Map Literature Contradictions

Map Literature Contradictions

Discipline And Interview Setup

Read discipline-routing.md, then only the selected discipline profile. Reuse the user's existing brief; do not restart intake at each stage. For journalism/communication or information science, use that profile's concepts, extraction fields, and appraisal checks instead of a generic social-science template.

When the user requests "grill me", 追问, or 模拟答辩, read research-interview.md. Ask one decision-relevant question per turn and wait; do not generate the stage deliverable until enough decisions are settled or the user asks to proceed. In quick/no-question mode, record assumptions and continue without inventing answers.

For evidence extraction or drafting, read review-workflow.md. With no sources, give a plan or outline, not a purported synthesis. If reference files are unavailable, retain the source guardrails below and state that discipline guidance is missing.

Goal

Use contradiction as information. Instead of avoiding inconsistent findings, map what researchers agree on, where they diverge, and what hidden conditions or method limits explain the disagreement.

Workflow: Comb Method

1. Assess convergence within the available corpus, accounting for study quality and overlapping reports. If none is supported, say so; repetition or citation count alone is not consensus.
2. Separate contradictions from incomparable questions or measures: name the exact conflicting claim only when the studies address a comparable object. Otherwise describe the difference without manufacturing disagreement.
3. Compare study conditions: inspect object, sample, indicator, method, time window, and context.
4. Infer why the difference matters: connect the comparison to theory, mechanism, or validity.
5. Mark explanation boundaries: identify what current methods or data cannot answer.
6. Convert the boundary into a possible research contribution.

Contradiction Matrix

Use this table when the user provides multiple papers or claims. Leave the status column honest: if the user gave no source for a claim, write "未提供出处" and do not fill in a citation; if a check found nothing, write "未查到".

markdownCopy code
| Study/claim | Finding | Object/sample | Method | Indicator | Time/context | Possible reason for difference | Source / verification status (出处/核验状态) |
|---|---|---|---|---|---|---|---|
| ... | ... | ... | ... | ... | ... | ... | Author (year) · 用户提供/已核验/未核验 · how verified |

Use the following pattern only when the evidence supports a conditional explanation. Otherwise state that the disagreement is unresolved or the studies are not comparable; do not force this conclusion:

textCopy code
The disagreement is less about whether [factor] matters and more about when it matters. Studies using [sample/method/indicator] tend to find [result], while studies in [context/time] find [different result]. This suggests the effect may be conditional on [boundary], rather than universally positive or negative.

Difference Dimensions To Check

• Research object: region, industry, population, organization type, development stage.
• Sample: size, selection method, subgroup, representativeness.
• Indicator: how the key variable or outcome is measured.
• Method: cross-sectional, longitudinal, interview, experiment, case, meta-analysis.
• Time window: policy period, crisis period, before/after intervention, short vs long term.
• Theoretical lens: rational choice, institutional, social network, capability, cultural, ecological, etc.

Output Format

Provide:

markdownCopy code
## Consensus

- ...

## Main Contradictions

1. ...

## Contradiction Matrix

| Study/claim | Finding | Object/sample | Method | Indicator | Time/context | Possible reason | Source / verification status (出处/核验状态) |
|---|---|---|---|---|---|---|---|

## 段落草稿（需作者改写并核对来源）

[Draft synthesis paragraph. Start with: "以下为草稿，请改写成你自己的论证，并逐句核对来源后再使用。" Tag every literature-dependent sentence with source and status, or [需补来源]]

## Explanation Boundaries

- Current literature cannot yet explain ...

## 待核实清单

[Unverified sources, "未查到" results, sentences marked [需补来源], and inferences that still need checking; write "无" if empty]

## 给作者的问题

[Judgment calls that belong to the author; write "无" if empty]

The output sections above, and the two closing sections, keep fixed names and order; see references/output-contract.md (a copy bundled in this skill folder).

Guardrails

• Do not force agreement by selecting only consistent studies.
• Do not claim a contradiction is solved unless the provided evidence supports the explanation.
• Distinguish observed differences from inferred importance; label inferences clearly.
• Source integrity: follow references/source-integrity.md. Every row of the contradiction matrix needs a source and status; do not invent rows to fill the table. Use only sources the user provided or that are verified, tag each with a status (用户提供 / 已核验 / 未核验), write "未查到" instead of completing authors, year, journal, or pages, and put anything unverified in ## 待核实清单. If that file is not available, still apply these rules.