scan-v5.json (24582B)
1 { 2 "scan_version": 5, 3 "paper_type": "empirical", 4 "paper": { 5 "title": "Interpretive Cultures: Resonance, randomness, and negotiated meaning for AI-assisted tarot divination", 6 "authors": [ 7 "Matthew Kieran Prock", 8 "Ziv Epstein", 9 "Hope Schroeder", 10 "Amy Smith", 11 "Cassandra Lee", 12 "Vana Goblot", 13 "Farnaz Jahanbakhsh" 14 ], 15 "year": 2026, 16 "venue": "CHI '26", 17 "arxiv_id": "2602.11367", 18 "doi": "10.1145/3772318.3791571" 19 }, 20 "checklist": { 21 "claims_and_evidence": { 22 "abstract_claims_supported": { 23 "applies": true, 24 "answer": true, 25 "justification": "Abstract claims about identifying distinct practitioner workflows (navigating uncertainty, exploring perspectives, streamlining practice) are substantiated by the thematic analysis of 12 interviews reported in Section 5.", 26 "source": "haiku" 27 }, 28 "causal_claims_justified": { 29 "applies": false, 30 "answer": false, 31 "justification": "The paper makes no causal claims requiring experimental design; findings are exploratory and descriptive, characterizing how practitioners use AI rather than claiming AI causes specific outcomes.", 32 "source": "haiku" 33 }, 34 "generalization_bounded": { 35 "applies": true, 36 "answer": false, 37 "justification": "Section 7 acknowledges generalizability limits, but Section 6.1 design recommendations ('Designers should consider...', 'designers could intentionally design...') generalize broadly beyond the 12 North American, AI-positive participants studied.", 38 "source": "haiku" 39 }, 40 "alternative_explanations_discussed": { 41 "applies": true, 42 "answer": true, 43 "justification": "The paper acknowledges that participants are self-selected for AI positivity and are unrepresentative of the broader tarot community; it also notes that trust in AI and experience level shape findings in ways requiring further investigation.", 44 "source": "haiku" 45 }, 46 "proxy_outcome_distinction": { 47 "applies": true, 48 "answer": true, 49 "justification": "Interview themes about AI use patterns are presented as findings about AI use patterns — no conflation of measured constructs with unmeasured claims occurs.", 50 "source": "haiku" 51 } 52 }, 53 "limitations_and_scope": { 54 "limitations_section_present": { 55 "applies": true, 56 "answer": true, 57 "justification": "Section 7 'Limitations and Future Work' is a dedicated section covering sample size, geographic restriction, and selection effects.", 58 "source": "haiku" 59 }, 60 "threats_to_validity_specific": { 61 "applies": true, 62 "answer": true, 63 "justification": "Specific threats are named: n=12 too small for subgroup analysis, North American only due to IRB/GDPR constraints, participants inherently AI-positive by inclusion criteria, stigma limiting recruitment channels.", 64 "source": "haiku" 65 }, 66 "scope_boundaries_stated": { 67 "applies": true, 68 "answer": true, 69 "justification": "The paper explicitly notes it studied tarot readers (not consumers), excludes EU/UK populations, and acknowledges that results may not extend to practitioners who reject AI or to non-English-speaking communities.", 70 "source": "haiku" 71 } 72 }, 73 "conflicts_of_interest": { 74 "funding_disclosed": { 75 "applies": true, 76 "answer": false, 77 "justification": "The Acknowledgments section thanks only participants; no funding source is disclosed anywhere in the paper.", 78 "source": "haiku" 79 }, 80 "affiliations_disclosed": { 81 "applies": true, 82 "answer": true, 83 "justification": "All seven authors list their institutional affiliations (UMich, MIT, Queen Mary, Goldsmiths) on the title page.", 84 "source": "haiku" 85 }, 86 "funder_independent_of_outcome": { 87 "applies": false, 88 "answer": false, 89 "justification": "No funder is disclosed, so independence cannot be evaluated.", 90 "source": "haiku" 91 }, 92 "financial_interests_declared": { 93 "applies": true, 94 "answer": false, 95 "justification": "No competing interests or financial interests statement appears anywhere in the paper.", 96 "source": "haiku" 97 } 98 }, 99 "scope_and_framing": { 100 "key_terms_defined": { 101 "applies": true, 102 "answer": true, 103 "justification": "Key terms are defined: 'interpretive work' (Section 1), Rosa's four axes of resonance (Section 3, with explicit definitions of internal/horizontal/diagonal/vertical), and 'meta-semiotic system' (Section 2.4).", 104 "source": "haiku" 105 }, 106 "intended_contribution_clear": { 107 "applies": true, 108 "answer": true, 109 "justification": "The paper enumerates three