Bridging consciousness and AI: ChatGPT-assisted phenomenological analysis.
David Martínez-pernía, Alejandro Troncoso, Sergio E Chaigneau, Nicolás Marchant, Antonia Zepeda, Kevin A Blanco-Madariaga
Frontiers in psychology January 1, 2025 DOI: 10.3389/fpsyg.2025.1520186 via PubMed
Summary
AI-generated from the abstractChatGPT can process large qualitative datasets for phenomenological analysis while preserving depth and nuance. The tool follows four stages: preparing phenomenological data, individual analysis highlighting experiential nuances, global analysis synthesizing narratives, and structuring shared experience components. Custom prompts ensure alignment and precision. ChatGPT organizes themes reflecting sensation intensity and variations in empathetic encounters, transforming raw input into detailed phenomenological accounts. Its proficiency combines precision with scalability for consciousness studies. Further research is needed to understand AI's capacity in phenomenological analysis and strengthen the methodological framework.
Study at a glance
| Characteristics | Methodological study Qualitative Peer reviewed |
|---|---|
| Topics | Philosophy of mind |
| Keywords | Chatgpt-assisted phenomenological analysis Artificial intelligence Experimental phenomenology Mixed-methods studies Neurophenomenology |
| Citations | 4 |
| Key finding | ChatGPT can effectively perform phenomenological analysis by organizing themes that reflect the intensity of sensations and variations in empathetic encounters, transforming raw input into detailed phenomenological accounts. |
Abstract
Mixed-method studies require adaptation to the era of big data in quantitative research, seeking scalable approaches that can analyze extensive qualitative datasets while preserving the depth and nuance inherent in the study of consciousness on a broader scale. This study aimed to leverage ChatGPT, renowned for its descriptive generation proficiency, to perform a phenomenological analysis. Our research followed four key stages: (1) Preparation of Phenomenological Data, where transcriptions were refined to align with the research question; (2) Individual Analysis, where ChatGPT highlighted experiential nuances from each participant; (3) Global Analysis, synthesizing insights from individual narratives temporally and transversally; and (4) Structure of the Experience, which synthesized the elemental components of shared experiences. Custom prompts, tailored for each stage, ensured alignment and precision in capturing the experience dimensions. ChatGPT showcased a sophisticated processing capability of human experiences, effectively organizing themes that reflect the intensity of sensations and variations in empathetic encounters. The tool's proficiency in thematic organization provided a phenomenologically-grounded processing of data, highlighting how individuals engage with and are affected by stimuli. Our findings highlight ChatGPT's potential in consciousness studies, transforming raw input into detailed phenomenological accounts. ChatGPT combines precision with scalability, making it a compelling tool for researchers exploring the intricacies of human experiences. Further research is essential to better understand AI's capacity in phenomenological analysis and to strengthen the methodological framework, ensuring it effectively captures the nuances and depth of phenomenological inquiry.