Generative models, linguistic communication and active inference
Open accessThe DOI matches its Crossref record; 53 of the 53 references it cites were found in a registry or an index.
Abstract
This paper presents a biologically plausible generative model and inference scheme that is capable of simulating communication between synthetic subjects who talk to each other. Building on active inference formulations of dyadic interactions, we simulate linguistic exchange to explore generative models that support dialogues. These models employ high-order interactions among abstract (discrete) states in deep (hierarchical) models. The sequential nature of language processing mandates generative models with a particular factorial structure-necessary to accommodate the rich combinatorics of language. We illustrate linguistic communication by simulating a synthetic subject who can play the 'Twenty Questions' game. In this game, synthetic subjects take the role of the questioner or answerer, using the same generative model. This simulation setup is used to illustrate some key architectural points and demonstrate that many behavioural and neurophysiological correlates of linguistic communication emerge under variational (marginal) message passing, given the right kind of generative model. For example, we show that theta-gamma coupling is an emergent property of belief updating, when listening to another.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2018) · arXiv · doi:10.48550/arxiv.1810.04805 · Cited by 73,173
The free-energy principle: a unified brain theory?
Friston, K. (2010) · Nature Reviews Neuroscience · doi:10.1038/nrn2787 · Cited by 8,120 · cited 2×
“…ideas about how the brain works; from Kant through Helmholtz (Helmholtz, 1878 (1971)), from analysis by synthesis (Yuille and Kersten, 2006) to perception as hypothesis testing (Gregory, 1980), from the Helmholtz machine (Dayan et al., 1995) to the free energy principle (Friston, 2010).”
Self-citationRizzolatti, G., & Craighero, L. (2004) · Annual Review of Neuroscience · doi:10.1146/annurev.neuro.27.070203.144230 · Cited by 7,433 · cited 1×
“, visual search in active vision) to an interrogation of the world via semantics and semiotics, acquired by encultured learning (Constant et al., 2019; Creanza and Feldman, 2014; Penn et al., 2008; Rizzolatti and Craighero, 2004).”
Factor graphs and the sum-product algorithm
Kschischang, F. R., Frey, B. J., & Loeliger, H.-A. (2001) · IEEE Transactions on Information Theory · doi:10.1109/18.910572 · Cited by 6,486 · cited 1×
“The use of factors is known as a mean field approximation in the variational machine learning literature (Jaakkola and Jordan, 1998; Kschischang et al., 2001; Sallans and Hinton, 2004; Zhang et al., 2018) and is important for simplifying the form of the generative model and ensuing inference.”
The faculty of language: what is it, who has it, and how did it evolve?
Hauser, M. D., Chomsky, N., & Fitch, W. T. (2002) · Science · doi:10.1126/science.298.5598.1569 · Cited by 5,189
SCHEGLOFF, E. A., & SACKS, H. (1973) · Semiotica · doi:10.1515/semi.1973.8.4.289 · Cited by 4,589
Hoeting, J. A., Madigan, D., Raftery, A. E., & Volinsky, C. T. (1999) · Statistical Science · doi:10.1214/ss/1009212519 · Cited by 4,118 · cited 1×
“In Bayesian model selection, this phenomena is known as evidence dilution (Hoeting et al., 1999).”
Is the P300 component a manifestation of context updating?
Donchin, E., & Coles, M. G. H. (1988) · Behavioral and Brain Sciences · doi:10.1017/s0140525x00058027 · Cited by 3,585 · cited 7×
“Our simulations also predict electrophysiological violation responses, of the sorts observed in P300 and N400 studies.”
BackgroundInfluentialRecent Trends in Deep Learning Based Natural Language Processing [Review Article]
Young, T., Hazarika, D., Poria, S., & Cambria, E. (2018) · IEEE Computational Intelligence Magazine · doi:10.1109/mci.2018.2840738 · Cited by 2,921
Linguistic complexity: locality of syntactic dependencies
Gibson, E. (1998) · Cognition · doi:10.1016/s0010-0277(98)00034-1 · Cited by 2,537 · cited 1×
“There is a substantial literature on cognitive models of syntax processing (for a recent review, see Demberg and Keller, 2019) and how listeners deal with semantic ambiguity (Altmann and Steedman, 1988; Bever, 1970; Gibson, 1998, 2000).”
