Beliefs aren’t real…
Yeah, I know, the great wisdom of countless self-development and psychotherapeutic approaches. Beliefs aren’t real, they’re all in your head, the map is not the territory.
Big deal! This is such a common approach… Wasn’t this blog supposed to be about more advanced stuff?
Well, yes. It was.
So when I’m saying “beliefs aren’t real”, I don’t mean “your beliefs are all in your head”.
I’m saying: the things you think of as “beliefs” do not exist at all. Your “beliefs” are a cognitive illusion, an experience of something which does not actually occur in the brain in any manner similar to how we’d learned to think about it.
And if I’m even partially right, this is a huge gamechanger for things like psychotherapy or persuasion. Because so much of these things concentrate on changing beliefs and attitudes. And if these things don’t exist, if they aren’t there to change… It actually explains so very much…
I’ve started to dig into this topic through a combination of two sets of influences. The first one was Lisa Feldman Barrett and her work on constructed emotions (as opposed to essentialst, “inbuilt” emotions). Now that work was fascinating and it did change how I understood a lot of things. But it also made me refresh my knowledge of a lot of other constructivist cognitive experiences we have – everything from our vision, through our conscious awareness, to our memories seems to be, in fact, a thing that is constructed in real-time, based on some previous experiences and categorizations, on context and predictive cues.
And soon after that, one of the (mostly useless, to be honest) coaching groups I’ve participated in had a discussion of “what’s the difference between a belief and a value”. Aside from the obvious issue of a false division there, it did get me into thinking about the structure of beliefs in more detail…
Which is where I realized that the constructivist model does seem to fit beliefs very well.
So beliefs wouldn’t be something we “had” and “could change”.
They are endproducts of a real-time data processing mechanism. We don’t have beliefs. We create them, when and as needed, only to have them disappear seconds later.
The difference might not seem significant at first. After all, one way or another, you do get to experience something like “I believe X”. The end result seems to be identical, whether we actually have some beliefs, or simply whip them up as necessary, right?
Not quite, actually.
Because if you look at pretty much all of the tools designed to engage with beliefs, verify them and change them, they have one thing in common. They are all designed to engage with that endproduct. But if the endproduct is an ephemeral result of deeper processes, and not a separate structure, then any changes to it would be pretty much useless. Even if you did manage to change it somehow, it would still be discarded in a few moments… Only to be recreated again and again, in it’s original form.
In fact, when you would be engaging with someones belief or attitude, you would most likely be engaging only with a rationalization of a set, unchangeable endproduct. So even if you successfully challenged that rationalization, the person would still stick to the original belief, only finding a new explanation for it, or, if unable to do so, grow frustrated and agitated. (This reminds me of people blowing up in some of Ramachandran’s experiments with anosagnosia, when they were forcefully confronted with the half-paralysis they attempted to deny and rationalize.)
These kinds of reactions are, in fact, well known to coaches, therapists and people engaging in persuasion and attitude change. People commonly react to belief change techniques with “well, now I know it isn’t true, but I still react that way”. They rationalize (and the more intelligent they are, the more effectively they do so) and stick to their rationalizations. They blow up when they cannot keep rationalizing anymore.
This is not to say that effective belief changework doesn’t happen. But permanent, long-lasting belief change is far from the norm in using techniques from coaching, cognitive psychotherapy, and the like. It is even less common in everyday discussion and persuasion. In fact, research shows that people have significant problems in changing even such beliefs, which were based on facts falsified mere minutes afterwards. Clearly then, the typical cognitive processes aren’t all that effective in this issue.
I would therefore suggest that when belief changework does work successfully, it works somewhat by accident. In attempting to engage with the final result, the belief-result formed through the activity of a certain cognitive mechanism, such changework manages to touch upon the process of belief (re)formation and modify some of the building blocks on which belief change is based. Therefore, new belief-results formed out of this modified mechanism come out differently then the belief-results formed previously.
If this assumption is correct, it would lead us to significant changes in both therapy and persuasion in general. We would no longer look at and engage with belief-results and their rationalizations as we do now. Rather, we would attempt to engage with the building blocks of such belief-results, in order to modify the generating mechanism.
Again, the difference might seem trivial on the outside. In the end we start with someone experiencing one belief and its results, and change the situation so that the person experiences a different belief and its results. However, if the difference is real, then it has a huge impact on change effectiveness. Instead of sometimes randomly impacting the structures we need to change, we would now have a direct way to impact the key elements of the mechanism.
What are the elements of this mechanism? This is still something I’m working on, but some clear contenders are certainly here:
a) the categorization process, with the different possible categories is clearly a significant part of belief formation
b) the prediction/simulation system will be engaged in some way, although I’m not sure if it will impact belief-formation, or simply work on the end-result (“this belief-result impacts prediction in this way)
c) previous experiences certainly impact the categorization process, but do they have a part in the formation of belief-results?
d) is physiological state a participating factor, as in the case of emotions? Here I have my doubts.
e) is situational context a factor?
Also, when talking about beliefs, it’s important to differentiate between the verbalized statements, and the emotional experience of “belief” or “doubt”. Since we now know that a pure, unconnected feeling of sheer belief can be experienced (through, among other things, migraines and TMS), it’s important to differentiate between the feeling of believing, and beliefs as endproducts of the cognitive categorization process. Now such endproducts will probably be supported by raw feeling of belief, as part of the whole experience,but it’s important to understand that the feelings themselves are a separate entity.
In other words, while categories are a certain factor and work with them will be significant here, are there any other elements impacting the process? This is something I will need to look into more, possibly through doing a structural review of any semi-effective belief-change techniques and trying to note what the underlying structural processes might be at work there.
This is, of course, assuming the general claim of beliefs being just belief-results of a more basic process is true. While it does seem to fit with the available data quite well and while the only real basis for beliefs is the philosophical concept they are historically and culturally based on, there is still a lot of research to be done before this can be verified. Forming specific, predictible hypothesis is key here. One thing which I believe would be a possible marker of this model being correct would be if, when belief change does happen, there would be an observable delay, an incongruency between the old belief and a new one, with an increased neural activity in the meantime, when a new belief-result would be formed. In other words, we shouldn’t expect a seamless transition between and old belief-result and a new-one. There should be other testable claims here, though.
Of course, alternatively, a way of falsifying this model would be to be able to point to a repeatably activated, predictable representation of specific beliefs, which could be stimulated to bring forth a specific belief, as is possible with physical sensation of pieces of memory. Should that be possible, it would be clear that beliefs are in fact a separate structure, and not a temporary creation. No suggestion of such beliefs has been observed so far, however, at least to my knowledge.