Jump to content

Slave Owner Game/Diagnostic Inversion Test: Difference between revisions

From The Sovereign Games
← Older editNewer edit →

Sovereign (talk | contribs)

3 user groups
3,617 edits
VisualWikitext
m deleted extra gamemodule
Edit Summary Changes Made: Tightened and clarified the opening paragraph (removed repetition). Sharpened several Failure Modes descriptions for precision (especially Status Bias and State-Dependent Bias). Improved flow and clarity in the “How to Run the Test in Practice” section. Made the Substitution Fairness Criteria more prominent and easier to reference. Minor wording improvements throughout for readability and instrument-like tone. Updated version number and Admin Page Status rationale. W
Line 8: Line 8:
| Functional Layer = Diagnostic
| Functional Layer = Diagnostic
| Application Layer = Multi-Layer
| Application Layer = Multi-Layer
| Version = 0.3
| Version = 0.4
| Maturity = Development
| Maturity = Development
| Last Updated = {{CURRENTYEAR}}-{{CURRENTMONTH}}-{{CURRENTDAY}}
| Last Updated = {{CURRENTYEAR}}-{{CURRENTMONTH}}-{{CURRENTDAY}}
Line 18: Line 18:
= Diagnostic Inversion Test =
= Diagnostic Inversion Test =


'''The Diagnostic Inversion Test is not specifically a test for tribalism.''' It is a general symmetry check: does this diagnosis get applied using the same standard of evidence, regardless of who is being diagnosed? Tribalism — treating your own group's behavior more leniently than an out-group's identical behavior — is the '''most common and most visible''' failure of this symmetry, which is why it's often used as the example. But it is one instance of a broader failure, not the definition of the test.
The **Diagnostic Inversion Test** is a symmetry check. It asks whether a diagnosis is being applied using the same standard of evidence, regardless of who is being diagnosed. Tribalism (treating your own group more leniently than an out-group for identical behavior) is the most common and visible failure of this symmetry, which is why it is often used as the primary example. However, it is only one instance of a broader problem.


'''Passing this test is necessary, not sufficient.''' A symmetric conclusion is not thereby a correct one — this test only rules out one specific failure mode (diagnostic asymmetry). The actual mechanism test (see [[Observable Behavior Rule]] and [[Slave Owner Game/False Positives|False Positives]]) still has to be run and still has to be passed. Treat this test as a necessary gate, not a substitute for the underlying diagnosis.
**Passing this test is necessary, but not sufficient.** A symmetric conclusion is not automatically correct. This test only rules out one specific failure mode (diagnostic asymmetry). The actual mechanism test (see [[Observable Behavior Rule]] and [[Slave Owner Game/False Positives|False Positives]]) must still be run and passed.


== The Test ==
== The Test ==


Take the specific claim you're about to make — "this person/group/institution is running the Slave Owner Game" — and ask: '''would you reach the same conclusion, using the same evidence and the same standard, if you swapped in someone or something you feel differently about?''' If the answer changes based on identity rather than evidence, the diagnosis is not calibrated — it's motivated reasoning wearing this framework's vocabulary.
Take the specific claim you are about to make and ask:
**Would I reach the same conclusion, using the same evidence and the same standard, if I substituted someone or something I feel differently about?**


== Failure Modes Beyond Tribalism ==
If the conclusion changes based on identity rather than evidence, the diagnosis is not calibrated.


Diagnostic asymmetry shows up in several distinct ways. Tribalism is the most visible; the others are just as real and easier to miss precisely because they don't look like group bias.
== Failure Modes ==


* '''Tribal Bias''' — applying a looser standard to your own in-group, and a stricter one to an out-group, for behavior that is otherwise identical.
Diagnostic asymmetry can appear in several forms. Tribalism is the most visible, but the others are equally real and often harder to detect because they do not obviously involve group identity.
* '''Familiarity Bias''' — applying a looser standard to someone you know personally, and a stricter one to a stranger, independent of any group identity at all.
* '''Recency/Sympathy Bias''' — applying a looser standard to someone you currently sympathize with (perhaps because they were recently wronged themselves), even when the specific behavior under evaluation is unrelated to that sympathy.
* '''State-Dependent Bias''' — applying a stricter standard when you are currently angry, hurt, or personally invested in an outcome, and a looser one when you are calm and uninvolved — same evidence, different verdict, driven entirely by your own emotional state at the time of judgment.
* '''Status Bias''' — applying a looser standard to someone with more social, institutional, or professional power, on the assumption that their position itself is evidence of legitimacy, or the reverse: a stricter standard applied specifically because someone is powerful, regardless of the actual evidence.


