On Modeling

Models

1. "Getting it right, not making mistakes, has been of paramount importance to every living thing on this planet for more than three billion years, and so these organisms have evolved thousands of different ways of finding out about the world they live in."  - Daniel Dennett
   
2. "How are we going to create an ethical practice in the Anthropocene, this time of ours in which futures, of human and nonhuman kinds, are increasingly entangled, and interdependent in their mutual uncertainty?" - Eduardo Kohn
   
3. "Robert Rosen's work on modeling should be seen as relevant to societies and the functioning of democracies. What is required is an interrogation of the models that societies have of themselves and their relationship to their ambience or environments." - Arran Gare
   
4. Modeling "is the art of bringing entailment structures into congruence... It is an art, just as surely as poetry, music, and painting are. Indeed, the essence of art is that, at root, it rests on the unentailed, on the intuitive leap". - Robert Rosen
   
5. "The environment is acting upon you and you are acting upon the environment. It's a dance, a dialogue. Your action upon the world becomes the world's way of perceiving you. And the world acts upon you, through your perception of the world. There's a beautiful symmetry there." - Karl Friston
   
6. Models are closely tied to stories. We explain models by telling stories, and when we tell stories, listeners form models - mental pictures of the actors, how they are related, and how they behave. Models provide a basis for trust, shared understanding, agreement, and group action. They shape what we see - what we let ourselves notice - telling us what is important, what counts, and what to look for. Old models often resist new ones and can limit our ability to see new evidence and understand changing situations. That’s why organizations need to expose the fundamental models that guide them and periodically challenge those models. Creating or revising a model is meta-activity, taking us outside the primary activity in which we were engaged. It requires attention, energy, and time.
   
7. "Those who are well versed in the facts alone will treat each fact as a fact and no more. Those who are well versed in the Way will unify their treatment of the facts... since they examine and compare the facts, their perception will be clear." - Xunzi
8. A framework for describing modeling processes can provide the tools for model comparison, analysis, and selection. It can also allow us to characterize how systems become de-coupled and find ways to re-establish symbiotic harmony. The benefits of using better models include the ability to identify: problem areas, goals and the tasks relevant to addressing them, unmet needs and possible organizational improvements, spot inefficiencies, anticipate disruptions before they occur, facilitate adaptive changes, and offer a descriptive analysis of existing conditions. 

Personal models

9. When we compare 'what is' with 'what is desired', we are comparing reality and our desired model of reality. The discrepancies we observe by means of this comparison can lead to anxiety, rumination, and depression.
10. The mindfulness approach of recognizing and accepting these mental events as comparative modeling processes can be therapeutic by providing a position from which to manage the discrepancy between ‘what is’ and ‘what is desired’ and shed light on the implicit models we use that influence our choices, behaviors, and mental health every day, allowing us to work with them in a healthier way.
    
11. Carefully choose what to pay attention to. Without constraints, the volume of evidence that is available to us today can overwhelm our attentional capacities. Well-established boundaries increase efficiency and agility while minimizing the free energy of distractions and avoidable uncertainty. We have to relearn how to attend, what to attend to, and what to ignore. "How can the mind understand the Way? Because it is empty, unified, and still."
    
12. Contemplative practices of loving kindness meditation entail the explicit enrichment and effortful rehearsal of one’s mental models of others, which eventually become automatic through practice. The linguistic (narrative) elaboration of these models may be essential to their extension to include members of out-groups, the whole of humanity, or even to all sentient beings, regardless of spatial or temporal location.
13. Democritus exhorts us to always maintain a holistic perspective and recall that we are but a small, inseparable part of much larger community of life (as did Wang Yangming). It is our ignorance of the greater whole that causes our emotional disposition to depart from equanimity. From understanding holism there is empathy, and from empathy there is equanimity. Holism not only binds one person to the next, but one day to the next as well.
    
