Past neuroscience and psychology studies have shown that people’s expectations of the world can influence their perceptions, either by directing their attention to expected stimuli or by reducing their sensitivity (i.e., perceptual acuity) to variations within the categories of stimuli we expect to be exposed to.

While the effects of expectations on perceptions are now well-documented, their neural underpinnings remain poorly understood.

Researchers at University of California San Diego (UC San Diego) carried out a study involving songbirds aimed at better understanding how expectation-fueled biases in perception shape brain activity and behavior.

Their findings, published in Nature Neuroscience, suggest that the perceptions of songbirds, like those of humans, are influenced by expectations, with peripheral sensory systems utilizing expectations to enhance sensory perception and retain high-fidelity representations of the world.

“This work was inspired by an observation about human speech, namely that listeners are able to comprehend speech even though there is a great degree of variability in the sound entering their ears,” Tim Sainburg, first author of the paper, told Medical Xpress.

“Not only are we tasked with understanding speech in noisy environments, but we also have to deal with variability in the actual speech signal.”

Human speakers are known to have different voices, while also pronouncing many words differently. Past studies suggest that the human brain possesses robust underlying mechanisms designed to address these differences, by grouping speech sounds into stable perceptual categories, a process referred to as “categorical perception.”

“One of these mechanisms is that we use context to cue and bias our perception,” said Sainburg. “The goal of our study was to understand how that bias works in behavior and in the brain.”

Timothy Q. Gentner’s lab at UC San Diego, which Sainburg is a part of, often examines the vocal behavior and perceptions of songbirds. This is because songbirds are known to share many similarities with humans in terms of their vocal behavior, thus studying them can help to better understand human speech and speech-related perceptions.

“Behaviorally, we were interested in how expectation biases perception in songbirds,” explained Sainburg.

“To study this, we needed to be able to synthesize birdsongs, so we developed generative neural network models that could create synthetic songs. We then played these songs back to the bird, and modulated their expectations over what they would hear.”

The team’s initial experiments utilizing synthesized birdsongs showed that, similarly to humans who are listening to others speak, the perceptions of songbirds while listening to birdsongs are biased by their expectations. Subsequently, Sainburg, Gentner and their colleagues carried out further experiments aimed at understanding the brain processes involved in the expectation-guided perceptions they observed.

“We investigated the neural basis by recording the electrical activity of populations of neurons in the brain while birds listened to and classified the synthetic songs,” said Sainburg. “We wanted to understand whether the sensory brain showed the same signs of bias as in behavior, or if that bias occurs elsewhere.”

Overall, this study confirmed the hypothesis that the song perceptions of songbirds closely resemble the speech perceptions of humans. Specifically, it gathered strong evidence suggesting that the vocal perceptions of songbirds are also biased and influenced by expectations.

“This is important because many properties of human speech are unique,” said Sainburg.

“The songbird song system is really the closest animal model we have to human language in many ways. For example, many songbirds are vocal learners and produce complex vocal syntax. In this paper, we found another property of human speech which is shared with songbirds.”

The second important finding of this study emerged from the team’s second experiment probing the neural basis of context-dependent categorical perception in songbirds. While their first experiment showed that the birds’ expectations influenced how they classified songs, the second was aimed at determining whether the birds’ sensory systems reflected this shift in perception.

“Some researchers hypothesized that the sensory brain would integrate expectations with the incoming sound, and so it was one of our hypotheses going in,” said Sainburg.

“To our surprise, we found this not to be the case at all. Instead, the sensory brain appears to use expectation in a more clever way, by rededicating neural responses to focusing on relevant, expected signals, improving perceptual acuity.

“It then leaves the bias to downstream processing like motor and decision-making regions of the brain. In this way, the brain can retain high-fidelity, unbiased, representations of the world, while still incorporating bias to make optimal decisions.”

The results gathered by Sainburg, Gentner and their colleagues could soon inspire further research focusing on the neural underpinnings of expectation-driven shifts in perception, both in songbirds and humans. These efforts could shed further light on the intricacies of speech and vocal perception, as well as their supporting brain processes.

“In this work, we’ve shown that songbirds probabilistically integrate expectation in their song perception, but that their sensory brains remain unbiased,” added Sainburg.

“We still need to understand if this is a property specific to the songbird auditory system, or if we see it in other modalities and species, like human vision and language.”

© 2025 Science X Network