'''None of these require an out-group to occur.''' A person can apply this diagnostic asymmetrically to two members of their own family, their own political side, or their own close friends, based purely on which one they currently favor.
* **Tribal Bias** — Applying a looser standard to one’s own in-group and a stricter standard to an out-group for behavior that is otherwise identical.
* **Familiarity Bias** — Applying a looser standard to someone known personally and a stricter standard to a stranger, even within the same group or category.
* **Recency/Sympathy Bias** — Applying a looser standard to someone currently sympathized with (for example, because they were recently harmed), even when the behavior being evaluated is unrelated.
* **State-Dependent Bias** — Applying a stricter standard when angry, hurt, or personally invested, and a looser standard when calm — producing different verdicts on the same evidence depending on the evaluator’s emotional state.
* **Status Bias** — Applying a looser standard to someone with greater social, institutional, or professional power (assuming their position signals legitimacy), or conversely, applying a stricter standard specifically because someone holds power.
 
None of these require an out-group. A person can apply this diagnostic asymmetrically to two members of their own family, political side, or friend group based purely on current favor or emotional state.


== Meta-Bias Warning: Gaming the Test ==
== Meta-Bias Warning: Gaming the Test ==


'''This test can itself be gamed.''' A person can satisfy its letter while defeating its purpose by choosing a substitution deliberately designed to be easy to distinguish from the real case — proving nothing, while claiming the appearance of having checked their own bias. Example: substituting "a stranger you've never met" for "your close friend" is a weak test, since almost any asymmetry there is explainable by legitimate differences in available evidence, not bias. A genuine substitution holds context, power differential, and available evidence roughly constant, and only varies the identity/relationship variable actually being checked.
This test can itself be gamed. A person can satisfy its surface requirements while defeating its purpose by choosing a substitution deliberately designed to be easy to distinguish from the real case.


'''If a person cannot articulate why their chosen substitution was fair before running the test — not after, when the answer is already known — the test has likely been gamed rather than genuinely applied.'''
**Example:** Substituting “a stranger you’ve never met” for “your close friend” is usually a weak test, because differences in available evidence can explain asymmetry without requiring bias.
 
A genuine substitution holds power differential, context, and available evidence roughly constant, and only varies the identity or relationship being checked.
 
**If a person cannot clearly articulate *why* their chosen substitution is fair *before* running the test (not after the result is known), the test has likely been gamed rather than genuinely applied.**


== Substitution Fairness Criteria ==
== Substitution Fairness Criteria ==


A valid substitution should hold roughly constant:
A valid substitution should hold roughly constant:
* '''Power differential''' — comparable relative power between the parties, not a case with a vastly different power imbalance.
* '''Context''' — a comparable relationship type or setting (don't substitute a national government for a toddler; don't substitute a stranger for a lifelong close relationship).
* '''Stakes and evidence available''' — a comparable amount and quality of evidence, not a case where one side has far more documented behavior than the other.


If a substitution fails these criteria, the test has not actually been run — it has been performed.
* **Power differential** — Comparable relative power between the parties.
* **Context** — Comparable relationship type or setting.
* **Stakes and evidence available** — Comparable amount and quality of evidence.
 
If a substitution fails these criteria, the test has not actually been run — it has only been performed.


== How to Run the Test in Practice ==
== How to Run the Test in Practice ==


# State the specific claim and the specific evidence supporting it.
1. State the specific claim and the specific evidence supporting it.
# Identify who is being diagnosed.
2. Identify who is being diagnosed.
# Choose a substitution that satisfies the Substitution Fairness Criteria above — and be able to state why it's fair '''before''' running the test, not after.
3. Choose a substitution that satisfies the Substitution Fairness Criteria above. Be able to explain *why* it is fair **before** running the test.
# Substitute, holding the evidence constant. Ask honestly: does the conclusion change? If yes, identify which of the failure modes above is producing that change.
4. Substitute while holding the evidence constant. Ask honestly: Does the conclusion change?
# If the conclusion holds regardless of a fair substitution, the diagnosis has passed this test — it has '''not''' thereby been proven correct, only shown to not be failing '''this particular''' check. See [[Observable Behavior Rule]] and [[Slave Owner Game/False Positives|False Positives]] for the actual content tests still required.
5. If the conclusion changes, identify which failure mode is producing the difference.
6. If the conclusion holds under a fair substitution, the diagnosis has passed this test. It has **not** thereby been proven correct — only shown to not be failing this particular check.