explicit contributions at the end of Section 1: (1) interdisciplinary theorizing via Resonance Theory, (2) empirical understanding of AI-assisted meaning-making, (3) design recommendations.", 110 "source": "haiku" 111 }, 112 "engagement_with_prior_work": { 113 "applies": true, 114 "answer": true, 115 "justification": "Section 2 engages substantively with prior HCI work on tarot design, AI and subjective labor, interpretive labor, and perspectivist NLP, explaining how this work extends and differs from each strand.", 116 "source": "haiku" 117 } 118 } 119 }, 120 "type_checklist": { 121 "empirical": { 122 "artifacts": { 123 "code_released": { 124 "applies": false, 125 "answer": false, 126 "justification": "This is a qualitative interview study; no computational code exists to release.", 127 "source": "haiku" 128 }, 129 "data_released": { 130 "applies": true, 131 "answer": false, 132 "justification": "Interview transcripts are not publicly released; recordings were deleted post-cleaning per protocol, and no transcript dataset is deposited.", 133 "source": "haiku" 134 }, 135 "environment_specified": { 136 "applies": false, 137 "answer": false, 138 "justification": "No computational environment to specify; Atlas.ti was used for coding but no reproducible software environment is required.", 139 "source": "haiku" 140 }, 141 "reproduction_instructions": { 142 "applies": true, 143 "answer": true, 144 "justification": "Appendix A provides full screening survey questions and all 30+ interview questions, giving sufficient detail to replicate the interview protocol.", 145 "source": "haiku" 146 } 147 }, 148 "statistical_methodology": { 149 "confidence_intervals_or_error_bars": { 150 "applies": false, 151 "answer": false, 152 "justification": "This is a qualitative study with no inferential statistical analysis; percentages cited (e.g., 76.7% of recruitment requests) are descriptive counts with no uncertainty quantification.", 153 "source": "haiku" 154 }, 155 "significance_tests": { 156 "applies": false, 157 "answer": false, 158 "justification": "No comparative quantitative claims are made that would require significance testing.", 159 "source": "haiku" 160 }, 161 "effect_sizes_reported": { 162 "applies": false, 163 "answer": false, 164 "justification": "Qualitative study; effect sizes are not applicable.", 165 "source": "haiku" 166 }, 167 "sample_size_justified": { 168 "applies": true, 169 "answer": true, 170 "justification": "The paper justifies n=12 through thematic saturation — the 12th interview produced no new themes and both co-first authors independently confirmed code saturation (Section 4.1).", 171 "source": "haiku" 172 }, 173 "variance_reported": { 174 "applies": false, 175 "answer": false, 176 "justification": "No quantitative outcomes are measured; variance reporting is not applicable.", 177 "source": "haiku" 178 } 179 }, 180 "evaluation_design": { 181 "baselines_included": { 182 "applies": false, 183 "answer": false, 184 "justification": "Qualitative interview study; no system is evaluated against a baseline.", 185 "source": "haiku" 186 }, 187 "baselines_contemporary": { 188 "applies": false, 189 "answer": false, 190 "justification": "Not applicable — no baseline comparison.", 191 "source": "haiku" 192 }, 193 "ablation_study": { 194 "applies": false, 195 "answer": false, 196 "justification": "Not applicable to a qualitative interview study.", 197 "source": "haiku" 198 }, 199 "multiple_metrics": { 200 "applies": false, 201 "answer": false, 202 "justification": "Qualitative thematic analysis; no quantitative metrics are used.", 203 "source": "haiku" 204 }, 205 "human_evaluation": { 206 "applies": true, 207 "answer": true, 208 "justification": "The study is constituted by human evaluation — 12 practitioners evaluate AI-generated tarot interpretations in naturalistic practice, and their assessments are the primary data.", 209 "source": "haiku" 210 }, 211 "held_out_test_set": { 212 "applies": false, 213 "answer": false, 214 "justification": "Not a prediction task; no test set is relevant.", 215 "source": "haiku" 216 }, 217 "per_category_breakdown": { 218 "applies": true, 219 "answer": true, 220 "justification": "Findings are organized along four axes of Rosa's Resonance Theory (internal, horizontal, diagonal, vertical) in Section 5.2, providing structured thematic breakdowns.", 221 "source": "haiku" 222 }, 223 "failure_cases_discussed": { 224 "applies": true, 225 "answer": true, 226 "justification": "The paper discusses cases where AI inhibits intuition, exhibits sycophancy, and