Deep Collaboration Methodology 2.0

Abstract

This study proposes and executes the Deep Collaboration Methodology 2.0 (DCM 2.0), taking responsibility attribution perception in contexts of AI use as its research object. Street-intercept interviews were conducted at Da'an Forest Park in Taipei and at several university campuses, producing 328 valid interviews spanning 33 nationalities (including Taiwan, with 32 foreign nationalities). Field execution ran from December 22, 2025 to April 15, 2026. The study employs a Convergent Mixed-Methods Design, using two fixed core questions as quantitative anchors: Q1 asks "when AI output is erroneous, to whom does responsibility belong?" (the level of moral attribution), and Q2 asks "when a loss results from acting on an AI-generated answer, who bears the loss responsibility?" (the level of practical allocation). Ten rotating question sets (Q3/Q4) were used to explore respondent reaction patterns across different AI social issue contexts. The principal findings are organized across three levels. First, the user internalization of loss responsibility in Q2 exhibits overwhelming cross-group robustness (79% overall), remaining stable across stratifications by gender, age, nationality, and tool choice for most background characteristics — indicating that "when a loss results from acting on an AI-generated answer, the user should bear responsibility" constitutes a cross-group ethical intuition about tool use. Second, the Q1 responsibility framework is systematically moderated by usage experience, forming a continuous gradient from non-AI users (AI attribution 52%) to short-term users (33%) to long-term users (24%), with the core mechanism being the accumulation of error-encounter experience. Third, a structural epistemic gap exists between Q1 and Q2: Q1's tripartite dispersion (user attribution 25%, shared responsibility 37%, AI attribution 31%) collapses abruptly upon entering Q2's loss responsibility framework — even among respondents who attributed Q1 responsibility to AI, 71% still pointed to the user in Q2 — reflecting the systematic fusion that occurs when two distinct ethical questions ("who caused the error" versus "who should bear the consequences") are presented in sequential order. At the methodological level, DCM 2.0 fills structural gaps in existing AI cognition research across three dimensions — ecological validity, cross-cultural coverage, and sample diversity — providing a traceable, replicable, and institutionally unaffiliated street-intercept mixed-methods protocol. The construction of DCM as an independent methodological framework is presented as a core academic contribution alongside the substantive findings of this study. Note: The PDF in this deposit references an earlier DOI(10.5281/zenodo.20280701). Both records refer to the same work.Please cite this record: https://doi.org/10.5281/zenodo.20280700 Correction (May 20, 2026): Author name corrected in Section 2.2 and reference list — Makoana et al. → Kakraba et al. (2026). No changes to research findings or methodology. Version 3 Correction and Revision Note (2026/06/25) Corrections: 1.Journal name: Branda and Ciccozzi (2026) was incorrectly listed as published in eBioMedicine. The correct journal is Artificial Intelligence in the Life Sciences. DOI updated to 10.1016/j.ailsci.2026.100158. Revisions: 2. Section 1.4: Inline meta-commentary and parenthetical glosses have been removed. Substantive findings and analytical content are unchanged. 3. Section 3.2.1: A paragraph discussing the structural sampling advantages of street-intercept design in the context of rising digital tool adoption has been removed. Core design logic is unchanged. 4. Section 3.2.2: Paragraphs explaining the rationale for withholding question sequence information and the open-ended design of Q1 have been removed. Interview structure and question wording are unchanged. 5. Section 8.7: Fully rewritten. The revised version reframes the researcher's epistemological position in methodological rather than personal terms, and restructures the section into three analytical components: outlier status, structural institutional exclusion, and execution-level operational logic. The substantive findings, analytical conclusions, and all other content remain unchanged. Version 3 Correction and Revision Note (2026/07/24) 1. Section 2.2 — Definitional Origins of Epistemic SovereigntyAdded an opening paragraph tracing the problem consciousness underlying this study's conception of Epistemic Sovereignty to the author's earlier case study, The Automation Paradox (Fan, 2025a), positioning it as the conceptual point of origin preceding the two 2026 academic definitions reviewed in this section. 