Simulating PET/fMRI studies of visual and auditory pattern recognition using biologically realistic large-scale neural models

Formidable conceptual problems exist in interpreting human functional neuroimaging data in terms of the underlying neural activity. To surmount these difficulties, we have developed two neurobiologically realistic models (one for vision, one for audition) of the object recognition pathway in human neocortex in which data at multiple spatiotemporal levels can be simulated and cross-validated by multiple disciplines, including PET and fMRI. Our models, based on neurophysiological and neuroanatomical data from primate and human studies, enable us to simultaneously simulate cellular electrophysiological and PET/fMRI activities in multiple, interconnected brain regions. These types of models enable us to quantitatively combine multiple data sets so that a coherent account of human cognition can be generated We illustrate this approach using delayed match-to-sample (DMS) tasks for visual shape and for auditory tonal patterns