The functional role of cross-frequency coupling
Canolty, R. T., & Knight, R. T. (2010) · Trends in Cognitive Sciences · doi:10.1016/j.tics.2010.09.001 · Cited by 2,101 · cited 1×
“This nesting naturally leads to phase-amplitude coupling—which is the most commonly studied type of cross frequency coupling (Canolty and Knight, 2010).”
BackgroundHow to grow a mind: Statistics, structure, and abstraction
Tenenbaum, J. B., Kemp, C., Griffiths, T. L., & Goodman, N. D. (2011) · Science · doi:10.1126/science.1192788 · Cited by 1,815
On a measure of the information provided by an experiment
Lindley, D. V. (1956) · The Annals of Mathematical Statistics · doi:10.1214/aoms/1177728069 · Cited by 1,509
Bayesian model selection for group studies
Stephan, K. E., Penny, W. D., Daunizeau, J., Moran, R. J., & Friston, K. J. (2009) · NeuroImage · doi:10.1016/j.neuroimage.2009.03.025 · Cited by 1,423
Self-citationVinyals, O., & Le, Q. (2015) · arXiv · doi:10.48550/arxiv.1506.05869 · Cited by 1,400
Deconstructing episodic memory with construction
Hassabis, D., & Maguire, E. A. (2007) · Trends in Cognitive Sciences · doi:10.1016/j.tics.2007.05.001 · Cited by 1,229
Bayesian surprise attracts human attention
Itti, L., & Baldi, P. (2009) · Vision Research · doi:10.1016/j.visres.2008.09.007 · Cited by 1,224
Darwin's mistake: Explaining the discontinuity between human and nonhuman minds
Penn, D. C., Holyoak, K. J., & Povinelli, D. J. (2008) · Behavioral and Brain Sciences · doi:10.1017/s0140525x08003543 · Cited by 1,174 · cited 2×
“…(Young et al., 2018): e.g., word embeddings derived from learning predictive relationships (Collobert et al., 2011; Mikolov, 2010; Mikolov et al., 2013; Pennington et al., 2014) and fully contextualized word representations (Devlin et al., 2018; Radford et al., 2019; Vaswani et al., 2017).”
Active inference: a process theory
Friston, K., FitzGerald, T., Rigoli, F., Schwartenbeck, P., & Pezzulo, G. (2017) · Neural Computation · doi:10.1162/neco_a_00912 · Cited by 1,156
Self-citationA probabilistic earley parser as a psycholinguistic model
Hale, J. (2001) · Second meeting of the North American Chapter of the Association for Computational Linguistics on Language technologies 2001 - NAACL '01 · doi:10.3115/1073336.1073357 · Cited by 1,150 · cited 1×
“Although the idea of ‘surprisal’ has become increasingly prevalent in the literature (Hale, 2001; Levy, 2008), this usually refers to the magnitude of ‘surprise’ conveyed by individual words, such that expected semantics are simply an amalgamation of the semantics conveyed by all preceding words.”
BackgroundCortical oscillations and sensory predictions
Arnal, L. H., & Giraud, A.-L. (2012) · Trends in Cognitive Sciences · doi:10.1016/j.tics.2012.05.003 · Cited by 1,114 · cited 2×
“See (Arnal and Giraud, 2012; Giraud and Poeppel, 2012) for a discussion of related phenomena.”
Situated simulation in the human conceptual system
Barsalou, L. (2003) · Language and Cognitive Processes · doi:10.1080/01690960344000026 · Cited by 979 · cited 2×
“use simulations to test hypotheses about communication; such as generation and understanding of linguistic phrases, in relation to conceptual knowledge (Barsalou, 2003; Yufik, 1998, 2019), the use of a shared narrative (Mar et al.”
