Is Your Pain Real?

Pain is both a sensation and an affect. Affects are subjective reactions to sensations. That’s why the same stimulus can evoke excruciating pain in one person and mild discomfort in another. In this respect, pain is no different from fear: A stimulus that’s terrifying to one person is thrilling to another.

The objective aspect of pain—its sensation—is registered mainly in the forebrain (which maps where the pain is coming from) and the subjective aspect of pain—its affect—is registered mainly in the brainstem (which measures how you feel about it). The same applies to fear: Perception tells you that someone is running towards you; affect tells you that you’re scared.

Needs are deviations from homeostatic set-points, e.g. we need to remain between 97.7 and 99.5 °F, and as we deviate from this range, we feel hot or cold. (Like pain, body temperature produces both sensation and affect.) In the case of pain, we need to not suffer tissue damage. If we do suffer it, we feel pain. But the affective aspect registers how you feel about the noxious stimulus, not the stimulus itself.

Each affect has both valence—a measure of the degree of deviation from the setpoint—and quality—a "flavor" or "color." Valence is a continuous variable and quality is a categorical one.

Valence measures the degree and direction of homeostatic error signals, registered as pleasure/unpleasure. Deviations away from the set-point are felt as unpleasure ("badness"); returning towards it is felt as pleasure ("goodness"). Staying within the set range feels like nothing.

Quality isn’t a measurement. The quality of unpleasure (e.g., pain vs. fear vs. overheating) tells us which category of need demands work. "Work" entails doing something about the need (e.g., if you’re too hot, you must leave the kitchen.)

You can’t feel all your needs at once, because you can’t do everything at once (e.g., you can’t eat and sleep simultaneously.) So, our various needs must be prioritized. This function is performed in the brainstem, by the periaqueductal grey (PAG) in conjunction with the superior colliculi (located immediately behind it). The PAG is a well-known site for pain modulation, but in fact, it's where all affects are generated. All our multiple homeostats send their error signals to the PAG and the multiple sensory systems send summaries of "where things stand now" to the superior colliculi. At any given moment, then, the state of your current (subjective) needs is registered by the PAG while the state of your current (objective) opportunities is registered by the superior colliculi.

Needs must be prioritized not only in terms of their relative valence but also in terms of their relative chances of being met. E.g., if you’re hungry and sleepy, you’re more likely to sleep first and eat later when your fridge is empty than when it's full; but if you’re very hungry, you might nevertheless prioritize eating and do some late-night shopping.

Selection of the prioritized error signal is determined by a statistical calculation called precision optimization. "Precision" is the degree of confidence attached to each error signal. The signal with the highest precision value is the one that’s felt.

The prioritized need "flavors" or "colors" your conscious state—you feel either hungry or sleepy (or in pain)—and this determines what you do next. Then the scene of action shifts from the brainstem to the forebrain.

Let’s assume you prioritize sleep. Now you set in motion an action plan: Drawing on your (forebrain) long-term memory systems, you predict that if you do X, Y and Z, you’ll fall asleep. Then feelings of sleepiness tell you how well the plan is working. That’s what feelings are for. They tell you how you’re doing within a prioritized category of need. They do this by palpating the sensory consequences of your actions with feeling. If the plan is working (i.e., if the sensory consequences are the expected ones), it feels good, but if things are not turning out as predicted, it feels bad. On this basis, you adjust your plans. "Voluntary behaviour" entails the making of ongoing choices guided by feelings.

This entire process is governed by precision optimization. Increasing precision in an error signal means decreasing confidence in the current action plan; decreasing precision in an error signal means increasing confidence in the plan.

It’s evident that error signals are incoming and action plans are outgoing. The outgoing plans predict the incoming signals. (“If I do this, I expect that will happen.”) The difference between the two is called "prediction error." The aim of voluntary action is to minimize prediction error.

Intelligence Essential Reads

Beyond High IQ: Why Giftedness Is Not Just a Number

What Primitive, Ancient Animals Can Teach Us About Being Human

The relative weighting of the two signals is performed by the brain’s neuromodulatory systems. The best known of these is the reticular activating system (which deploys dopamine, serotonin, norepinephrine, acetylcholine and histamine) to modulate – i.e., to increase or decrease the precision in – various parameters of incoming and outgoing signals. There are many other hormones (e.g., estrogen) and peptides (e.g., opioids) which contribute to neuromodulation, some not even sourced in the CNS (e.g., adrenal glands). The signaling itself is called "synaptic" transmission; the adjustment of precision is called "post-synaptic" modulation (a.k.a. post-synaptic "gain"). Most people know only about synaptic neurotransmission; but, as we’ve seen, the relative influence of incoming and outgoing signals is controlled by post-synaptic modulation. E.g., during sleep the precision in incoming signals is reduced almost to zero, when your confidence in the prediction “I will lose consciousness” is high, with the result that incoming signals are sequestered in your sensory epithelium.

Now we come to the crucial bit. If neuromodulators are the physiological vehicles of precision (of confidence), then they are the physiological vehicles of affect. That’s why most psychiatric drugs act upon the neuromodulators listed above. Neuromodulation is the physiological means by which the subjective side of the affect/stimulus equation is implemented.

So, neuromodulating drugs (like opiates) have an important role to play in pain relief. But what they modulate are predictions, based in beliefs. That is why the effects of psychological treatments (like psychotherapy and mindfulness) and behavioral interventions (like biofeedback and placebo) are no less "physiological" than drug therapies.

What I’ve described in this blog post is the neuropsychological mechanism of pain, and what I’ve emphasized is that it's influenced both by physiological and psychological interventions. Finally, I must add that the same applies to the etiology of pain. Anything that affects your confidence in your outgoing predictions (either negative or positive) over your incoming sensory signals can cause pain.

Here "anything" includes factors as diverse as precise error signals caused by trauma or infection, which lead to chronic expectations of pain; inflammatory changes, which sustain error signals in the absence of any objective stimulus; and depression or anxiety, which produce precise predictions of negative outcomes, and thereby yield error signals which would otherwise have been treated as noise.

The take-home message is this: In the current state of neuropsychological knowledge, it makes no sense to distinguish "psychological" from "physiological" pain. Both are equally real… and equally subjective.