Things have been really hectic around here lately and I have been meaning to post something on this for a while now.
Recently Michael Fanselow and Zachery Pennington, both at UCLA, have argued against the kind of position developed by LeDoux and Colleagues. This includes his paper with Daniel Pine who is a psychiatrist, his paper with me developing a higher-order theory of fear and anxiety, and his paper with Stefan Hofmann who is a cognitive behavioral therapist. The papers are linked to below.
LeDoux and Pine responded here:
There is also a a bit of a response in our recent general piece on these issues here:
I think the responses do a good job but there is one passage that I think needs more attention.
This is from the ‘Psychiatric Dark Ages’ paper where Fanselow and Pennington say,
5. A logical inconsistency within the two-system framework
The two-system framework formally states that fear as a subjective experience arises from the neural circuitry that gives rise to working memory and conscious recollection, and more specifically, to episodic memory (LeDoux & Brown, 2017; LeDoux, 2017). As an example of an episodic memory, I can recall the what, where and when of yesterday’s breakfast. This includes my memory for the flavors I experienced. I can use this memory to flexibly guide today’s choices—yesterday I had bacon, better stick to oatmeal today. The neural circuits that support such episodic memories are also the neural systems that allow animals to take alternate paths when the one normally used is blocked. And in the two-systems framework, they support the subjective emotion of fear. The question then becomes what is unique about fear that differentiates it from other cognitions? The answer to this question is immediately apparent if one looks at LeDoux and colleagues’ schematics [Figure 1b (LeDoux & Pine, 2016), Figure 2a (LeDoux, 2017) and Figure 5 (LeDoux & Brown, 2017)]: it is the input from the subcortical defensive system, and in the case of LeDoux and Brown, feedback from the behavioral responses generated by the subcortical defensive circuits. In other words, the unique qualities of subjective fear in the two-system framework reduce to the more parsimonious single generator model, where conscious fear reflects one component of an integrated response. Indeed, the additional machinery needed to generate subjective report probably adds additional noise, rendering it, as many previous to us have suggested, a less pure and objective measure of fear.
The central argument here seems to be that since we allow that activity of the amygdala and other lower-order areas influence the subjective experience of fear then it is the case that you could have had that same subjective experience without the higher-order activity. This simply doesn’t follow.
On the view LeDoux and I developed the unique qualities of subjective fear come from the unique contents of certain higher-order representations. It is entirely plausible that activity from the subcortical defensive system may cause the appropriate higher-order representations to have a specific kind of content, which in turn results in a specific subjective experience. When that activity is missing and fear is still felt it may be subjectively different because of the missing causal contribution from the subcortical defensive circuit. This does not collapse the view into a first-order view.
What the HOROR view is committed to, though, is that if it were to be the case that we could, via some other means than normal, mimic the causal input of the subcortical circuits, then we could produce the higher-order state with the appropriate content without any activity in the defensive survival circuits and that would result in the exact same subjective experience of fear.