Generative models, brain function and neuroimaging

The representational capacity and inherent function of any neuron, neuronal population or cortical area in the brain is dynamic and context-sensitive. Functional integration, or interactions among brain systems, that employ d riving (bottom up) and backward (top-down) connections, mediate this adapti ve and contextual specialisation. A critical consequence is that neuronal r esponses, in any given cortical area, can represent different things at dif ferent times. This can have fundamental implications for the design of brai n imaging experiments and the interpretation of their results. Our arguments are developed under generative models of brain function, wher e higher-level systems provide a prediction of the inputs to lower-level re gions. Conflict between the two is resolved by changes in the higher-level representations, which are driven by the ensuing error in lower regions, un til the mismatch is "cancelled". From this perspective the specialisation o f any region is determined both by bottom-up driving inputs and by top-down predictions. Specialisation is therefore not an intrinsic property of any region but depends on both forward and backward connections with other area s. Because the latter have access to the context in which the inputs are ge nerated they are in a position to modulate the selectivity or specialisatio n of lower areas. The implications for classical models (e.g., classical receptive fields in electrophysiology, classical specialisation in neuroimaging and connectioni sm in cognitive models) are severe and suggest these models may provide inc omplete accounts of real brain architectures. Here we focus on the implicat ions for cognitive neuroscience in the context of neuroimaging.