produces 'algorithm over mystery' experiences that undermine divinatory resonance (Sections 5.1.2, 5.2.1, 6).", 227 "source": "haiku" 228 }, 229 "negative_results_reported": { 230 "applies": true, 231 "answer": true, 232 "justification": "Negative aspects are reported throughout: AI inhibits intuition (P3, P4, P5, P6, P9, P12), concerns about skill deterioration, sycophancy, and community stigma — not just positive findings.", 233 "source": "haiku" 234 } 235 }, 236 "setup_transparency": { 237 "model_versions_specified": { 238 "applies": true, 239 "answer": false, 240 "justification": "Participants mention ChatGPT, Midjourney, and DeepSeek by name but no specific model versions or snapshot dates are recorded or reported.", 241 "source": "haiku" 242 }, 243 "prompts_provided": { 244 "applies": true, 245 "answer": false, 246 "justification": "Participant quotes reveal example prompting strategies (asking for 3 interpretations, asking about blind spots) but the actual prompts used by participants are not systematically documented.", 247 "source": "haiku" 248 }, 249 "hyperparameters_reported": { 250 "applies": false, 251 "answer": false, 252 "justification": "The study observes naturalistic AI use by participants; researchers did not configure or control AI hyperparameters.", 253 "source": "haiku" 254 }, 255 "scaffolding_described": { 256 "applies": false, 257 "answer": false, 258 "justification": "No agentic scaffolding is involved; participants interact with commercial chatbots directly.", 259 "source": "haiku" 260 }, 261 "data_preprocessing_documented": { 262 "applies": true, 263 "answer": true, 264 "justification": "Section 4.1 documents the full preprocessing pipeline: Zoom-generated transcripts → manual cleaning by one researcher → deletion of recordings → Atlas.ti coding with anonymized spreadsheets.", 265 "source": "haiku" 266 } 267 }, 268 "data_integrity": { 269 "raw_data_available": { 270 "applies": true, 271 "answer": false, 272 "justification": "Interview transcripts are not publicly available; recordings were deleted after cleaning per protocol.", 273 "source": "haiku" 274 }, 275 "data_collection_described": { 276 "applies": true, 277 "answer": true, 278 "justification": "Section 4 describes Zoom interviews averaging 60 minutes, audio/video auto-saved, transcripts cleaned by the interviewing researcher and then deleted.", 279 "source": "haiku" 280 }, 281 "recruitment_methods_described": { 282 "applies": true, 283 "answer": true, 284 "justification": "Recruitment is described in detail: purposive sampling via Reddit/Discord/Instagram/TikTok keyword search, physical flyer in a tarot shop, direct outreach to AI+tarot posters, three-question screening survey with specific criteria.", 285 "source": "haiku" 286 }, 287 "data_pipeline_documented": { 288 "applies": true, 289 "answer": true, 290 "justification": "Section 4.1 documents the full pipeline: open coding → analytic memos → iterative team discussion → axial/selective coding → interpretive coding → saturation check on 12th interview.", 291 "source": "haiku" 292 } 293 }, 294 "contamination": { 295 "training_cutoff_stated": { 296 "applies": false, 297 "answer": false, 298 "justification": "The paper does not evaluate model capabilities on benchmarks; it studies naturalistic user behavior.", 299 "source": "haiku" 300 }, 301 "train_test_overlap_discussed": { 302 "applies": false, 303 "answer": false, 304 "justification": "Not applicable — no benchmark evaluation.", 305 "source": "haiku" 306 }, 307 "benchmark_contamination_addressed": { 308 "applies": false, 309 "answer": false, 310 "justification": "Not applicable — no benchmark evaluation.", 311 "source": "haiku" 312 } 313 }, 314 "human_studies": { 315 "pre_registered": { 316 "applies": true, 317 "answer": false, 318 "justification": "No pre-registration is mentioned anywhere in the paper.", 319 "source": "haiku" 320 }, 321 "irb_or_ethics_approval": { 322 "applies": true, 323 "answer": true, 324 "justification": "Section 4 explicitly states: 'This study was approved by our Institutional Review Board (IRB).'", 325 "source": "haiku" 326 }, 327 "demographics_reported": { 328 "applies": true, 329 "answer": true, 330 "justification": "Table 1 reports age range, gender identity, education level, length of tarot practice, tarot skill level, and country for all 12 participants.", 331 "source": "haiku" 332 }, 333 "inclusion_exclusion_criteria": { 334 "applies": true, 335 "answer": true, 336 "justification": "Section 4 states explicit criteria: 18+ years old, practice