Epistemic Agency in Self-Tracking: AI Qualitative Methods

This post explains how to evaluate epistemic agency in self-tracking (the primary keyword: epistemic agency in self-tracking) and how AI-enabled qualitative analysis can accelerate personal science. According to the Frontiersin.org review (published 18 August 2026), a structured four-phase framework (Questioning, Observing, Reasoning, Discovering) plus three agency levels meaningfully distinguishes low-agency data collection from high-agency personal science. The post translates the Frontiers findings into concrete qualitative research workflows and shows where AI tools speed up synthesis for researchers, patient innovators, and UX teams.

Key Takeaways

According to the Frontiersin.org article (published 18 August 2026), an agency-based framework applied to the literature separates self-tracking that is merely measurement from self-tracking that is self-directed knowledge production.

The Frontiersin.org review (April–May 2025 search) screened 721 records, retained 241 full papers, and analyzed 46 papers that used self-tracking as the data-collection method.

721 records were identified in Web of Science and PubMed in April–May 2025 and screened, yielding 241 retained papers for full review (Frontiersin.org).
Of the 241 full papers, 46 used self-tracking as the data-collection method and were classified into 17 low-agency, 10 mixed-agency, and 19 high-agency studies (Frontiersin.org).
The review’s authors emphasize the epistemic hinge: “who owns the question, ” and quote Catherine Elgin that “Epistemic agents should think of themselves as, and act as, legislating members of a realm of epistemic ends” (Frontiersin.org).
The Frontiersin.org paper (18 Aug 2026) proposes a four-phase personal science cycle: Questioning, Observing, Reasoning, Discovering, which the authors used to operationalize agency.

What Happened and How the Framework Works

Answer: The Frontiersin.org review applied a four-phase personal science framework to classify the epistemic agency present in peer-reviewed literature on the Quantified Self.

According to the Frontiersin.org article (published 18 August 2026), the authors searched Web of Science and PubMed in April–May 2025 for the term “quantified self, ” returned 721 unique records, and after staged screening retained 241 full papers for analysis.

According to the Frontiersin.org methods section, only 46 of those 241 papers used self-tracking as the primary data-collection method and were therefore eligible for agency coding using the four phases: Questioning, Observing, Reasoning, Discovering.

According to the Frontiersin.org results, the 46 self-tracking papers distributed across agency levels as 17 low-agency (participants only observed), 10 mixed-agency (participants engaged partly), and 19 high-agency (participants conducted all four phases, often as co-authors).

Findings Snapshot

Date / Period	Metric	Value	Implication
April–May 2025	Records returned by search for “quantified self”	721	Large and heterogeneous literature corpus that required conceptual filtering (Frontiersin.org).
April–May 2025	Full papers retained after screening	241	Corpus used to map representations of quantified self in health literature (Frontiersin.org).
N/A (subset)	Papers using self-tracking as data collection	46	Eligible for agency coding across the four-phase framework (Frontiersin.org).
Through 2024 (publication years)	Distribution by agency among 46 papers	17 low / 10 mixed / 19 high	Shows that high-agency personal science appears frequently in the literature but coexists with researcher-driven studies (Frontiersin.org).
2018 and 2024 (publication counts)	Peak and uptick	2018: 38 articles; 2024: 13 articles	Publication pattern suggests early peak and renewed interest; terminology shifts affect counts (Frontiersin.org).

Implications for Qualitative Researchers and Patient-Led Research

Answer: Researchers should distinguish measurement-driven self-tracking from high-agency personal science when designing studies or synthesizing literature.

According to the Frontiersin.org discussion (18 August 2026), low-agency papers treat participants as sensor platforms, while high-agency papers show individuals owning the question and often co-authoring publications.

Qualitative researchers should code for agency explicitly in study protocols and literature reviews, for example tagging whether Questioning, Reasoning, and Discovering phases were performed by participants (Frontiersin.org).

Patient-led research programs and participatory health initiatives should plan for infrastructure needs named by the Frontiersin.org authors: documentation templates, training in reasoning methods, and ethical guidance for personal science.

