Gabriel Kreiman, Ph.D.

Department of Ophthalmology
Swartz Center for Theoretical Neuroscience
Center for Brain Science
Harvard Medical School/ Children's Hospital
Karp Family Research Building, 11th Floor, Room 217
1 Blackfan Circle, Boston, MA 02115

tel: (617) 919-2530; fax: (617) 432-7828
email: Gabriel.kreiman@tch.harvard.edu
Web: http://klab.tch.harvard.edu

Research Interests and Directions:

The Kreiman Lab is interested in the neuronal circuits and computational mechanisms involved in visual recognition, learning and memory formation. In a fraction of a second, our brains are capable of solving challenging pattern recognition problems such as identifying an object. By combining multi-electrode neurophysiological recordings, theory and computational modeling, we aim to shed light on the magic transformation from pixel-like input patterns in the retina to our complex, transformation-invariant and selective visual perception. Specific research themes include: (1) the role of top-down feedback signals in visual recognition; (2) how the ventral visual stream transforms signals to achieve visual selectivity and invariance; (3) applications to biologically inspired computer vision and (4) applications to brain-machine interfaces.

Another problem that we are interested in centers around the biophysical rules and neuronal mechanisms underlying learning.  Our brains show a remarkable capability to learn and to abstract. This ability to generalize arguably represents a central aspect of intelligence and a major driving force in the evolution of neocortex. We combine neurophysiology and biophysically plausible models to investigate learning and generalization. Specific research themes include: (1) the interactions between neocortex and the hippocampus during memory consolidation; (2) learning to generalize during visual recognition; (3) biophysical rules for spike-timing dependent synaptic plasticity and (4) applications to brain-machine interfaces.

 

Selected Publications:

Burbank, K.S., Kreiman, G.K. (2012). Depression-Biased Reverse Plasticity Rule Is Required for Stable Learning at Top-Down Connections. PLoS Computational Biology 8(3):1-16.

Fried, I., Mukamel, R., Kreiman, G. (2011). Internally Generated Preactivation of Single Neurons in Human Medial Frontal Cortex Predicts Volition. Neuron. 69: 548-562.

Agam, Y., Liu, H., Pappanastassiou, A., Buia, C., Golby, A.J., Madsen, J.R., Kreiman, G. (2010). Robust selectivity to two-object images in human visual cortex. Current Biology 20:872-879.

Kim, T.K.*, Hemberg, M.*, Gray, J.M., Harmin, D.A., Kuersten, S., Costa, A., Haley, K., Papadimitriou, E., Kreiman, G., Greenberg, M.E.  (2010).  Widespread transcription at thousands of enhancers during activity-dependent gene expression in neurons.  Nature  465(7295): 182-7.

Liu, H., Agam, Y., Madsen, J., Kreiman, G. (2009).  Timing, timing, timing: Fast decoding of object information from intracranial field potentials in visual cortex.  Neuron 62: 281-290.

Rasch, M., Logothetis, N.K., Kreiman, G.  (2009). From neurons to circuits: linear estimation of local field potentials.  Journal of Neuroscience  29: 13785-13796.

 

Page created and maintained by Xaq Pitkow