tarot/divination, use AI for personal divination; EU/UK participants excluded due to GDPR/IRB constraints.", 337 "source": "haiku" 338 }, 339 "randomization_described": { 340 "applies": false, 341 "answer": false, 342 "justification": "Qualitative interview study using purposive sampling; no experimental randomization is involved.", 343 "source": "haiku" 344 }, 345 "blinding_described": { 346 "applies": false, 347 "answer": false, 348 "justification": "Not feasible or applicable for qualitative semi-structured interviews.", 349 "source": "haiku" 350 }, 351 "attrition_reported": { 352 "applies": false, 353 "answer": false, 354 "justification": "All 12 enrolled participants completed the study; the 12th interview served as a saturation check, and no dropout is mentioned.", 355 "source": "haiku" 356 } 357 }, 358 "cost_and_practicality": { 359 "inference_cost_reported": { 360 "applies": false, 361 "answer": false, 362 "justification": "The researchers did not run AI inference; participants used commercial tools in naturalistic settings.", 363 "source": "haiku" 364 }, 365 "compute_budget_stated": { 366 "applies": false, 367 "answer": false, 368 "justification": "No computational budget is relevant to this qualitative interview study.", 369 "source": "haiku" 370 } 371 } 372 } 373 }, 374 "claims": [ 375 { 376 "claim": "Practitioners use AI primarily to navigate inner uncertainty and self-doubt during readings.", 377 "evidence": "Multiple participants (P1, P2, P5, P6, P8, P9, P12) describe turning to AI when readings feel opaque or emotionally difficult; P1 and P2 quoted directly on AI resolving second-guessing.", 378 "supported": "strong" 379 }, 380 { 381 "claim": "AI largely inhibits intuition but can also help practitioners navigate ambiguity.", 382 "evidence": "P3, P4, P5, P6, P9, P12 report AI inhibiting intuition; P2, P10, P11 report exploiting AI dialectically to deepen self-connection — both patterns documented in Section 5.2.1.", 383 "supported": "moderate" 384 }, 385 { 386 "claim": "More experienced tarot practitioners are more critical of AI integration than beginners.", 387 "evidence": "Section 5.1.4 reports that advanced participants worry about intuition deterioration and AI's impact on the art industry; newer practitioners (P4, P7, P8, P10, P12) are 'more unequivocally positive.'", 388 "supported": "moderate" 389 }, 390 { 391 "claim": "The tarot community broadly harbors negative sentiment toward AI use.", 392 "evidence": "76.7% of recruitment requests received no response; 70% of responses were rejections; 57% of rejections explicitly cited AI as the reason (Section 5.1.1).", 393 "supported": "moderate" 394 }, 395 { 396 "claim": "AI helps extend existing divinatory practices by enabling card pulls, streamlining interpretation, and journaling.", 397 "evidence": "P3, P6, P9, P12 use AI for card draws; P4, P10, P12 delegate full interpretation to AI; P1, P9 use AI for journaling and feedback on divination system development (Section 5.1.4).", 398 "supported": "strong" 399 }, 400 { 401 "claim": "AI functions as a source of alternative perspectives and 'objective' outside viewpoints.", 402 "evidence": "Nearly all participants (P1–P12) use AI to generate alternative interpretations; P2, P3, P4, P6 describe AI as providing a view 'without stake' in the personal situation (Section 5.1.3).", 403 "supported": "strong" 404 } 405 ], 406 "methodology_tags": [ 407 "qualitative", 408 "case-study" 409 ], 410 "key_findings": "Twelve tarot practitioners who use AI report three primary usage patterns: navigating uncertainty and self-doubt, gaining alternative perspectives, and streamlining the interpretive workflow. AI mostly inhibits intuitive engagement but can deepen reflection when practitioners retain interpretive agency. More experienced practitioners are more skeptical, while beginners use AI more uncritically. Selection bias is substantial — the tarot community at large is largely hostile to AI, making the sample inherently unrepresentative. The study offers design recommendations centered on preserving user agency, productive friction, and re-injecting randomness into AI systems.", 411 "red_flags": [ 412 { 413 "flag": "Tiny, self-selected sample", 414 "detail": "n=12 participants who already use AI for tarot; explicitly unrepresentative of the broader tarot community which largely rejects AI, as documented by recruitment rejection rates." 415 }, 416 { 417 "flag": "No inter-rater reliability statistic", 418 "detail": "Coding reliability is described qualitatively (consensus discussion, double-coding, saturation check) but no Cohen's kappa or percent agreement figure is reported." 