2. Section 2.4 — Research Gaps and DCM 2.0's PositioningAdded a paragraph preceding the section's closing statement, detailing the methodological discontinuity between DCM 2.0 and the earlier exploratory study DCM 1.6 (Fan, 2025b): differences in interview-guide fixity, the absence of the ten-set Q3/Q4 rotation design, sample scale (n = 36 vs. n = 328), absence of refusal records, restricted-layer field structure (23 fields in 2.0), and the absence of quantitative analysis and literature citation in 1.6. This addition also states explicitly, within the body text rather than solely in Zenodo metadata, the rationale for publishing DCM 2.0 under an independent DOI rather than as a revised version of DCM 1.6. 3. Section 3.8 — Research LimitationsAdded a bridging sentence to the "Consistency limitations of single-person execution" entry, cross-referencing Section 3.7.2 to clarify the division of labor between the epistemological legitimacy of the participant-researcher's overlapping roles (discussed in 3.7.2) and the operational limitation of the absence of cross-interviewer verification (discussed here), preempting a reading of internal contradiction between the two sections. 4.Reference formattingCorrected APA citation format for the two self-authored Zenodo works, and disambiguated same-year, same-author citations with a/b suffixes ordered by title: Fan, C.-C. (2025a). Case study: The Automation Paradox and cognitive sustainability. https://doi.org/10.5281/zenodo.18102508 Fan, C.-C. (2025b). DCM (Deep Collaboration Methodology): A recursive framework for human-AI symbiosis and epistemic sovereignty. https://doi.org/10.5281/zenodo.18124731 Version 4 Correction and Revision Note (2026/08/09) This revision addresses a previously undisclosed gap in the study's consent infrastructure, discloses GPS data gaps across three record categories, adds a new fieldwork spatial distribution figure, adds a formal rigor-benchmarking section for the study's data protection design, and corrects a citation-year error in the reference list. No changes were made to the study's core findings, sample, or statistical results. 1. Section 3.6.1 (Design Choices for Informed Consent) — revised.Added disclosure that during the earliest phase of fieldwork (December 22–29, 2025; six respondents), the post-interview business-card distribution practice had not yet been standardized: cards were provided only to respondents who expressed strong interest in the research. The remaining respondents in this subgroup therefore had no channel to identify the researcher afterward and could not exercise the withdrawal right described in this section. This limitation was present in the original fieldwork practice but had not previously been stated in the manuscript. 2. Section 3.3.2 (Dynamic Adjustment of Execution Strategy) — closing sentence added.Added a summary sentence noting that business-card distribution timing and GPS logging for refusals were each adjusted at different points during the 114-day field period, with pointers to Section 3.6.1 (business-card timing) and Section 3.5.2 (GPS logging) respectively. 3. Section 3.5.2 (An Honest Assessment of Data Quality) — revised.Added disclosure of 22 cases lacking recorded GPS coordinates: 3 valid interviews (GPS malfunction on the day of execution; all other recording elements preserved), 17 refusals, and 2 exclusions (refuser/exclusion-case location was not yet being systematically recorded during an earlier stage of the field protocol, standardized in subsequent fieldwork). This gap was present in the original field records but had not previously been itemized. 4. Figure 1 — new figure added.Added a fieldwork spatial distribution figure to Section 3.3.2 (N = 665 markers, coordinates generalized to ~11 m), showing participant, refusal, exclusion, empirical-data, and site-survey locations across the main fieldwork zone (Da'an District) and full geographic extent. The caption discloses the 22 cases (see item 3 above) not represented in the figure and cross-references Section 3.5.2. 5. Section 3.6.2 (Design Principles for Privacy Protection) — closing sentence revised.The section's closing line ("The law is the floor, not the ceiling...") now cross-references Section 3.6.4 as the location of a more formal benchmarking treatment, rather than standing as an unelaborated claim. 6. Section 3.6.4 (GDPR Benchmarking as a Rigor Standard) — new section added.Added a new subsection following 3.6.3, formally benchmarking the study's data architecture against GDPR principles (Article 3 territorial scope, Article 9 special-category data, Article 5(1)(c) data minimization) as a voluntary rigor standard rather than a compliance claim. This section elaborates, with regulatory grounding, on design commitments already described narratively in Section 3.6.2 (absence of special-category data collection, the tiered public/restricted architecture, and the absence of respondent photographs); it introduces no new empirical claims. 7. Section 3.8 (Research Limitations) — new sixth item added.Added "Consent and post-hoc contact limitations," documenting the same gap described in the revision to 3.6.1, and cross-referencing the data-minimization measures in Section 3.6.2 as a partial constra