Post Doctoral Fellow

Neurobiology

Yale University (New Haven, CT)

Xiao-Jing Wang

I am interested in how the brain maintains the memory of past experiences and how it processes those memories to initiate future actions. Studies in cognitive neurobiology may help understanding how neurons integrates sensory stimuli to adjust behavior according to the internal goals and desires. My methods of investigation fall within the field of computational neuroscience, and include a combination of analysis and modeling of neuronal and behavioral data. My background is primarily theoretical physics, but I also use methods from statistical inference and machine learning.

I am particularly interested in memory and learning. During my graduate studies, I built a neurobiological network model of working memory for sequential image presentation: each image activates a separate population of neurons, representing the memory of that image, which inactivates when the memory is lost. Our model has inspired a series of experiments. However, recent neurophysiological results seems to confute our model, and I'm currently working on its modification to account for the novel findings.

During my Phd, I investigated the possibility that long-term memories are modified by learning and experience: some memories are lost and new memories are created. In addition, different memories may be erroneously intermixed or merged into a single memory. I constructed a neural network model trying to explain the biological mechanisms underlying the processes of long-term memory modifications. While the biological predictions of the model have not been tested, it has provided an explanation for different observations in human psychophysical experiments.

During my postdoctoral research, I studied the memory of stimuli that are fundamental for survival, the primary reinforcers (e.g. food and water). In neurophysiological studies, food and water rewards represent the main force driving the actions of tested animals. I studied the short-term memory of reward events and actions, and I found that the memory of different neurons decays according to different timescales. Those memories of obtained rewards and performed actions may be integrated by the animal to adjust the behavioral strategy and solve different tasks.

2002-2005 INFM fellowship (Institute of Condensed Matter), Rome

1998 E.Persico prize, Accademia dei Lincei, Rome