Affect is a positive-to-negative experience in consciousness that has eluded a consensus understanding within psychology and neuroscience. The contemporary neuroscience perspectives of predictive processing (Clark, 2022; Rao & Ballard, 1999) and active inference (Parr et al., 2022) have shown a lot of promise in refining the concept of affect over the past three decades.

As recently outlined by Velasco and Loev (2021), there are two families of theories stemming from predictive processing and active inference that attempt to account for “the mark of the affective” (i.e., valence).

The first they describe as interoceptive inference theories (IITs; Barrett & Simmons, 2015; Seth, 2013; Seth & Friston, 2016), which suggest that affective valence reflects the always-ongoing generation of interoceptive predictions to match the context of what’s happening in the environment. In IITs, the constant interoceptive predictions regulating the internal milieu of the body are a necessary part of allostasis, the process of achieving stability at all times through context-based change (Sterling, 2012).

Velasco and Loev (2021) offer a few critiques of this perspective as an explanation of affect, including that not every conscious experience that has a bodily component seems to be inherently evaluative in the way that affect is (e.g., awareness of air going in and out of the lungs). Furthermore, they point out that there are many experiences that can sway affect (e.g., listening to a piece of instrumental music) that don’t have obvious implications for allostasis (i.e., unlike running from a threat).

In contrast to this account of affective valence is the family of theories that Velasco and Loev (2021) call error........

© Psychology Today