Cosyne 2008 Workshops

March 3-4, 2008

Snow Bird, Utah

Workshop Title

What can functional imaging tell us about population coding in sensory systems?: Bridging computation, single neurons and imaging


Justin Gardner (primary contact) (Center for Neural Science / Department of Psychology, NYU):
Alex Huk (Neurobiology / Center for Perceptual Systems / Imaging Research Center, The University of Texas at Austin):
Denis Schluppeck (School of Psychology, University of Nottingham):


Any given sensory stimulus is typically encoded by the simultaneous activity of a large number of neurons in the brain. What types of measurements can experimentalists make that will most reveal the basis of the population codes that are used by these neurons to represent the sensory world? The great majority of our knowledge is based on measurements that have been made from single or small numbers of neurons at a time-- but do these measurements miss the forest for the trees? Functional imaging techniques such as fMRI and optical imaging are capable of measuring activity over a much larger area of cortex than single electrodes; either directly through voltage sensitive dyes or indirectly through measurements of changes in local concentrations of deoxyhemoglobin. Both fMRI and optical imaging are currently capable of making measurements at the scale of cortical columns and in principle can be used to glean knowledge of sensory population coding that other techniques might not be sensitive to.

In this workshop we aim to critically evaluate what we have learned and what we could potentially learn from functional imaging studies of sensory population coding.

This workshop will facilitate the debate over how imaging techniques can complement electrophysiological measurements to characterize population coding. The talks will cover the following topics: What aspects of population activity might functional imaging be better at assessing than electrophysiological measurements (particularly single electrode measurements)? What biases (e.g. large output neurons) might electrode measurements be biased towards that could skew knowledge of populations? What biases (e.g. metabolic activity related to synaptic inputs) is functional imaging susceptible to? How might different techniques be used in conjunction to overcome each other’s limitations? Can imaging techniques tell us about higher order representations that are missed by looking at neurons one-by-one? What spatial and temporal resolution does functional imaging need to make useful inferences? What are the fundamental physiological limitations to measuring indirectly from vascular signals? Can clever experimental design and data analysis (e.g., adaptation, classification schemes) be used to overcome some of these limitations? What is the role of interneuronal correlation in population-level computations, and what is the best technique for assessing them?

This workshop is aimed at scientists of diverse backgrounds who have a common interest in population coding: Theorists working on modeling of population coding and dynamics who have a perspective on what types of measurements are needed to distinguish different theoretical proposals. Experimentalists with measurements of population activity that can guide theory. Physiologists who currently use electrophysiological measurements from single units or using multi-electrode arrays who are interested in complementary measurements of population activity with functional imaging. Those who are using functional imaging, either with fMRI or optical imaging, who are interested in applying these techniques to find a principled way of attacking problems related to population activity.


Matteo Carandini (Smith-Kettlewell) Imaging the dynamics of population responses in visual cortex
Manabu Tanifuji (RIKEN Brain Science Institute) Object representation in IT cortex at a columnar level - Comparison between columnar and single cell level representation
Eyal Seidemann (UT-Austin) Statistical properties of neural population responses in the visual cortex: Consequences for efficient neural decoding
Chou Hung (National Yang Ming University) Weak signal or no signal? How imaging and firing rate signals missed a robust functional circuit for brightness
Adam Kohn (Albert Einstein) Adaptation and population coding in visual cortex
Matt Smith (CMU) Spatial and temporal scales of correlation in spikes and the local field potential
Chris Tyler (Smith-Kettlewell) Nonlinear dynamic forward modeling of the metabolic coupling between neurons and BOLD
Denis Schluppeck (Nottingham) Imaging visual and somatosensory cortex with high resolution (f)MRI at 7T
Justin Gardner (NYU)
Alex Huk (UT-Austin)

Cosyne 16
Meeting program
Mailing lists
Travel grants

Cosyne 15
Cosyne 14
Cosyne 13
Cosyne 12
Cosyne 11
Cosyne 10
Cosyne 09
Cosyne 08
Cosyne 07
Cosyne 06
Cosyne 05
Cosyne 04