The concept of the brain as a prediction machine has enjoyed a resurgence in the context of the Bayesian brain and predictive coding approaches within cognitive science. To date, this perspective has been applied primarily to exteroceptive perception (e.g., vision, audition), and action. Here, I describe a predictive, inferential perspective on interoception: 'interoceptive inference' conceives of subjective feeling states (emotions) as arising from actively-inferred generative (predictive) models of the causes of interoceptive afferents. The model generalizes 'appraisal' theories that view emotions as emerging from cognitive evaluations of physiological changes, and it sheds new light on the neurocognitive mechanisms that underlie the experience of body ownership and conscious selfhood in health and in neuropsychiatric illness.
Interoception refers to the sensing of internal bodily changes. Interoception interacts with cognition and emotion, making measurement of individual differences in interoceptive ability broadly relevant to neuropsychology. However, inconsistency in how interoception is defined and quantified led to a three-dimensional model. Here, we provide empirical support for dissociation between dimensions of: (1) interoceptive accuracy (performance on objective behavioural tests of heartbeat detection), (2) interoceptive sensibility (self-evaluated assessment of subjective interoception, gauged using interviews/questionnaires) and (3) interoceptive awareness (metacognitive awareness of interoceptive accuracy, e.g. confidence-accuracy correspondence). In a normative sample (N=80), all three dimensions were distinct and dissociable. Interoceptive accuracy was only partly predicted by interoceptive awareness and interoceptive sensibility. Significant correspondence between dimensions emerged only within the sub-group of individuals with greatest interoceptive accuracy. These findings set the context for defining how the relative balance of accuracy, sensibility and awareness dimensions explain cognitive, emotional and clinical associations of interoceptive ability.
Granger causality is a statistical notion of causal influence based on prediction via vector autoregression. Developed originally in the field of econometrics, it has since found application in a broader arena, particularly in neuroscience. More recently transfer entropy, an information-theoretic measure of time-directed information transfer between jointly dependent processes, has gained traction in a similarly wide field. While it has been recognized that the two concepts must be related, the exact relationship has until now not been formally described. Here we show that for Gaussian variables, Granger causality and transfer entropy are entirely equivalent, thus bridging autoregressive and information-theoretic approaches to data-driven causal inference.
1 February 2014 in Journal of Neuroscience Methods
Using a set of hydrodynamical simulations of nine galaxy clusters with masses in the range 1.5 × 1014 < Mvir < 3.4 × 1015 M⊙, we have studied the density, temperature and X-ray surface brightness profiles of the intracluster medium in the regions around the virial radius. We have analysed the profiles in the radial range well above the cluster core, the physics of which are still unclear and matter of tension between simulated and observed properties, and up to the virial radius and beyond, where present observations are unable to provide any constraints. We have modelled the radial profiles between 0.3R200 and 3R200 with power laws with one index, two indexes and a rolling index. The simulated temperature and [0.5–2] keV surface brightness profiles well reproduce the observed behaviours outside the core. The shape of all these profiles in the radial range considered depends mainly on the activity of the gravitational collapse, with no significant difference among models including extraphysics. The profiles steepen in the outskirts, with the slope of the power-law fit that changes from −2.5 to −3.4 in the gas density, from −0.5 to −1.8 in the gas temperature and from −3.5 to −5.0 in the X-ray soft surface brightness. We predict that the gas density, temperature and [0.5–2] keV surface brightness values at R200 are, on average, 0.05, 0.60, 0.008 times the measured values at 0.3R200. At 2R200, these values decrease by an order of magnitude in the gas density and surface brightness, by a factor of 2 in the temperature, putting stringent limits on the detectable properties of the intracluster-medium (ICM) in the virial regions.