TY - JOUR T1 - Predictive coding with neural transmission delays: a real-time temporal alignment hypothesis JF - eneuro JO - eNeuro DO - 10.1523/ENEURO.0412-18.2019 SP - ENEURO.0412-18.2019 AU - Hinze Hogendoorn AU - Anthony N Burkitt Y1 - 2019/03/28 UR - http://www.eneuro.org/content/early/2019/03/28/ENEURO.0412-18.2019.abstract N2 - Hierarchical predictive coding is an influential model of cortical organization, in which sequential hierarchical levels are connected by backward connections carrying predictions, as well as forward connections carrying prediction errors. To date, however, predictive coding models have largely neglected to take into account that neural transmission itself takes time. For a time-varying stimulus, such as a moving object, this means that backward predictions become misaligned with new sensory input. We present an extended model implementing both forward and backward extrapolation mechanisms that realigns backward predictions to minimize prediction error. This realignment has the consequence that neural representations across all hierarchical levels become aligned in real-time. Using visual motion as an example, we show that the model is neurally plausible, that it is consistent with evidence of extrapolation mechanisms throughout the visual hierarchy, that it predicts several known motion-position illusions in human observers, and that it provides a solution to the temporal binding problem.Significance Statement Despite the enormous scientific interest in predictive coding as model of cortical processing, most models of predictive coding do not take into account that neural processing itself takes time. We show that when the framework is extended to allow for neural delays, a model emerges that provides a natural, parsimonious explanation for a wide range of experimental findings. It is consistent with neurophysiological data from animals, predicts a range of well-known visual illusions in humans, and provides a principled solution to the temporal binding problem. Altogether, it explains how predictive coding mechanisms cause different brain areas to align in time despite their different delays, and in so doing explains how cortical hierarchies functions in real-time. ER -