== Relationship to Other Pages ==
== Relationship to Other Pages ==


This test is referenced by, and load-bearing for, several other pages that already depend on symmetric application to function correctly:
This test supports several other pages that depend on consistent application:
* [[Slave Owner Game/Legitimate Authority]] — the Symmetry Table there is a direct application of this test to one specific case.
 
* [[Slave Owner Game/False Positives]] — misdiagnosis is far more likely when this test hasn't been run first.
* [[Slave Owner Game/Legitimate Authority]] — The Symmetry Table is a direct application of this test.
* [[Observable Behavior Rule]] — this test and that rule are closely related but distinct: Observable Behavior Rule governs '''what counts as evidence''' (actions and consequences, not motive); this test governs whether that evidence is being '''applied consistently''' once gathered. A diagnosis can pass Observable Behavior Rule (it's genuinely about actions, not motive) and still fail this test (the same action gets diagnosed differently depending on who did it).
* [[Slave Owner Game/False Positives]] — Misdiagnosis becomes far more likely when this test has not been run.
* [[Observable Behavior Rule]] — These two are related but distinct. Observable Behavior Rule governs *what counts as evidence*. This test governs whether that evidence is being *applied consistently*.


== Development Breadcrumb ==
== Development Breadcrumb ==


The five failure modes above are a first-pass taxonomy, not validated against real cases. Future calibration should test whether these are genuinely distinct categories or overlapping descriptions of the same underlying bias, whether additional failure modes exist beyond these five, and whether the Substitution Fairness Criteria are themselves sufficient to prevent the test from being gamed, or whether gaming can occur even within technically "fair" substitutions.
The five failure modes listed above are a first-pass taxonomy. Future calibration should test:
- Whether these categories are genuinely distinct or overlapping.
- Whether additional failure modes exist.
- Whether the Substitution Fairness Criteria are sufficient to prevent gaming, or whether gaming can still occur within technically “fair” substitutions.


