The Predictive Dynamics of Happiness and Well-Being

Abstract

We offer an account of mental health and well-being using the Predictive Processing Framework (PPF). According to this framework, the difference between mental health and psychopathology can be located in the goodness of the predictive model as a regulator of action. What is crucial for avoiding the rigid patterns of thinking, feeling and acting associated with psychopathology is the regulation of action based on the valence of affective states. In PPF valence is modelled as error dynamics - the change in prediction errors over time. Our aim in this paper is to show how error dynamics can account for both momentary happiness and longer-term well-being. What will emerge is a new neurocomputational framework for making sense of human flourishing.

Full text

1
Title: The Predictive Dynamics of Happiness and Well-Being
Authors:
Mark Miller 1 (markmiller@chain.hokudai.ac.jp)
Erik Rietveld 2, 3 (d.w.rietveld@amsterdamumc.nl)
Julian Kiverstein 2 (j.d.kiverstein@amsterdamumc.nl)
Affiliations:
1. Center for Human Nature, Artificial Intelligence and Neuroscience, Hokkaido University,
Japan.
2. Department of Psychiatry, Amsterdam UMC, University of Amsterdam, Meibergdreef 9,
Amsterdam, Netherlands.
3. Department of Philosophy, University of Twente, Ravelijn 4232, P.O. Box 217, 7500 AE
Enschede, Netherlands.
Acknowledgements:
We are grateful to Abby Tabor, Mahault Albarracin, Brett Anderson, Ines Hipolito and Andy
Clark for helpful comments and discussions along the way. Mark Miller carried out this work
with the support of Horizon 2020 European Union ERC Advanced Grant XSPECT - DLV-
692739. Julian Kiverstein and Erik Rietveld are supported by the European Research Council in
the form of ERC Starting Grant 679190 (EU Horizon 2020) for the project AFFORDS-HIGHER
awarded to Erik Rietveld, and by a project grant from the Amsterdam Brain and Cognition
research group at the University of Amsterdam.
Abstract:
We offer an account of mental health and well-being using the Predictive Processing Framework
(PPF). According to this framework, the difference between mental health and psychopathology
can be located in the goodness of the predictive model as a regulator of action. What is crucial
for avoiding the rigid patterns of thinking, feeling and acting associated with psychopathology is
the regulation of action based on the valence of affective states. In PPF valence is modelled as
error dynamics - the change in prediction errors over time. Our aim in this paper is to show how
error dynamics can account for both momentary happiness and longer-term well-being. What
will emerge is a new neurocomputational framework for making sense of human flourishing.
Key words: predictive processing, error dynamics, valence, happiness, reward, well-being

2
“Whatever Is Flexible and Flowing Will Tend to Grow, Whatever Is Rigid and Blocked Will
Wither and Die”. Tao Te Ching
Introduction
The predictive processing framework (henceforth “PPF”) has recently been proposed as a
unifying theory of the embodied brain and its cognitive functions (Friston 2010; Hohwy 2013;
Clark 2013, 2016). The central idea behind PPF is that the embodied brain is a predictive model
of the body in the world. This model is used to generate predictions of, among other things, the
sensory outcomes of the organism’s actions in the world. These predictions can be compared
with the sensory states the organism actually visits when it acts. So long as the organism keeps
the error in its predictions to a minimum over time, the organism will typically succeed in
achieving the valued outcomes it aims for in acting. PPF is an increasingly influential theoretical
framework for studying mental illness in computational psychiatry.
1
However, so far little
attention has been given to the question of mental health and well-being from the perspective of
the PPF. We take first steps in this direction in our paper.
We take as our starting point the proposal that to be mentally healthy an organism must be a
good predictor of the hidden causes (environmental and bodily) of its sensory states. Such an
organism will tend to behave in ways that maintain homeostasis at each moment in time. While
we take this to be an important part of the story, we use the example of substance addiction to
explain why moment by moment prediction-error minimisation is probably not sufficient for
mental well-being (Miller, Kiverstein & Rietveld 2020). What goes wrong in addiction offers
clues about what it means for a human being to be well. Addiction is an example of a sub-
optimal strategy that an agent can pursue for reducing prediction error (i.e., bring about a certain
goal) by relying on an over-learned, habitual form of behaviour. What is harmful to an agent here
is the ways in which drugs of addiction engender a rigidity in thinking and acting (see Miller et
al. 2020, Barrett & Simmons 2017). Drugs of addiction do this by acting on dopaminergic
systems that strengthen drug seeking and using policies at the expense of alternatives.
We will argue that what is crucial for avoiding pathological forms of rigidity as seen in addiction
is managing error based on sensitivity to error dynamics - the change in the rate of error
reduction (Joffily & Corricelli 2013; Van de Cruys 2017; Kiverstein et al. 2017). Change in rate
means that error reduction has either unfolded worse or better than expected. When agents do
1
See e.g. Fletcher & Frith 2009; Seth, Suzuki & Critchley 2012; Edwards et al 2014; Friston et al 2014; Corlett &
Fletcher 2014; Seth & Friston 2016; Barrett et al 2017; Badcock et al 2017; Smith et al. 2020; Linson & Friston
2019; Paulus et al. 2019; Barca & Pezzulo 2020; Ciaunica et al. 2020; Gerrans & Gadsby 2020; Prosser 2018;
Hipolito et al. forthcoming;; Fotopolou et al. 201x; Linson & Friston 2019; Wilkinson (delusions); Linson, Parr &
Friston 2020; Parr, Rees & Friston 2018; Adams et al. 2013; Schwartenbeck & Friston 2016; Smith, Badcock &
Friston 2020; Ramsteade et al. forthcoming; Miller et al. 2020; Kiverstein et al. 2020; Dean, Miller & Wilkinson
2020).

3
better or worse than expected this is registered in the body as positively or negatively valenced
affect. Agents that use these affective states to regulate their behaviour will be driven to
continuously make progress in error reduction. As we will show, this will require them to
sometimes disrupt their own habits of thinking and acting in ways that temporarily lead to
increases in error and uncertainty but that in the long-run allow them to make progress in
learning. That is, they will sometimes perform actions that temporarily lead to an increase in
uncertainty if doing so will help them to do better at reducing error in the long-run. This is
precisely what does not happen in long term addicts. What turns out to be important for
optimality is being attuned to opportunities for making progress in error reduction. We will
characterise this attunement in terms of metastable dynamics
2
or what we will call metastable
attunement. We will argue that metastable attunement is conducive to well-being because it
allows an agent to remain in touch with and integrate their various cares and concerns over a life-
time.
1. A Predictive Processing Account of Mental Health
The definition of well-being has proven to be controversial among psychologists. The debate has
turned upon different conceptions of “the good life”, with some psychologists favouring a
hedonic view that understands well-being to consist of a life of positive experiences such as
pleasure and happiness (Kahneman et al. 1999). The other tradition has drawn upon ancient ideas
of eudaimonia understanding well-being to consist of fulfilling or realising one’s potential as a
human being (Ryff 1995). What makes a person better-off according to the eudaimonia tradition
need not include pleasure or the satisfaction of their desires. In what follows we will have
something to say about the computational differences between momentary subjective happiness
and overall well-being. However, for the most part we set aside the debate concerning the nature
of psychological well-being. Instead, we take as our starting point the well-established
conceptual connection between mental health and well-being.
3
There is currently a growing literature within the field of computational psychiatry that applies
the predictive processing framework (PPF) to model various psychopathologies including
schizophrenia, depersonalisation, autism spectrum disorder, obsessive compulsive disorder,
major depression, eating disorders, post-traumatic stress disorder, among others.
4
These
psychopathologies are characterised by diverse behavioural, cognitive and emotional symptoms
that manifest differently across individuals. Computational psychiatry provides a formal
2
Interestingly, metastability has been shown to be related to well-being elsewhere in the literature on the
neurobiology of eudaimonia (Kringelbach & Berridge 2017).
3
This connection is enshrined for instance in the 1948 Constitution of the World Health Organisation health is
defined as “a state of complete physical, mental and social well-being.” We will assume the WHO definition of
health is roughly along the right lines and health does indeed consist in a state of complete well-being. Our overall
argumentative strategy will be to consider whether this account of mental health can also help us to understand the
state of “complete well-being” in naturalistic terms as a biological condition of persons.
4
For a non-exhaustive sampling of this literature see the references in footnote 1.

