# Evidence and sources

Research checked 10 September 2026. This is a focused research synthesis, not an exhaustive systematic review. The ten workflows are original practice designs; none has been tested as an AI intervention. Academic citations do not imply endorsement.

## What the evidence supports
Trust concerns differ, coordination needs shared information and workable structures, and relationship work has costs. Asking someone for their perspective can add information that imagination alone does not. Repair depends on the harm, the response and the history; no script guarantees forgiveness. These propositions inform practical drafts, not diagnoses of a stakeholder.

## Source register and evidence-to-step map

| ID and academic source | Design and access used | Finding used and workflow step | Limit or counterpoint |
|---|---|---|---|
| R1. Colquitt, Scott & LePine (2007). *Trust, trustworthiness, and trust propensity: A meta-analytic test of their unique relationships with risk taking and job performance*. Journal of Applied Psychology, 92, 909–927. [DOI](https://doi.org/10.1037/0021-9010.92.4.909) | Meta-analysis, 132 independent samples; abstract and article opening accessible. | Distinguishes willingness to rely from perceptions of ability, benevolence and integrity. Trust Check steps 1–4 identify a specific concern rather than assign a global score. | Mostly association; a manager’s judgment of a trust dimension is not an objective character test. |
| R2. Bolton, Logan & Gittell (2021). *Revisiting Relational Coordination: A Systematic Review*. Journal of Applied Behavioral Science, 57, 290–322. [DOI](https://doi.org/10.1177/0021886321991597) | Systematic review of empirical work published 1991–2019; publisher abstract. | Relational and communication processes connect across role boundaries. Interface Agreement steps 1–6 and Updates steps 1–3 combine shared work with usable communication. | Mixed study designs and substantial health/human-services relevance; a review of a construct does not validate this pack’s agreement template. |
| R3. Lan, Li, Wang & Meng (2022). *Benefits and costs of employee boundary-spanning behavior: A meta-analytic review*. Acta Psychologica Sinica, 54, 665–683. [DOI](https://doi.org/10.3724/SP.J.1041.2022.00665) | 52 independent samples from 50 studies, N = 14,366; publisher English abstract. | Boundary work relates to useful work outcomes and role stress. Mapping step 3, Updates step 6, Boundaries steps 2–3 and Review step 6 examine who carries the effort. | Association and heterogeneous roles; do not assume more networking or more meetings always helps. |
| R4. Eyal, Steffel & Epley (2018). *Perspective mistaking: Accurately understanding the mind of another requires getting perspective, not taking perspective*. Journal of Personality and Social Psychology, 114, 547–571. [DOI](https://doi.org/10.1037/pspa0000115) | 25 experiments; PubMed abstract. | Imagining another perspective did not consistently improve accuracy; direct conversation improved accuracy in the final experiment. Get Their Perspective steps 2–5 obtain and check real input. | Varied judgment tasks are not a field trial of client relationship repair. Perspective-taking may serve aims other than predictive accuracy. |
| R5. Colquitt, Conlon, Wesson, Porter & Ng (2001). *Justice at the millennium: A meta-analytic review of 25 years of organizational justice research*. Journal of Applied Psychology, 86, 425–445. [DOI](https://doi.org/10.1037/0021-9010.86.3.425) | Meta-analysis of 183 studies; PubMed abstract and article opening. | Fairness has distinct outcome, procedure, interpersonal and informational dimensions. Interests steps 3–5 and Interface steps 2–4 inspect process as well as outcome. | Organizational associations do not prove that a particular process will satisfy all external stakeholders. |
| R6. Shaw, Wild & Colquitt (2003). *To justify or excuse? A meta-analytic review of the effects of explanations*. Journal of Applied Psychology, 88, 444–458. [DOI](https://doi.org/10.1037/0021-9010.88.3.444) | Meta-analyses of 54 independent samples; indexed metadata and PubMed abstract. | Explanation provision and adequacy matter to justice and responses, with contextual differences. Updates steps 3–5 explain constraints and impacts clearly. | “Excuse” is a research category, not permission to evade responsibility. This pack requires truthful explanations and does not prescribe the statistically favored category mechanically. |