14. The 'Deep Social Mind' theory describes how we are adapted to read the minds of trusted others while at the same time assisting those others in reading our own minds. Our minds mutually interpenetrate. 'Mind' in the human sense is not locked inside this or that skull but instead is relational, stretching between us.
    
15. A primary distinguishing feature among people are the beliefs under which they are operating, and which guide adaptive behaviour. Provided that we have a model of someone else's beliefs, then we can leverage our own action (policy) selection mechanisms according to the beliefs of another person. For a species such as Homo sapiens that evolved to rely upon cooperative and highly elaborate coordinated action, expectations about folk psychology (inferences about the way other people think and reason and what they expect of the world) are at least as important as, if not more important than, expectations about the physical world itself.
    
16. False inferences and maladaptive beliefs (cultural norms, religious dogma, political ideology, or economic systems) often persist longer when associated with the archetypal (and often charismatic) identities we emulate, and a failure of reliable models to spread from one community to another can result in polarization. 
    
17. Con artists exploit our need to locate and learn from the best models. In extreme cases, as Robert Rosen wrote, "The identification of one's self with one's models explains, perhaps, why human beings are so often willing to die; i.e. to suffer biological extinction, rather than change their models." But as Karl Popper noted, we can look upon them critically and "let our conjectures, our theories, die in our stead".

Suprapersonal models

18. Cells, organs, brains, people, societies, eco-niches - anything that self-assembles and maintains its structural and functional integrity - all of these can be cast as having their own individual identity (model).
    
19. Social systems should be seen as autopoietic, organism-like agents that, via their human constituents, actively counteract any deviation from their organization, so as to ensure the continuation and self-regeneration of the system.
    
20. On the one hand, the relationship between individual and social system is one of symbiosis or mutual benefit, with social systems providing means for reducing free energy to the individual through coordination of action and prevention of conflicts. On the other hand, social systems can also veer into dogmatism, radicalism, and mind control that suppresses individual expression, creativity, and well-being.
    
21. The observation that particular suboptimal or maladaptive norms obtain is not a problem but an observation that forces us to explore the impact of initial and boundary conditions and historical trajectories on the constitution of humans and their cultural niches.
    
22. “Our world is complex; how different things are related is hard to understand. Our leaders don't fully use good models for what to invest time and money into." While symptoms are addressed, root causes tend to be ignored.
    
23. A pluralist democracy can only prosper in the long-term by designing an infrastructure (institutions, laws, rules, procedures, and norms) that allows it to look ahead or effectively plan, to implement a coherent multi-decade policy program to both prepare for and prevent the worst of climate change.
    
24. "All of us have some roles to play within some institutions, even if that's our family or community or workplace, let alone national institutions and politics and the economy. We each have to say, given my role here, what's my responsibility? What should I do in this situation? Not just what do I want, not just what would look good, but given my role here, what should I do? It is a question you ask when you take the institutions that you're part of seriously." - Yuval Levin
25. "To be awakened is to realize that one has a role in the harmonious development of all and to strive to fulfill that role." - A.T. Ariyaratne, who started the Sarvodaya Shramadana Movement, meaning "welfare for all through our shared labor".
    
26. Popular movements do not generate social change when there isn't a clear demand of what the alternative to the failed model is. - Naomi Klein
27. Our models of the world share more in common than we would readily admit, and people can demonstrate flexibility when reasoning about complex issues. If we begin our conversations from this place of shared understanding we can work together much more effectively in pursuit of our common goals.
    
28. A shared generative model allows for the emergence of communication and cultural dynamics. It is our ability to construct, share, and compare these sort of detailed models that allows us to effectively plan and act in an uncertain future. This offers a way to account for both social stabilization and social change, and indeed for human historicity itself.
29. When you can write down the generative model, you can create the kind of steady state system it describes. The big problem, however, is writing down the generative model. We need to have models that are fit for purpose for the kind of world in which we live.

