Published on November 12, 2012  by Allen Institute for Brain Science

Dr. Cori Bargmann, recent winner of the Kavli Prize in Neuroscience and a pioneer in methods of looking at C. elegans to uncover how neural circuits operate, presented the idea that particular classes of genes — neuropeptides to be specific — are good places to look for the genetic origin of behavior. Because much of the genome is conserved across species and throughout time, new behaviors may be created by redeploying old genes in different ways.

Uncovering the basic building blocks of behavior, she believes, is an unsolved question in neuroscience that is now becoming solvable.

“The question is not whether calbindin is expressed in the hippocampus, but whether it has something to do with the unique functions of human memory capabilities,” Bargmann said. “That is a hard question.” Certain kinds of molecules relate to innate behaviors that are shared across ...

Published on October 12, 2012  by Allen Institute for Brain Science

“Brain rhythms and cognition”

From physics to electrophysiology and imaging, Dr. Miller and his lab focus on broad and far-reaching approaches to neuroscience questions. This symposium talk focused on questions similarly posed by Sabine Kastner, namely the relationship between action potential timing and brain function. By recording neural activity in monkeys as they switch among two tasks, Miller found that over half of the recording sites in the network for one task also appear to participate in the network for the other task. This suggests that circuitry in the prefrontal cortex may overlap and that oscillations are the key to selecting appropriate networks for the task that needs to be performed. Miller showed that neural ensembles, or networks, in close proximity oscillate out of phase with one another to avoid being simultaneously activated. That is, the oscillations bind neural ensembles ...

“Deciphering the mammalian brain transcriptome”

Published on November 12, 2012  by Allen Institute for Brain Science

The development, structure and function of our brains are guided by selective usage of the 20,000-odd genes in our genomes. Taking advantage of the Allen Institute’s anatomically and genomically comprehensive atlases of gene expression in the developing brain in species from mouse to human, Dr. Lein explores the molecular logic of gene expression in the brain. How do gene expression profiles relate to the functional and cellular architecture of the brain, and what makes the human brain unique? Focusing on the human neocortex, the outermost layer of the brain, Lein showed that molecular similarities reflect spatial proximity across the neocortex; such that neighboring cortical regions are more similar to one another than to more distant regions. Surprisingly, these molecular similarities vary across the cortex in a graded fashion, in contrast to models of cortical architecture ...

“The role of thalamo-cortical interactions in spatial attention”

Similar to the questions posed by Ila Fiete and Earl Miller, Dr. Kastner asks: How do large-scale networks achieve cognition? How do the connections between molecules and neurons give rise to complex visual processing that we as primates enjoy? The basic premise of her work relies on the assumption that spike timing may play a critical role in brain function, particularly for higher cognitive functions, and that how neurons are connected may have important implications on the function of their networks. By simultaneously recording neural activity from three areas in the brain’s attentional network, Kastner has developed a hypothesis that identifies one area as performing a synchronizing function between the other two in order to optimize the transmission of information between them. She maintains that the pulvinar, a collection of nuclei in the back of the thalamus, is strongly connected to the ...

“A new class of neural population codes”

Dr. Fiete is a theoretician, the only member of such discipline at this year’s symposium, whose work relates closely to various approaches of other speakers. An Alfred P. Sloan Foundation Fellow, a Searle Scholar, and a McKnight Scholar, Fiete studies coding of sensory or memory information in neural populations, ultimately seeking to unravel the meanings in such patterns. She aims to reduce noise-induced errors in measurements and recording of brain activity by searching for a type of neural population code that, for over six decades now, has been known to exist theoretically but has not been unambiguously described. Most neural codes in the sensory and motor cortices tend to follow a classical population code, where accuracy increases with population size. These, however, are considered weak codes. The brain makes up for this, posits Fiete, by organizing in an interleaving system she compared to ...

“Neurotechnology: Fixing broken brains by decoding cortex”

Dr. Donoghue’s lab works in both basic and applied neuroscience, studying network computation on one end, and restoring lost motor cortical function on the other end. His work has culminated in BrainGate, a human brain-computer interface that senses neural activity in the motor cortex and decodes it to instruct a robotic arm or a cursor, reconnecting the cortex to lost action in the case of spinal cord injury or paralyzing disease such as amyotrophic lateral sclerosis. Donoghue showed video clips from his lab featuring long-paralyzed patients successfully navigating a robotic arm to drink their favorite beverage or guiding a computer cursor to communicate with loved ones. Among other noteworthy aspects of this work are the two notions that 1) the program actually decodes activity in the motor cortex rather than using a binary measure of activity (e.g., on or off) and 2) individual neurons in the motor cortex retain activity and function, even in quadriplegics who have not used this area of the brain in over a dozen years. While the functional utility of this work is self-evident, the value and applications to basic neuroscience are no less important. In going from a plan ...

“Using human genetics and stem cells to study brain development”

Ricardo Dolmetsch, recently joining the Allen Institute team from Stanford University, discussed his plans for a research program in molecular networks. The goals of the program, he described, are to 1) identify the molecular networks that control neuronal development and function, 2) understand the cellular and molecular basis of neurdevelopmental diseases, and 3) generate public resources that facilitate the study of brain development and developmental diseases. In a landscape of increasing data collection and “big science”, Dolmetsch suggested a refined and potentially more efficient tactic towards reaching these goals: use human genetic diversity to prioritize data collection. The idea is to genotype human cells to identify critical signaling pathways and molecular hubs that regulate brain development and are affected in disease. The neuroscience community has been uncovering mountains of copy number variants (CNVs) and rare mutations (RMs) that tip the ...

Functional magnetic resonance imaging (fMRI) detects the location of functions in the brain better than any other method we have today. While localization is necessary, it is not sufficient for understanding how the brain works. Dr. Gallant suggests the reverse approach – to search for functional maps. That is, he uses brain activity to determine or reconstruct what a subject was looking at. To this end the Gallant lab has constructed the WordNet model, which is able to predict what an observer is seeing from 2,000 nouns and verbs. The process uses brain activity in fMRI to predict from semantic models while an observer watches a video, and the results are remarkably accurate. Dr. Gallant explains how encoding models, decoding models, and functional maps of the brain are all closely related. “Once you have encoding, you get decoding for free,” he proclaims.

Published on November 13, ...