The Bayesian Brain Hypothesis considers the brain as a statistical organ of hierarchical inference that predicts current and future events on the basis of past experience.
The Bayesian Brain Hypothesis considers the brain as a statistical organ of hierarchical inference that predicts current and future events on the basis of past experience.
IF YOU GET EASILY BORED IR DISTRACTED, THEN SCROLL DOWN TO THE IMAGE OF EINSTEIN AND BEGIN THERE!
There are many computational perspectives that could be called upon to characterize psychopathology. These range from neural network and dynamical systems theory to reinforcement learning and game theory. A recent paradigm shift in cognitive neuroscience provides exactly the right sort of theory that allows one to talk about false beliefs – and understand how these arise from synaptic pathophysiology. Cognitive neuroscientists now view the brain as a statistical organ that generates hypotheses or fantasies that are tested against sensory evidence. This perspective can be traced back to Helmholtz and the notion of unconscious inference (Helmholtz, 1866/1962). In the past decades this approach has been formalized to cover deep or hierarchical Bayesian inference – about the causes of our sensations – and how these inferences induce beliefs and behavior. (Friston et al., 2016)
According to this theory, the mind makes sense of the world by assigning probabilities to hypotheses that best explain (usually sparse and ambiguous) sensory data – and continually updating these hypotheses according to standard probabilistic rules of inference. This fine-tuning (optimization) of perception and action operates under the single imperative of minimizing surprise (free energy) and uncertainty; thereby maximizing statistical and thermodynamic efficiency. Learning in the Bayesian brain differs from reinforcement (and machine) learning because it occurs with understanding. Mental models of past experience use these experiences to anticipate new experiences, as opposed to being shaped by them. Continual optimisation of the models also enables efficient exchange with the environment in a self-organised, self-evidencing and unsupervised fashion.
Illustrated graphically, this looks like so …
Modern formulations of Helmholtz’s ideas usually appeal to theories such as predictive coding. Predictive coding describes how the brain processes sensory information by optimizing explanations for its sensations. In predictive coding, neuronal representations in higher levels of cortical hierarchies generate predictions of representations in lower levels. These top-down predictions are compared with representations at the lower level to form a prediction error (associated with the activity of superficial pyramidal cells). The ensuing mismatch signal is passed back up the hierarchy, to update higher representations (associated with the activity of deep pyramidal cells). This recursive exchange of signals suppresses prediction error at each and every level to provide a hierarchical explanation for sensory inputs. In computational terms, neuronal activity is thought to encode beliefs about states of the world that cause sensations. The simplest encoding corresponds to the expected value or expectation of a (hidden) cause. These causes are referred to as hidden because they have to be inferred from their sensory consequences. (Friston et al., 2016)
In short, predictive coding represents a biologically plausible scheme for updating beliefs about the world using sensory samples. The figure here, tries to convey the basic idea behind predictive coding in terms of minimizing prediction errors.
Predictive Coding deals with the problem of inferring the causes of sparse and ambiguous sensory inputs. This is illustrated in terms of a sensory signal that can be regarded as a sensory impression. A plausible explanation for what this input could be is then predicted by the Brain’s Internal Model. Predictive Coding assumes that the brain has an internal model that generates predictions of sensory input, given a hypothesis or expectation about how that input was caused. (Friston et al., 2016)
The expectation [prediction] is denoted by y and the sensory prediction it generates is summarized with p(y). The prediction error is the difference between the input and predictions of that input. This prediction error is then used to update or revise the expectation, until prediction error is minimized (p=y/x). At this point, the expectation provides the best explanation or inference for the causes of sensations. Note that this inference does not have to be veridical. If the actual cause of sensations was as predicted; however, the beholder may never know the true causes – provided that we minimize our prediction errors consistently (p=x/y, our model of the world will be sufficient to infer plausible causes in the outside world that are hidden behind a veil of sensations (x).
Albert Einstein was not the first, nor the last, to reflect on this conundrum, because how can something which seems so innately counterintuitive be right? We all believe that, that which we experience as reality, that must surely be reality, so what’s this business about ‘reality is just an illusion’? That is what we will try to get to the bottom of in this endeavor.
First, to align our beliefs about what is real, and what is not, let us agree upon some basic presumptions:
a) The Sun rises in the East and settles in the West
b) The Sun is at the center of our solar system
c) Our solar system is a part of the Milky-Way Galaxy
d) Everything else we perceive as factual, or reality, is largely based on inferred beliefs based on the subjective interpretations of empirical scientific evidence.
What I am saying here is not, that the empirical scientific evidence is somehow untrue, but rather that they are the product of our subjective interpretations of the probable causality for the observable phenomenon we are experiencing. In other words, what we infer the causality to be, is based on observed empirical evidence (like those stated about the Sun above), are still biased by ‘that which the Observer is able to perceive’.
On October 12, 1492, Italian explorer Christopher Columbus made landfall in what is now the Bahamas. The ‘New World’ which he had then discovered, was believed to be India, since the purpose of his journey West into the vast unknown of the Atlantic Ocean, was to find a shorter route from Europe to India. The peoples he encountered in Bahamas, he then inferred to be … Indians …
This an excellent analogy to the complicated, yet comprehendible topic for today’s exploration: The Bayesian Brain Hypothesis. Since 2010, the research in Neuroscience into the depth of how the human brain computes the information it receives, have made extraordinary landfall, especially in the area of Perception and Action.
The history that leads up to this timeframe, goes as follows: An approach to visual perception that bridges this apparent divide, proposed indeed more than a century ago by Helmholtz (1866/1962), puts an emphasis on the formation of a percept within a process of evaluation. On Helmholtz’s suggestion, the evaluation involves a test of a hypothesis about what is being seen based on “inductive inferences” gained from “sensations”. By inductive inference Helmholtz meant that perceptions are conclusions based not only on present sensations but also with reference to past sensations of the objects perceived. Latent within this conceptualisation is the idea that the perceived image is at least partly the outcome of stored information – a stored representation, that is a memory – of that object or of similar objects in similar contexts. This was potentially the first proposal of a top-down influence in perception. It regards perception not primarily as a sensory phenomenon but as perceptual inference relying on internal models built through past experience. Helmholtz’s idea of perceptual inference has been revived by computational models of perception relying on statistical inference. (Aggelopoulos, 2015)
Between Helmholtz’s proposal and its recent revival, several other theories of perception were advanced. A contemporary of, and indeed, student of Helmholtz, was William James, the American Physiologist turned Psychologist. During 1867–68 James went to Germany for courses with the physicist and physiologist Hermann von Helmholtz, who formulated the law of the conservation of energy. This trip sparked a flame in James, and he spent the next 25 years dedicated to decoding the human psyche. This resulted in the 2 volume, 1,200 page long ‘The Principles of Psychology’ which he published in 1890, after having toiled with its conception for 10 years, trying to get it finished. When the book was published in 1890, it became an instant success among the growing populous with interest in the new field of Psychology.