Prediction and Intervention

30. The causal structure of spacetime reveals that some outcomes are more probable than others. Learning this 'generative process' allows organisms to make useful inferences (or predictions) about future events.
    
31. When an organism's 'generative model' corresponds to the 'generative process' of the world it inhabits, then goal oriented (teleonomic) behaviors become possible. "Beliefs about outcomes in the distal future influence beliefs about states in the proximal future and present. That these beliefs then drive policy selection suggests that (beliefs about) the future can indeed cause the past." 
    
32. Living organisms construct and use generative models. These map beliefs about causes to observed consequences and characterizes the mismatch between organism and environment. Organisms constantly tune them so that they fit the generative processes ‘out there in the world’ that cause their sensory data. 
33. The goal is to learn a model that allows survival and adaptation. It doesn't have to be a completely accurate world model. But it does have to be tailored to a specific set of prior preferences and regularities for agent-environment interaction. "Prediction-error minimization is not the ultimate aim for an organism, but rather a means by which a better grip is achieved."
34. If an organism can determine the causal structure of a new threat and incorporate that within an updated generative model, then (under the assumption that its generative model is close to the generative process) it may be able to actively intervene in the world so as to bring about preferred states and prevent or mitigate possible threats. However, because coupled organism-environment systems are deeply entangled, the inertia of the former systemic model, with its habitual, conditioned responses, may actively resist any reconfiguration and present obstacles to the necessary changes.
    
35. Adaptation that relies primarily on conscious behavioral interventions (will power alone) can be slow, difficult, or impossible, especially at larger scales. 'Resetting' our models often requires making corresponding changes in our environments so that new habits have a supportive context within which to take root and grow. "And no man putteth new wine into old bottles; else the new wine will burst the bottles, and be spilled, and the bottles shall perish. But new wine must be put into new bottles; and both are preserved." - Luke 5:37-38
    
36. "Given that we live in a highly dynamic world, being metastably poised to switch to multiple new relevant action-readiness patterns is itself an adaptive trait, and necessary for an agent to tend toward grip on its interactions with the environment." - Bruineberg and Rietveld
    
37. Society at all levels, from local to global, must model its environment and it's impacts thereupon, so that it may anticipate future change and take proactive measures to ensure survival. "Look and listen for the welfare of the whole people and have always in view not only the present, but also the coming generations, even those whose faces are yet beneath the surface of the ground - the unborn of the future Nation." - Iroquois Constitution
    
38. Through establishing congruence between our econiche and our system of anticipations, models allow us to navigate exchanges with the environment. We can model not just present world configurations, but also simulate possible world configurations based on future (counterfactual) actions.
    
39. Models give us agency to identify and calibrate our actions with the goal of shaping our future. We can do this by pruning catastrophic branches of a tree of possibilities that lies before us. Models provide a way to see our potential futures ahead of time, and how that interacts with the choices we make today.
40. "Animals that deploy a model-based system are thus able to 'navigate into the future' rather than remaining 'driven by the past'." - Andy Clark
41. Evolution equipped organisms with progressively more effective types of regulation, able to use information from the environment about causes as the source of regulatory actions. For example, the nervous system of a cow is able to sense at the skin when the cause of a potential future fall in blood temperature has occurred, perhaps from a sudden gust of ice cold air, and then predict a need to regulate proactively, before the error even happens, raising the blood temperature without any preliminary fall. Conant and Ashby use this example to support their conclusion that the living brain, so far as it is to be successful and efficient as a regulator for survival, must proceed, in learning, by the formation of a model (or models) of its environment so that it can anticipate change before it occurs. They write “the best regulator of a system is one which is a model of that system as seen through a mapping... any regulator must model what it regulates.”
42. The primary function of these models is to make a prediction. And we use that prediction as the basis for selecting actions that will best preserve and protect our lives. For this reason models or our world must be formed and operate accurately. "Ecology needs to be a predictive science." - E.O. Wilson
    
