Abstract

AbstractThis thesis is divided into six chapters (following Introduction). Each chapter was intended to be self-contained, so they do not have to be read in the order they are presented. Second chapter (Sec. 2) contains a detailed description of experimental techniques: surgery, recording, and training techniques we used in awake head-fixed rats. We have also included a detailed description of all sets of stimuli we used to probe neurons, analytical methods used to analyze data, and description of computational models used in other parts of the thesis.Third chapter (Sec. 3) focuses on description of single-neuron responses in primary auditory cortex of awake head-fixed rats. The primary emphasis of this part is on the sparse representation of various auditory stimuli we used to probe neurons, and the heterogeneity of responses of single neurons. To characterize population responses to sound in the auditory cortex we asked the question "What is the typical response to acoustic stimuli?" instead of what is usually asked "What is the stimulus that evokes a response?" We found that the population response was sparse, with many unresponsive neurons. In addition, the responsive neurons showed a great variety of responses. This heterogeneity of neuronal responses ("response zoo," courtesy of Anthony M. Zádor) was, however, surprisingly well characterized by lognormal distribution of firing rates. The observation that firing rates in awake auditory cortex were lognormally distributed was even more interesting given the observation of lognormal distribution of synaptic weights in the cerebral cortex.The fourth chapter (Sec. 4) focuses on mechanisms which could give rise to lognormal distribution of firing rates, as well as synaptic weights. We proposed specific types of correlations among synaptic connections, and formulated a multiplicative learning rule which led to the observed distributions. We were also able to characterize intracellular activity of neurons in awake auditory cortex. The fifth chapter (Sec. 5) contains analysis of so-called up and down states in awake auditory cortex. We show that up and down states—the "signature" subthreshold dynamics so often described in various cortical areas of anesthetized animals—were rare in the primary auditory cortex of awake rats, instead, subthreshold dynamics was consisted of brief, infrequent fluctuations of membrane potential.The experiments described and analyzed in chapters 2—-4 were conducted in naïve awake rats. As behavior or attention can influence neuronal activity even in primary sensory areas, we developed a setup for head-fixed behavior. In the sixth chapter (Sec. 6) we describe the sound discrimination task we have used to study behavior in head-fixed rats. We present a comparison of basic behavioral parameters between restrained and unrestrained rats, as well as evidence of nonauditory modulations of single neuron activity in auditory cortex.

Highlights

  • The brain is the most complex computational device known to Man

  • We show that up and down states—the “signature” subthreshold dynamics so often described in various cortical areas of anesthetized animals—were rare in the primary auditory cortex of awake rats, instead, subthreshold dynamics was consisted of brief, infrequent fluctuations of membrane potential

  • To create a corresponding Gaussian distribution of firing rates we matched the mean firing rate and entropy of the lognormal distribution, which corresponded to a Gaussian with a mean of 4.2 sp/s and a standard deviation of 5.2 sp/s, on a linear scale, with negative firing rates replaced by rates drawn again from the same distribution until the distribution contained only non-negative firing rates

Read more

Summary

Introduction

Does it mediate our orientation in both external (physical) and internal worlds, but—even more astonishingly—the brain enables study of itself This amazing device is composed of only a limited set of neurons and their connections. External inputs (light, sound, touch, etc.) are detected by sensory receptors, translated into internal representations, which in turn are interpreted into percepts, and eventually lead to (motor) actions. Such transformations— especially in the beginning of the processing chain—are the main topics of sensory neuroscience

Objectives
Results
Discussion
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call