# Research Appendixes — The Searcher's Paradox

The Searcher's Paradox — Research Context, Sources, and Evidence Analysis.

These appendixes document the research architecture behind each chapter of The Searcher's Paradox. They show which claims are supported directly by peer-reviewed evidence, which are supported by adjacent research, and which represent the author's strategic synthesis. Each appendix includes a claim-to-evidence analysis, recommended references with live source links, a source-strength assessment, and reflective questions for practitioners.

Classification system: Foundational scholarly sources (peer-reviewed research establishing durable concepts), Current empirical evidence (recent studies testing AI-mediated search behavior), Technical and platform documentation (primary documentation for named systems), Industry observation (practitioner analysis of market behavior), and Author interpretation (Jean Dorff's synthesis, framework, and conclusions).

## Appendix — Prologue: Research and Endnote Recommendation

Editorial recommendation: keep the prologue free of parenthetical citations; use numbered endnotes for three research-supported additions and retain a selective research appendix.

Core prologue endnotes: Huang, Goyal, Saha, & Chandrasekharan (2026), Answer Bubbles: Information Exposure in AI-Mediated Search (arXiv:2603.16138) — most direct support for source-selection bias and premature closure. Choi & Rieh (2026), Interest-Driven Search in AI-Mediated Information Environments: An Audio Diary Study (CHIIR 2026, 453–457) — tension between easy beginnings and sustained inquiry. Afgiansyah, Arlusi, Sakaril, & Fasta (2026), AI-Mediated Communication Beyond Human-AI Dyads (JCLC 6(1), 40–62, https://doi.org/10.33093/jclc.2026.6.1.3) — human–AI–human mediation framework. Background: Gao et al. (2023), Retrieval-Augmented Generation for LLMs: A Survey (arXiv:2312.10997). Supporting: Dai et al. (2024), Bias and Unfairness in Information Retrieval Systems. Excluded from scholarly endnotes: Substack posts, vendor explainers, Forbes commentary, access-denied pages, Semantic Scholar discovery links.

## Appendix — Chapter 1: Sources, Research Pathways, and Conceptual Continuities

Historical anchors: British Museum, Barnham fire-making research, deliberate fire-making c. 400,000 years ago (https://www.britishmuseum.org/about-us/press/press-releases/groundbreaking-discovery-shows-earliest-evidence-fire-making). Polynesian Voyaging Society, Hōkūleʻa wayfinding (https://hokulea.com/polynesian-wayfinding/). Library of Congress, deciphering scribal abbreviations (https://guides.loc.gov/manuscript-facsimiles/deciphering-scribal-abbreviations). British Library manuscripts, scribes and palimpsests (https://www.bl.uk/manuscripts/). Ann M. Blair, Too Much to Know (Yale University Press, 2010, https://yalebooks.yale.edu/book/9780300165395/too-much-to-know/) — early modern information overload. Conceptual continuity with Jean Dorff's Voice Intelligence: orientation as recognizing a present internal state before expression becomes reliable.

## Appendix — Chapter 2: Educational Design, Inquiry Expansion, and Receiver Participation

National Research Council, How People Learn (National Academies Press, 2000). Tomlinson, The Differentiated Classroom (ASCD, 2014). Chi & Wylie, The ICAP Framework (Educational Psychologist 49(4), 2014, 219–243). Sanders & Stappers, Co-creation and the New Landscapes of Design (CoDesign 4(1), 2008, 5–18). Kapur, Productive Failure (Cognition and Instruction 26(3), 2008, 379–424; Wiley, 2024). Texas Instruments graphing-calculator research; Burrill et al. (2002); Ellington, meta-analysis of calculator effects (JRME 34(5), 2003, 433–463). NIST AI Risk Management Framework (2023); UNESCO Guidance for Generative AI in Education and Research (2023); Shneiderman, Human-Centered AI (Oxford, 2022).

## Appendix — Chapter 3: AI-Mediated Search, Business Strategy, and the Visibility Paradox

Claims: AI-mediated search changes the route to a provider (Pagadala, SSRN 2026, https://doi.org/10.2139/ssrn.6512878; Pham et al., SAGE Open 14(4), 2024, https://doi.org/10.1177/21582440241300007). AI participates in orientation before a human provider enters (Angus et al., JAMA 334(18), 2025, https://doi.org/10.1001/jama.2025.18490). Synthesis narrows source exposure (Huang et al., Answer Bubbles, 11,000 queries, https://doi.org/10.48550/arXiv.2603.16138). Organizational priorities shape AI outcomes (MIT SMR AI and Business Strategy, https://sloanreview.mit.edu/big-ideas/artificial-intelligence-business-strategy/; Wang & Zhang, Int. J. Production Research 63(18), 2025, https://doi.org/10.1080/00207543.2025.2485318). "The visibility trap" and "machine-legible without becoming machine-shaped" are original author syntheses.

## Appendix — Chapter 4: The Seeker-Source Triangle, Recursive Expertise, and Platform-Mediated Visibility

Claims: providers participate in the systems they advise on (Google, OpenAI, Microsoft guidance). Mediators select and reshape the information environment (Zhang et al., 55,936 queries, https://doi.org/10.48550/arXiv.2512.09483). Conversational mediators change how seekers search (Sharma, Liao, & Xiao, CHI 2024 Best Paper, https://doi.org/10.1145/3613904.3642459). Counterfactual explanations make mediation scrutable (Ghazimatin et al., PRINCE, WSDM 2020, https://doi.org/10.1145/3336191.3371824). Generative engine optimization (Aggarwal et al., KDD 2024, https://doi.org/10.1145/3637528.3671900). Calibrated trust (NIST AI RMF 1.0, https://doi.org/10.6028/NIST.AI.100-1; Generative AI Profile, https://doi.org/10.6028/NIST.AI.600-1). Platform documentation: Google Search Central AI optimization guide (https://developers.google.com/search/docs/fundamentals/ai-optimization-guide); Microsoft Bing AI Performance in Webmaster Tools (https://blogs.bing.com/webmaster/February-2026/Introducing-AI-Performance-in-Bing-Webmaster-Tools-Public-Preview); OpenAI publishers and developers FAQ (https://help.openai.com/en/articles/12627856). The seeker-source-mediator triangle is an original organizing model.