43. Security and protection are about avoiding harm. It's the reason fear is a motivating emotion, and why care for the vulnerable and defenseless is culturally valued. We protect ourselves and those we love.
44. Our ability to protect depends entirely upon our ability to predict. If we cannot correctly predict which things may later threaten us or cause us future suffering, then we cannot protect at all. Prediction is hard. We must not only accurately determine what threatens us, but also correctly predict the means by which the threat can be averted.
45. Let's encourage the natural desire people have to protect. We must only ensure their predictions follow from accurate models, and avoid the failures of poorer models. Whether intentional or accidental, a failure to predict comes at a high cost.

Selection

46. Biological systems encode a model of their environment implicit in their phenotype. This model includes a set of prior beliefs (such as the ability to avoid phase-transitions that would irreversibly alter their structure, possibly resulting in death) that distinguish it from the external environment. These beliefs can only be realized through the ability to predict and act, giving rise to the apparently purposeful and autopoietic behavior of life. “Only when we think in teleonomic terms, and regard the structure as end-directed, does it make sense to speak of ‘selection’ at all.”
    
47. "It is possible to consider the physiological make-up of a fish, as a model of the fluid dynamics and other elements that constitute its aquatic environment — its internal dynamics depends on the dynamics of the eco-niche in which it lives. It is in this embodied sense that an organism is a model of the world it lives in. And by virtue of being a model, it is able to encode or represent that world (within a mental model)." - Kirchhoff et al.
48. “Life is the manifestation of a certain kind of (relational) model. A particular material system is living if it realizes this model.” - Robert Rosen
49. "There is no information unless there is something that can be informed. This something is a model." - John Campbell
50. Organisms themselves are the implicit model for which they gather evidence, resulting in the interpretation that they produce evidence for their own existence. Therefore, we are now in a position to interpret processes of adaptation as collecting model evidence and, by extension, to cast natural selection as a form of model selection. (To Karl Friston, evolutionary theory is basically a process of inference.) What sorts of models are fit for an eco-niche? 
    
51. One method that has proved useful has been the application of phylogenetic analyses to the constituent elements of a model. These analyses clearly reveal how the constituent components in a diversity of designs have recombined (through Darwinian processes of variation, differential selection, and heredity) into the models and policies we see today.
52. Viewed in context, how do the models used by an individual, society, or superorganism further shape it's development and evolution? Do they help the whole system tend toward grip on its ecological niche? Or has the system lost its grip? Where are the branching points that shift future trajectories? The resulting comparison, analysis, and selection or modification could guide an ecological process for future planning on any scale, from hours to aeons.
53. "All models are evaluated in relation to each other, in terms of their relative evidence. ...having the ability to compare different models means that one has the ability to explore a model space." - Karl Friston
54. "The goal of regulation is not constancy, but coordinated variation to optimize performance at the least cost. In short, natural selection ensures prediction down to the limit set by physical laws." - Peter Sterling

Attunement

55. The inside has to have some form of synchronization with the outside. "To hold and fill a cup to overflowing is not as good as to stop in time."
56. We do not have a model of the world—we are a model of that world we seek to bring forth. The synchrony between ourselves and the world is crucial because a flourishing future requires that we are able to both anticipate and work to bring about the conditions that are compatible with our future existence and ways of flourishing.
57. "We are in the world and the world is in us." - Whitehead
    
58. "If we can have a coherent approach to reality, then reality will respond coherently to us" - David Bohm
59. "When hungry, I eat. When tired, I sleep. Fools laugh at me. But the wise understand." Appropriate action (wu wei) follows from being in tune with one's body and environment.
60. The ongoing tuning and maintenance of the generative model by active inference entails the dynamic entanglement of the agent and environment.
61. Coupled systems form larger systems via mutual entrainment. This coupling relationship is one of mutual modeling and collaborative inference.
62. The free-energy principle can be regarded as a systematic attempt to understand the ‘fit’ between an embodied agent and its niche, where the quantity of free-energy is a measure for the ‘misfit’ or disattunement between agent and environment.
    