4
framework for relating symptom expression to neurocomputational mechanisms based upon a
general theory of inference and control in biological systems (e.g., Montague et al. 2012). What
seems to be common to psychopathologies are abnormal beliefs of various kinds and their
behavioural consequences (Friston et al. 2014). Thus, it makes sense to seek an explanation of
the expression of complex symptoms characteristic of a given psychopathology in terms of the
inferential mechanisms that lead to the formation of these abnormal beliefs, and the control
processes that result in pathological behaviours.
According to the PPF, agents infer beliefs and control their actions by maintaining and
continuously updating a hierarchically structured generative model of their environment. The
generative model is used to approximate Bayesian probabilistic inferences under conditions of
uncertainty. When the agent gathers new sensory evidence, it must combine a likelihood function
(a probabilistic mapping from hidden states of the world and their dynamics x to sensory inputs
y) with its prior beliefs (a probability distribution that predicts possible states of the world over
time x). These two probability distributions (the prior beliefs and the likelihood function) are
referred to as the “generative model” (Ramstead et al. 2020). The likelihood and prior beliefs are
described as a generative model because they can be interpreted as mapping how sensory inputs
y are believed to be generated by states x of the environment. Given some sensory observations
the generative model is used to compute the posterior probability of a possible state of the world
that is the cause of those observations.
The generative model, as instantiated in humans, has a deep temporal structure that tracks the
sensory consequences of actions over multiple timescales. Higher cortical layers of the model
track regularities that unfold over longer time scales while lower layers of the generative model
track faster changing events such as the sensory consequences of motor movements or the
control of homeostatic setpoints (Kiebel et al. 2008). Minimising prediction errors in the long-
run requires predicting the sensory outcomes of sequences of actions, sometimes reaching far
into the future. Think for instance of organising your summer vacation. This calls for inferring
plans that, when acted on in the future, bring about the preferred outcomes that are predicted -
you’re visiting a holiday destination in Greece. Prediction error here signals a mismatch between
the predicted outcomes that one is aiming at in the future and the actual outcomes of one’s
actions. The generative model has temporal depth insofar as it aims to control the future
outcomes of actions or expected prediction error.
The goodness of a model m can be measured by the model-evidence E, where E is the probability
of sensory observations the agent samples when it acts, given the model m it uses to control its
actions. Model-evidence is identical to negative surprise of a sensory observation y under a
model m. A good model is one that minimises “surprise” understood in the technical
information-theoretic sense of the negative log probability of a sensory observation given a
model. Given some new sensory observations (prediction errors) that do not fit with the model’s

5
prior predictions, the agent should update its model so as to infer a new posterior prediction that
better explains its sensory observations. A model that minimises surprise in the long-run
(understood as long-term prediction error) qualifies as a good model. Our initial proposal is
therefore to understand mental health very broadly, in terms of the “goodness” of a generative
model the agent uses to form beliefs about the world and to control its actions.
According to the PPF, the abnormal beliefs that arise in various psychopathologies are
hypothesised to be the consequence of an agent making use of a generative model whose prior
predictions persistently fail to match with its sensory observations (Friston et al. 2014, for
additional references see footnote 1). In order to avoid persistent surprise an agent will need
some means of assessing the uncertainty of its prior predictions and of the likelihood function in
a given context. These estimations of uncertainty can then be used to modulate the influence of
new sensory evidence (prediction errors) on the model’s subsequent predictions. The agent’s
uncertainty is referred to as precision, a measure of the reliability of information. The weighting
that is given to the likelihood relative to past learning is referred to as the precision of the
prediction error where precision refers to the inverse of the variance of a probability distribution.
We can think of the precision of the prediction error as equivalent to the learning rate. Thus,
precision of the prediction error is high when the likelihood is estimated to be precise, but
decreases with precision of the prior predictions. The result of this kind of precision weighting is
that inferential processes rely on past learning when new sensory information is weighed as
imprecise and unreliable.
5
Aberrant precision estimation is what leads to abnormal beliefs of the kind seen in
psychopathology. The PPF claims that this failure to find the right balance between the precision
of prior beliefs and current sensory evidence may be common to many different
psychopathologies.
6
When too much, or too little, precision is given to prediction error signals,
the agent will operate with a model whose predictions will come to diverge substantially from
the sensory states it samples. Decreasing precision, for example, can lead to prior predictions
dominating, as happens in schizophrenic delusion (Fletcher & Frith 2008; Corlett et al. 2010;
Corlett & Fletcher 2014). By contrast, in Autism Spectrum Disorder too much precision is given
to prediction errors relative to prior predictions (Pellicano & Burr 2012; Lawson et al. 2014;
Palmer, Lawson & Hohwy 2017, Karvelis et al. 2018). People with autism are hypothesised to
5
Three different sources of uncertainty can be distinguished (Parr & Friston 2017). The first source of uncertainty is
the likelihood that may be excessively precise or imprecise. The result of estimating the precision of the likelihood
function is a weighing of the reliability of a prediction error signal. A second source of uncertainty are the priors that
map the dynamics of environmental causes. Volatility and noise may make for a high degree of unreliability in such
mappings. A third source of uncertainty concerns the sensory states the agent has control over through their actions.
The agent can be more or less confident that an action policy (a sequence of actions) will lead to the sensory states it
predicts.
6
See Hipolito et a. (forthcoming) for a computationally rich account of how “insulated” internal states, states that
fail to be updated relative to incoming information, show psychotic (maladaptive) behaviours due to inevitable
increases in deluded beliefs.

6
rely too much on current sensory information and only weakly on prior beliefs in making
inferences about the state of the world over time.
Interestingly, the pathological behaviours that ensue are the result of processes that approximate
Bayesian inference (Schwartenbeck et al. 2015). What makes the agent’s behaviour pathological
and sub-optimal is the generative model, and the prediction errors the agent repeatedly
encounters, when they use the predictions of this model to control their actions. In the next
section, we will show how health more generally can be tied to processes of allostatic control
that ensure the body has the necessary metabolic resources available to meet the challenges of its
environment. We will show how in the PPF, allostasis can be modelled as a process of prediction
error minimisation.
2. Health and Allostatic Control
In this section, we expand on the claim made above, in part based on this literature in
computational psychiatry, that the difference between mental health and psychopathology can be
located in the goodness of the generative model as the regulator of the agent’s behaviour. This
proposal is related to what Conant and Ashby (1970) called the good regulator theorem, which
states that an agent is only able to effectively regulate or control the states of its environment if it
is a “good” model of its environment. We have seen above that the goodness of a generative
model derives from its model-evidence. A good model is a model that maximises model-
evidence or minimises surprise. Recall that surprise is to be understood as a mathematical
measure of the unexpectedness of sampling a sensory state given a model. Maximising model
evidence is identical to minimising surprise because the evidence the agent gathers for a model
comes in the form of the sensory states the agent samples when they act to test the model’s
predictions. So long as the agent keeps the surprise of its sensory states to a minimum, they will
succeed in maximising the evidence for the predictions of their model. In order for a generative
model to minimise surprise, the sensory states the generative model predicts must be equal to the
number of ways the environment can influence the agent, combined with the number of ways the
agent can influence the environment through its actions. A generative model whose predictions
systematically diverge from the states of the world, and their dynamics will fail to function as a
good regulator of the agent’s behaviour.
Of all the possible sensory states the organism can find itself in, a small subset will prove to be
consistent with the organism remaining well-adapted to its changing environment. Sensory states
belonging to this subset will include internal states of the body sensed through interoception and
vital to the organism’s continued existence (e.g., respiratory rate, blood acidity, glucose levels,
bodily temperature, and plasma osmolality). These states of the body are maintained within a
tight range of values compatible with the organism’s viability through feedback control.
Whenever the organism senses a deviation from these set-points, processes of physiological,

