Event Date and Time
1103 Bioscience Research Building

Investigating effects of stimulus type and selective attention on the subcortical encoding of sounds

The human auditory system is remarkable in its ability to derive meaning and even emotion from the vibrations of air molecules. This process depends on many linked subprocesses, which begin with the encoding of a sound’s acoustics, the extraction of features like pitch, location, and timbre, and further abstraction from there. Additionally, these processes are shaped by a listener’s experience and goals, such as musical training or when a person decides to focus on one person talking at a party, ignoring all other sounds. Many studies have revealed strong effects of these higher order factors on the encoding of sounds in the cortex, but subcortical studies are fewer and results are more mixed. The potential for such effects exists: there are as many efferent (feedback) connections to subcortical auditory areas as there are afferent (feedforward) ones. In this talk I will first present a framework my lab has developed for studying the subcortical encoding of natural sounds like speech and music. I will then discuss the results of two studies using that framework. In the first study we presented listeners with several different genres of music and speech stimuli and measured subcortical and cortical responses to investigate where differentiated encoding of the two stimulus classes begins. In the second study we presented listeners with competing audiobook stimuli and asked them to attend to one. We then assessed the effect of selective attention across the auditory system, from the auditory nerve through the brainstem and cortex. The results of both studies are suggestive that role of subcortical areas is mostly to extract the features on which higher order, cognitively-driven cortical effects are based.

Dr. Ross Maddox is an Associate Professor at the University of Michigan.


Dr. William Hodos NACS Seminars are free and open to the public.

Picture of Dr. Maddox