63. There is a continuous feedback loop, in which what the agent does changes the environment, which changes what the agent perceives, which changes the expectations of the agent, which in turn changes what the agent does (to change the environment). The convergence of both agent and environment is a function of their reciprocal interaction. Because actions change the environment, they tend to make it a good mirror of the agent. 
64. The active inference formulation offers a symmetrical view of exchanges between agent and environment. The effect of the agent on the environment can be understood as the environment `learning’ about the agent through the accumulation of ecological legacies. "The environment is trying to learn about you as much as you are trying to learn about the environment." - Karl Friston
65. Active inference rests upon the coupling between a generative process (i.e., environment) and a generative model of that process (i.e., agent). The mutual adaptation between the process and model means that there is a common phenotypic space that is shared by the environment and agent. The metaphor of the agent and environment `driving’ each other through phenotypic space is in line with the 'extended evolutionary synthesis'.
66. Convergence or (generalized) synchronization emerges as the agent and environment ‘get to know each other’. The rate at which the agent learns about the environment and vice versa, and the degree to which their respective expectations converge, depends in a sensitive way on the relative confidence placed in prior beliefs.
67. One can think of an environment consisting of multiple agents, an 'ecology of minds', where the sensory states of one agent are generated by the action of the other agents. Over time (via sympoiesis), the agents mutually constrain each other until an attracting (synchronization) manifold is reached. Coupled agent-environment systems contain mutually predictive information.
    

Active Inference

68. "Models aim to engage the world, rather than to depict it in some action-neutral fashion. They are rooted in the patterns of organism-environment interaction. The role of these models is to deliver an efficient, context-sensitive grip upon a world of multiple competing affordances for action. ...A grip on the patterns that matter for the interactions that matter. A model that helps maintain the integrity and viability of a system by enabling it to minimize prediction error and thus avoid compromising (possibly fatal) encounters with the environment. A model is a structured meaningful realm apt for perception, thought, imagination, and action." - Andy Clark
69. Karl Friston has described the process of adaptation as "active inference", a theory that underpins the way we perceive and act in the world and provides an increasingly influential account of action and choice behavior. It connects and extends ideas such as the good regulator theorem (Conant and Ashby), perceptual control theory (Powers), and the notion of entailment in theoretical biology (Rosen).
    
70. Active inference is a framework that casts perception, learning, and action as essentially being in the same game: that of gathering evidence for the model that underwrites the existence of the agent. In this sense, active inference casts living and cognitive processes as self-fulfilling prophecies, which gather evidence for an implicit (generative) model that the agent embodies and enacts. Generative models are so called because they are models of the generative process ‘out there in the world’ that cause (or generate) our sensory data. The active inference agents themselves, of course, do not have direct access to the generative process, and must deploy inference and action to guess-timate its structure.
71. Selecting which actions to deploy makes active inference all about policy selection.
72. Active inference provides an account for how our ‘ecology of mind’ becomes, or fails to become, attuned to the world.
73. "Active inference is a process theory for complex adaptive systems consisting of (inter)dependent dynamics. Variation is the norm - not static pictures." - Marco Lin
74. "In active inference, everything that can change, changes to minimise the mismatch between organism and environment." - Veissière et al.
75. Agents have to optimize a model of their environment and, given that model, minimize surprise about the states they find themselves in. This is equivalent to maximizing the evidence for their model of the world.
76. If our planning horizon is insufficient to contemplate distal outcomes, we can easily get stuck as we pursue our goals. A chess player must evaluate their actions in light of possible outcomes, and think several moves ahead.
77. With a greater planning horizon we are able to plan and execute the shortest path to our ultimate goal, which often involves excursions through state (and belief) space that point away from it.
78. Creatures like us may be characterised by deep generative models that see far into the future, enabling the capacity to select courses of action that consider long term consequences.
79. Economic models run the spectrum from microeconomics (a job, a career of several jobs) to macroeconomics (a job sector, or transitions between economic systems). We must use a thick temporal model to contemplate the more distal outcomes.
80. A belief-based (active inference) formulation of choice, based on surprise minimization, is different from the classical economic treatment of decision-making as maximizing value or utility. "This may provide important insights into how agents ‘make sense out of the world’ and how certain aberrancies in these processes may induce psychopathological behavior."
81. Based on this view, belief, as opposed to reward, based formulations of choice behavior are attracting interest. This further implies that valuable behavior is the accumulation of evidence for our beliefs about, that is our models of, the world.