The Principles, which was recognized at once as both definitive and innovating in its field, established the functional point of view in psychology. It assimilated mental science to the biological disciplines and treated thinking and knowledge as instruments in the struggle to live. At one and the same time it made the fullest use of principles of psychophysics (the study of the effect of physical processes upon the mental processes of an organism) and defended, without embracing, free will. [Encyclopedia Britannica]
In 1890 William James wrote: “Whilst part of what we perceive comes from the object before us, another part (and it may be the larger part) always comes out of our own head”, a statement sometimes referred to as William James’ Law of Perception.
Perceptual inference refers to the ability to infer sensory stimuli from predictions that result from internal neural representations built through prior experience. (Aggelopoulos, 2015)
Perceptual Inference was a term first used by (Ballard et al., 1983), marking a re-emergence of the Helmholtzian view of perception as inductive inference, a notion re-articulated by (Dayan et al., 1995) as the “Helmholtz Machine”. These models described perception as hypothesis testing using the Bayes rule, the latter incorporating ideas on hierarchical coding from neuroscience. Expectation was introduced by (Mumford, 1992) and predictive coding by (Rao & Ballard, 1997, 1999). A subsequent theoretical framework of “active inference” was developed by (Friston, 2010) from the ideas of statistical decision theory and predictive coding in a series of recent publications.
The Bayesian statistical models were developed partially in physics, artificial vision, and artificial intelligence and partially in relation to experimental psychophysics, where the model observer has to passively predict an input and the world acts as an instructor. Information flows unidirectionally from the world to the brain. In these simplified situations, the visual stimulus was until recently regarded as a feedforward input. The classical view in experimental studies of the sensorimotor system has been quite the reverse. The sensory input has been traditionally regarded as feedback while the top-down motor commands have been regarded as the feedforward action that causes an interaction with the world. The result of that interaction between the motor command and the sensory feedback is the minimisation of corrections (essentially the same as the minimisation of error or optimisation of precision). (Aggelopoulos, 2015)
Active inference seeks to explain brain function in terms of predictive coding, the brain always seeking to minimise prediction error or “free energy” through the optimisation of precision (prediction accuracy) in attention and in motor commands.
An overview of the theoretical literature based on generative statistical models can be found in these (Clark, 2013b, 2013a; Friston, 2012; Friston, Adams, et al., 2012; Friston, Thornton, et al., 2012; Hohwy, 2012; Palmer et al., 2015, 2017; Paton et al., 2013; Solms, 2019; Stein et al., 2014). Introductions to the modern statistical models and theories can be found in Clark (2013).
Perception arises in prediction error minimization where the brain’s hypotheses about the world are stepwise brought closer to the flow of sensory input caused by things in the world. This is an elegant idea because it gives the brain all the tools it needs to extract the causal regularities in the world and use them to predict what comes next in a way that is sensitive to what is currently delivered to the senses. This idea can be explicated in more complex terms of minimizing surprisal to ensure agents sample sensory inputs that are characteristic of their phenotype. This can be cast in terms of minimizing the divergence between hypotheses or probabilistic representations of the world and the true posterior probability, given sensory evidence—a minimization that necessarily invokes a Bayesian brain perspective on perception and places the role of probabilistic representations centre stage. This perspective provides an account in terms of the overall way prediction error bounds the creature’s surprisal. This idea of a bound on surprise is something we will return to a number of times. (Hohwy, 2013)
The Constructed Mind approach is an extension of the Theory of Constructed Emotion, which itself began as a more modest theoretical proposal, called the Conceptual Act Theory. Built from psychological and social construction approaches, the conceptual act theory proposed that the human mind transforms feelings of affect into instances of emotion by categorizing them with situation-specific, embodied emotion concepts. Following publication of the initial papers outlining the conceptual act theory, however, a deeper understanding of nervous system structure and function suggested that instances of emotions do not arise from categorizing affect. Instead, they emerge in a brain as it continually makes meaning of sense data from its body and the world by categorizing those data with situation-specific concepts, thereby constructing experience and guiding action. (Shaffer et al., 2022)
The operationalization of the framework of the above mentioned core theories can be summarized and explained using the Theory of The Constructed Mind developed and refined from 2000 and onwards by a series of publications from (Barrett, 2017; Barrett et al., 2007, 2016; Barrett & Quigley, 2021; Barrett & Satpute, 2013; Feldman Barrett & Bliss-Moreau, 2009; Feldman Barrett & Kyle Simmons, 2015; Hoemann et al., 2019; Hutchinson & Barrett, 2019; Kleckner et al., 2017; Kuppens et al., 2013; Lindquist et al., 2015; Quigley et al., 2021a, 2021b; J. A. Russell, 2003, 2021; J. Russell & Barrett, 2015; Sennesh et al., 2022; Shaffer et al., 2022; Wormwood et al., 2021)
THE MIND IN PREDICTION: The mind exists in prediction. Our perceptual experience of the world arises in our attempts at predicting our own current sensory input. This notion spreads to attention and agency. Perception, attention, and agency are three different ways of doing the same thing: accounting for sensory input as well as we can from inside the confines of the skull. We are good at this, mostly, but it is a precarious and fragile process because we are hostages to our prior beliefs, our noisy brains, the uncertain sensory deliverances from the world, and to the brain’s urge to rid itself efficiently of prediction error. The mind is shaped by how we manage these predictive efforts. We continually need to adjust, regulate, and revisit the balances and checks on prediction. The way we do this determines how we bind sensory attributes and how much our preconceptions can penetrate experience; more chronic, systematic failures to manage prediction can tip us into mental illness. The predictive mind has extreme explanatory reach. Conscious unity, emotion, self, and introspection can all seemingly be brought under the prediction error minimization mechanism that maintains and shapes the mind. With this mechanism we can see ourselves as mere cogs in nature’s causal machinery and also as mental islands set over against the world, which is hidden behind the veil of sensory input. (Hohwy, 2013)
The goal of The Constructed Mind Theory is not to reduce every mental phenomenon to energy regulation but rather to highlight energy regulation as a key element in the state space of a brain:
Based on the empirical evidence presented so far, I will now explain how the Bayesian Brain Hypothesis along with The Constructed Mind, explains why ‘reality is just an illusion’, as Einstein stated.
