Project Summary
In order to accurately perceive the world and respond accordingly, the brain has to deal with noise
inherent in sensory cues. One method is for the brain to learn which cues are reliable across contexts and rely
on these cues in the future. Previous studies have demonstrated that sensory systems are able to actively
learn recent statistics underlying cue reliability and adapt to short term changes. However, relatively little is
known about whether and how sensory systems adapt to natural statistics of sensory cues that are expected to
be permanent. While humans display biases based on long term patterns of sensory cue statistics, the neural
mechanisms underlying this are unknown. To address these knowledge gaps, this proposal will investigate
whether and how anticipated reliability of binaural spatial cues determines sound localization of barn owls.
Barn owls compute the interaural time difference (ITD) to determine azimuthal location. Previous work showed
that the signal-to-noise ratio of ITD varies across frequencies in a location dependent manner, based on the
acoustical properties of the head. Thus, certain frequencies convey more reliable ITD cues for sounds from a
given location, and the neural tuning in the midbrain is optimized to reflect this pattern. This provides a system
where the reliability of natural acoustic cues is represented, allowing for the investigation of the effect of
anticipating sensory statistics on perceptual discriminability and how it is learned. Recent work by our group
has shown evidence that human auditory perception is also shaped by ITD statistics, potentially allowing the
properties of neural coding investigated in this project to explain the bases of these findings. The overall
hypothesis is that barn owls learn to anticipate cue reliability during juvenile critical periods, leading to
biases in neural and behavioral processing. Behavioral and electrophysiological approaches along with
computational methodology will be used to test this hypothesis. Experiments outlined in Aim 1 will determine
whether barn owls show better sound discrimination for frequencies anticipated to be reliable. Aim 2 will
investigate whether and how the neural population displaying the pattern of ITD-frequency tuning described in
the midbrain explains spatial discriminability. Novel multi-unit recordings in the midbrain in vivo along with
computational techniques will be used to address this. Aim 3 will elucidate the capacity of the barn owl to learn
altered forms of cue reliability, by changing the acoustical properties of the owl’s head at different stages of
development. Altogether, this proposal will provide understanding on the mechanisms used to deal with long
term patterns of cue reliability. The data will offer insight into how the auditory system learns the natural
statistics underlying sensory noise to optimize perception, as well as its capacity to adapt to unanticipated
changes.
