Dissertation
Information processing in the auditory cortex: The case studies of perceptual alternation in auditory streaming and laminar origin of surface evoked response to sounds
University of Iowa
Doctor of Philosophy (PhD), University of Iowa
Autumn 2019
DOI: 10.17077/etd.005258
Abstract
The auditory system plays a critical role in human life, providing us useful information about the surroundings. The sound waves entering our ears are composed into frequency components by the cochlear nucleus and then transduced into electrical signals by hair cells in the inner ear. The information is sent to numerous brain areas including the primary auditory cortex (A1), then the information is further processed and integrated with inputs from other sensory systems. Although the initial representations of sound in the early stage such as cochlear nucleus is well understood, much less is known about information processing in the auditory cortex. For instance, it is believed that A1 plays a critical role in formation of auditory objects or perceptual organization. There is also evidence that supports the role of A1 in the early stage of sound transformation. This doctoral thesis presents results from two research projects that study the role of A1 in representation of sound stimuli and perception.
First, I investigate how acoustic events are grouped over time to form internal mental representations of sound streams. A set of stimuli that has been used extensively in behavioral experiments consists of sequences of alternating high (A) and low (B) pure tones presented as repeated triplets, ABA-ABA-.... Depending on the frequency separation (DF) between the two tones, subjects report either of two percepts: “integration” (a single, coherent stream of high and low tones) and “segregation” (two parallel distinct streams). After the stimulus onset, it takes several seconds for the probability of stream segregation to build up, and this probability depends on DF. Multi-unit recordings from monkeys’ A1 show similarities between build-up functions of stream segregation obtained from psychophysical experiments (the psychometric functions) and those inferred from neural data (the neurometric functions). After the buildup phase, the probability of stream segregation stabilizes and alternation between percepts emerges. Here, we propose a model that implements evidence accumulation over time at the readout of spiking activity from A1 to capture buildup and perceptual alternation characteristics, during both first and subsequent percepts. Our model provides insights into the dynamical mechanisms of bistable auditory perception. In particular, the model operates within noise-induced switching regime. Bifurcation diagrams of noise and activity target levels reveal varying properties of perceptual switching, and only a small domain within that space can capture behavioral data. Our model demonstrates the interplay between external sensory evidence and internal neuronal state in shaping characteristics of mean durations. The model accounts qualitatively and quantitatively for key features of behavioral data at three DF values: the psychometric buildup; first and subsequent mean dominance durations; and first and subsequent percept duration distributions.
In the second part of my thesis I investigate A1 at the hierarchical functional level. There is evidence that shows connectivity of A1 differs greatly from those of other sensory cortices, for example the visual cortex and somatosensory cortex. Here I study how complex, coordinated activities in cortical columns can give rise to converging evoked cortical surface electrical potentials (CSEP). To this end, we used recordings from micro electrocorticographic (micro ECoG) arrays at the surface and laminar polytrode through the cortical depth of A1. They were generously provided by Dr. Bouchard's lab (based at Lawrence Berkeley National Laboratory, LBNL, California) and were obtained from experiments conducted on anesthetized female rats with discrete tone pip stimuli. In this study we model superficial response profiles as a function of discrete laminar contributions using the multiple linear regression setup. Output from the regression model is root-mean squared across time and averaged across depth (layers I-VI, respectively) to reveal depth contribution of surface. We found that spiking activity in layer I (L1) contributes to surface spiking activity. Also, surface bands such as beta, gamma, high-gamma show contribution of depth spiking activity aggregating across all layers. Further, we identified some contribution from layers IV and V (L4, L5), but more data is needed for a definite conclusion. This coupling relationship is not entirely due to correlated activity pattern and tuning similarity between depth and surface; it weakens as recordings sites are located further away from the respective cortical column.
Details
- Title: Subtitle
- Information processing in the auditory cortex: The case studies of perceptual alternation in auditory streaming and laminar origin of surface evoked response to sounds
- Creators
- Quynh-Anh Phuoc Nguyen
- Contributors
- Rodica Curtu (Advisor)Kristofer Bouchard (Committee Member)Colleen Mitchell (Committee Member)Zahra Aminzare (Committee Member)Lihe Wang (Committee Member)
- Resource Type
- Dissertation
- Degree Awarded
- Doctor of Philosophy (PhD), University of Iowa
- Degree in
- Applied Mathematical and Computational Sciences
- Date degree season
- Autumn 2019
- DOI
- 10.17077/etd.005258
- Publisher
- University of Iowa
- Number of pages
- xv, 152 pages
- Copyright
- Copyright 2019 Quynh-Anh Phuoc Nguyen
- Language
- English
- Description illustrations
- color illustrations
- Description bibliographic
- Includes bibliographical references (pages 147-152)
- Public Abstract (ETD)
In daily life, the auditory system provides us with useful information about the surroundings. The sound waves entering our ears are composed into frequency components by the cochlear nucleus and transduced into electrical signals by hair cells in the inner ear. The information is sent to several brain areas including the primary auditory cortex (Al). Then information is processed and integrated with inputs from other sensory systems. Although the initial representations of sound in the early stage such as in the cochlear nucleus are well understood, much less is known about the subsequent stages of the auditory system. The latter process high level functions that enrich our daily life such as perception, speech and music. For instance, perceptual formation of auditory objects enables us to sort the mixture of sounds from different sources into specific acoustic information, a process called auditory scene analysis. Experimental and modeling studies have showed that Al plays an important role in the early stage of sound transformation. However, the columnar organization of Al is not well understood. A question of interest is how the columnar activity in Al contributes to surface signals from different frequency bands with various functionality. In this thesis I present results from two projects that study the role of Al in percept formation and the respective contribution of columnar auditory microcircuits to surface signals in Al.
- Academic Unit
- Interdisciplinary Graduate Program in Applied Mathematical & Computational Sciences
- Record Identifier
- 9983779899702771
Metrics
30 File views/ downloads
120 Record Views