Fact: The BRAIN is a physical organ which resides inside the Skull, isolated from the physical External World
The Central Nervous System (CNS) consists of the BRAIN and the Spinal Cord (SC). Since the BRAIN have no direct way of interacting with the external, physical world, it is entirely dependent on accessing it via the BODY, which it uses as a mechanism that allow the BRAIN to experience, explore, and exploit, the environment in which it exists.
The CNSis connected to the Peripheral Nervous System (PNS) through which it interacts with both the internal world [called Interoception] and the external world [called Exteroception] through the Somatic Nervous System (SNS), as well as receiving information of the current State-of-Affairs of the BODY through the Autonomic Nervous System (ANS).
The result of this intricate system is an immense, unprioritized amount of the information, which the BRAIN somehow have to makes sense out of, from inside its ‘DARK-ROOM’ context of the Skull. For this specific purpose Nature have. throughout evolution, developed the MIND.
Fact: The MIND is a non-physical entity, which is conceptualized as consisting of various neural networks within the BRAIN, which in unison gives rise to consciousness, or the SELF.
The SELF is what you and I experience as ‘the feeling of being ourselves’.
Fact: The SELF is a non-physical, mentalization of the BODY in the world, as defined by the MIND
Therefore, our SELF is referred to as being our EMBODIED SELF and we experience our EMBODIED SELF through the mentalization functions within the MIND, called COGNITION.
Cognition: all forms of knowing and awareness, such as perceiving, conceiving, remembering, reasoning, judging, imagining, and problem solving. Along with affect and conation, it is one of the three traditionally identified components of mind. [APA Dictionary]
Cognition, as it is used here, broadly encompasses every mechanism of mind including (but not limited to) perception, attention, motivation, planning, deliberation, metacognition, action selection, and motor control, as well as the embodiment of all of these activities. “Cognition” then is meant to cover the entirety of the agent’s mental life including its embodiment and embodied actions. (Kronsted et al., 2022)
The paradigm shift toward an action-oriented view stresses that cognition permits meaningful interactions with a dynamic environment and cannot be reduced to thinking-related mental representations. (Senden et al., 2020)
Embodied agents need to be able to interact with their environment autonomously and adaptively. One highly complex cognitive aspect of sensorimotor integration, involving the recruitment and concerted interplay among a large number of cortical and subcortical brain regions, is action selection. The cerebellum is a key structure for sensorimotor control, as it coordinates voluntary movements through prediction and sensory feedback. Evolution does not act on static, but rather on plastic systems learning from experiences in their environment. This highlights the significance of neuron dynamics. The balance between energy income, expenditure and availability determine neural dynamics to a significant extent. The effects of these factors manifest themselves at all levels from molecular to behavioral. Constructing state-of-the-art embodied systems that are able to intelligently interact with their environment in a closed loop, requires the development of large-scale architectures incorporating several structural as well as functional components. (Senden et al., 2020)
Fact: Consciousness is the result of the Executive Functions of COGNITION, used by the MIND to construct our Subjective, Experienced, Embodied Sense-of-Self.
This is the functional diagram which depicts the complex interactions of physical, physiological, and neurocognitive elements, that in its overarching mechanism, the MIND, by which we construct of Embodied Sense-of-Self.
To summarize, based upon the collective knowledge of all of the above mentioned concepts and theories, which are all based on the empirical scientific evidence, or ‘the truth’ as we know it to, as of now, have led me to develop a descriptive conceptual model for explaining this in Layman’s Terms, and I call this the Subjective, Experienced, Embodied Sense-of-Self Model (SEES).
In order for me to explain the SEES Model, we need to have some basic biological facts aligned.
Fact: ANTICIPATION is the ACT of using INFORMATION about the PAST and PRESENT to make PREDICTIONS about FUTURE scenarios
It is important to note that the terms anticipation and prediction will be used synonymously throughout this article. Some of the phenomena discussed may be described by other authors as “predictive,” but there is certainly no need to create unnecessary semantical obstacles in the discussion of like topics. The term “anticipatory,” however, has been specifically selected instead of “predictive” to describe the aforementioned biological processes because its connotation emphasizes a purposiveness that has clear biological and evolutionary relevance. This language also emphasizes the importance of the predicted state over the potentially inconsequential act of making a prediction. (Deans, 2021)
As a concept, it is predominantly associated with the psychology of the human mind; however, there is accumulating evidence that diverse taxa without complex neural systems, and even biochemical networks themselves, can respond to perceived future conditions. Although anticipatory processes, such as circadian rhythms, stress priming, and cephalic responses, have been extensively studied over the last three centuries, newer research on anticipatory genetic networks in microbial species shows that anticipatory processes are widespread, evolutionarily old, and not simply reserved for neurological complex organisms. Overall, data suggest that anticipatory responses represent a unique type of biological processes that can be distinguished based on their organizational properties and mechanisms. Unfortunately, an empirically based biologically explicit framework for describing anticipatory processes does not currently exist. This review attempts to fill this void by discussing the existing examples of anticipatory processes in non-cognitive organisms, providing potential criteria for defining anticipatory processes, as well as their putative mechanisms, and drawing attention to the often-overlooked role of anticipation in the evolution of physiological systems. Ultimately, a case is made for incorporating an anticipatory framework into the existing physiological paradigm to advance our understanding of complex biological processes. (Deans, 2021)
Fact: Since EVOLUTION does not develop or maintain unnecessary systems, any ORGANISM whose environmental conditions would have allowed it to survive and reproduce, will not have invested resources to develop and maintain more advanced and costly brains.
