An overview of neurophenomenological approaches to meditation and their relevance to clinical research

A review of recent advances in neurophenomenology, a research program that rigorously examines subjective experience using first-person methods inspired by phenomenology and contemplative practices. The review covers three areas: new multidimensional tools for capturing dynamic changes in consciousness during meditation; empirical studies using experienced meditators to deconstruct aversive and self-related processes, revealing markers for pain regulation, self-dissolution, and acceptance of mortality; and a deep computational neurophenomenology framework that uses deep parametric active inference to naturalize phenomenology. These innovations suggest that mutual constraints among phenomenological, computational, and neurophysiological domains can contribute to an integrated understanding of mental illness and its treatment.

Study at a glance

Abstract

There is a renewed interest in taking phenomenology seriously in consciousness research, contemporary psychiatry, and neurocomputation. The neurophenomenology research program, pioneered by Varela (1996), rigorously examines subjective experience using first-person methodologies inspired by phenomenology and contemplative practices. This review explores recent advancements in neurophenomenological approaches, particularly their application to meditation practices and potential clinical research translations. We first examine innovative multi-dimensional phenomenological assessment tools designed to capture subtle, dynamic shifts in experiential contents and structures of consciousness during meditation. These experience sampling approaches allow shedding new light on the mechanisms and dynamic trajectories of meditation practice and retreat. Secondly, we highlight how empirical studies in neurophenomenology leverage the expertise of experienced meditators to deconstruct aversive and self-related processes, providing detailed first-person reports that guide researchers in identifying novel behavioral and neurodynamic markers associated with pain regulation, self-dissolution and acceptance of mortality. Finally, we discuss a recent framework, deep computational neurophenomenology, which updates the theoretical ambitions of neurophenomenology to “naturalize phenomenology” (Varela, 1997). This framework uses the formalism of deep parametric active inference, where parametric depth refers to a property of generative models that can form beliefs about the parameters of their own modeling process. Collectively, these methodological innovations, centered around rigorous first-person investigation, highlight the potential of epistemologically beneficial mutual constraints among phenomenological, computational, and neurophysiological domains. This could contribute to an integrated understanding of the biological basis of mental illness, its treatment and its tight connections to the lived experience of the patient.