419 }, 420 { 421 "flag": "Design recommendations overgeneralize", 422 "detail": "Section 6.1 offers broad design prescriptions ('Designers should consider...') for AI interpretive systems generally, unsupported by the 12-person North American sample." 423 }, 424 { 425 "flag": "No funding disclosure", 426 "detail": "Acknowledgments thank only participants; no funding source, grant number, or competing interests statement is present." 427 }, 428 { 429 "flag": "AI tools not documented", 430 "detail": "Participants used ChatGPT, Midjourney, DeepSeek and other tools, but model versions, dates of use, and usage contexts are not systematically recorded." 431 } 432 ], 433 "cited_papers": [ 434 { 435 "title": "Large Language Models in Qualitative Research: Uses, Tensions, and Intentions", 436 "relevance": "Directly relevant — examines tensions when using LLMs in qualitative/interpretive research, a closely analogous domain to this paper's AI-assisted meaning-making." 437 }, 438 { 439 "title": "Just Put a Human in the Loop? Investigating LLM-Assisted Annotation for Subjective Tasks", 440 "relevance": "Shows AI assistance homogenizes subjective annotation outputs and slows users — supports this paper's finding about AI's double-edged role in interpretive tasks." 441 }, 442 { 443 "title": "Homogenization Effects of Large Language Models on Human Creative Ideation", 444 "relevance": "Documents AI reducing diversity of outputs in creative tasks — parallel concern to this paper's worry about AI homogenizing tarot interpretation." 445 }, 446 { 447 "title": "Towards understanding sycophancy in language models", 448 "relevance": "Cited to explain AI chatbot sycophancy, which participants identify as a key limitation of AI tarot interpretation." 449 }, 450 { 451 "title": "From explainability to ineffability? ML tarot and the possibility of inspiriting design", 452 "relevance": "Prior HCI work directly using ML-generated tarot cards to explore AI explainability — foundational for this paper's design recommendations." 453 }, 454 { 455 "title": "Resonance: A Sociology of Our Relationship to the World", 456 "relevance": "Rosa's Resonance Theory is the primary theoretical framework applied throughout this paper." 457 }, 458 { 459 "title": "Generative artificial intelligence enhances individual creativity but reduces the collective diversity of novel content", 460 "relevance": "Provides empirical evidence for the creativity/homogenization tradeoff discussed in the related work section." 461 }, 462 { 463 "title": "AI Séance: Recounts from designing artificial intelligence for transcendence, interpretive lenses and chance", 464 "relevance": "Direct precursor — the same research group's practice-based design work on AI divination, grounding the design recommendations." 465 } 466 ], 467 "engagement_factors": { 468 "practical_relevance": { 469 "score": 1, 470 "justification": "Design recommendations are speculative and niche; practitioners and AI designers may find the framework interesting but actionable guidance is thin." 471 }, 472 "surprise_contrarian": { 473 "score": 2, 474 "justification": "The finding that AI 'hallucinates' meaning usefully in a non-causal context — and that practitioners deliberately exploit AI randomness as a divinatory mechanism — is genuinely counterintuitive." 475 }, 476 "fear_safety": { 477 "score": 0, 478 "justification": "No AI safety or risk concerns are raised; the stakes are personal meaning-making, not harm." 479 }, 480 "drama_conflict": { 481 "score": 1, 482 "justification": "Community hostility to AI in tarot (70% recruitment rejection rate) provides a low-grade conflict angle but is not the paper's focus." 483 }, 484 "demo_ability": { 485 "score": 2, 486 "justification": "Anyone can immediately try asking ChatGPT to interpret a tarot spread, making the phenomenon directly experiential for readers." 487 }, 488 "brand_recognition": { 489 "score": 2, 490 "justification": "MIT Media Lab affiliation and CHI venue are recognizable; no famous lab product but strong academic brand." 491 } 492 }, 493 "hn_data": { 494 "threads": [ 495 { 496 "hn_id": "47023905", 497 "title": "Retrieval-Aware Distillation for Transformer-SSM Hybrids", 498 "points": 2, 499 "comments": 0, 500 "url": "https://news.ycombinator.com/item?id=47023905", 501 "created_at": "2026-02-15T14:24:02Z" 502 } 503 ], 504 "top_points": 2, 505 "total_points": 2, 506 "total_comments": 0 507 } 508 }