## Appendix — Chapter 5: Binary Decisions, Conditional Relevance, and AI-Mediated Judgment

Claims: binary categories simplify decisions but do not exhaust reality (Voxco, https://www.voxco.com/resources/the-pitfalls-of-binary-thinking-in-research-and-marketing; HBR IdeaCast, https://hbr.org/podcast/2022/08/decisions-dont-have-to-be-either-or). Analytical capability does not eliminate situated judgment (Szukits, Journal of Management Control 33, 2022, https://doi.org/10.1007/s00187-022-00343-w). Communicated AI confidence changes human reliance (Ishizu et al., Applied Sciences 14(16), 2024, https://doi.org/10.3390/app14167192). Explanation can increase over-reliance (He, Aishwarya, & Gadiraju, ACM IUI 2025, https://doi.org/10.1145/3708359.3712133). Algorithmic curation can create an illusion of autonomy (Ponce-Rojo et al., Frontiers in Communication 10, 2026, https://doi.org/10.3389/fcomm.2025.1684370). Decisions remain conditional (Deshmukh et al., PLOS ONE 20(2), 2025, https://doi.org/10.1371/journal.pone.0316805; Stanghellini, European Psychiatry 68(1), 2025, https://doi.org/10.1192/j.eurpsy.2025.10054). "Conditional positioning" is an original formulation.

## Appendix — Chapter 6: Authority, Fit, and the Boundaries of Expertise

The full research appendix follows the manuscript structure; see the original appendix document for the complete claim-to-evidence analysis, citation notes, recommended references, and source-strength assessment.

## Appendix — Chapter 7: Fear, Urgency, and Decisions Under Uncertainty

Claims: organizational FOMO can be present without proving bias (Mari, Mandelli, & Algesheimer, UZH Working Paper 401, 2024, https://doi.org/10.2139/ssrn.4736154). Manufactured urgency is visible in ordinary marketing practice (Mailchimp, https://mailchimp.com/resources/fomo-marketing/). FOMO can alter choice (Hasan et al., JIPD 8(9), 2024, https://doi.org/10.24294/jipd.v8i9.6419; Mahmud et al., IJPBR 8(11), 2023, https://doi.org/10.26668/businessreview/2023.v8i11.3834; Alvita, DIJDBM 6(6), 2025, https://doi.org/10.38035/dijdbm.v6i6.5676). AI adoption is conditioned by trust, access, and integration (ESCP, https://escp.eu/thechoice/tomorrow-choices/how-artificial-intelligence-will-transform-decision-making/). Online reviews can buffer FOMO effects (Oktaga et al., ICCMS 2(1), 2025, https://doi.org/10.70062/iccms.v2i1.125). Correctable decisions align with adaptation research (Hennes et al., npj Climate Action 3, 2024, https://doi.org/10.1038/s44168-024-00116-2). "Fear should sharpen orientation, not surrender choice" is the author's formulation.

## Appendix — Chapter 8: Agency, Participation, and Responsible Orientation

Claims: agency is partly a design responsibility (Monge Roffarello et al., ACM TOCHI 32(4), 2025, https://doi.org/10.1145/3725215). Participation depends on where and how AI intervenes (Reza et al., PACMHCI 9(7), 2025, https://doi.org/10.1145/3757566). Human-in-the-loop control is not the whole of agency (Barak-Medina, HHAI Workshops 2025, https://ceur-ws.org/Vol-4074/short9-2.pdf). Contradiction can become material for learning (Liu et al., Technology in Society 85, 2026, https://doi.org/10.1016/j.techsoc.2025.103196). Visibility preserves informed control but is not participation (Hodjat & Wiener, World Economic Forum, 2025, https://www.weforum.org/stories/emerging-technologies/rethinking-the-user-experience-in-the-age-of-multi-agent-ai/). "Do not force the answer — improve the orientation from which the person chooses the next step" is an original formulation.

## Appendix — Closing: Evidence, Conditions, and the Search That Continues

Claims: discovery is an extended process (Schickore, Stanford Encyclopedia of Philosophy, https://plato.stanford.edu/archives/fall2024/entries/scientific-discovery/). Direct answers change how people search (Wu et al., ACM CIKM 2020, https://doi.org/10.1145/3340531.3412017). Conditional findings clarify when conditions are examined (Dramé & Godaert, Nutrients 15(7), 2023, https://doi.org/10.3390/nu15071780; Liu et al., Obesity Reviews 24(2), 2023, https://doi.org/10.1111/obr.13534). Expertise has a maintenance requirement (Fröhlich et al., Cardiology 147(5-6), 2022, https://doi.org/10.1159/000527332). Conditional relevance as procedural analogy (Federal Rule of Evidence 104, https://www.law.cornell.edu/rules/fre/rule_104). The responsible triangle (searcher, provider, mediator) is the book's own synthesis.

Author: Jean Dorff. Publisher: Structural Insight Publishing. Published 2026-08-09. Book page: /books/searchers-paradox