Posted on Leave a comment

Why Pain Experience is not a Controlled Hallucination of the Body


This is where a Le Pub Scientifique can lead to. Professor Mick Thacker and Dr. Julian Kieverstein just published this paper. A must read!


This paper aims to provide an account of the subjective character of pain experience in terms of predictive processing. The PP theory is often taken to support a view of perceptual experience as a controlled hallucination of the external world. Transposed to pain this would have the consequence that pain is a controlled hallucination of the body. The PP theory would have the consequence that the body that is in pain is just another hidden cause of sensory input that stands in need of inference and control by the brain. We argue that pain experience cannot be a controlled hallucination of the body since the predictive machinery that constitutes pain experience is not brain bound. The subject’s pain experience is physically realised in a system that is spread
across the body as a whole. This system comprises the immune system, the endocrine system, and the autonomic system in continuous causal interaction with pathways spread across the whole neural axis. We will argue that these systems function in a coordinated and coherent manner as a single complex adaptive system to maintain homeostasis. This system, which we refer to as the neural-endocrine-immune (NEI) system maintains homeostasis through the process of prediction error minimisation. We go on to propose a view of the NEI system as a multiscale nesting of Markov blankets that integrates the smallest scale of the cell to the largest scale of the embodied person in pain. The NEI system is the embodied subject’s first-person perspective on the world. The PP theory, we will argue, can therefore make sense of how a living body that acts to minimise prediction error can also be a lived body, the subject’s embodied point of view on their surrounding world.

Key Points

We have shown how the predictive processing theory of pain is best understood against a backdrop of a view of the whole NEI ensemble as working predictively. Pain experience cannot be reduced to nociception or decomposed into sensory-discriminatory, affective-motivational and cognitive-evaluative elements. Instead we have argued pain is the outcome of predictive processing that takes place in the whole neural axis in continuous reciprocal interaction with the immune system, the neuroendocrine and the autonomic system. All of these systems are working together as an integrated whole. Pain occurs when all of these systems together conclude that the prediction of body integrity is likely to be disconfirmed. Body integrity here means the states of the body the organism should return to under a wide variety of different conditions that are necessary if it is to continue to exist. Pain is an allostatic process that aims to maintain the body in these states under conditions of constant change, providing the organism with feedback that it is diverging in potentially dangerous ways from these (adjustable) setpoints.  


We have argued that the predictive processing theory of pain does not support a conception of pain as controlled hallucination of the body. Proponents of the PP theory often present the body as just another hidden cause of sensory input no different from anything else in the environment external to the body. We have argued this is a mistake. Predictive processing takes place in all of the systems that maintain the homeostasis of the body. What travels up the neural axis to the brain is prediction error from these systems, not a sensory signal originating from an external hidden cause.


In place of a view of the brain separated from the body by an evidentiary boundary we have proposed to think of the predictive processing that takes place within each of the systems that makes up the nervous system as producing and maintaining a nesting of Markov blankets. At the smallest scale to the largest scale processes of prediction error minimisation play out that separates but also connects and integrates these systems. This nesting of Markov blankets makes sense of how the body is presented in pain not as an object but as the subject’s point of view on the world. Pain is not only in the brain. Pain is a state of the whole body that prioritises the actions the organism needs to undertake to return the body to the state of healthy flourishing that is expected.

Research papers

Posted on Leave a comment

The value of uncertainty

Brilliant essay on uncertainty – co-authored by a previous Le Pub panellist Mark Miller.

Quick summary:

– We can deal with a bit of uncertainty in our lives, in fact it is healthy

– However, uncertainty takes many forms and when our worlds are turned upside down (with no predictable end) this can be unhealthy

– Engaging in activities that question or tease our expectations may be helpful, particularly when they’ve become entrenched (predictable). For the inner geek relating to chronic pain and volatility:”Sustained exposure to such volatile situations and environments [i.e. ongoing pain]

– where the outcomes of actions appear inherently unpredictable

– leads to an inevitable decrease in confidence in one’s ability to bring about the outcomes they expect. At that point, our predictive brains begin to infer an inability to exert successful control, and this then forms a damaging part of the model that guides our future actions.”

– perhaps leading to a state of learned helplessness?!

Check it out – read and re-read – there’s an incredible depth of insight and knowledge.