Kathryn Nave

Karl Friston’s Free Energy Principle has been proposed as a definition of existence from which “everything of interest about life and the universe can be derived”. Despite pretensions to a theory of every ‘thing’, focus has largely been on the first of these and the attempt to “unify all adaptive autopoietic and self-organizing behaviour under one simple imperative; avoid surprises and you will last longer”. By analysing biological existence in terms of the stability of a set of essential variables and relations that are taken to define a particular organism, then redescribing this stability amid perturbation in statistical terms as ‘surprisal-minimization’, the FEP proposes to connect this biological imperative up to an account of higher-level cognition as involving the operation of hierarchical predictive models. At first glance then, it appears the ideal partner to the bioenactivist seeking to ground the content of our mental lives in this basic, biological intentionality. The potential compatibility is misleading. By reducing autonomy to the kind of stability amid perturbation that could be shared by pendulums and people alike, the FEF makes no distinction between a homeostatic mechanism and a self-producing agent, thereby failing to account for why the latter alone should be credited with active engagement in its own continuation. The free energy theorist is left to choose between instrumentalism about intentional attribution, or mathematically-motivated animism. Neither is an option for the bioenactivist who seeks a naturalistic account of intentionality and immanent teleology as real features distinct to living systems alone. This does not, however, mean that the FEP has inadvertently undermined the bioenactivist program by revealing there to be no difference in kind between life and non-life. The problem lies instead in its underlying assumption of a machine-substantialist ontology, which erases this distinction from the get-go by presuming the existence of a thing can be reduced to a set of invariant parts and the invariant rules that govern its dynamics. Organisms are constrained by neither. Unlike machines, we not only persist through but depend upon both material turnover and unprestatable change in our space of possible behaviours. This, as a number of theoretical biologists and complexity scientists have argued, makes it impossible to derive any invariant form or equations that could fix the possible trajectories of a living system in advance. The failure of the FEF as a theory of life is not just due to the inadequacy of surprisal-minimization as a particular principle, but due to the broader inappropriateness of a machine-substance ontology for analysing biological existence. The FEF’s failure is not the bioenactivist’s victory. For this we need to establish not only that living systems are different from machines, but also to show how these differences grant the former alone the status of intentional agents. The prevailing enactivist definition of autonomy as operational closure among precarious processes does not achieve this, for it fails to capture the mutual dependence between structure and dynamics that characterizes the self-determination of organisms. Instead, I draw upon Montévil, Moreno, and Mossio’s account of closure among constraints, to show how this describes a system that is intrinsically active in seeking out a continual supply of energy with which to rebuild the network of constraints that both embody that system and which would irretrievably disintegrate without such activity. Only such systems, I argue, can truly be said to have their own existence as the goal of their operations. This will likely not convince the eliminativist, determined to reduce subjectivity to atoms and the void, nor the anthropocentrist who jealously withholds agency to his species alone. But for the person who feels that there is a vital difference between her strivings and the operations of a machine, I believe that the place to look is not in the rational powers that might set us apart from our humbler biological relations, but in the peculiar form of precariousness that we share.

Predictive processing accounts of neural function view the brain as a kind of prediction machine that forms models of its environment in order to anticipate the upcoming stream of sensory stimulation. These models are then continuously updated in light of incoming error signals. Predictive processing has offered a powerful new perspective on cognition, action, and perception. In this chapter we apply the insights from predictive processing to the study of emotions. The upshot is a picture of emotion as inseparable from perception and cognition, and a key feature of the embodied mind.

Among the exciting prospects raised by advocates of predictive processing [PP] is the offer of a systematic description of our neural activity suitable for drawing explanatory bridges to the structure of conscious experience. Yet the gulf to cross seems wide. For, as critics of PP have argued, our visual experience certainly doesn’t seem probabilistic.While Clark proposes a means to make PP compatible with the experience of a determinate world, I argue that we should not rush to do so. Two notions of determinacy are conflated in the claim that perception is determinate: ‘univocality’ and ‘full detail’. The former, as Clark argues, is only to be expected in any PP agent that models its world for the purpose of acting on it. But as Husserl argued, and as perceptual psychology has borne out, we significantly overestimate the degree of detail with which we perceive a univocal world.This second form of indeterminacy is due not to the probabilistic nature of PP’s model, but rather to its hierarchical structure, with increasingly coarse-grained representations as we move further from the sensory periphery. A PP system may, or may not, deliver a univocal hypothesis at each of these levels. An action-oriented PP system would only be expected to do so only at the level needed for successful action guidance. A naïve reporter’s overestimation of the degree of determinate detail in their visual experience can thereby be accounted for with a more gradual version of the ‘refrigerator light’ effect: we experience determinate details just to the degree that they’re needed – immediately as they’re needed.

Predictive processing has begun to offer new insights into the nature of conscious experience—but the link is not straightforward. A wide variety of systems may be described as predictive machines, raising the question: what differentiates those for which it makes sense to talk about conscious experience? One possible answer lies in the involvement of a higher-order form of prediction error, termed expected free energy. In this paper we explore under what conditions the minimization of this new quantity might underpin conscious experience. Our suggestion is that the minimisation of Expected Free Energy is not in itself sufficient for the occurrence of conscious experience. Instead, it is relevant only insofar as it helps deliver what Ward et al. ( 2011 ) have previously described as a _sense of our own poise over an action space_. Perceptual experience, we will argue, is nothing other than the process that puts current actions in contact with goals and intentions, enabling some creatures to know the space of options that their current situation makes available. This proposal fits with recent work suggesting a deep link between conscious contents and contents computed at an ‘intermediate’ level of processing, apt for controlling action.