7
hormonal and immunological regulation ensure that internal bodily states swiftly return to the
set-points consistent with the organism’s continued existence.
However, many of the brain and body’s regulatory responses are not reactive but anticipatory. A
predicted deviation from so-called “homeostatic setpoints” is avoided by taking preparatory
action in advance of the deviation’s occurrence. This process is referred to as “allostasis”,
meaning the stability of the internal conditions of the body through change (Sterling & Eyer
1988; McEwen & Stellar 1993; McEwen 2000).
7
Allostasis is a form of predictive regulation
where anticipatory actions are selected that ensure that the organism’s needs are prioritised and
opportunities are weighed against dangers.
8
Examples of allostatic systems include hormonal,
autonomic, and immune systems. The responsiveness of these systems is optimal when the brain
is able to predict and accommodate the demands on the body before they arise. For instance,
blood pressure varies continuously throughout the day. When the individual can let their guard
down (e.g., during sleep) blood pressure drops sharply (Young et al. 2004; Lightman et al. 2020).
When we wake in the morning, blood pressure ramps up in anticipation of stress and the need to
remain vigilant. A comparable increase in blood pressure occurs during sexual intercourse.
Fluctuations above and below an average state occur throughout the day depending on the need
for the organism to maintain a state of vigilant arousal. Blood pressure is thus regulated to match
the demands of a dynamically changing environment. The result of this matching of the body’s
resources to the predicted demands on its physiology and metabolism is the efficient regulation
of the body’s responsiveness to its environment.
If the body encounters constant high demand, this can result in the body adapting its predictions
and remaining in a state of high arousal. Chronic stress arising from poverty, physical and
emotional abuse, or loneliness leads the body to predict constant environmental challenges
(McEwen 1998, 2000; Seeman & McEwen 1996; Adler & Ostrove 1999; Sterling 2011; Seeman
et al. 2010; Cacciopo et al. 2015; Wilkinson & Pickett 2010). Just as muscles can learn to
anticipate exercise, so also the body can learn to anticipate stress. This regulatory circuit can
eventually enter into a pathological feedback loop. Arteries thicken and harden, consequently
requiring higher pressure which further reinforces their stiffness (Sterling 2018: p.9). Chronically
high blood pressure leads to inflammation of the nervous system and eventually to heart disease
or stroke. So long as the body continues to predict the need for high blood pressure (an example
of high “allostatic load”), the cycle will be very difficult to break.
7
Cf. Sterling 2011; Power & Shulkin (2012). The latter defines allostasis as the means by which the body
reestablishes homeostasis in the face of a challenge (p.25, cited by Corcoran & Hohwy (2018: p.4)). McEwen
conceives of allostasis as anticipatory physiological responses aimed at restoring homeostatic variables to the range
of values that allow for the maintaining of the organism’s biological viability
8
Sterling distinguishes allostasis from homeostasis on the grounds that the latter is reactive relying on negative
feedback however the distinction cannot be drawn in this way if one thinks of homeostasis as working through
active inference. Both processes are equally anticipatory, proactive and predictive. For a discussion of the relation
between homeostasis and allostasis in the PPF see Stephan et al. 2017; Corcoran & Hohwy 2018.

8
We propose that health can be understood in terms of processes that forecast the likely demands
on the organism’s body by maintaining a generative model. This model is used to predict how
signals arising internally and externally to the body are likely to evolve over time. Predictions
track the likelihood that actions will maintain the body within the range of physiological,
hormonal and immunological values consistent with its remaining well-adapted to the challenges
of its environment. The proposal we explore in the remainder of this paper accounts for mental
health and well-being in terms of a generative model that works in the service of allostatic
control. In the next section we will take up the idea introduced above that maintaining a good
model depends on the agent being able to estimate their own uncertainty in relation to who they
are, what they are doing, and the world around them. Failure to accurately estimate uncertainty is
thought to underlie various pathologies including addiction, a point we will return to in section
four.
3. Reward, Error Dynamics and Momentary Happiness
We have seen in the previous section how mental health depends on estimations of uncertainty.
Assigning too much precision, or too little precision, to prediction errors can result in abnormal
beliefs and a generative model that fails to get a good grip on incoming sensory information. In
this section we will show how precision predictions are maintained in part by tracking the rate of
change in error reduction.
According to PPF, the outcomes of actions that are preferred and valued are highly expected, and
the agent selects actions that fulfill those expectations (den Ouden et al., 2010, Friston et al.,
2009, Clark, 2015, FitzGerald et al., 2014, Friston et al., 2012; Kiverstein, Miller & Rietveld
2017). Dopaminergic discharges (and other neuromodulatory chemicals such as serotonin,
oxytocin, and norepinephrine) weigh the precision of a belief that an action policy will bring
about expected outcomes (Friston et al., 2012; Schwartenbeck et al. 2014; Linson et al., 2018;
Parr & Friston 2017). When we do worse than expected, the unexpected sensory and
physiological states are punishing because they are states the outcomes which were not well
predicted by the agent, perhaps because the agent does not have a good grip on the volatility of
the environment or because they are acting on a high risk policy. Doing better than expected at
reducing error indicates by contrast that there is less volatility or risk than one expected. One is
therefore able to do better than expected at bringing about the valuable sensory states that one
predicts to be the consequences of one’s actions.
In recent work we have highlighted the role of doing better than expected at error reduction in
contributing to precision estimation (Kiverstein, Miller & Rietveld 2017, 2020; see also Hesp et
al. 2021). Unexpected increases or decreases in volatility are good information for the agent
about how confident they can be that an action policy will lead to expected outcomes.
Unexpected decreases in the rate of error reduction informs the organism that a belief in an