References:

1. Daniel Dennett, "Darwin's Dangerous Idea"
2. Eduardo Kohn, "Eduardo Kohn on How Forests Think"
3. Arran Gare, "The Philosophical Foundations of Ecological Civilization"
4. Robert Rosen, "Life Itself"
5. Karl Friston, "Embodied Cognition"
6. Hugh Dubberly, "Models of Models"
7. Xunzi, "Dispelling Obsession"
8. Niels Röling, "Gateway to the Global Garden"; Mel Andrews, "Is the FEP Epistemologically Applicable?"
9. Bishop et al., "Mindfulness"
10. Bishop et al., "Mindfulness"
11. Karl Friston, "The Mind at Work"; Xunzi, "Dispelling Obsession"
12. Veissière et al., "Thinking Through Other Minds"; Ed Yong, "Empathy With Your Future Self"
13. Democritus, fragment 191; Wang Yangming, "Inquiry on The Great Learning"
    
14. Wikipedia, "Deep social mind"
15. Veissière et al., "Thinking Through Other Minds"
16. David Roberts, "Arguments are beside the point"; O’Connor and Weatherall, "Misinformation Is About Who You Trust"; Chris Mooney, "Why we don't believe science"
    
17. Joseph Henrich, "The Secret of Our Success"; Robert Rosen, "Life Itself"; Karl Popper, "Natural Selection and the Emergence of Mind"
18. Karl Friston, "The Mind at Work"; Ramstead et al., "Multiscale integration"
    
19. Van de Cruys et al., "The Dark Side of TTOM"
20. Van de Cruys et al., "The Dark Side of TTOM"
21. Veissière et al., "TTOM in Action"
22. Allen Tien, Tweet
23. David Roberts, "YouTube has a big climate misinformation problem"
24. Yuval Levin, "When Institutions Are Used As Stages, People Lose Trust"; Henry Rosemont Jr., "Cooperating Interrelated Role-bearing Persons"
    
25. A.T. Ariyaratne, "Sarvodaya"
    
26. Naomi Klein, "The Case for a Green New Deal"
27. Rebecca Onie, "What Americans agree on when it comes to health"; Pärnamets and Van Bavel, "How Political Opinions Change"
    
28. Hesp et al., "A Multi-scale View of the Emergent Complexity of Life"; Veissière et al., "TTOM in Action"
29. Karl Friston "Neuroscience and the Free Energy Principle"
30. Bruineberg et al., "Free-energy minimization in joint agent-environment systems";
    
31. Parr and Friston, "Generalised free energy and active inference"
    
32. Tschantz et al., "Learning action-oriented models through active inference"; Ramstead et al., "Multiscale integration"
33. Bruineberg et al., "The anticipating brain is not a scientist"
34. Dolega and Dewhurst, "Fame in the predictive brain"; Bruineberg et al., "Free-energy minimization in joint agent-environment systems"
35. Luke 5:37-38
36. Bruineberg and Rietveld, "What’s Inside Your Head Once You’ve Figured Out What Your Head’s Inside Of"
37. Conant and Ashby, "Every Good Regulator of a System must be a Model of that System"; Iroquois Constitution
38. Adam Safron, "Integrated World Modeling Theory"
39. Zeynep Tufekci, "Don’t Believe the COVID-19 Models"
40. Andy Clark, "Surfing Uncertainty"; Schafer and Schiller, "The Brain’s Social Road Maps"
    
