Principled Limitations on Self-Representation for Generic Physical Systems

Abstract

The ideas of self-observation and self-representation, and the concomitant idea of self-control, pervade both the cognitive and life sciences, arising in domains as diverse as immunology and robotics. Here, we ask in a very general way whether, and to what extent, these ideas make sense. Using a generic model of physical interactions, we prove a theorem and several corollaries that severely restrict applicable notions of self-observation, self-representation, and self-control. We show, in particular, that adding observational, representational, or control capabilities to a meta-level component of a system cannot, even in principle, lead to a complete meta-level representation of the system as a whole. We conclude that self-representation can at best be heuristic, and that self models cannot, in general, be empirically tested by the systems that implement them.

Forgetting ourselves in flow: an active inference account of flow states and how we experience ourselves within them

“…Nevertheless, planning need not involve an explicit sense of the conceptually-represented-self-as-object; rather, in cases of planning where there is an intended sense of self, the cognitive system implicitly utilises the beliefs that underwrite the self-concept (except in cases of self-delusion; cf., Marchi and Newen, 2022 ) to select a policy in line with its preferences, projecting a non-propositional and non-historicised self-image into the past and future to assess the validity of possible policies. Conversely, a historicised conceptually-represented-self-as-object emerges from a post-hoc and higher-order propositional inference over these beliefs, engendering the hypothesis that there is a fixed me , which, notably, can exhibit and has exhibited acts of mental autonomy, such as planning ( Metzinger, 2017 ; Fields et al, 2024 ). As mentioned above, such a case of self-conceptualisation is one in which Bayesian beliefs converge with (but do not collapse into) psychological beliefs.…”

Forgetting Ourselves in Flow: An Active Inference Account of Flow States and How We Experience Ourselves Within Them

“…Nevertheless, planning need not involve an explicit sense of the conceptually-represented-selfas-object; rather, in cases of planning where there is an intended sense of self, the cognitive system implicitly utilises the beliefs that underwrite the self-concept (except in cases of selfdelusion; cf., Marchi & Newen, 2022) to select a policy in line with its preferences, projecting a non-propositional and non-historicised self-image into the past and future to assess the validity of possible policies. Conversely, a historicised conceptually-represented-self-as-object emerges from a post-hoc and higher-order propositional inference over these beliefs, engendering the hypothesis that there is a fixed me, which, notably, can exhibit and has exhibited acts of mental autonomy, such as planning (Fields et al, 2024;Metzinger, 2017). As mentioned above, such a case of self-conceptualisation is one in which Bayesian beliefs converge with (but do not collapse into) psychological beliefs.…”