9
action policy should be assigned lower confidence. An unexpected increase in rate of error
reduction informs the organism that things are going better than expected. Precision, then, is
adjusted on action policies not only based on the amount of error or error reduction occurring in
the system, but also the rate at which error is managed over time (Kiverstein, Miller & Rietveld
2017, 2020; Hesp et al. 2021)
Error dynamics - the rate of change in error reduction - are registered by the organism as
embodied affective states (Kiverstein, Miller & Rietveld 2017, 2020; Joffily & Coricelli 2013;
Van de Cruys 2017; Hesp et al. 2021; Haar et al., 2020). We can think of an agent’s performance
in reducing error in terms of a slope that plots the various speeds that prediction errors are being
accommodated relative to their expectations. Positively and negatively valenced affective states
are a reflection of better than or worse than expected error reduction, respectively. Valence refers
to the organism’s evaluation of how it is faring in its engagement with the environment (i.e., how
well or badly things are going for the organism). Think, for example, of the frustration and
agitation that commuters feel when their train is late, and they have an urgent meeting to attend.
These negative feelings are, in part, the body informing the system that some relevant source of
error was expected to have been reduced by now but is not. The unexpected rise in error at the
train's tardiness is felt in the body as an unpleasant tension. That tension may provoke the agent
to check the transit authority for delays or find an alternative (more reliable) means of transport
such as a taxi in order to reduce the felt tension - to catch back up to their previous slope of error
reduction. We will henceforth describe optimally functioning agent’s as being motivated to seek
out good slopes of error reduction.
From this perspective, momentary subjective happiness is the result of unexpectedly reducing
prediction error. This feels good because we have done better than expected at improving our
predictive grip on the environment, something our very health depends upon (Sterling 2018,
2020). There are already a number of well-established approaches to understanding the
neurobiology of momentary happiness that point in a similar direction. For example, Ruttledge
and colleagues have, over a number of brain imaging experiments, demonstrated a strong
relationship between subjective feelings of happiness and better than expected performance
(2014, 2015).
9
Positively charged affect plays an important role in the predictive system. It ups
the learning rates for situations in which there is a prime opportunity to learn how to adapt to the
demands of the environment more efficiently, which is the modus operandi of the predictive
system. We will see in the next section, however, that while this is no doubt an important part of
9
Rutledge and colleagues had subjects engage in a probabilistic reward task, where they selected between various
risky monetary options. Participants were asked between trials: “how happy are you right now?”. Rutledge and
colleagues showed that the feeling of happiness comes not when participants received a monetary payoff but when
they did better than expected relative to their previous performance. This tracking of better-than-expected gains
shows up in the brain in reward-related midbrain dopaminergic activity (Rutledge et al. 2014, p.12255). Instead of
taking dopamine to track reward prediction errors, we have suggested that dopamine may track the rate of change in
reduction of prediction error.

10
what it is to be well as a human being, it is not the whole story PP has to offer. Addicts can
maximise their momentary subjective happiness but still find themselves in sub-optimal modes
of engaging with their environment.
4. Bad Bootstraps and Sub-Optimal Grip
There are various dangers and difficulties that can arise in the optimisation of a generative
model. The central role that prediction plays in generating perception and action means that
hidden biases have tremendous power to direct behaviours in ways that tend to produce the
outcomes that confirm just those biases. Relative to a predictive model, the agent can find
themselves acting in ways that confirm their predictions, thus allowing them to minimise
prediction error. Thus, having a generative model that succeeds in minimising prediction error is
thus no guarantee of optimal psychological functioning.
Take as an instructive example long term substance addiction. Substances of addiction impact on
the midbrain dopaminergic systems in the same way as unexpected rewards.
10
This has the effect
of training expectations about the rate of error reduction both in the present moment, and over
the longer term (Miller et al. 2020). The drug user comes to expect a tremendous reduction in
error each time they use a substance. The continued release of dopamine that accompanies the
use of the substance makes it seem as if the addictive substance is always and endlessly
rewarding. The agent learns that nothing else in their life can reduce error in such a dramatic
fashion. As a consequence, the agent neglects other policies that could serve the agent’s goals.
They get caught in a vicious cycle in which they act to fulfill the prediction that the drug seeking
and drug using action policies are the best opportunity for realising their preferred and valued
outcomes. So strong is the pull of the policy to use the addictive substance that the person pays
no attention to other action policies that may also be of relevance to them. As they lose touch
with their other cares and concerns error inevitably begins to build (e.g., health begins to
degrade, relationships fall apart, jobs are lost), which in turn motivates the drug seeking and
taking behaviours as a means of regulating the increasingly unmanageable levels of error.
A recent agent-based model showed that in order to optimise a model of the environment an
agent must strike the right balance between epistemic actions that explore the environment for
new policies, and pragmatic actions that exploit existing policies (Tschantz et al. 2020). A model
that generates only pragmatic actions, like we see in the addition example above, will lead an
agent to an overly rigid, sub-optimal course of behaviour we will henceforth refer to as a “bad
bootstrap” (following Tschantz and colleagues). A model that generates only epistemic actions
will be accurate and comprehensive, but it will fail to guide behaviour towards relevant
10
Psychostimulants (e.g., cocaine, and amphetamines) act directly on this system producing a burst of dopamine as
if the organism was encountering something which is needed. Opiates (e.g., heroin and morphine) inhibit
GABAergic neurons leading to the disinhibition of dopamine neurons (Khoshbouei et al 2003).

11
possibilities for action in a dynamically changing environment. Agents learn an optimal model
through strategies for balancing exploratory epistemic actions with exploiting what is already
known for the purpose of pragmatic action. One way that organisms strike this optimal balance is
by setting precision over action policies using their sensitivity to error dynamics. We will
suggest it is negotiating this explore-exploit trade-off by means of sensitivity to error dynamics
that is key to well-being. First, we use substance addiction to provide an illustration of how the
prediction-minimising agent can get trapped in bad bootstraps.
Substance addiction is an example of a bad bootstrap because precision estimation over action
policies is context-insensitive. Addicts choose the familiar option of seeking and using the drug,
and continue to do so even when the outcomes are negative. In order to learn an optimal
generative model an agent must flexibly update the estimation of precision on action policies
with changes in context. PP theorists see addiction as a problem that arises when the higher-
levels of the hierarchy (which is where the person’s longer-term goals are encoded) are no longer
assigned precision) (Pezzulo, Rigoli & Friston 2015; Clark 2019). Addiction, then, can be
thought of as the result of a loss of contextualization between higher (cortical) and lower
(subcortical) neural behavioural controllers. Goal-directed control at higher cortical levels
provides the context for simpler habit-based and sensorimotor forms of control at lower-levels of
cortical hierarchy. As drug-related habits become increasingly powerful, all the other goals that
matter to the agent such as going to the gym or pursuing a promotion at work come to be
neglected. Pathological forms of addiction arise when goal-directed and habit-based control
come into conflict. The result of this conflict is a buildup of error in the person’s life. Predictions
related to goal-directed control at higher layers in the cortical hierarchy are trumped by highly
precise prediction errors associated with drug-seeking and using behaviours. Instead of habit-
based forms of control working in the service of fulfilling predictions arising from longer-term
goals and concerns, habit-based control comes to drive action in isolation from goal-based
predictions.
The key question the brain must settle is whether the agent is in a context in which habits can be
relied upon to bring about valuable outcomes. Should the agent instead invest effort to explore
for more valuable outcomes that do a better job of fulfilling long-term goals? To settle this
question, however, requires the context-sensitive updating of precision estimation, which is
exactly what fails to happen in pathological cases of addiction. People struggling with addiction
tend not to gather more evidence that might lead them to change their behaviour. At least, they
fail to do so until they are able to see through the illusion of error reduction induced by the
effects of substances of addiction on the systems that estimate the precision of action policies.
The failure of this context-sensitive adjustment of precision leads the global dynamics of the
brain to get trapped in fixed-point attractors that lead to a single attractive outcome. Fixed point
attractors are contrasted with itinerant policies that allow for epistemic actions, and the

