More than one hundred years ago, Hermann von Helmholtz proposed that perception was a process of inference on the external causes of sensory signals.  Recently this view has become increasingly powerful, and is now often known as ‘the Bayesian Brain’ or ‘predictive coding’.  At heart, this view suggests that the expectations, or predictions, are what determines the contents of perception – not simply the ‘bottom-up’ processing of sensory signals.  At the Sackler Centre, we have pioneered the application of this framework to interoception, the sense of the internal state of the body.  Put simply, ‘interoceptive predictive coding’ is a theoretical framework within which emotions – or subjective feeling states – are taken to arise from the brain’s predictions about the internal state of its own body.  We are now testing this framework using novel combinations of visual-cardiac feedback and augmented reality, and we believe that the work carries important implications for understanding neuropsychiatric conditions such as depersonalization, derealisation, as well as anxiety disorders.  This work is part of an EU integrated project CEEDS (www.ceeds-project.eu).

 People: Anil Seth, Keisuke Suzuki, Hugo Critchley

Reference:

Seth, A.K, Suzuki, K., and Critchley, H.D.  (2011). An interoceptive predictive coding model of conscious presence. Frontiers in Psychology: Consciousness Research. 2:e395