The Neural Bases of Social Cognition and Story Comprehension
Mar, R. A. (2011) · Annual Review of Psychology · doi:10.1146/annurev-psych-120709-145406 · Cited by 950
Variational free energy and the Laplace approximation
Friston, K., Mattout, J., Trujillo-Barreto, N., Ashburner, J., & Penny, W. (2007) · NeuroImage · doi:10.1016/j.neuroimage.2006.08.035 · Cited by 880 · cited 1×
“, the standard error on some parameter estimates (Friston et al., 2007).”
Self-citationVision as Bayesian inference: analysis by synthesis?
Yuille, A., & Kersten, D. (2006) · Trends in Cognitive Sciences · doi:10.1016/j.tics.2006.05.002 · Cited by 849
Gregory, R. L. (1980) · Philosophical Transactions of the Royal Society of London. B, Biological Sciences · doi:10.1098/rstb.1980.0090 · Cited by 789
Predictions not commands: Active inference in the motor system
Adams, R. A., Shipp, S., & Friston, K. J. (2013) · Brain Structure and Function · doi:10.1007/s00429-012-0475-5 · Cited by 765 · cited 1×
“This sort of outcome is formally related to motor output under active inference; namely, the fulfilment of proprioceptive predictions by classical reflexes (Adams et al., 2013; Shipp et al., 2013).”
Self-citationFriston, K. (2013) · Journal of The Royal Society Interface · doi:10.1098/rsif.2013.0475 · Cited by 754 · cited 1×
“The procedures used here assume that the brain restricts itself to a limited number of characteristic states (Friston, 2013)—a property that all sentient systems must possess.”
Self-citationZhao, T., Zhao, R., & Eskenazi, M. (2017) · Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/p17-1061 · Cited by 736 · cited 2×
“In contrast, response generation—including conversational dialogue agents—have been framed as either deep reinforcement learning (Li et al., 2016; Zhao et al., 2017) or inference problems (Liu et al.”
Friston, K., FitzGerald, T., Rigoli, F., Schwartenbeck, P., ODoherty, J., & Pezzulo, G. (2016) · Neuroscience & Biobehavioral Reviews · doi:10.1016/j.neubiorev.2016.06.022 · Cited by 674
Self-citation(2005) · Edinburgh Research Explorer (University of Edinburgh) · doi:10.5555/1046920.1088695 · Cited by 650
Hohwy, J. (2016) · Noûs · doi:10.1111/nous.12062 · Cited by 578 · cited 2×
“Technically, this can be formulated in terms of maximising the marginal likelihood for models of the lived world—that is neatly summarized as self-evidencing (Clark, 2016; Hohwy, 2016); in other words, gathering sensory evidence for our generative models.”
Dehaene, S., Meyniel, F., Wacongne, C., Wang, L., & Pallier, C. (2015) · Neuron · doi:10.1016/j.neuron.2015.09.019 · Cited by 478
On the semantics and pragmatics of linguistic feedback
ALLWOOD, J., NIVRE, J., & AHLSÉN, E. (1992) · Journal of Semantics · doi:10.1093/jos/9.1.1 · Cited by 442 · cited 2×
“First, we will assume that language is for communication, which immediately implies a shared forward-looking narrative (Allwood et al., 1992; Brown and Brune, 2012; Friston and Frith, 2015a; Mar et al., 2011; Schegloff and Sacks, 1973; Specht, 2014).”
Cognitive control over learning: Creating, clustering, and generalizing task-set structure.
Collins, A. G. E., & Frank, M. J. (2013) · Psychological Review · doi:10.1037/a0030852 · Cited by 426 · cited 1×
“, 2016) that is elegantly addressed using nonparametric Bayesian methods (Collins and Frank, 2013; Goldwater, 2006; Teh et al., 2006), such as those found in computational linguistics (please see below).”