12
exploration of sets of attractive states (Friston 2012; Zarghami & Friston 2020). Any given
neural region can perform multiple functions over time depending on the patterns of effective
connectivity it forms with other neural regions.
11
This multifunctional profile allows for task-
specific coalitions to be configured on the fly as and when they are needed in a context-
dependent manner (Anderson 2014; Clark 2016, ch.5). Recall that it is by means of the constant
adjustment of precision estimations that patterns of effective connectivity in the brain emerge
and change from moment to moment (Zarghami & Friston 2020). We’ve suggested above that
neurotransmitters track the rate of change in error reduction (amongst other things). Positive and
negative changes in the rate of error reduction are sensed in the body as positive and negatively
charged affective states. We suggest these affective states (when all is going well) serve as an
endogenous source of instability ensuring that neural coalitions form, dissolve, and reform in the
brain in a context and task-dependent manner. In bad bootstraps rigid affect can have the
opposite effect, trapping the global dynamics of the brain in sub-optimal patterns of engagement.
Bad bootstraps can be conceived of in dynamical systems terms as the loss of metastable
dynamics.
12
Metastability is the consequence of two competing tendencies of the parts of a
system to separate and express their intrinsic dynamics and to integrate and coordinate to create
new dynamics (Kelso 1995; 2012). In a metastable system, there is “attractiveness but, strictly
speaking, no attractor" (Kelso & Engström, 2006, p. 172; cf. Araújo et al. 2014). Attractor states
describe the states in a system’s phase space that the system tends to converge on when
contextually perturbed. Metastable systems transit between regions of their state space
spontaneously without the need for external perturbation. The organisation of a metastable
system is therefore transient. For short periods, coordination among the parts emerges reflecting
the tendency of the parts of the system to integrate. However, due to the tendency of the same
parts to segregate, a recurring destruction of this coordination can also be observed as the
behaviour of the component parts escapes from each other’s orbit of influence. In the brain we
see this creation and destruction of coordination in large-scale global patterns of synchronous
and desynchronised activation across neuronal ensembles (Friston 1997, 2000; Varela 1999;
Varela et al. 2001; Deco & Kringelbach 2016; Zarghami & Friston 2020). The brain as a
metastable system is typically poised between stability (coordination of parts) and instability
(segregation of parts) remaining close to a critical state from which the system can spontaneously
shift from a coordinated to a disordered state and back again. We will close our paper by
explaining why this poise between stability and instability might be necessary for well-being.
5. Metastable Attunement and Wellbeing
11
“Effective connectivity” refers to the short-term moment to moment patterns of causal influence between neurons
modelling by Dynamic Causal Modelling (Kiebel et al. 2009).
12
Friston (2012) proposes that “metastability” is jeopardized in addiction by precision weighing being set too high
on a certain set of sensory errors. This in turn specifically impedes itinerant wandering policies characteristic of
metastable dynamics - the visiting of a succession of unstable fixed points in a phase space (Rabinovich et al. 2008).
For more on this point, see section 5 below.

13
Agents like us that live in complex dynamic environments will benefit from remaining at the
edge of criticality between order and disorder, between what is well known (and reliable) and the
unknown (and potentially more optimal). Frequenting this edge of criticality requires that
predictive organisms are prepared to disrupt their own fixed-point attractors (habitual policies
and homeostatic setpoints) in order to explore just-uncertain-enough environments that are ripe
for learning about their engagements. When things are going well, and they are on good slopes of
error reduction, they should continue on the same path. When, however, a niche is so well
predicted that there ceases to be good slopes of error reduction available, agents should begin to
explore for opportunities to do better. Rate of error reduction is continuously changing. We will
argue that if an agent uses error dynamics to set precision on action policies this will have the
consequence that they avoid getting stuck in any attractor state. We will refer to this dynamical
state of remaining metastably poised as a state of “metastable attunement”. By tracking the
changing rate of error reduction, such an agent will be attuned to opportunities to continually
improve in error reduction.
Metastable attunement moves the agent in such a way that they find the balance between
exploiting existing action policies and performing information-seeking epistemic actions that aim
at reducing uncertainty. We have seen above how slower dynamics at higher layers of the
hierarchical generative model provide the context that constrains the faster changing dynamics at
lower layers of the generative model (Friston et al. 2020). The patterns of effective connectivity
that form between higher and lower layers of the model are transient, changing each moment on
the basis of precision assigned to policies. These patterns form, we have suggested, because of
the role of valence in sculpting patterns of effective connectivity. Given the connection between
valence and error dynamics, large-scale neural coalitions change from moment to moment in
ways that reflect changes in the rate of error reduction. When a particular niche ceases to yield
productive error slopes negative valence signals to the agent that they ought to destroy their own
fixed-point attractors in favor of more itinerant wandering policies of exploration. Patterns of
effective connectivity emerge and dissolve due to both environmental conditions and changes in
our own internal states and behaviours. However, we also have a tendency to actively destroy
these attractor states, thereby inducing instabilities and creating peripatetic or itinerant
(wandering) dynamics (Friston, Breakspear, and Deco 2012). Alternatively, when errors
accumulate, due to our frequenting spaces where there is an unmanageable complexity or
volatility, the negative valence then tunes the agent to fall back on opportunities for action that
are already well known and highly reliable. Notice, when all goes well such slope-chasing agents
will be constantly moved by their valenced affective states (via changes in error dynamics)
towards this edge of criticality, where error is neither to complex nor too easily predicted that the

14
agent no longer has anything to learn (Kiverstein, Miller & Rietveld 2017; Anderson et al.
2020).
13
Being attuned in this way to the edge of criticality makes for a resilient agent, one that can
readily adapt to environmental challenges in a way that we have seen is necessary for allostasis.
Systems that frequent this edge of criticality have fitness advantages over other more strictly
ordered or chaotic systems because they strike an optimal balance between efficiency and
degeneracy (Sajid et al. 2020). Such systems are able to respond efficiently to particular contexts
of activity while also remaining open to exploring a wide variety of other possible contexts to
bring about their goals (degeneracy) (Roli et al. 2018). This is precisely what people suffering
from long term addiction tend to fail at - highly precise drug seeking and taking behaviours
overwhelm the system leading it to inflexibly select those drug related policies even when other
more beneficial policies may be available. Bad bootstraps like addiction create fragility in a
dynamical system due to their making the system rigid and so less adaptable to a changing
environment.
We have seen that metastable attunement allows the agent to remain poised over a multiplicity of
possible actions. To put this in a different vocabulary from ecological dynamics: agents that are
metastably attuned are able to maintain grip on a field of affordances as a whole (Bruineberg &
Rietveld 2014; Rietveld, Denys & van Westen, 2018). This is because an agent that is able to
remain at the edge of order and disorder will combine flexibility with robustness. Think of the
boxer finding an optimal distance from the boxing bag where she is ready for all the relevant
affordances the bag offers (Chow et al. 2011; Hristovski et al. 2009). She is ready to make jabs,
uppercuts and hooks based on her distance from the bag. Given this bodily readiness, a random
fluctuation of the bag then contributes to the selection of which action unfolds and which
affordance she engages first. Systems that maintain metastable attunement are poised in a way
that allows them to make the most of the affordances relevant to them, and to learn the most
about the environments they frequent (see for example, Shew & Plenz 2013; Shew et al. 2011;
Gautam et al 2015).
We suggest a distinction is therefore needed between local error dynamics that allow for the
tuning of precision in relation to a particular action policy, and global error dynamics that track
how well the agent is doing overall given the many affordances that are relevant to them.
14
Local
13
Prediction errors that are neither too complex for a model to resolve nor too simple for the model to learn anything
from we have called “consumable errors” (Miller et al. forthcoming). As slope-chasers we are motivated to seek out
just the right quantities of manageable error that allow for the improvement of a model’s predictions (Oudeyer &
Smith 2016; Oudeyer, Kaplan & Hafner 2007; Kidd et al 2012; Andersen & Roepstorff, under review; cf. Berlyne
1970). Too many error signals that an environment is unmanageably volatile, while too little error means the
environment is too well known for the predictive mind to learn.
14
See Sandved-Smith et al 2020 for discussions about the structure of higher-level policies governing the allocation
of precisions over lower-level tasks. In this ‘deep parametric’ generative modelling framework it becomes possible

