Christopher G. Davey

I wrote this as a companion piece to the paper I published in Neuroscience and Biobehavioral Reviews in the same year. I had written way too much for that paper, and realised it had gotten unwieldly. I saved the treatment implications of the active inference framework for depression for this paper, helped with the perspectives of my friend and colleague, Paul Badcock. One of the themes I keep returning to is the tendency of people in the field to apply a dualistic framework to mental illnesses like depression: it is either seen as a psychological response to the social environment or as a brain disorder. Treatments should by either psychosocial or biological. I believe very strongly that this dichotomy is a false one.

This paper represents my thinking about how the active inference framework can explain the clinical presentation of depression. I have long thought that depression is fundamentally somatic at the same time as being embedded in the social environment. I didn't have a good theoretical framework to make sense of that intuition. The active inference framework filled that gap for me. I don't know that I say much that is new here: it is more a synthesis of the ideas of others, including Barrett, Seth, Stephan and Solms. If there is an original contribution, it is that depression is especially a disorder of interoceptive processes that have extended time courses, such as those that underlie sleep, appetite and energy levels.

As editor of the Australian & New Zealand Journal of Psychiatry, I rotated the task of writing the editorials among the editors. Leading up to this editorial, I had decided to step down from the editorship, and this was to be my last one. The deadline for it was shortly after I had given the Beattie Smith lecture at the University of Melbourne, and I thought it was a good idea to use my lecture notes for the piece. It was perhaps a little self-indulgent to publish my lecture, but I am glad I took the opportunity. I had been reading Karl Jaspers's work, and at the same time diving into Karl Friston's active inference framework. The links between the ideas of these two psychiatrists, both such innovative influences on the theories of their day, jumped out at me.

I had been working with a diverse bunch of people with lived experience in different contexts (in research, clinical and government meetings) at the time I wrote this. I was confronted at one meeting by the idea that clinicians couldn't represent clinical experience and lived experience at the same time, even if they had experienced their own mental illnesses. I have come to understand the reasons for that – lived experience experts have undertaken training and have adopted a professional role – but there is still much that can be learnt from people whose experiences span both. This article reflected on that.

The paper had a long gestation. Ben and I had been talking about the self and how it relates to the default mode for years, discussing how our research could be elaborated into a model of the self in depression. We eventually got to expressing that in writing: and I feel we captured the essence of those discussions in a way that makes sense of our work and (and of others).

I was engaged in establishing a ketamine clinic for depression at the time wrote this, and also running a ketamine trial. At the same time, the Australian regulators had downscheduled MDMA to allow it to be used to treat PTSD, and I had written a grant with my PTSD colleagues that aimed to examine how we might examine its combination with exposure therapy (not funded, unfortunately). I reflected on how quickly these drugs had moved from recreational use to clinical credibility, and wanted to capture something of that pathway.

I had somehow come across Joseph Wortis in the months before writing this, and was fascinated by his career. He had had a brief analysis with Freud in the 1930s, and then returned to the US to become a champion of biological psychiatry – and became the inaugural editor of the journal of that name. It seemed to mirror the journey of psychiatry.

I assumed the editorship of the Australian & New Zealand Journal of Psychiatry in 2021, and this was my first editorial. I took it as an opportunity to set out my framework for understanding psychiatry. It captures the pragmatism of my clinical philosophy, and how I want to see psychiatry move on from old debates.

I was encouraged to write this article by one of my mentors, Pat McGorry. He was keen that we establish the concept of early intervention for depression in the way that he and others had for psychotic disorders. Depression is a little different because there is a long history of child psychiatrists writing about depression in adolescence. But I do think that the concept of early intervention can be usefully applied, especially to young people with severe and complex depression, who are at high risk of poor outcomes without good care.

I was invited to write this article by the editor of World Psychiatry. I enjoyed the challenge of condensing the ideas about the self that Ben and I had been discussing into a concise piece.

I was invited to write this article; I can’t recall why. It caused a bit of kerfuffle when it was published, because The Age newspaper in Melbourne wrote a front-page story based on it, claiming that our research showed that antidepressants didn’t work. I hadn’t written that and don’t believe it to be true (and there was no original research in the paper) – but I do wish antidepressants were better and that more people responded to them, and also that they weren’t relied on so often as the only treatment.

I wrote this while I was doing my PhD, and drew on ideas that one of my supervisors, Nick Allen, was keen on. This was the theory that depression was related to reduced capacity to represent and experience reward. The article argues that the capacity to represent reward into the future, especially in the context of our social relationships, emerges in adolescence, around the same time that depression starts to increase in prevalence. It has occurred to me only recently that I have arrived at similar ideas in grappling with active inference: depression seems to involve a difficulty in representing the future as having anything positive in it.