The graphical brain: Belief propagation and active inference
Friston, K. J., Parr, T., & de Vries, B. (2017) · Network Neuroscience · doi:10.1162/netn_a_00018 · Cited by 415
Self-citationTowards a mathematical theory of cortical micro-circuits
George, D., & Hawkins, J. (2009) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1000532 · Cited by 404 · cited 1×
“In deep models of this sort, deeper hierarchical levels are constituted by factors that change over progressively longer timescales (Friston et al., 2017d; George and Hawkins, 2009; Kiebel et al., 2009).”
Joint Action, Interactive Alignment, and Dialog
Garrod, S., & Pickering, M. J. (2009) · Topics in Cognitive Science · doi:10.1111/j.1756-8765.2009.01020.x · Cited by 402
Supra-personal cognitive control and metacognition
Shea, N., Boldt, A., Bang, D., Yeung, N., Heyes, C., & Frith, C. D. (2014) · Trends in Cognitive Sciences · doi:10.1016/j.tics.2014.01.006 · Cited by 368 · cited 1×
“, inference about) agency; namely, metacognitive capability (Fleming et al., 2012; Shea et al., 2014).”
Friston, K., & Frith, C. (2015) · Consciousness and Cognition · doi:10.1016/j.concog.2014.12.003 · Cited by 359
Self-citationMetacognition: computation, biology and function
Fleming, S. M., Dolan, R. J., & Frith, C. D. (2012) · Philosophical Transactions of the Royal Society B: Biological Sciences · doi:10.1098/rstb.2012.0021 · Cited by 344 · cited 1×
“, inference about) agency; namely, metacognitive capability (Fleming et al., 2012; Shea et al., 2014).”
Bayesian Spiking Neurons I: Inference
Deneve, S. (2008) · Neural Computation · doi:10.1162/neco.2008.20.1.91 · Cited by 313 · cited 2×
“This has some construct validity in relation to theoretical proposals and empirical work on evidence accumulation (de Lafuente et al., 2015; Kira et al., 2015) and the neuronal encoding of probabilities (Deneve, 2008).”
The anatomy of choice: dopamine and decision-making.
Friston, K., Schwartenbeck, P., FitzGerald, T., Moutoussis, M., Behrens, T., & Dolan, R. J. (2014) · Philosophical Transactions of the Royal Society B: Biological Sciences · doi:10.1098/rstb.2013.0481 · Cited by 289
Self-citationActive inference, communication and hermeneutics
Friston, K. J., & Frith, C. D. (2015) · Cortex · doi:10.1016/j.cortex.2015.03.025 · Cited by 266
Self-citationKnowing one's place: a free-energy approach to pattern regulation
Friston, K., Levin, M., Sengupta, B., & Pezzulo, G. (2015) · Journal of The Royal Society Interface · doi:10.1098/rsif.2014.1383 · Cited by 234
Self-citationThe dopaminergic midbrain encodes the expected certainty about desired outcomes
Schwartenbeck, P., FitzGerald, T. H. B., Mathys, C., Dolan, R., & Friston, K. (2015) · Cerebral Cortex · doi:10.1093/cercor/bhu159 · Cited by 219
Self-citationThe role of prediction in social neuroscience
Brown, E. C., & Brüne, M. (2012) · Frontiers in Human Neuroscience · doi:10.3389/fnhum.2012.00147 · Cited by 190 · cited 1×
“First, we will assume that language is for communication, which immediately implies a shared forward-looking narrative (Allwood et al., 1992; Brown and Brune, 2012; Friston and Frith, 2015a; Mar et al., 2011; Schegloff and Sacks, 1973; Specht, 2014).”
Prediction, Bayesian inference and feedback in speech recognition
Norris, D., McQueen, J. M., & Cutler, A. (2016) · Language, Cognition and Neuroscience · doi:10.1080/23273798.2015.1081703 · Cited by 170 · cited 1×
“In a similar vein, any exciting advances in computational neurolinguistics (Barlow, 1974; Lightfoot, 1979; MacKay and Peto, 1995; Norris et al., 2016; Rosenfeld, 2000) that do not deal explicitly with belief states updating cannot be used to create artefacts that communicate.”