15
success in error reduction is not sufficient for overall well-being. To see why not consider how a
teenager might achieve this kind of improvement in their skills by spending their days playing
computer games.
15
The computer game could provide them with just enough of a challenge to
ensure that they are continually making progress in reducing prediction errors. We can suppose
that the computer game would be designed to provide the player with just the right amount of
prediction error - neither too much so that they find themselves frustrated, nor too little so that
they quickly master the game and become bored of playing it. We can imagine that the game
would create just enough novelty to keep the player engaged. But as with the example of
substance addiction, this continued engagement would come at the expense of everything else in
their lives. They may begin to neglect their friendships, schoolwork, and overall fitness in order
to spend more time playing the game. Such an individual could not reasonably be said to be
flourishing even though they may experience positive affect so long as they are playing the
game.
Given that the agent has many cares and concerns, there will, on any given occasion, be multiple
affordances of relevance to them. An important part of the optimisation of the generative model
are apt predictions about how best to deploy precision in relation to any relevant affordances of
concern to them. Changes in how well these predictions about precision fare can be used in much
the same way as local error dynamics, helping to tune the agent in ways that keep them in touch
with the best slopes of error reduction. However, instead of the slopes of prediction error
management having to do with improvements in a specific domain, the high levels of the
generative model that track global error dynamics pertain to the system’s overall ability to
manage volatility across multiple domains. The time scale of global error dynamics is longer
than local error dynamics pertaining to how the general trend of error reduction is going into the
future. For this reason, we suggest that the levels of the hierarchical generative model that
control the deployment of precision are likely to be higher levels that deal with processes that
unfold over long intervals of time.
Global error dynamics are important for psychological well-being because they allow an agent to
maintain metastable poise over the field of relevant affordances as a whole. So long as the agent
uses global error dynamics to adjust precision estimations, they will tend to act in ways that
reflect their multiple cares and concerns. When an activity does not go as anticipated (say you
are learning a musical instrument and struggling to play a piece of music) you can fall back on
other projects or concerns that you also care about (such as your family relationships). You can
switch from one activity to doing something else that is also expected to lead to valued
outcomes. The result is that an agent can be failing to predict well in some local activity, but
succeeding at predicting how to get into valued sensory states elsewhere, thus resulting in overall
to appreciate the precisions over these higher-level policies themselves, creating a nested hierarchy of error
dynamics corresponding to local vs global considerations. We thank Sandved-Smith for discussions on this point.
15
Our thanks to Andy Clark for pressing us on this point. For discussion of related examples see Clark (2018).

16
predictive success. Such an agent will continually make progress in learning, growing and
broadening their field of relevant affordances, which will, in turn, increase their confidence in
managing unexpected volatility as it arises over the whole of their lives. Since agents that make
use of global error dynamics will do best at reducing error in the long run, they will tend also to
occupy positively valenced affective states. (this follows from the explanation we have given of
positive valence in terms of error dynamics.) This is to say they will tend to experience a positive
hedonic sense of wellbeing over the course of their lives. They will experience a background
mood of positive well-being - feedback that they are succeeding at deploying precision in an
optimal way.
A key component of psychological well-being is therefore continual progress in learning that
metastable attunement makes possible (cf. Kaplan & Oudeyer 2007; Oudeyer & Smith 2016;
Kidd et al. 2012; Clark 2018). Metastable attunement doesn’t just underwrite resilience, it also
allows for the additional possibility of growth or improvement. Finding the right balance
between pragmatic and epistemic actions, which is made possible by metastable attunement, is
key. Doing so means that the agent will be able to optimally reduce long-term uncertainty. The
result is an agent that will sometimes actively induce temporary stress in the form of increased
uncertainty so that they can grow and improve in their skills.
There are certain human activities that increase the likelihood of metastable attunement.
Interestingly these are also arguably activities that contribute to eudaimonic well-being. There
are well established correlations between increased well-being over a lifetime and a focus on
non-zero-sum goals and activities such as altruism, the development of virtue, social activism, a
commitment to family and friends (Headey 2008; Garland et al. 2010). In contrast, pursuit of
zero-sum activities, such as purely financial gains, has been found to be detrimental to life-long
well-being (Headey 2005, 2008). The development of skills and abilities for engaging in non-
zero-sum activities seems to be especially important for creating and sustaining lifelong
satisfaction - or what is traditionally referred to as eudaimonia.
16
Why is this the case? Consider
someone who approaches life as a zero-sum game. They will tend to develop skills and abilities
that are socially antagonistic (Różycka-Tran et al. 2019). One side effect of this approach to life
is that it can lead to missed opportunities for collaboration and social complexifications that
often support long term success or happiness. A zero-sum approach to life tends to reduce or
restrict one of our richest sources for reducing meaningingful prediction-errors: other people. In
contrast, non-zero-sum activities encourage cooperation and collaboration, and therefore
conducive to metastable attunement. These sorts of activities support a continuous opening to
new possibilities and affordances. While the goal of buying a car comes to an end upon
purchasing that car, the goal of being more mindful or compassionate, of being a better partner,
16
Garland and colleagues (2010, 2015) have developed an account of how eudaimonic activities support well-being
by encouraging upward spirals of psychological resilience and flourishing through forwardly-progressing and self-
reinforcing cycles of positive affect and cognition.

17
or serving one's community are all goals that are potentially never finished. These are activities
that allow for the continuous broadening of the field of relevant affordances we described above.
The more one engages with non-zero-sum activities the more opportunities for development
emerge - new skills to hone, new qualities to develop, new people to engage and collaborate
with.
Conclusion
For prediction error minimizing agents like ourselves, optimality refers to our development of a
generative model capable of successfully managing the volatility of our environments over the
long term. Part of that optimization relies on the continual development and refinement of our
various niche-appropriate skills and abilities. As we’ve seen, agents that are behaviourally tuned
by changes in how well or poorly they are doing at reducing prediction error will be attracted to
that critical edge where the most error can be resolved. The most resolvable error tends to be
encountered just above the level of our current skillfulness - not so complex that we cannot get a
good predictive grip and not so well known that there are no productive errors left to resolve.
Momentary subjective happiness signals that our generative model is improving in its
predictions. A system that is tuned by momentary subjective happiness, as we are, naturally
becomes a better predictor of its environment over time. However, while this continuous
progression in prediction is necessary for optimal well-being, it is not sufficient. We only have to
reflect on the various ways that our current designer culture has manufactured for generating
local predictive successes while diminishing our longer-term optimizations. Addictive activities
as a whole are examples of this.
Optimal psychological functioning requires that we are able to continually develop in our various
local projects and balance our metabolic expenditures between those activities in ways that
provide good predictive dividends. Computationally speaking, this balancing occurs when the
predictive system is able to make good predictions about how precision is being allocated to
beliefs about action policies. When those predictions are good the agent is able to optimize the
balance between exploiting well-learned policies and exploring new policies (even when doing
so temporarily leads to increases in error).
We have proposed that optimal psychological function should be thought of as emerging from
maintaining a metastable poise. A system that is sensitive to how it deploys precision, and so is
able to juggle multiple cares and concerns in an optimal way, will also be a system that is best
able to meet and resolve unexpected uncertainty. It is this continual growth of skills and abilities
and the optimal balancing of resources between those domains of learning that produces this
optimal control. And it is this optimal control that is experienced by the agent as a background
feeling of well-being - the felt experience that the system is set up to handle life’s many
challenges.

