Author: tio

Awais Aftab’s blog about the Sunday Times article on my new book, Chemically Imbalanced, was predictable. Like previous reactions to our serotonin paper, it illustrates how elements of the psychiatric profession attempt to control the message that gets out to the public. Aftab even subtitled his blog ‘British journalists and editors this is for you’. This process shapes what becomes accepted as scientific ‘knowledge’ in line with professional and commercial interests.

After a few comments characterising me as a ‘contrarian,’ Aftab’s basic point about the science is simply this: although we haven’t found them yet, depression might be associated with specific brain processes, including those involving serotonin. And because it might be, we should assume it is. Hence he is perturbed by what he sees as a ‘general dismissal of the neurobiology of depression’.

This argument also underpinned many of the original responses to our serotonin paper, but it inverts the most basic precepts of science. An idea or theory is unproven until it is proven, not the other way round. This has to be the case because anyone can propose anything- and they do. There are scores of theories about links between this or that biological process or chemical and depression. A psychiatrist posted a list of 59 of them on Twitter in 2022, and there are more.  

Just as a quick aside, by questioning that there is a ‘neurobiological’ basis to depression, I am not suggesting that nothing is going on in the brain when people are depressed. The question is whether there is a specific brain mechanism that produces depression in the same way that there is a specific process that produces Parkinson’s disease, for example.

What Aftab presents as alternative ways of understanding the relationship between serotonin and depression are not even testable theories, although the sprinkling of technical jargon (‘signalling’, ‘dysfunctions’) makes them sound impressive. He proposes, for example, that one way to think of the relationship is that ‘depression generally, or in subset of patients, involves alterations in the serotonin signalling system (e.g. in the distribution or sensitivity or certain sorts of serotonin receptors).’ But what subset of patients, and what sort of alterations of receptors? There is no agreement about whether there are certain subtypes of depression, let alone what they are, and no body of research that has tested serotonin functions in such groups.

Then he suggests that ‘the serotonin system mechanistically links depressive symptoms and neurobiological dysfunctions in other aspects of brain functioning (e.g. neurogenesis or neuroplasticity)’. But how exactly? And, although there has been a lot of talk about the role of neurogenesis and neuroplasticity in depression of late, especially since we debunked the serotonin hypothesis, there is no real agreement about what these terms refer to or how you would even test them, let alone a convincing body of evidence.

In other words, Aftab is putting forward unsubstantiated speculations and suggesting these are a good enough basis to accept the idea that depression is a neurobiological condition. In Chemically Imbalanced I call this ‘wishful thinking dressed up in scientific terminology.’

Later on Aftab suggests that endocrine and inflammatory processes may be involved in depression- and again they might be, but there is little evidence that they actually are. I summarise research on the main contenders to the serotonin theory in the book, and conclude that ‘as soon as one theory is discredited, the advocates of the biological paradigm turn to another, putting forward a new set of ropey, inconclusive and ambiguous studies as putative evidence. Challenging the biological model of depression feels like a game of whack-a-mole: as soon as you put one theory to bed, another one sprouts up.’

In relation to serotonin and depression, Aftab admits that ‘nothing conclusive has emerged that commands a strong consensus’ and that links are ‘not conclusively established’ and not ‘robust’. But he thinks there is animal research demonstrating an association between serotonin and mood, citing a review paper that I describe in the book. In fact, this paper shows that animal experiments are highly inconsistent – different studies produce contradictory results. However, you wouldn’t know this unless you read the paper carefully because the authors present a convoluted hypothesis to explain the inconsistencies that appears to suggest a role for serotonin, but is simply another example of ‘well, it might be this’.

It is not true, therefore, that we ‘know’ that serotonin is involved in mood or behaviour. In the book I describe how there is evidence that serotonin has a detrimental effect on sexual behaviour, but that is about all we can say about its behavioural functions.

When Aftab suggests that whether there is an alteration of the serotonin system in depression is still ‘an open scientific problem,’ he is technically correct. You cannot prove a negative in science. It’s worth remembering, however, that there have been six decades of research on the serotonin system in depression, and they have not revealed anything remotely conclusive. If you can’t give up on a theory after that, when can you? 

Why does any of this matter? Because the idea that depression is a neurochemical or neurobiological condition creates the impression that treatments that modify the brain, such as antidepressants, ‘work’ by targeting and correcting the underlying mechanism (whether this mechanism is regarded as ‘abnormal’ or not is unimportant). This is the rationale that persuaded people that taking antidepressants is a sensible thing to do.   

Aftab wants us to continue to assume this is the case, but we do not know whether such a mechanism exists and if so what it is. And for this reason, we do not know how antidepressants produce their effects.

As I have said before, I expect the very small difference between antidepressants and placebos in clinical trials is not a pharmacological effect but a result of an amplified placebo effect due to people being able to guess what they are taking (as we showed in the TADS study of fluoxetine in adolescents). But let’s assume it is a pharmacological effect for a moment. Antidepressants are drugs that enter the brain and change our normal brain chemistry and activity – this is not in dispute – and these changes inevitably impact on our mental states. In other words, antidepressants have brain and mind-altering properties. If you give someone with depression a dose of a mind-altering drug, like heroin, they would most likely feel less depressed for a while. Antidepressants are not the same as heroin, and the alterations they produce are usually quite subtle, but most of them numb or restrict emotions to some degree, which might reduce depression scores.

I am using ‘might’ here too, but this scenario needs to be ruled out before we can conclude that antidepressants ‘work’ by affecting a putative biological mechanism. The burden of proof needs to be on those who suggest that as well as their brain and mind-altering effects, antidepressants also target depression mechanisms. This view has quite different implications from taking a drug that numbs emotions by interfering with normal brain activity. The latter situation should rightly worry us because we haven’t properly researched what antidepressants do to the brain exactly or what the consequences are.

Aftab finishes with some philosophical points. He characterises my views of depression as being a ‘normal reaction to adverse circumstances’ and asks ‘what is normal?’ I don’t believe I have ever said this because, like him, I recognise that some people have extreme emotional reactions that are out of the ordinary. It doesn’t follow, however, as he seems to suggest, that because something is out of the ordinary it should therefore be classified as a medical condition.

He then asks whether it is valid to distinguish between the idea that ‘circumstances make us depressed’ and ‘chemistry makes us depressed’ and refers to a ‘rich body of philosophical literature’ on the topic. Indeed, I have written papers on this subject myself in the past, to which Aftab responded. I noted that although the ‘biopsychosocial model’ sounds appealing, when a specific biological process is really involved, as in Parkinson’s disease and other recognised brain diseases, it trumps other influences. I argued that ‘physiological states are different from meaningful states like beliefs, emotions and moods’. People can read my original paper here, Aftab’s response to me here, and my response to Aftab here. I concluded that ‘People with what we call mental disorders are trying to negotiate their individual circumstances in various human ways. They are not walking representations of “neurological mechanisms.”’