrows and swamp sparrows supports this prediction, although we've found that reliability isn't necessarily compromised by bluffing (i.e., when a signaler exaggerates its aggressive intent), but rather by "low-balling," in which an individual under-signals its likelihood of subsequent attack.

Here are some of our papers in this area:

Soha JA, Peters S, Anderson RC, Searcy WA & Nowicki S. 2019. Performance on tests of cognitive ability is not repeatable across years in a songbird. Animal Behaviour doi:10.1016/j.anbehav.2019.09.020

Sewall KB, Anderson RC, Soha JA, Peters S & Nowicki S. 2018. Early life conditions that impact song learning in male zebra finches also impact neural and behavioral responses to song in females. Developmental Neurobiology doi:10.1002/dneu.22600

Lachlan RF, Anderson RC, Peters S, Searcy WA & Nowicki S. 2014. Typical versions of learned swamp sparrow song types are more effective signals than are less typical versions. Proceedings of the Royal Society, London B 281: 20140252.

Searcy WA, Anderson RC, Ballentine B & Nowicki S. 2013. Limits to reliability in avian aggressive signals. Behaviour 150: 1129-1145.

Lachlan RF & Nowicki S. 2012. How reliable is song learning accuracy as a signal of male early condition? American Naturalist 180: 751-761.

Searcy WA & Nowicki S. 2008. Bird song and the problem of signal reliability. American Scientist 96: 114-121.

And our book on assessment signaling and the evolution of reliability in animal communication in general:

Searcy WA & Nowicki S. 2005. The Evolution of Animal Communication: Reliability and Deception in Signaling Systems. Princeton University Press: Princeton, NJ.


Nowicki Lab
Department of Biology
Box 90338, Duke University
Durham, NC  27708-0338  USA
Lab phone: 919-684-6950