To know the principles on which we found our modeling, we need to analyze what the different adaptive systems share among the different species that they currently own or have owned in the past. To do so, we must consider that, whenever a new species appears, all inherited systems and tactics must face new conditions, environments, constraints, and requirements for survival and reproduction that will test the limits of its operating range. From a Darwinian point of view, we can say that, within their habitats, each species is a new experiment that tests all functional elements against the filter of natural selection. If we take into account that 99% of species that have existed since the origin of life have become extinct, we can postulate that the greater the number of species and the longer they retain a certain adaptive system, the more necessary, evolved, and versatile this system should be. This argument allows us to articulate three fundamental and hierarchical principles on which we will base our reasoning.(Garcés & Finkel, 2019)
Necessity principle: To be able to adapt to certain conditions, be they environmental, ecological, sexual, or otherwise, a living organism requires a system or a set of systems capable of detecting and evaluating those conditions, and of identifying or developing one or more appropriate responses to address them, choose the best available, and most importantly, implement a response, acting on the stimulus to use, avoid, or modify it. If the individual has no such system, or exaptation, nor the ability to adapt existing ones, and the stimulus does not disappear spontaneously, the challenge will not be met. This fact is even more evident if a change affecting a system is shared among numerous highly diverse species, as it will be exposed to very different selection conditions, we call this principle the Necessity Principle. (Garcés & Finkel, 2019)
Efficacy principle: The environmental conditions to which species are exposed within an ecological niche can change dramatically over evolutionary time periods, thus testing the responsiveness of different adaptive systems. If effective, that is, if species successfully resolve the situation for which they were selected, the individual survives and reproduces, and the system is conserved. If not or if they cease to be effective, individuals perish and disappear; we call this conditioning the Efficacy Principle. An example of the application of this principle may be the extinction of the large dinosaurs. After more than 150 million years, all their adaptive systems failed (ceased to be effective) when a series of dramatic global changes converged in a short period of time. On the contrary, birds, the living descendants of dinosaurs, and mammals survived. (Garcés & Finkel, 2019)
Efficiency principle: Success in survival is not defined by effectiveness alone. As we have already seen, energy is the key component to maintaining the structural order of a living being. All adaptive systems have an implicit energy cost. The body is thus forced to permanently devote a variable amount of resources to maintain it. Throughout evolutionary periods, all the resources required to maintain different adaptive systems have not always been available. Nature will, therefore, have preserved only the most efficient: those that maintain their resolving ability with as few resources as possible. We call this the Efficiency Principle, and it can also be observed at different scales. The importance of this principle is exemplified by the human brain’s ability to fulfill all human functions with just the power of a 25 W light bulb. (Garcés & Finkel, 2019)
The Nervous System: Responses integrate and summarize all processed information, from lower sensory levels to decision-making and behavior. The success or failure of the nervous system’s adaptive capacity ultimately depends on the quality of the responses it is capable of generating and the quality of the execution of those responses. The very existence of the following biological mechanisms could be considered a confirmation of the importance of these three variables: Activation Threshold, Reaction Time, and Accuracy. (Garcés & Finkel, 2019)
Thus, mechanisms such as Memory are able to encode, store, and quickly retrieve previously processed information, making it suitable for being efficiently incorporated in new processes and reused. Pattern Recognition enables information to be shared, encoding it with fewer connections, thereby saving resources more quickly and perhaps reusing ready-made responses. Predictive Systems can recognize patterns that occur at different points in time or in sequences that, according to our reasoning, are closely related to memory capacity. Feedforward uses a prediction from predictive systems and is able to activate in advance neural and physiological components of the responses, thus creating faster circuits to send activation information along the shortest paths. Feedback acts as a regulating element, allowing the nervous system to dynamically adjust its operation by checking the effectiveness of its own responses and the effects they exert on the eliciting stimuli. For instance, Efferent Copy which, combined with inverse models, gives way to corollary discharge, allows us to explain, for example, why we cannot tickle ourselves. (Garcés & Finkel, 2019)
The Mirror System makes it possible to anticipate—and imitate—the actions of others, thus triggering advanced social interactions and behaviors. And Mental Imagery is a high-level mechanism for optimizing critical variables. If the information developed through predictive systems is re-fed through sensory circuits, it can be managed as new self-generated stimuli, which in turn can elicit new responses, either neural or physiological. In turn, this self-generated information could form the basis of self and social interactions, which is a good example of an advanced system that emerges as a combination of simpler ones. (Garcés & Finkel, 2019) Interdependency: If we want to improve accuracy, we need to spend more time generating and exploring more alternatives. However, if we delay, when we finally find the best response it may no longer be needed, either because the predator has devoured us, or because our potential partner has found another. Also, if we reduce the reaction time, the quality of response suffers and may no longer be accurate enough to successfully resolve the stimulus that elicited it, thus becoming ineffective. If we display unnecessary responses, albeit accurate and fast, we may waste our energy and time solving insignificant problems, thus diminishing the availability of resources to address other and more important tasks. (Garcés & Finkel, 2019)
Human survival is secured via adapting and automatizing its responses through an optimization of the three critical variables: Activation Threshold (when to act), Reaction Time (how quickly), Accuracy (in what way). These three variables are critical to assessing the quality of nervous system responses, and they are interdependent. When one of the three critical variables is modified, the others will be affected by the change.
Fact: Automaticity is the process by which the neural pathway associated with a response reaches the optimal balance of interconnection between its elements, thus providing the best possible relationship between the three critical variables that characterize its quality
Based on this definition, we outline the Automaticity Principle as follows: as a result of its own mechanisms of growth and development, and in order to fully optimize their effectiveness and efficiency, the nervous system will automate, as much as possible, the new circuits and neural networks that encode a stimulus recognition, calculation, and execution of the response associated with it (Garcés & Finkel, 2019)
Levels of Response: However, this does not mean that an automated response is the best possible response to solve a particular stimulus. When the best response available within the limitations of individual capacities in a given context is found, the neural network that encodes it is optimized to do three things: recognize the stimulative pattern, compute the response, and run it as quickly and accurately as possible. Thus, the automaticity concept refers to the response execution quality. The three levels of response are: the Automatic, Mentalized, and Automated.
