Functional integration and inference in the brain - PubMed (original) (raw)
Review
Functional integration and inference in the brain
Karl Friston. Prog Neurobiol. 2002 Oct.
Abstract
Self-supervised models of how the brain represents and categorises the causes of its sensory input can be divided into two classes: those that minimise the mutual information (i.e. redundancy) among evoked responses and those that minimise the prediction error. Although these models have similar goals, the way they are attained, and the functional architectures employed, can be fundamentally different. This review describes the two classes of models and their implications for the functional anatomy of sensory cortical hierarchies in the brain. We then consider how empirical evidence can be used to disambiguate between architectures that are sufficient for perceptual learning and synthesis. Most models of representational learning require prior assumptions about the distribution of sensory causes. Using the notion of empirical Bayes, we show that these assumptions are not necessary and that priors can be learned in a hierarchical context. Furthermore, we try to show that learning can be implemented in a biologically plausible way. The main point made in this review is that backward connections, mediating internal or generative models of how sensory inputs are caused, are essential if the process generating inputs cannot be inverted. Because these processes are dynamical in nature, sensory inputs correspond to a non-invertible nonlinear convolution of causes. This enforces an explicit parameterisation of generative models (i.e. backward connections) to enable approximate recognition and suggests that feedforward architectures, on their own, are not sufficient. Moreover, nonlinearities in generative models, that induce a dependence on backward connections, require these connections to be modulatory; so that estimated causes in higher cortical levels can interact to predict responses in lower levels. This is important in relation to functional asymmetries in forward and backward connections that have been demonstrated empirically. To ascertain whether backward influences are expressed functionally requires measurements of functional integration among brain systems. This review summarises approaches to integration in terms of effective connectivity and proceeds to address the question posed by the theoretical considerations above. In short, it will be shown that functional neuroimaging can be used to test for interactions between bottom-up and top-down inputs to an area. The conclusion of these studies points toward the prevalence of top-down influences and the plausibility of generative models of sensory brain function.
Similar articles
- Learning and inference in the brain.
Friston K. Friston K. Neural Netw. 2003 Nov;16(9):1325-52. doi: 10.1016/j.neunet.2003.06.005. Neural Netw. 2003. PMID: 14622888 - Beyond phrenology: what can neuroimaging tell us about distributed circuitry?
Friston K. Friston K. Annu Rev Neurosci. 2002;25:221-50. doi: 10.1146/annurev.neuro.25.112701.142846. Epub 2002 Mar 19. Annu Rev Neurosci. 2002. PMID: 12052909 Review. - A theory of cortical responses.
Friston K. Friston K. Philos Trans R Soc Lond B Biol Sci. 2005 Apr 29;360(1456):815-36. doi: 10.1098/rstb.2005.1622. Philos Trans R Soc Lond B Biol Sci. 2005. PMID: 15937014 Free PMC article. Review. - Forward and backward connections in the brain: a DCM study of functional asymmetries.
Chen CC, Henson RN, Stephan KE, Kilner JM, Friston KJ. Chen CC, et al. Neuroimage. 2009 Apr 1;45(2):453-62. doi: 10.1016/j.neuroimage.2008.12.041. Epub 2008 Dec 31. Neuroimage. 2009. PMID: 19162203 - Acquisition of nonlinear forward optics in generative models: two-stage "downside-up" learning for occluded vision.
Tajima S, Watanabe M. Tajima S, et al. Neural Netw. 2011 Mar;24(2):148-58. doi: 10.1016/j.neunet.2010.10.004. Epub 2010 Oct 27. Neural Netw. 2011. PMID: 21094592 Review.
Cited by
- Evolution of Telencephalon Anterior-Posterior Patterning through Core Endogenous Network Bifurcation.
Sun C, Yao M, Xiong R, Su Y, Zhu B, Chen YC, Ao P. Sun C, et al. Entropy (Basel). 2024 Jul 26;26(8):631. doi: 10.3390/e26080631. Entropy (Basel). 2024. PMID: 39202101 Free PMC article. - Assembloid learning: opportunities and challenges for personalized approaches to brain functioning in health and disease.
Mencattini A, Daprati E, Della-Morte D, Guadagni F, Sangiuolo F, Martinelli E. Mencattini A, et al. Front Artif Intell. 2024 Apr 19;7:1385871. doi: 10.3389/frai.2024.1385871. eCollection 2024. Front Artif Intell. 2024. PMID: 38708094 Free PMC article. No abstract available. - Markov Blankets and Mirror Symmetries-Free Energy Minimization and Mesocortical Anatomy.
Wright J, Bourke P. Wright J, et al. Entropy (Basel). 2024 Mar 27;26(4):287. doi: 10.3390/e26040287. Entropy (Basel). 2024. PMID: 38667842 Free PMC article. - Emergence of complex oscillatory dynamics in the neuronal networks with long activity time of inhibitory synapses.
Khanjanianpak M, Azimi-Tafreshi N, Valizadeh A. Khanjanianpak M, et al. iScience. 2024 Mar 4;27(4):109401. doi: 10.1016/j.isci.2024.109401. eCollection 2024 Apr 19. iScience. 2024. PMID: 38532887 Free PMC article. - Fundamental behavioral and neurophysiologic relationships between sensory processing, intolerance of uncertainty, and autistic traits in children: A hybrid approach.
Cardon G, Buckhannon M, Rojas D. Cardon G, et al. Biol Psychol. 2023 Nov;184:108712. doi: 10.1016/j.biopsycho.2023.108712. Epub 2023 Oct 13. Biol Psychol. 2023. PMID: 37839521