How Evidano Helps: AI workflows for epistemic agency

Problem: Scattered, heterogeneous qualitative evidence

Answer: Heterogeneous QS literature mixes low-, mixed-, and high-agency reports making synthesis slow and error prone.

Evidence: The Frontiersin.org review screened 721 records (April–May 2025) and retained 241 papers, of which only 46 used self-tracking as the data method, illustrating the filtering burden.

Solution: Thematic synthesis and agency coding with Evidano

Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interviews, open-ended surveys, and documents.

Evidano automates transcript ingestion, applies codebooks, and generates thematic, frequency, and cross-segment analyses that map to the four-phase framework (Questioning, Observing, Reasoning, Discovering).

Practical mapping: Problem (Slow synthesis → Feature) AI-assisted thematic extraction and cross-segment comparisons; Problem (Missing participant voice → Feature) AI chat over documents and citation-ready extracts; Problem (PII or domain terms → Feature) custom dictionary transcription and translation (see Evidano features).

Solution: Turn N-of-1 narratives into extractable evidence

Answer: Evidano converts personal science case reports into structured codes and frequency tables so reviewers can compare agency indicators across studies.

Using Evidano’s content and co-occurrence analyses, teams can quantify how often Questioning is participant-driven versus researcher-driven across a corpus like the 46 self-tracking studies described by Frontiersin.org.

Operational help: transcription, privacy, and reproducible outputs

Answer: Evidano supports transcription with custom dictionaries and PII redaction and produces reproducible visualizations for publication and policy briefs.

Users can export codebooks, excerpts, and AI-chat transcripts that document decisions about agency coding for peer review and ethics committees (see Evidano speech-to-text and Evidano data-security).

FAQ: epistemic agency in self-tracking

What is epistemic agency in self-tracking?

Epistemic agency in self-tracking means the individual exercises control over inquiry: posing questions, collecting data, analyzing results, and deciding how to act or share findings.

According to the Frontiersin.org review (18 August 2026) this concept is operationalized by asking whether participants performed the phases Questioning, Observing, Reasoning, and Discovering.

How did the Frontiers review measure agency across papers?

The review coded agency by assessing who performed each of four phases: Questioning, Observing, Reasoning, and Discovering.

According to the Frontiersin.org methods, two authors independently classified articles and resolved conflicts by discussion, with 13% disagreement during abstract screening and 17% disagreement during full-text coding resolved by consensus.

Why does agency matter for qualitative synthesis?

Agency matters because it changes the interpretive stance: participant-owned questions produce situated, actionable knowledge while researcher-owned questions prioritize generalizability.

The Frontiersin.org discussion argues the key distinction is “who owns the question, ” which alters what counts as evidence and how findings should be synthesized.

Can AI safely assist with coding agency in personal science studies?

Yes, AI can assist but human oversight is required for validity and ethical considerations.

Evidano’s platform pairs AI thematic extraction with reviewer-led codebook refinement and supports PII redaction and provenance tracking to document decisions for ethics review.

Conclusion & Next Steps

Answer: Operationalizing epistemic agency in self-tracking makes personal science auditable and comparable across studies and AI-enabled qualitative tools make that practical at scale.

The Frontiersin.org review (18 August 2026) provides concrete numbers: 721 initial records, 241 retained papers, and 46 self-tracking studies coded into 17 low-, 10 mixed-, and 19 high-agency cases; these counts show both the promise and heterogeneity of the field.

Researchers and patient innovators can adopt the Questioning–Observing–Reasoning–Discovering framework as an explicit codebook, then use AI-assisted platforms to extract themes, frequency counts, and co-occurrence networks for rapid synthesis.

If you want to convert N-of-1 narratives and QS case reports into reproducible qualitative evidence, Try Evidano for free.

Topics

epistemic agency in self-tracking
personal science qualitative analysis
quantified self analysis
AI qualitative research tools

Keep reading

Browse all articles