Automatic Responses: We have already seen that the first kind of responses available to the nervous system to react to stimuli are Innate Responses. The same stimulus will produce the same response. That the origin of this type of response is genetic indicates that it has been preserved by species over generations. In turn, this tells us that it has been useful in solving certain very specific, ancestral, frequent, and repetitive stimuli, including crying, coughing, pupillary dilation to changes in light, sweat secretion, heart rate control, and breathing. Within innate responses we include Reflex Arcs. As Automatic Reponses are also included Fixed Action Patterns, or Instincts, defined as “patterns of behavior that are fully functional from the first time they are executed, even if the individual has had no previous experience with the stimuli that elicit the response”.
Mentalized Responses: The nervous system must develop new responses from the elements available. This level of cognitive response is called Mentalized Response. They form a broad set of more or less advanced tools which enable the body to create new solutions to address the most diverse stimuli. These mechanisms are highly flexible but have the disadvantage of requiring more time and resources to find or develop, select, and apply a response, thus reducing biological fitness. Once the brain finds the best possible response to a repetitive stimulus within its own capacity, it activates the automaticity principle in an attempt to create the most optimal pathway to process and execute the response, when necessary, as quickly and accurately as possible.
Automated Responses: The results of a Mentalized Response is then encoded and stored, as a third type of response called Automated Responses.
NOTE: It is important to note the difference between Automatic Responses and Automated Responses.
While all Automatic Responses are genetic and therefore automatic from the outset, Automated Responses do not initially exist, they are cognitively constructed via the Mentalized Response, and optimized via the Automaticity Principle, and stored as Empirical Priors within the Internal Model of The Bayesian Brain.
Regardless of the level to which a response may belong, it include two types of complementary components that, may or may not activate simultaneously:
a) a physiological component that includes autonomic and somatic systems with corresponding motor elements, endocrine, heart rate, blood pressure, etc. directed at allowing the body to perform the necessary physiological activation and physical actions to face the stimulus.
b) a neural component that will trigger the activation or regulation of other neural networks, thus initiating new complementary brain processes.
Fact: Automaticity is thus, the process by which that response is wired in a new specific neural network or pathway.
Optimization has been the keystone for all our reasoning until now:The Nervous System must generate a response that effectively addresses the stimulus, only if necessary, as quickly and accurately as possible, and with the least consumption of resources. To improve performance, the Nervous System has developed multiple biological mechanisms in architecture and dynamics (memory, pattern recognition, predictive systems, feedback, feedforward, mirror system, automaticity). The three levels of response (automatic, mentalized, and automated) allows the Nervous System to optimize the three critical variables (activation threshold, reaction time and accuracy) and their interdependence.
Thus, and according to SEES model,
Fact: The Emotional System is always Active and Controls the Dynamics of Attention, which in turn Regulates and Prioritizes the Access of Stimuli Information to Advanced Cognitive Systems
Cognitive systems are able to develop cognitive responses, which in turn also modulate emotional responses, thereby closing a circular, complementary, dynamic, and interdependent architecture. According to this, emotion and cognition do not compete but collaborate, mutually complementing each other to achieve a complete and most efficient way to resolve the challenges faced by the individual.
As the observant reader might already have discovered, I have introduced some novel concepts in the illustration of the SEES model. These are called Mesaception, Metaception, and Supraception. Likewise, I have introduced to concepts of Allostasis and Homeostasis. These last two concepts are already established and defined in the literature, so I will simply refer to (Caforio et al., 2020; McEwen, 2017; McEwen & Wingfield, 2003; Ramsay & Woods, 2014; Sennesh et al., 2022; Sterling, 2012, 2014) for more in-depth descriptions of the these.
Each of these 3 levels of abstraction (Mesaception, Metaception, and Supraception), I’ve coined as Perceptual Levels of Consciousness. Each level contributes to the final construction of what is implemented with the MIND as that which becomes our Subjective, Experienced, Embodied Sense–of–Self. Allostasis and Homeostasis are the guardians which ensures that things don’t run into the ditch, and they do so by following two key indicators: Circadian Rhythmicity, and the current State-of-Affairs, known as Core Affect. Circadian Rhythmicity is hardcoded into every cell in our BODY, and it contains the blueprint for ‘how to run a BODY efficiently’ as dictated by Nature through endless iterations of evolutionary generations, and is thus stored within are DNA. Core Affect is the nonconscious emotional valuation system, that dictates our Gut-feeling.
At the heart of emotion, mood, and any other emotionally charged event are states experienced as simply feeling good or bad, energized or enervated. These states— called Core Affect—influence reflexes, perception, cognition, and behavior and are influenced by many causes internal and external, but people have no direct access to these causal connections. Core Affect can therefore be experienced as free-floating (mood) or can be attributed to some cause (and thereby begin an emotional episode). These basic processes spawn a broad framework that includes perception of the core-affect-altering properties of stimuli, motives, empathy, emotional meta-experience, and affect versus emotion regulation; it accounts for prototypical emotional episodes, such as fear and anger, as Core Affect attributed to something plus various nonemotional processes. (J. A. Russell, 2003)
Core Affect is a continuous assessment of one’s current state, and it affects other psychological processes accordingly. A change in Core Affect evokes a search for its cause and therefore facilitates attention to and accessibility of like-valenced material. Core Affect thus guides cognitive processing according to the principle of mood congruency. The more positive Core Affect is, the more positive events encountered or remembered or envisioned seem— provided that the Core Affect is not attributed elsewhere. Core Affect is part of the information used to estimate affective quality and thus is implicated in incidental acquisition of preferences and attitudes. Core Affect influences behavior from reflexes to complex decision making. One can seek to alter or maintain Core Affect directly—affect regulation—from the morning coffee to the evening brandy. People generally (but not always) seek behavioral options that maximize pleasure and minimize displeasure. Decisions thus involve predictions of future Core Affect. Core Affect is involved in motivation, reward, and reinforcement. (J. A. Russell, 2003)
In short, Core Affect is the Nonconscious State-of-Affairs (Sense-of-Self) which determines the optimal response to handling the information presented by Trigger Events that arise from the Outside-In, as Sensory stimuli, from the Inside-Out as Visceral stimuli, and Top-Down as Neuronal stimuli. Sensory Stimuli arises from Exteroception as Semi-Conscious Feelings, Visceral Stimuli arises from Interoception as Nonconscious Sensations, and Neuronal Stimuli arises from Cognition as Conscious Emotions. These 3 information streams converge into Core Affect, which in turn processes and calculates the Expected Value, Expected Utility, and Hedonic Value of the Object of Affordance presented by the Trigger Event. The State-of Affairs in Core Affect thus gives rise to Perception and Action. The State-of-Affairs in Core Affect can be illustrated as depicted, where each level is color-coded according to the corresponding levels with the same color in the conceptual illustration of the Subjective, Experienced, Embodied Sense–of–Self Model.
