One approach for understanding the nature of neuronal computations involves the study of coherent spatio-temporal patterns that encompass many or all of the neurons in the intact brain. A powerful tool that facilitates this approach is the use of molecular indicators of the cell membrane potential to report the electrical dynamics of populations of neurons. When applied to invertebrates and lower vertebrates, studies that utilized this technical approach have shown that nervous systems represent sensory information and intended motor output on the scale of hundreds of neurons. This representation takes the form of traveling waves and other patterns whose form cannot be predicted on the basis of anatomy or conventional electrical recordings. Technology for the extension of such studies to the mammalian brain, on a routine basis, is the major goal of this work. The investigators propose to develop and build a multi-channel optical detector whose novel features, design flexibility and cost effectiveness should significantly advance the use of molecular indicators to report neuronal function. Numerical techniques for the removal of physiological noise contributions to the dynamic imaging data will also be further developed. There is a need for improved detection in the area of voltage-sensitive dye signals. These signals appear as small changes in light levels (11 part in 10(-5) to 10(-3). Further, they are modulated by both rhythmic and fluctuating physiological changes in the intrinsic optical properties of cortex. The proposed multi-site sensor will have the sensitivity, dynamic range, resolution and bandwidth to resolve the electrical contributions to the dye signal on top of these fluctuations. There is a general need for a modular, robust, and readily available multi- site detector in the cell biology and neurobiology communities. The proposed device can be used for the study of ions as well as electrical activity in a variety of cellular and tissue systems, both in vitro and in vivo, in which large dynamic range and quantum limited noise is essential. Present systems are mainly limited to the measurement of signals whose amplitude is well above theoretical noise limitations. There is a need for robust algorithms that will aid in the real-time analysis of single-trial dynamic brain images. These include techniques to estimate the contributions from vasomotion, respiration, and cardiac pulsation from the data and techniques to extract the weakly correlated components of functional activity. The proposed device is a research tool for the study of brain function in both the normal state and pathological states. These proposed studies of stimulus representation in the rat vibrissa sensorimotor pathway will serve as a scientific test-bed for the proposed technology.
| Taylor, Adam L; Cottrell, Garrison W; Kleinfeld, David et al. (2003) Imaging reveals synaptic targets of a swim-terminating neuron in the leech CNS. J Neurosci 23:11402-10 |
| Lev-Ram, Varda; Mehta, Samar B; Kleinfeld, David et al. (2003) Reversing cerebellar long-term depression. Proc Natl Acad Sci U S A 100:15989-93 |