41. Conant and Ashby, "Every Good Regulator of a System must be a Model of that System"
42. Noam Chomsky, "Coronavirus pandemic could have been prevented"; Stewart Brand, "Whole Earth Discipline" (p267)
    
43. Hannah Pickard, "CCL March 2020 Monthly Speaker"
44. Hannah Pickard, "CCL March 2020 Monthly Speaker"
45. Hannah Pickard, "CCL March 2020 Monthly Speaker"
46. Karl Friston, "Free Energy and the Brain"; Tschantz et al., "Learning action-oriented models through active inference"; Grace de Laguna, “The Role of Teleonomy in Evolution”
    
47. Kirchhoff et al., "The Markov blankets of life"
    
48. Robert Rosen, "Life Itself"
49. John Campbell, "Universal Darwinism As a Process of Bayesian Inference"
50. Hesp et al., "A Multi-scale View of the Emergent Complexity of Life"
51. Muthukrishna and Henrich, "Innovation in the collective brain"; Ramstead, Tweet
52. Bruineberg and Rietveld, "What’s Inside Your Head Once You’ve Figured Out What Your Head’s Inside Of"
53. Karl Friston, "Free Energy and the Brain"
54. Peter Sterling, "Principles of Allostasis"
55. Karl Friston, "A Neuroscientist’s Theory of Everything"; Tao Te Ching
56. Karl Friston, "Active inference and free energy"
57. Alfred North Whitehead, "Modes of Thought"
58. David Bohm, "Art Meets Science & Spirituality in a Changing Economy"
59. Nanyue Mingzan, "Enjoying the Way"
60. Kiverstein and Kirchhoff, "Third-wave extended mind and predictive processing"
61. Adam Safron, "Integrated World Modeling Theory"
62. Bruineberg et al., "Free-energy minimization in joint agent-environment systems"
63. Bruineberg et al., "Free-energy minimization in joint agent-environment systems"; Veissière et al., "Thinking Through Other Minds"
64. Bruineberg et al., "Free-energy minimization in joint agent-environment systems"; Karl Friston, "I Am Therefore I Think”
65. Bruineberg et al., "Free-energy minimization in joint agent-environment systems"
66. Bruineberg et al., "Free-energy minimization in joint agent-environment systems"
67. Bruineberg et al., "Free-energy minimization in joint agent-environment systems"; Gregory Bateson, "Steps to an Ecology of Mind"; Adam Safron, "Integrated World Modeling Theory"; Donna Haraway, "Staying with the Trouble"
    
68. Andy Clark, "Surfing Uncertainty"
69. Schwartenbeck et al., "Evidence for surprise minimization over value maximization in choice behavior"; Baltieri et al., "Predictions in the eye of the beholder"
    
70. Smith et al., "Towards a formal neurophenomenology of metacognition"
71. Smith et al., "Towards a formal neurophenomenology of metacognition"
72. Claire Wardle, "Misinformation Has Created a New World Disorder"
73. Marco Lin, "Sense-making in Stereo"
74. Veissière et al., "Thinking Through Other Minds"
75. Schwartenbeck et al., "Evidence for surprise minimization over value maximization in choice behavior"
76. Friston et al., "Sophisticated Inference"
77. Friston et al., "Sophisticated Inference"
78. Friston et al., "Sentience and the Origins of Consciousness"
79. Karl Friston, "The mathematics of mind-time"
80. Schwartenbeck et al., "Evidence for surprise minimization over value maximization in choice behavior"
81. Friston et al., "What is value - accumulated reward or evidence?"