“Alice meets the Growling Dog …”
“Alice sees the Growling Dog, her heart begins to race, her breath becomes heavy, and her hands begin to tremble …”
Allice is now in a Nonconscious State-of-Affairs in Core Affect [High Activation and High Negative Affect = High Negative Displeasure] which can be seen on the Core Affect Circumplex above.
Below is the same Core Affect Circumplex, but here I’ve indicated which State-of-Affairs one most often finds oneself in, when in a situation of experiencing High Activation [High Arousal], High Negative Affect [Low Valence] which is felt close to a Semi-conscious State-of-Affairs known as ‘Jittery, Nervous‘.
This change in Core Affect have been brought forth due to the process of Attribution of Affective Quality toward the Object.
Attribution of Affective Quality of the Object: As depicted below, the Nonconscious Valuation Process [Attribution of Affective Quality] of the Object (the Growling Dog) is quickly evaluated as
a) Important vs. Unimportant (or Neutral)? (safe or unsafe, valuable or non-valuable, relevant or irrelevant)
b) Good vs. Bad? (Pleasure or Displeasure, Activating or Deactivating)
c) Approach vs. Avoid? (Freeze, Fight, or Flight)
This State-of-Affairs gives rise to the Perception of Danger, which the MIND then constructs an instance of an emotional episode of the Subjective, Experienced, Embodied Sense-of-Self, as … “Alice is afraid of the Growling Dog, because I know that dog that growl just before the bite, and I don’t like to be bitten and feel the pain’ …”
This is done by cognitive processing of the Salience of the Object as depicted in the Cognitive Control Processes framework …
The Salience is calculated by weighting Expected Value [what’s it worth, objectively], and Expected Utility [what’s it worth to me, subjectively] against the Expectancy [what did Alice anticipate and expect to see, when she entered the woods, not knowing that the Growling Dog was in there]. The result of this calculation is measured in Uncertainty.
Uncertainty is the result of the calculation of all of the above information, and it is used to optimize how the MIND responds to future Trigger Events which resembles this Growling Dog debacle. As we learned from our evolutionary principles, the MIND need to ensure that the three critical variables: Activation Threshold, Reaction Time, and Accuracy are all evaluated, before any ACTION is taken, so as to ensure maximum reward at minimum energy expenditure.
This is done using the Free-Energy Minimization Principle, which is a hardcore, nerdy mathematical mumbo-jumbo wizardry, that I myself personally don’t have the intellectual capacity to comprehend in the mathematical terms, but I do, however, understand and comprehend the idea of it, and this is actually quite simple …
If you have a high degree of Uncertainty, then you are unable to use your Automatic or Automated Responses to react to the Trigger Event, but must initiate the energy-heavy Mentalized Response instead. This means that a high degree of Uncertainty = high energy consumption, whereas a low degree of Uncertainty is quickly and energy-efficiently handled using the Automatic or Automated Responses = low energy consumption.
Since energy costs energy to obtain, the Free-Energy Minimization Principle is aimed at optimizing the degree of Uncertainty as best possible.
What is the easiest way of reducing Uncertainty? The answer have been found, and it is:
The equation above states, that:
the value of what you believe [BELIEF] that you will expect to experience in the next moment, is measured in how good you are in anticipating what is most likely going to happen next, based on the information you have, right now [INPUT], and the information you have from your past experiences in a situation most like the current situation [PRIOR BELIEF], compared to the sum of = the probability of your guess is correct [Prediction] -(minus]- the probability of your guess being incorrect [Prediction Error] = Surprisal
Surprisal is what determines how well you Free-Energy Minimization is performing, which in turn is measured in something called Entropy.
The end goal of all this math-wizardry is to provide the scientific evidence for the theory stating that Nature have been doing advanced math all through human evolution, by ‘installing a system which is self-optimizing, adaptive, and energy-efficient’ so as to ensure optimal survival and reproduction for the species. This ‘system’ is today known as Bayes Optimal, and what is giving the name to The Bayesian Brain Hypothesis.
So, when Alice sees the Growling Dog, her MIND have already anticipated the Uncertainty of her meeting [or not meeting] a Growling Dog in those particular woods, at that time of year, on this time of day, based on its past experiences of ‘those woods’, and therefore it has already predicted, what Alice is most likely to experience next [based on the statistical probability], and therefore in advance sent instructions to the BODY to enact whatever ACTIONS that the MIND have calculated to be the optimal response to the expected situation [Fight, Freeze, or Flight], while simultaneously sending ACTION-PLANS to the BODY [called EFFERENT COPIES] which instructs the BODY on what to signal the muscles to get ready for executing – IF – the MIND finds that …
What it predicted [PREDICTION SIGNAL] was going to happen, is deemed the better solution as the response to the Trigger Event, than that incoming sensory signal [PREDICTION ERROR], and therefore chooses to ignore ‘reality’ and implement the MIND’s own version of reality [the ‘illusion’] instead, and then ‘releases the brakes’ on the Inhibitory Control so that the Motor Control can enact the instructions into actual verbal and/or nonverbal behavior. It is actually not that difficult to comprehend, as long as you are able to ACCEPT that even though it FEELS real, as Subjective, Experienced, Embodied Sense-of-Self, it is ‘just an illusion‘
To make this concept a little more simple to see from a Layman’s perspective, I’ve made this model, which I’ve coined: The ADDspeaker BODY-MIND Model
The model can be used to see the process from Stimuli to Mood (Sense-of-Self), and the clever feature is, that you can input both Sensory Signals [taste, smell, sound, sight, touch, and vestibular] as well as Neuronal Signals [thoughts or emotions] into the STIMULI box at the bottom of the model, and then just follow the steps in the process up the chain. This is possible due to the fact, that the MIND do not distinguish between what is ‘real’ and what is ‘imagined’ when evaluating the impact of the Object on the State-of-Affairs in Core Affect, it just handles its business regardless 😉
I have chosen to include some additional references as well, but as it would end up in a 250+ page long deep-dive into the nerdy details of the Bayesian Brain Hypothesis, I leave it up to the reader to interpret them as best possible, and refer the reader to the References section at the end of this document for a lifetime worth of reading material containing information on the topics I’ve based my model and concept upon.
