In her last post, Ema described in some more detail condition (2). Here, I want to focus on condition (1) by offering an example of the epistemic benefits that imperfect cognitions could have. Take delusions, one of the paradigmatic manifestations of irrationality, what Tony David on this blog called the "hallmark of madness". Delusions can be regarded as irrational in many ways: they are likely to conflict with other beliefs the subject has, they are implausible given what the subject knows, they are strenuously resistant to counter-evidence (as we discussed previously, DSM 5 defines them as "fixed beliefs that are not amenable to change in the light of conflicting evidence").
Can delusions have any epistemic benefit notwithstanding their many epistemic faults? Arguments to this effect can be put forward, based on the situation in which a subject finds herself prior to adopting a delusion. Consider the phase preceding the formation of a delusion in the context of schizophrenia. The aberrant salience hypothesis and the prediction-error theory (sketches can be found in the contributions by Phil Gerrans and Phil Corlett to this blog) tell us that the subject experiences reality in a distressing way. Random events around her seem to acquire a new significance, but they are like messages that cannot be deciphered and cause anxiety as a result, preventing the subject from responding to cues in the environment and acquiring true beliefs. The emergence of a delusional hypothesis puts an end to this.
Via the delusion, the subject attributes meaning to her anomalous experiences. The anxiety caused by hypersalience is relieved, although the often deeply disturbing content of the delusion can be a new source of anxiety. Some sense of control is regained, as temporary and illusory as this may be. In people with elaborated and systematic delusions, this sense of control can also translate into enhanced "sense of coherence", that is, the capacity to see the world as meaningful and predictable, often positively correlated to health and longevity (Bergstein et al. 2008). Arguably, the increased sense of purpose and the regained confidence can be instrumental to the pursuit of epistemic projects, although the outcomes of these are bound to be heavily influenced by the content of the delusion.
Other potential epistemic benefits will be discussed in future posts.The impression I have from the literature is that benefits are always accompanied by high costs in the context of a delusion, and thus the delusion remains an irrational belief (at least, for people who are doxastically inclined, like me). But what I would like to explore as part of the project is the possibility that the epistemic benefits associated with delusions are such that they could not be obtained otherwise, due to there not being relevant cognitions with fewer epistemic costs that would deliver the same benefits (e.g., release the subject from the perpetual delusional mood characterised by hypersalience).