In this paper I introduce a novel and comprehensive theory of truth grounded in modern cognitive science and the principles of mathematical modelling. This is achieved by proposing a general theory of cognition, where the beliefs of the brain represent a unified predictive model of how best to consider and act in the world in order to maximise future subjective wellbeing. The concept of wellbeing is grounded in a neurocognitive reward signal, well supported by current cognitive science, that dri…
Read moreIn this paper I introduce a novel and comprehensive theory of truth grounded in modern cognitive science and the principles of mathematical modelling. This is achieved by proposing a general theory of cognition, where the beliefs of the brain represent a unified predictive model of how best to consider and act in the world in order to maximise future subjective wellbeing. The concept of wellbeing is grounded in a neurocognitive reward signal, well supported by current cognitive science, that drives individuals toward evolutionarily motivated priorities and facilitates decision making. Truth in belief is simply the success condition of cognition, and is defined by how similar the predictive model of belief is to the structures of reality that determine the future reward that an individual experiences. In this way, the theory I propose allows for the naturalised unification of existing concepts such as the correspondence, coherence, and pragmatic theories of truth. Stable truths in this framework are produced by predictive belief structures that are consistent in their truthfulness over time and between subjective perspectives. Applied within a dynamic social context on the other hand, this theory also produces emergent phenomena similar to consensus, social constructionist, and post-structuralist ideas regarding truth. I detail the process of belief update in response to reward prediction error, demonstrating how this can increase the truthfulness of belief over time, and resolve any concerns regarding epistemological circularity. The metaphysical and ethical implications of this theory are briefly discussed, as is its relationship with Bayesian epistemology, and structurally related concepts in science and engineering such as predictive processing and reinforcement learning. This theory has the potential to act as a basic structural philosophical concept, which can be used to construct, evaluate, and refine more complex ideas, both in and outside of philosophy. By providing a stable epistemological and metaphysical foundation, this framework should facilitate increased agreement on truth between perspectives, and promote the required conditions for cooperation and progress in society.