Acoustic communication involves a signaller (the caller) and a receiver (the listener) – and their interests often conflict. In competition or mate choice, for example, signallers can reap huge benefits by deceptively exaggerating (or sometimes minimising) fitness-related traits like their true body size. At the same time, evolution should favour receivers who can detect such deceptive signals and recalibrate their behavioural responses. This honest signalling paradox has perplexed scientists for over 50 years, posing the critical question: how can deceptive signals exist in an evolutionarily stable signalling system?
In our study published in Nature Communications, we address this foundational question by combining our research expertise on size exaggeration in mammals (Reby) and vocal communication of body size in humans (Pisanski) to examine the production and perception of deceptive vocal signals.
Two major obstacles have traditionally limited the extent to which researchers could study deceptive signalling using animal models: uncertainty in what a signal is actually intended to convey, and uncertainty in what the receiver is attending to. For example, we knew too well from our previous research that while male deer often modulate their roars to ostensibly exaggerate their perceived body size (see video below) , it is very hard to assert that the acoustic features of the roar actually serve this function, and even harder to assess whether receivers detect and/or fall for such deception: we can’t just ask them!
While scientists have long studied animals to better understand our own species, we decided to reverse this traditional approach to show that studying the human animal can answer difficult-to-test questions about animal behaviour more generally.
We realised that we can get around these obstacles by asking human speakers to sound bigger or smaller (and also to speak naturally) analysing the acoustic structure of their voices in all three conditions, and then directly assessing the effects of these ‘deceptive’ and ‘honest’ vocal signals on human listeners in a series of controlled psychoacoustic experiments. Listeners were simply asked to judge the true heights of the vocalisers, and second, to judge whether they thought the vocalisers were speaking naturally or trying to exaggerate or attenuate their true size.
We show that listeners can often detect deceptive signals, and that correct detection of deception can indeed lead them to effectively recalibrate their height judgments, especially in the case of men judging the heights of other men. This reduces errors and thus can mitigate the potential costs of being deceived.
This research provides answers to long-standing questions about reliability in animal communication: Yes, honesty can be retained in deceptive signals. Yes, listeners can often correctly detect deceptive vocal signals, and yes, they recalibrate their height judgments when they correctly and concurrently detect a ‘cheater’. Finally, and importantly, yes, it still pays off for vocalisers to deceive. This is because deceit goes undetected about half of the time, so listeners’ height judgements remain biased by deception overall.
We hope that these insights will help to design future studies to test whether these findings generalise to nonhuman species and to other communication channels, such as deceptive visual signals.
Find out more in our freely available paper: Pisanski, K. & Reby, D. (2021). Efficacy in deceptive vocal exaggeration of human body size Nature Communications.
Katarzyna Pisanski & David Reby
www.ENESlab.com | CNRS & Jean Monnet University of Saint-Etienne, France