Conative inference as a mechanism of allostatic resilience: wanting to be and being to want.

Stability, mental stability in particular, is necessary for adaptive functioning. Dynamical systems existing in a changing environment must adapt to it by resolving the dialectical tension between stability and flexibility, which is captured in the concept of resilience. Allostasis is considered as a systemic mechanism of resilience and has recently been framed in terms of active inference, a theory of sentient behavior. Allostatic systems are thought to operate on the principle of self-organized criticality, which is related to free energy minimization, the central tenet of active inference theory. Based on this framework, I propose a functional brain hierarchy that supports allostasis. In it, desire and affect, framed in terms of affective inference, serve to integrate and mediate mutual constraints between cognitive and physiological homeostases. I further suggest that as allostatic organisms evolve into affective ones, they develop into conative agents, whose conation regulates their affective dynamics. At the top of this regulatory hierarchy, I propose raw desire, a domain-general global desire for positive affect (or to simply feel better). This functional hierarchy makes the raw desire-affective charge regulatory loop a core mechanism of allostasis and mental stability. It also implies an inherent vulnerability of conative organisms to affective dysregulation. Accordingly, much of psychopathology may be considered as compromised conative inference, which could be a target for therapeutic intervention.