Thank you for your time!
/Peter ‘ADDspeaker’ Vang
Aggelopoulos, N. C. (2015). Perceptual inference. Neuroscience and Biobehavioral Reviews, 55, 375–392. https://doi.org/10.1016/j.neubiorev.2015.05.001
Ballard, D. H., Hinton, G. E., & Sejnowski, T. J. (1983). Parallel visual computation. Nature, 306(5938), 21–26. https://doi.org/10.1038/306021A0
Barrett, L. F. (2017). The theory of constructed emotion: an active inference account of interoception and categorization. Social Cognitive and Affective Neuroscience, 12(1), 1–23. https://doi.org/10.1093/scan/nsw154
Barrett, L. F., Mesquita, B., Ochsner, K. N., & Gross, J. J. (2007). The experience of emotion. Annual Review of Psychology, 58, 373–403. https://doi.org/10.1146/ANNUREV.PSYCH.58.110405.085709
Barrett, L. F., & Quigley, K. S. (2021). Interoception: The Secret Ingredient. Cerebrum: The Dana Forum on Brain Science, 2021. /pmc/articles/PMC8493823/
Barrett, L. F., Quigley, K. S., & Hamilton, P. (2016). An active inference theory of allostasis and interoception in depression. Philosophical Transactions of the Royal Society B: Biological Sciences, 371(1708). https://doi.org/10.1098/RSTB.2016.0011
Barrett, L. F., & Satpute, A. B. (2013). Large-scale brain networks in affective and social neuroscience: towards an integrative functional architecture of the brain. Current Opinion in Neurobiology, 23(3), 361–372. https://doi.org/10.1016/j.conb.2012.12.012
Caforio, B. C., Silvestrin, M., & Biazoli, C. E. (2020). Embodied concepts, allostasis, and the origin of emotions. Developmental Psychology, 56(4), 841–842. https://doi.org/10.1037/dev0000890
Clark, A. (2013a). Are we predictive engines? Perils, prospects, and the puzzle of the porous perceiver. The Behavioral and Brain Sciences, 36(3), 233–253. https://doi.org/10.1017/S0140525X12002440
Clark, A. (2013b). Whatever next? Predictive brains, situated agents, and the future of cognitive science. Behavioral and Brain Sciences, 36(3), 181–204. https://doi.org/10.1017/S0140525X12000477
Dayan, P., Hinton, G. E., Neal, R. M., & Zemel, R. S. (1995). The Helmholtz machine. Neural Computation, 7(5), 889–904. https://doi.org/10.1162/neco.1995.7.5.889
Deans, C. (2021). Biological Prescience: The Role of Anticipation in Organismal Processes. Frontiers in Physiology, 12, 672457. https://doi.org/10.3389/FPHYS.2021.672457
Feldman Barrett, L., & Bliss-Moreau, E. (2009). Affect as a Psychological Primitive. Adv Exp Soc Psychol, 41, 167–218. https://doi.org/10.1016/S0065-2601(08)00404-8
Feldman Barrett, L., & Kyle Simmons, W. (2015). Interoceptive predictions in the brain. Nature Publishing Group, 16. https://doi.org/10.1038/nrn3950
Friston, K. (2010). The free-energy principle: a unified brain theory? Nature Reviews. Neuroscience, 11(2), 127–138. https://doi.org/10.1038/nrn2787
Friston, K. (2012). The history of the future of the Bayesian brain. Neuroimage, 62–248(2), 1230. https://doi.org/10.1016/J.NEUROIMAGE.2011.10.004
Friston, K., Adams, R. A., Perrinet, L., & Breakspear, M. (2012). Perceptions as Hypotheses: Saccades as Experiments. Frontiers in Psychology, 3(MAY). https://doi.org/10.3389/fpsyg.2012.00151
Friston, K., Brown, H. R., Siemerkus, J., & Stephan, K. E. (2016). The dysconnection hypothesis (2016). Schizophrenia Research, 176(2–3), 83–94. https://doi.org/10.1016/J.SCHRES.2016.07.014
Friston, K., Thornton, C., & Clark, A. (2012). Free-energy minimization and the dark-room problem. Frontiers in Psychology, 3(MAY). https://doi.org/10.3389/FPSYG.2012.00130
Garcés, M., & Finkel, L. (2019). Emotional Theory of Rationality. Frontiers in Integrative Neuroscience, 13. https://doi.org/10.3389/FNINT.2019.00011
Hoemann, K., Xu, F., & Barrett, L. F. (2019). Emotion words, emotion concepts, and emotional development in children: A constructionist hypothesis. Developmental Psychology, 55(9), 1830–1849. https://doi.org/10.1037/dev0000686
Hohwy, J. (2012). Attention and conscious perception in the hypothesis testing brain. Frontiers in Psychology, 3(APR). https://doi.org/10.3389/fpsyg.2012.00096
Hohwy, J. (2013). The Predictive Mind. Oxford University Press.
