Anna Schapiro
Anna Schapiro

Photo of Anna

I will be starting up a new lab at the University of Pennsylvania in July 2019.

I am a post doc in Bob Stickgold's lab at Beth Israel Deaconess Medical Center / Harvard Medical School. I am interested in how we learn and represent information in our environment, especially structured information with temporal or semantic relationships that require integration over time. My research has explored the initial stages of acquiring such information as well as the offline consolidation processes that shape initial memory representations for longer term storage and use. I have found in fMRI experiments and in a patient case study that the hippocampus is involved in the rapid learning of novel temporal regularities. These findings are, on the surface, difficult to reconcile with the idea that the hippocampus specializes in memorizing the specifics of individual experiences, keeping related episodes separate to avoid interference. I have shown using a neural network model of the hippocampus, however, that these functions may coexist in separate anatomical pathways within the hippocampus. In my work on memory consolidation, I have found that sleep and offline replay benefit new semantic memories, with a particular focus on weaker memories -- those most in need of further processing. I am now developing a neural network model of learning and hippocampal-cortical interactions during sleep.

I was a grad student at Princeton with Ken Norman, Matt Botvinick, and Nick Turk-Browne, and worked with Jay McClelland as an undergrad at Stanford.

| Google Scholar