Christina Weaver

  • Adjunct Instructor of Neuroscience
  • Mount Sinai School of Medicine
  • Assistant Professor of Mathematics
  • Department of Mathematics and Computer Science
  • Franklin & Marshall College
  • P.O. Box 3003
  • Lancaster, PA 17604-3003
  • Phone: 1-717-291-3872
  • Fax: 1-727-358-4507
  • Email: christina.weaver@fandm.edu

I have been a central member of CNIC since arriving at Mount Sinai in 2003. Since 2009 I have been an Assistant Professor of Mathematics at Franklin & Marshall College (Lancaster, PA), but I maintain very active collaborations within CNIC. Research interests in my laboratory include computational neuroscience and image analysis. In particular, we use computational modeling to study cellular mechanisms underlying working memory, and how those mechanisms are affected by aging and neurodegenerative disease. The models are constrained by morphologic and electrophysiological data collected within CNIC, resulting in multidisciplinary studies that demonstrate our unique approach to research.

In humans and nonhuman primates, higher cognitive function involving working memory depends critically upon the integrity of circuits in area 46 of the prefrontal cortex (PFC). At a more basic level, neurons of the precerebellar nucleus Area II in goldfish are necessary for eye velocity storage, a mechanism that displays persistent activity after extinguishing visual or vestibular stimuli. Persistent activity is found in many different brain regions, and is believed to be a mechanism responsible for working memory. We study how morphology and electrical membrane properties interact to shape firing patterns in these different working memory systems.

Members of my lab develop both simplified and morphologically faithful models of neurons that are consistent with experimental data. This requires methods for automated parameter optimization, a research area that we actively pursue. We design fitness functions capable of capturing a range of neuronal dynamics, used to quantify salient differences between experimental data and model output, and carefully consider which optimization methods to employ.

An intriguing outcome of our research is the development of novel methods for sensitivity analysis, to quantify how parameters and their interactions affect model output. Significantly, we can predict which electrical parameters to change, and by how much, will to compensate for a given age-, disease-, or development-related morphologic change, to restore normal function. We are extending these methods in new directions, and applying them to our current modeling problems to help us elucidate the mechanisms that shape neuronal function.