18
References:
Adler, N. E., & Ostrove, J. M. (1999). Socioeconomic status and health: what we know and what
we don't. Annals of the New York academy of Sciences, 896(1), 3-15
Alexandrova, A. (2017). A philosophy for the science of well-being. Oxford University Press.
Anderson, M. L. (2014). After phrenology: Neural reuse and the interactive brain. MIT Press
Badcock, P. B., Davey, C. G., Whittle, S., Allen, N. B., & Friston, K. J. (2017). The depressed
brain: An evolutionary systems theory. Trends in Cognitive Sciences, 21(3), 182-194
Sutton, R. S., & Barto, A. G. (1998). Introduction to reinforcement learning (Vol. 135).
Cambridge: MIT press
Barrett, L. F., & Simmons, W. K. (2015). Interoceptive predictions in the brain. Nature reviews
neuroscience, 16(7), 419-429.
Boorse, C. (1997). A rebuttal on health. In What is disease? (pp. 1-134). Humana Press, Totowa,
NJ
Bruineberg, J., & Rietveld, E. (2014). Self-organization, free energy minimization, and optimal
grip on a field of affordances. Frontiers in human neuroscience, 8, 599;
Cacioppo, J. T., Cacioppo, S., Capitanio, J. P., & Cole, S. W. (2015). The neuroendocrinology of
social isolation. Annual review of psychology, 66, 733-767
Chow, J.Y., Davids, K., Hristovski, R., Araújo, D., Passos, P. (2011). Nonlinear pedagogy:
Learning design for self-organising neurobiological systems. New Ideas in Psychology, 29: 189-
200.
Clark, A. (2013). Whatever next? Predictive brains, situated agents, and the future of cognitive
science. Behavioral and brain sciences, 36(3), 181-204
Clark, A. (2015). Surfing uncertainty: Prediction, action, and the embodied mind. Oxford
University Press
Clark, A. (2018). A nice surprise? Predictive processing and the active pursuit of novelty.
Phenomenology and the Cognitive Sciences, 17(3), 521-534.

19
Conant, R. C., & Ross Ashby, W. (1970). Every good regulator of a system must be a model of
that system. International journal of systems science, 1(2), 89-97
Constitution of the World Health Organisation 1948
Conference, I. H. (2002). Constitution of the World Health Organization. 1946. Bulletin of the
World Health Organization, 80(12), 983.
Corcoran, A. W., & Hohwy, J. (2018). Allostasis, interoception, and the free energy principle:
Feeling our way forward. In Tsakiris, M., & De Preester, H. (Eds.). (2018). The interoceptive
mind: from homeostasis to awareness. Oxford University Press
Corlett, P. R., & Fletcher, P. C. (2014). Computational psychiatry: a Rosetta Stone linking the
brain to mental illness. The Lancet Psychiatry, 1(5), 399-402
Corlett, P. R., Taylor, J. R., Wang, X. J., Fletcher, P. C., & Krystal, J. H. (2010). Toward a
neurobiology of delusions. Progress in neurobiology, 92(3), 345-369
Deco, G., & Kringelbach, M. L. (2016). Metastability and coherence: extending the
communication through coherence hypothesis using a whole-brain computational perspective.
Trends in neurosciences, 39(3), 125-135;
Edwards, M.J, Adams, R.A., Brown, H., Pareés, I. & Friston, K. (2014). A Bayesian account of
‘hysteria’. Brain 135 (11): 3495-3512.
Fletcher, P. & Frith, C. (2009). Perceiving is believing: A Bayesian approach to explaining the
positive symptoms of schizophrenia. Nature Reviews: Neuroscience, 10: 48-58.
Friston, K. (2010). The free-energy principle: a unified brain theory?. Nature reviews
neuroscience, 11(2), 127-138
Friston, K., Breakspear, M., & Deco, G. (2012). Perception and self-organized instability.
Frontiers in computational neuroscience, 6, 44
Friston, K. J., Fagerholm, E. D., Zarghami, T. S., Parr, T., Hipólito, I., Magrou, L., & Razi, A.
(2021). Parcels and particles: Markov blankets in the brain. Network Neuroscience, 5(1), 211-
251
Friston, K. J., Stephan, K. E., Montague, R., & Dolan, R. J. (2014). Computational psychiatry:
the brain as a phantastic organ. The Lancet Psychiatry, 1(2), 148-158.

20
Garland, E. L., Fredrickson, B., Kring, A. M., Johnson, D. P., Meyer, P. S., & Penn, D. L.
(2010). Upward spirals of positive emotions counter downward spirals of negativity: Insights
from the broaden-and-build theory and affective neuroscience on the treatment of emotion
dysfunctions and deficits in psychopathology. Clinical psychology review, 30(7), 849-864
Garland, E. L., Farb, N. A., R. Goldin, P., & Fredrickson, B. L. (2015). Mindfulness broadens
awareness and builds eudaimonic meaning: A process model of mindful positive emotion
regulation. Psychological inquiry, 26(4), 293-314
Gautam, S. H., Hoang, T. T., McClanahan, K., Grady, S. K., & Shew, W. L. (2015). Maximizing
Sensory Dynamic Range by Tuning the Cortical State to Criticality. PLOS Computational
Biology, 11(12), e1004576. https://doi.org/10.1371/journal.pcbi.1004576
Haar, A.J.H., Jain, A., Schoeller, F., & Maes, P. (2020). Augmenting aesthetic chills using a
wearable prosthesis improves their downstream effects on reward and social cognition. Nature:
Scientific Reports, 10: 21603.
Hesp, C., Smith, R., Parr, T., Allen, M., Friston, K. J., & Ramstead, M. J. (2021). Deeply felt
affect: The emergence of valence in deep active inference. Neural Computation, 33(1), 1-49
Hohwy, J. (2013). The predictive mind. Oxford University Press
Hristovski, R., Davids, K., & Araujo, D. (2009). Information for regulating action in sport:
metastability and emergence of tactical solutions under ecological constraints. Perspectives on
cognition and action in sport, 43-57
Joffily, M., & Coricelli, G. (2013). Emotional valence and the free-energy principle. PLoS
Comput Biol, 9(6), e1003094
Kahneman, D., Krueger, A. B., Schkade, D., Schwarz, N., & Stone, A. (2004). Toward national
well-being accounts. American Economic Review, 94(2), 429-434
Kelso, J. S. (1995). Dynamic patterns: The self-organization of brain and behavior. MIT press
Kelso, J.S. (2012) Multistability and metastability: understanding dynamic coordination in the
brain. Philosophical Transactions of the Royal Society B: 367: 906-918.
Kelso, J. S., Engstrom, D. A., & Engstrom, D. (2006). The complementary nature. MIT press