Hutchinson, J. B., & Barrett, L. F. (2019). The power of predictions: An emerging paradigm for psychological research. Current Directions in Psychological Science, 28(3), 280. https://doi.org/10.1177/0963721419831992
Kleckner, I. R., Zhang, J., Touroutoglou, A., Chanes, L., Xia, C., Simmons, W. K., Quigley, K. S., Dickerson, B. C., & Feldman Barrett, L. (2017). Evidence for a large-scale brain system supporting allostasis and interoception in humans. In Nature Human Behaviour (Vol. 1, Issue 5). Nature Publishing Group. https://doi.org/10.1038/s41562-017-0069
Kronsted, C., Kugele, S., Neemeh, Z. A., Ryan, K. J., & Franklin, S. (2022). Embodied Intelligence: Smooth Coping in the Learning Intelligent Decision Agent Cognitive Architecture. Frontiers in Psychology, 13. https://doi.org/10.3389/fpsyg.2022.846931
Kuppens, P., Tuerlinckx, F., Russell, J. A., & Barrett, L. F. (2013). The relation between valence and arousal in subjective experience. Psychological Bulletin, 139(4), 917–940. https://doi.org/10.1037/A0030811
Lindquist, K. A., Satpute, A. B., Wager, T. D., Weber, J., & Feldman Barrett, L. (2015). The Brain Basis of Positive and Negative Affect: Evidence from a Meta-Analysis of the Human Neuroimaging Literature. https://doi.org/10.1093/cercor/bhv001
McEwen, B. S. (2017). Stress: Homeostasis, Rheostasis, Reactive Scope, Allostasis and Allostatic Load☆. In Reference Module in Neuroscience and Biobehavioral Psychology (pp. 557–561). Elsevier. https://doi.org/10.1016/B978-0-12-809324-5.02867-4
McEwen, B. S., & Wingfield, J. C. (2003). The concept of allostasis in biology and biomedicine. Hormones and Behavior, 43(1), 2–15. https://doi.org/10.1016/S0018-506X(02)00024-7
Mumford, D. (1992). On the computational architecture of the neocortex – II The role of cortico-cortical loops. Biological Cybernetics, 66(3), 241–251. https://doi.org/10.1007/BF00198477
Palmer, C. J., Lawson, R. P., & Hohwy, J. (2017). Bayesian approaches to autism: Towards volatility, action, and behavior. Psychological Bulletin, 143(5). https://doi.org/10.1037/bul0000097
Palmer, C. J., Seth, A. K., & Hohwy, J. (2015). The felt presence of other minds: Predictive processing, counterfactual predictions, and mentalising in autism. Consciousness and Cognition, 36, 376–389. https://doi.org/10.1016/J.CONCOG.2015.04.007
Paton, B., Skewes, J., Frith, C., & Hohwy, J. (2013). Skull-bound perception and precision optimization through culture. The Behavioral and Brain Sciences, 36(3), 222. https://doi.org/10.1017/S0140525X12002191
Quigley, K. S., Kanoski, S., Grill, W. M., Barrett, L. F., & Tsakiris, M. (2021a). Functions of Interoception: From Energy Regulation to Experience of the Self. Trends in Neurosciences, 44(1), 29–38. https://doi.org/10.1016/j.tins.2020.09.008
Quigley, K. S., Kanoski, S., Grill, W. M., Barrett, L. F., & Tsakiris, M. (2021b). Functions of Interoception: From Energy Regulation to Experience of the Self. Trends in Neurosciences, 44(1), 29. https://doi.org/10.1016/J.TINS.2020.09.008
Ramsay, D. S., & Woods, S. C. (2014). Clarifying the Roles of Homeostasis and Allostasis in Physiological Regulation. Psychological Review, 121(2), 225. https://doi.org/10.1037/A0035942
Rao, R. P. N., & Ballard, D. H. (1997). Dynamic model of visual recognition predicts neural response properties in the visual cortex. In Neural Computation (Vol. 9, Issue 4). https://doi.org/10.1162/neco.1997.9.4.721
Rao, R. P. N., & Ballard, D. H. (1999). Hierarchical Predictive Coding Model Hierarchical Predictive Coding of Natural Images. In Nature neuroscience (Vol. 2, Issue 1). http://neurosci.nature.com
Russell, J. A. (2003). Core Affect and the Psychological Construction of Emotion. Psychological Review, 110(1), 145–172. https://doi.org/10.1037/0033-295X.110.1.145
Russell, J. A. (2021). Psychological construction of episodes called emotions. History of Psychology, 24(2), 116–120. https://doi.org/10.1037/HOP0000169
Russell, J., & Barrett, L. F. (2015). The Psychological Construction Of Emotion.
Senden, M., Peters, J., Röhrbein, F., Deco, G., & Goebel, R. (2020). Editorial: The Embodied Brain: Computational Mechanisms of Integrated Sensorimotor Interactions With a Dynamic Environment. Frontiers in Computational Neuroscience, 14, 53. https://doi.org/10.3389/FNCOM.2020.00053
Sennesh, E., Theriault, J., Brooks, D., van de Meent, J. W., Barrett, L. F., & Quigley, K. S. (2022). Interoception as modeling, allostasis as control. Biological Psychology, 167, 108242. https://doi.org/10.1016/J.BIOPSYCHO.2021.108242
Shaffer, C., Westlin, C., Quigley, K. S., Whitfield-Gabrieli, S., & Barrett, L. F. (2022). Allostasis, Action, and Affect in Depression: Insights from the Theory of Constructed Emotion. Annual Review of Clinical Psychology, 18(1), 553–580. https://doi.org/10.1146/annurev-clinpsy-081219-115627
Solms, M. (2019). The Hard Problem of Consciousness and the Free Energy Principle. Frontiers in Psychology, 9. https://doi.org/10.3389/fpsyg.2018.02714
Stein, D. J., Solms, M., & van Honk, J. (2014). The Cognitive-Affective Neuroscience of the Unconscious. CNS Spectrums, 11(8), 580–583. https://doi.org/10.1017/S1092852900013626
Sterling, P. (2012). Allostasis: A model of predictive regulation. Physiology and Behavior, 106(1), 5–15. https://doi.org/10.1016/j.physbeh.2011.06.004
Sterling, P. (2014). Homeostasis vs allostasis implications for brain function and mental disorders. In JAMA Psychiatry (Vol. 71, Issue 10, pp. 1192–1193). American Medical Association. https://doi.org/10.1001/jamapsychiatry.2014.1043
Wormwood, J. B., Quigley, K. S., & Barrett, L. F. (2021). Emotion and threat detection: The roles of affect and conceptual knowledge. Emotion (Washington, D.C.). https://doi.org/10.1037/EMO0000884
This website uses cookies.
Tilføj din kommentar her - Feedback er altid velkomment!