Scene Construction, Visual Foraging, and Active Inference
Mirza, M. B., Adams, R. A., Mathys, C. D., & Friston, K. J. (2016) · Frontiers in Computational Neuroscience · doi:10.3389/fncom.2016.00056 · Cited by 165
Self-citationRepresentation of accumulating evidence for a decision in two parietal areas
de Lafuente, V., Jazayeri, M., & Shadlen, M. N. (2015) · The Journal of Neuroscience · doi:10.1523/jneurosci.2451-14.2015 · Cited by 162
A place for time: The spatiotemporal structure of neural dynamics during natural audition
Stephens, G. J., Honey, C. J., & Hasson, U. (2013) · Journal of Neurophysiology · doi:10.1152/jn.00268.2013 · Cited by 141 · cited 1×
“This means higher level factors are attributes of a sentence or phrase, while lower level factors may change from word to word (Chien and Honey, 2020; Davis and Johnsrude, 2003; Demirtaş et al., 2019; Specht, 2014; Stephens et al., 2013).”
Neuronal message passing using mean-field, bethe, and marginal approximations
Parr, T., Markovic, D., Kiebel, S. J., & Friston, K. J. (2019) · Scientific Reports · doi:10.1038/s41598-018-38246-3 · Cited by 128
Self-citationToward the neural implementation of structure learning
Tervo, D. G. R., Tenenbaum, J. B., & Gershman, S. J. (2016) · Current Opinion in Neurobiology · doi:10.1016/j.conb.2016.01.014 · Cited by 106
Cited by
The default mode network: where the idiosyncratic self meets the shared social world
Yeshurun, Y., Nguyen, M., & Hasson, U. (2021) · Nature Reviews Neuroscience · doi:10.1038/s41583-020-00420-w · Cited by 682
Laying down a forking path: tensions between enaction and the free energy principle
Di Paolo, E., Thompson, E., & Beer, R. (2022) · Philosophy and the Mind Sciences · doi:10.33735/phimisci.2022.9187 · Cited by 94
Generating meaning: active inference and the scope and limits of passive AI
Pezzulo, G., Parr, T., Cisek, P., Clark, A., & Friston, K. (2024) · Trends in Cognitive Sciences · doi:10.1016/j.tics.2023.10.002 · Cited by 80
Self-citationFederated inference and belief sharing
Friston, K. J., Parr, T., Heins, C., Constant, A., Friedman, D., Isomura, T., Fields, C., Verbelen, T., Ramstead, M., Clippinger, J., et al. (2024) · Neuroscience & Biobehavioral Reviews · doi:10.1016/j.neubiorev.2023.105500 · Cited by 30
“This follows because each agent is defined by its Markov blanket, which precludes reciprocal message passing with other agents (Heins et al., 2023; Palacios et al., 2017; Parr et al., 2020; Pellet and Elisseeff, 2008).”
BackgroundSelf-citationOn the minimal theory of consciousness implicit in active inference
Whyte, C. J., Corcoran, A. W., Robinson, J., Smith, R., Moran, R. J., Parr, T., Friston, K. J., Seth, A. K., & Hohwy, J. (2026) · Physics of Life Reviews · doi:10.1016/j.plrev.2025.11.002 · Cited by 15
Self-citation
Metadata
- Authors
- Friston, K. J., Parr, T., Yufik, Y., Sajid, N., Price, C. J., & Holmes, E.
- Venue
- Neuroscience & Biobehavioral Reviews
- Publisher
- Elsevier BV
- Published
- 2020-11
- Volume
- 118
- Pages
- 42-64
- ISSN
- 0149-7634 (ISSN-L)
- OpenAlex
- W3042228302
- Cited by
- 108
- References
- 53
- FWCI
- 7.39
- Type
- journal article
- Language
- en
- Open access
- hybrid
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.