| R7. Kim, Ferrin, Cooper & Dirks (2004). *Removing the shadow of suspicion: The effects of apology versus denial for repairing competence- versus integrity-based trust violations*. Journal of Applied Psychology, 89, 104–118. [DOI](https://doi.org/10.1037/0021-9010.89.1.104) | Two experimental studies; indexed record and PubMed abstract. | Response effects depend on violation type and subsequent evidence of guilt or innocence. Trust Check step 3 and Repair steps 1–2 separate established from disputed facts. | Laboratory findings must never become advice to falsely deny integrity violations. Honesty and evidence constrain the response. |
| R8. Haesevoets, Reinders Folmer & Van Hiel (2015). *Is Trust for Sale? The Effectiveness of Financial Compensation for Repairing Competence- versus Integrity-Based Trust Violations*. PLOS ONE, 10, e0145952. [DOI](https://doi.org/10.1371/journal.pone.0145952) | Scenario study plus laboratory experiment; methods/results/discussion passages available through web retrieval; index body retrieval unavailable. | Compensation did not restore trust equally across violation types, and overcompensation added no clear benefit over equivalent compensation. Repair step 3 distinguishes redress from restored confidence. | Artificial settings and financially compensable harm; the result does not determine fair compensation or contractual remedies in a live dispute. |
| R9. Lount, Zhong, Sivanathan & Murnighan (2008). *Getting off on the wrong foot: The timing of a breach and the restoration of trust*. Personality and Social Psychology Bulletin, 34, 1601–1612. [DOI](https://doi.org/10.1177/0146167208324512) | Two iterated-prisoner’s-dilemma experiments; indexed metadata and accessible article opening/method passages. | Timing and prior cooperation shaped recovery; later apparent recovery could still leave distrust. Reliability Trial step 6 and Review step 3 avoid declaring repair from one success. | Short game interactions do not predict the duration or possibility of recovery in an actual working relationship. |
| R10. Yuan, Dong & Li (2017). *The trust repair effect of apology: A systematic review and meta-analysis*. Advances in Psychological Science, 25, 1103–1113. [DOI](https://doi.org/10.3724/SP.J.1042.2017.01103) | 18 papers, 36 effect sizes, 4,731 participants; publisher English abstract. | Apologies showed an average positive effect, moderated by violation type and comparison condition. Repair steps 2–5 include acknowledgment with realistic expectations. | Average effects do not guarantee recovery; this synthesis is older and does not test our full repair workflow. |
| R11. Ferrin, Kim, Cooper & Dirks (2007). *Silence speaks volumes: The effectiveness of reticence in comparison to apology and denial for responding to integrity- and competence-based trust violations*. Journal of Applied Psychology, 92, 893–908. [DOI](https://doi.org/10.1037/0021-9010.92.4.893) | Two laboratory studies in different country contexts; PubMed abstract. | Reticence can leave important questions unresolved. Repair steps 1–2 give a truthful account of what is known and what remains under review. | Does not justify forced statements, disclosure of confidential investigations, or treating silence as proof of guilt. |
| R12. Leicht-Deobald, Backmann, de Vries, Weiss, Hohmann, Walter, van der Vegt & Hoegl (2025). *A Contingency Framework for the Performance Consequences of Team Boundary Management: A Meta-Analysis of 30 Years of Research*. Journal of Management, 51, 704–747. [DOI](https://doi.org/10.1177/01492063231206107) | 85 primary studies, 10,848 teams; indexed metadata and web body passages on boundary-activity definitions. Online first 2023, issue 2025. | Boundary activities have different functions and contingent value. Map step 5 and Interface step 1 distinguish coordination needs from protecting capacity. | We verified scope and activity distinctions, not all moderator estimates. Associations and resource-allocation theory do not establish a universal optimum. |

## What we designed
The map, six-step procedures, output formats, question limits, trial lengths, review prompts and safety routing are original product choices. They are not named academic instruments. None produces a validated trust score. Two-week or one-cycle trials are practical defaults to adjust, not experimentally established doses.

The package combines several levels of evidence: individual trust experiments, organizational justice studies, and team coordination syntheses. Moving between those levels is a reason for caution, not a license to generalize a numeric effect. We deliberately do not claim effect sizes for the pack.

## Use and evaluation
Before a trial, agree one work outcome and one burden indicator, such as returned briefs and reporting time. At review, ask both parties what changed and record alternative explanations. Do not quietly collect personal ratings or compare employees. A small useful local result is evidence about that trial, not proof of general effectiveness.