Understanding, Explanation, and Active Inference
Open accessThe DOI matches its Crossref record; 34 of the 34 references it cites were found in a registry or an index.
Abstract
While machine learning techniques have been transformative in solving a range of problems, an important challenge is to understand why they arrive at the decisions they output. Some have argued that this necessitates augmenting machine intelligence with understanding such that, when queried, a machine is able to explain its behaviour (i.e., explainable AI). In this article, we address the issue of machine understanding from the perspective of active inference. This paradigm enables decision making based upon a model of how data are generated. The generative model contains those variables required to explain sensory data, and its inversion may be seen as an attempt to explain the causes of these data. Here we are interested in explanations of one’s own actions. This implies a deep generative model that includes a model of the world, used to infer policies, and a higher-level model that attempts to predict which policies will be selected based upon a space of hypothetical (i.e., counterfactual) explanations—and which can subsequently be used to provide (retrospective) explanations about the policies pursued. We illustrate the construct validity of this notion of understanding in relation to human understanding by highlighting the similarities in computational architecture and the consequences of its dysfunction.
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
Causal inference in statistics: An overview
Pearl, J. (2009) · Statistics Surveys · doi:10.1214/09-ss057 · Cited by 2,336 · cited 1×
“An influential formalism designed to address causality (Pearl, 2009, 2010) rests upon the idea of interventions.”
BackgroundPredictive coding under the free-energy principle
Friston, K., & Kiebel, S. (2009) · Philosophical Transactions of the Royal Society B: Biological Sciences · doi:10.1098/rstb.2008.0300 · Cited by 1,579
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 · cited 2×
“Similar approaches have been employed for working memory tasks (Parr and Friston, 2017), enabling the maintenance of information ‘‘in mind’’ throughout a delay period (Funahashi et al., 1989).”
MethodSelf-citationTemporally structured replay of awake hippocampal ensemble activity during rapid eye movement sleep
Louie, K., & Wilson, M. A. (2001) · Neuron · doi:10.1016/s0896-6273(01)00186-6 · Cited by 1,148 · cited 1×
“Replay of this sort has been identified physiologically in rodents in the same hippocampal cells that signal sequences of locations while behaving (Louie and Wilson, 2001; Foster, 2017; Pezzulo et al., 2017), hinting that the mechanisms that solve this generative model may also be at work in…”
BackgroundAwake hippocampal sharp-wave ripples support spatial memory
Jadhav, S. P., Kemere, C., German, P. W., & Frank, L. M. (2012) · Science · doi:10.1126/science.1217230 · Cited by 924 · cited 1×
“That these mechanisms play a functional role is evidenced by the fact that interrupting hippocampal sharp-wave ripple activity (during which replays often occur) impairs memory-guided navigational choices (Jadhav et al., 2012).”
BackgroundOptimal control of Markov processes with incomplete state information
Åström, K. J. (1965) · Journal of Mathematical Analysis and Applications · doi:10.1016/0022-247x(65)90154-x · Cited by 854 · cited 1×
“First, we select decisions whose sensory consequences cohere with the data anticipated under our model (Åström, 1965; Pezzulo et al., 2018).”
MethodRelating Introspective Accuracy to Individual Differences in Brain Structure
Fleming, S. M., Weil, R. S., Nagy, Z., Dolan, R. J., & Rees, G. (2010) · Science · doi:10.1126/science.1191883 · Cited by 809
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×
“The idea is that by predicting the data we would anticipate given our beliefs, low-level reflexes of the sort found in the spinal cord or brainstem can correct deviations between our predictions and measurable data such that our predictions are fulfilled (Adams et al., 2013; Shipp et al., 2013).”
BackgroundThe neural basis of metacognitive ability
Fleming, S. M., & Dolan, R. J. (2012) · Philosophical Transactions of the Royal Society B: Biological Sciences · doi:10.1098/rstb.2011.0417 · Cited by 747 · cited 1×
“This concept is important in many fields, ranging from metacognition (Fleming and Dolan, 2012) to cognitive neurology (Ballard et al., 1997; Fotopoulou, 2012) and psychiatry (David, 1990), where some syndromes are characterised by a patient exhibiting a lack of insight into their own behaviour.”
Background
Cited by
Active inference as a theory of sentient behavior
Pezzulo, G., Parr, T., & Friston, K. (2024) · Biological Psychology · doi:10.1016/j.biopsycho.2023.108741 · Cited by 109
Self-citationGenerating 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
“Thistypeofmodeling isfundamentallydifferentfromLLMandcanbe usedtomodeldyadicinteractionswhereagentshaveepistemicskininthe game[138]andagentsthatcanexplainthemselves[139].”
Self-citationDi Paolo, L. D., White, B., Guénin-Carlut, A., Constant, A., & Clark, A. (2024) · Philosophical Transactions of the Royal Society B: Biological Sciences · doi:10.1098/rstb.2023.0148 · Cited by 14
“The production of systems able to describe the reasons for their decisions (i.e. explainability) in natural language, as a pragmatic premise to overt discourse, is a major avenue for AIs’ development [14–16].”
BackgroundGenerating Meaning: Active Inference and the Scope and Limits of Passive AI
Pezzulo, G., Parr, T., Cisek, P., Clark, A., & Friston, K. (2023) · doi:10.31234/osf.io/8xgzv · Cited by 8
Self-citationFormalizing Lacanian psychoanalysis through the free energy principle
Li, L., & Li, C. (2025) · Frontiers in Psychology · doi:10.3389/fpsyg.2025.1574650 · Cited by 4
Metadata
- Authors
- Parr, T., & Pezzulo, G.
- Venue
- Frontiers in Systems Neuroscience
- Publisher
- Frontiers Media SA
- Published
- 2021-11-05
- Volume
- 15
- Pages
- 772641-772641
- ISSN
- 1662-5137 (ISSN-L)
- OpenAlex
- W3211802703
- Cited by
- 36
- References
- 34
- FWCI
- 2.07
- Type
- journal article
- Language
- en
- Open access
- gold
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
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Understanding, Explanation, and Active Inference. https://app.cochat.ai/papers/10.3389/fnsys.2021.772641
- The paper
- Parr, T., & Pezzulo, G. (2021). Understanding, Explanation, and Active Inference. Frontiers in Systems Neuroscience. https://doi.org/10.3389/fnsys.2021.772641
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.