21
Kidd, C., Piantadosi, S. T., & Aslin, R. N. (2012). The Goldilocks effect: Human infants allocate
attention to visual sequences that are neither too simple nor too complex. PloS one, 7(5), e36399
Kiebel, S. J., Daunizeau, J., & Friston, K. J. (2008). A hierarchy of time-scales and the brain.
PLoS Comput Biol, 4(11), e1000209
Khoshbouei H, Wang H, Lechleiter JD, Javitch JA, Galli A (2003) Amphetamine-induced
dopamine efflux. A voltage-sensitive and intracellular Na+-dependent mechanism. J Biol Chem
278(14):12070–12077
Kringelbach, M. L., & Berridge, K. C. (2017). The affective core of emotion: linking pleasure,
subjective well-being, and optimal metastability in the brain. Emotion Review, 9(3), 191-199
Lawson, R. P., Rees, G., & Friston, K. J. (2014). An aberrant precision account of autism.
Frontiers in human neuroscience , 8 , 302. doi:10.3389/fnhum.2014.00302
Lewis, M. D., & Todd, R. M. (2007). The self-regulating brain: Cortical-subcortical feedback
and the development of intelligent action. Cognitive Development, 22(4), 406-430.
Lightman, S. L., Birnie, M. T., & Conway-Campbell, B. L. (2020). Dynamics of ACTH and
cortisol secretion and implications for disease. Endocrine reviews, 41(3), 470-490.
McClure, S. M., Berns, G. S., & Montague, P. R. (2003). Temporal prediction errors in a passive
learning task activate human striatum. Neuron, 38(2), 339-346
McEwen, B. S. (1998). Protective and damaging effects of stress mediators. New England
journal of medicine, 338(3), 171-179
McEwan, B.S. (2000). Allostasis and allostatic load: Implications for neuropsychopharmacology.
Neuropsychopharmacology 22(2): 108-124.
Miller, M., & Clark, A. (2018). Happily entangled: prediction, emotion, and the embodied mind.
Synthese, 195(6), 2559-2575
Montague, P. R., Dolan, R. J., Friston, K. J., & Dayan, P. (2012). Computational psychiatry.
Trends in cognitive sciences, 16(1), 72-80
Oudeyer, P. Y., & Smith, L. B. (2016). How evolution may work through curiosity‐driven
developmental process. Topics in Cognitive Science, 8(2), 492-502

22
Palmer, C. J., Lawson, R. P., & Hohwy, J. (2017). Bayesian approaches to autism: Towards
volatility, action, and behavior. Psychological bulletin, 143(5), 521
Parr, T., & Friston, K. J. (2017). Uncertainty, epistemics and active inference. Journal of The
Royal Society Interface, 14(136), 20170376
Pellicano, E., & Burr, D. (2012). When the world becomes ‘too real’: a Bayesian explanation of
autistic perception. Trends in cognitive sciences, 16(10), 504-510
Pezzulo, G., Rigoli, F., & Friston, K. (2015). Active Inference, homeostatic regulation and
adaptive behavioural control. Progress in neurobiology, 134, 17-35
Pezzulo, G., Rigoli, F., & Friston, K. J. (2018). Hierarchical active inference: a theory of
motivated control. Trends in cognitive sciences, 22(4), 294-306
Power, M. L., & Schulkin, J. (2012). Maternal obesity, metabolic disease, and allostatic load.
Physiology & Behavior, 106 (1), 22–28.
Rabinovich, M., Huerta, R., & Laurent, G. (2008). Transient dynamics for neural processing.
Science, 48-50
Rietveld, E., Denys, D., & Van Westen, M. (2018). Ecological-enactive cognition as engaging
with a field of relevant affordances: The skilled intentionality framework (SIF). In A. Newen, L.
De Bruin, & S. Gallagher (Eds.), The Oxford handbook of 4E (embodied, embedded, extended,
enactive) cognition. Oxford: Oxford University Press, pp.41-70.
Roli, A., Villani, M., Filisetti, A., & Serra, R. (2018). Dynamical criticality: overview and open
questions. Journal of Systems Science and Complexity, 31(3), 647-663
Różycka-Tran, J., Piotrowski, J. P., Żemojtel-Piotrowska, M., Jurek, P., Osin, E. N., Adams, B.
G., ... & Maltby, J. (2019). Belief in a zero-sum game and subjective well-being across 35
countries. Current Psychology, 1-10
Rutledge, R. B., Skandali, N., Dayan, P., & Dolan, R. J. (2015). Dopaminergic modulation of
decision making and subjective well-being. Journal of Neuroscience, 35(27), 9811-9822
Rutledge, R. B., Skandali, N., Dayan, P., & Dolan, R. J. (2014). A computational and neural
model of momentary subjective well-being. Proceedings of the National Academy of Sciences,
111(33), 12252-12257

23
Ryan, R. M., & Deci, E. L. (2001). On happiness and human potentials: A review of research on
hedonic and eudaimonic well-being. Annual review of psychology, 52(1), 141-166
Ryff, C. D., & Keyes, C. L. M. (1995). The structure of psychological well-being revisited.
Journal of personality and social psychology, 69(4), 719
Schultz, W., Dayan, P., & Montague, P. R. (1997). A neural substrate of prediction and reward.
Science, 275(5306), 1593-1599
Schwartenbeck, P., FitzGerald, T. H., Mathys, C., Dolan, R., Wurst, F., Kronbichler, M., &
Friston, K. (2015). Optimal inference with suboptimal models: addiction and active Bayesian
inference. Medical hypotheses, 84(2), 109-117
Seeman, T. E., & McEwen, B. S. (1996). Impact of social environment characteristics on
neuroendocrine regulation. Psychosomatic medicine, 58(5), 459-471
Seeman, T., Epel, E., Gruenewald, T., Karlamangla, A., & McEwen, B. S. (2010). Socio‐
economic differentials in peripheral biology: Cumulative allostatic load. Annals of the New York
Academy of Sciences, 1186(1), 223-239
Seth, A. K., & Friston, K. J. (2016). Active interoceptive inference and the emotional brain.
Philosophical Transactions of the Royal Society B: Biological Sciences, 371(1708), 20160007
Seth, A. K., Suzuki, K., & Critchley, H. D. (2012). An interoceptive predictive coding model of
conscious presence. Frontiers in psychology, 2, 395
Shew, W. L., & Plenz, D. (2013). The Functional Benefits of Criticality in the Cortex. The
Neuroscientist, 19(1), 88–100. https://doi.org/10.1177/1073858412445487
Shew, W. L., Yang, H., Yu, S., Roy, R., & Plenz, D. (2011). Information Capacity and
Transmission Are Maximized in Balanced Cortical Networks with Neuronal Avalanches. Journal
of Neuroscience, 31(1), 55–63. https://doi.org/10.1523/JNEUROSCI.4637-10.2011
Smith, L. S., Hesp, C., Lutz, A., Mattout, J., Friston, K., & Ramstead, M. (2020). Towards a
formal neurophenomenology of metacognition: modelling meta-awareness, mental action, and
attentional control with deep active inference
Sterling, P. (2012). Allostasis: a model of predictive regulation. Physiology & behavior, 106(1),
5-15

24
Sterling, P. (2018). Point of View: Predictive regulation and human design. Elife, 7, e36133
Sterling, P. (2020). What is Health?: Allostasis and the Evolution of Human Design. MIT Press
Sterling, P., Eyer, J., Fisher, S., & Reason, J. (1988). Handbook of life stress, cognition and
health. Allostasis; A new paradigm to explain arousal pathology. New York: Wiley, 629-649
Tschantz, A., Seth, A. K., & Buckley, C. L. (2020). Learning action-oriented models through
active inference. PLoS computational biology, 16(4), e1007805
Van de Cruys, S. (2017). Affective value in the predictive mind (pp. 0-0). MIND Group;
Frankfurt am Main.
Varela, F.J., Lachaux, J-P., Rodriguez, E. & Martinerie, J. (2001) The brain web:
Phase-synchronization and large brain integration. Nature Reviews Neuroscience 2, pp.229-239
Wilkinson, R. & Pickett, K. (2010) The Spirit Level: Why Equality is Better for Everyone.
London: Penguin.
Young, E. A., Abelson, J., & Lightman, S. L. (2004). Cortisol pulsatility and its role in stress
regulation and health. Frontiers in neuroendocrinology, 25(2), 69-76.
Zarghami, T. S., & Friston, K. J. (2020). Dynamic effective connectivity. Neuroimage, 207,
116453