Kamal Sen

Associate Professor, Neural coding of sound

  • Title Associate Professor, Neural coding of sound
  • Office ERB 414B
  • Phone (617) 353-5919
  • Education M.A., Ph.D., Physics, Brandeis University
    B.A., Physics, Bates College
  • Additional AffiliationsAssociate Chair for Graduate Studies, Biomedical Engineering
  • Areas of InterestNeural Coding of Natural Sounds: Theoretical methods for characterizing auditory receptive fields, statistics of natural sounds, neural encoding and decoding, information theoretic approaches to characterizing efficiency of coding
    Hierarchical Auditory Processing: Characterizing successive stages of auditory processing and changes in neural selectivity to natural sounds.
    Neural Discrimination: Quantifying neural discriminability of different classes of behaviorally relevant sounds, dynamics of discrimination.
    Population Coding of Natural Sounds: Coding of natural sounds in neural populations, quantitative methods for detecting patterns of population activity, dynamics of correlation, information in patterns of population activity.
    Learning in Single Neurons and Auditory Networks: Changes in receptive field structure during learning, dependence of receptive field structure on the statistics of natural sounds.
  • Research AreasHow do neurons in the brain encode complex natural sounds? What are the neural substrates of selectivity for and discrimination of different categories of natural sounds? Are these substrates innate or shaped by learning?Our laboratory investigates these questions with a focus on auditory cortex. Electrophysiological techniques are used to record neural responses from hierarchical stages of auditory processing. Theoretical methods from areas such as statistical signal processing, systems theory, probability theory, information theory and pattern recognition are applied to characterize how neurons in the brain encode natural sounds. Computational models are constructed to understand the processing of natural sounds both at the single neuron and the network level, to model neural selectivity and discrimination, and to explore the role of learning in shaping the neural code.

View all profiles