== Calibration Dependencies ==
== Calibration Dependencies ==
Line 101: Line 114:
| instrument_grade = Development
| instrument_grade = Development
| validation = Low
| validation = Low
| calibration_rationale = Stage 2 of round-robin development. ChatGPT identified and closed a real gap: the test as written (Stage 1) had no guard against being strategically gamed via an easy, non-representative substitution. Added Meta-Bias Warning, Substitution Fairness Criteria, and explicit "necessary not sufficient" framing near the top rather than buried in Step 5. Not yet reviewed by Grok — Stage 3 pending.
| calibration_rationale = Stage 3 revision. Tightened opening, sharpened Failure Modes descriptions (especially Status Bias and State-Dependent Bias), improved clarity and flow of "How to Run the Test in Practice," and made Substitution Fairness Criteria more prominent. Not yet round-robin reviewed.
| review_confidence = Moderate
| review_confidence = Moderate
| review_date = {{CURRENTYEAR}}-{{CURRENTMONTH}}-{{CURRENTDAY}}
| review_date = {{CURRENTYEAR}}-{{CURRENTMONTH}}-{{CURRENTDAY}}
Line 119: Line 132:
| drift_report_status = None open
| drift_report_status = None open
| depends_on = Slave Owner Game; Observable Behavior Rule; Slave Owner Game/False Positives; Slave Owner Game/Legitimate Authority
| depends_on = Slave Owner Game; Observable Behavior Rule; Slave Owner Game/False Positives; Slave Owner Game/Legitimate Authority
}}
}}
'''Canonical Question:''' Would you reach the same conclusion, using the same evidence, if the identity of the person or group being diagnosed were swapped for someone or something you feel differently about?
= Diagnostic Inversion Test =
'''The Diagnostic Inversion Test is not specifically a test for tribalism.''' It is a general symmetry check: does this diagnosis get applied using the same standard of evidence, regardless of who is being diagnosed? Tribalism — treating your own group's behavior more leniently than an out-group's identical behavior — is the '''most common and most visible''' failure of this symmetry, which is why it's often used as the example. But it is one instance of a broader failure, not the definition of the test.
== The Test ==
Take the specific claim you're about to make — "this person/group/institution is running the Slave Owner Game" — and ask: '''would you reach the same conclusion, using the same evidence and the same standard, if you swapped in someone or something you feel differently about?''' If the answer changes based on identity rather than evidence, the diagnosis is not calibrated — it's motivated reasoning wearing this framework's vocabulary.
== Failure Modes Beyond Tribalism ==
Diagnostic asymmetry shows up in several distinct ways. Tribalism is the most visible; the others are just as real and easier to miss precisely because they don't look like group bias.
* '''Tribal Bias''' — applying a looser standard to your own in-group, and a stricter one to an out-group, for behavior that is otherwise identical.
* '''Familiarity Bias''' — applying a looser standard to someone you know personally, and a stricter one to a stranger, independent of any group identity at all.
* '''Recency/Sympathy Bias''' — applying a looser standard to someone you currently sympathize with (perhaps because they were recently wronged themselves), even when the specific behavior under evaluation is unrelated to that sympathy.
* '''State-Dependent Bias''' — applying a stricter standard when you are currently angry, hurt, or personally invested in an outcome, and a looser one when you are calm and uninvolved — same evidence, different verdict, driven entirely by your own emotional state at the time of judgment.
* '''Status Bias''' — applying a looser standard to someone with more social, institutional, or professional power, on the assumption that their position itself is evidence of legitimacy, or the reverse: a stricter standard applied specifically because someone is powerful, regardless of the actual evidence.
'''None of these require an out-group to occur.''' A person can apply this diagnostic asymmetrically to two members of their own family, their own political side, or their own close friends, based purely on which one they currently favor.
== How to Run the Test in Practice ==
# State the specific claim and the specific evidence supporting it.
# Identify who is being diagnosed.
# Substitute a different person, group, or institution into the same claim, holding the evidence constant.
# Ask honestly: does the conclusion change? If yes, identify which of the failure modes above is producing that change.
# If the conclusion holds regardless of substitution, the diagnosis has passed this test — it has '''not''' thereby been proven correct, only shown to not be failing '''this particular''' check. See [[Observable Behavior Rule]] and [[Slave Owner Game/False Positives|False Positives]] for the actual content tests still required.
== Relationship to Other Pages ==
This test is referenced by, and load-bearing for, several other pages that already depend on symmetric application to function correctly:
* [[Slave Owner Game/Legitimate Authority]] — the Symmetry Table there is a direct application of this test to one specific case.
* [[Slave Owner Game/False Positives]] — misdiagnosis is far more likely when this test hasn't been run first.
* [[Observable Behavior Rule]] — this test and that rule are closely related but distinct: Observable Behavior Rule governs '''what counts as evidence''' (actions and consequences, not motive); this test governs whether that evidence is being '''applied consistently''' once gathered. A diagnosis can pass Observable Behavior Rule (it's genuinely about actions, not motive) and still fail this test (the same action gets diagnosed differently depending on who did it).
== Development Breadcrumb ==
The five failure modes above are a first-pass taxonomy, not validated against real cases. Future calibration should test whether these are genuinely distinct categories or overlapping descriptions of the same underlying bias, and whether additional failure modes exist beyond these five.
== Calibration Dependencies ==
* [[Slave Owner Game]]
* [[Observable Behavior Rule]]
* [[Slave Owner Game/False Positives]]
'''See the Game. Refuse the Game. Build Better.'''
{{Subpage Hub-Linked Transparency}}
{{Calibration Maintenance}}
{{Calibration Dependent}}
{{Calibration Dependencies Display}}
[[Category:Core Diagnostic]]
<div style="display:none;">
{{Resource
| Title = Slave Owner Game/Diagnostic Inversion Test
| URL = https://www.thesovereigngames.com/wiki/Slave_Owner_Game/Diagnostic_Inversion_Test
| Description = Tests whether this diagnostic is being applied symmetrically — the same evidence standard, regardless of who is being diagnosed. Tribalism is the most visible failure mode, but not the only one.
| Category = Core Diagnostic
}}
</div>
{{Admin Page Status
| categorization = Done
| calibration_review = Self-Assessment
| instrument_grade = Development
| validation = Low
| calibration_rationale = Reframed from a tribalism-specific test to the general symmetry principle it actually is, with tribalism demoted to one example failure mode among five (tribal, familiarity, recency/sympathy, state-dependent, status bias). This closes a real scope-narrowing gap — the underlying test applies cleanly to non-tribal asymmetry (favoring a friend over a stranger, judging harshly while angry) that a tribalism-only framing would have missed. Explicit relationship to Observable Behavior Rule clarified (evidence content vs. evidence application consistency — related but distinct). Not yet round-robin reviewed.
| review_confidence = Moderate
| review_date = {{CURRENTYEAR}}-{{CURRENTMONTH}}-{{CURRENTDAY}}
| reviewed_by = Sovereign
| priority = Supporting
| review_threshold = 30
| has_backlinks = Yes
| outbound_links_valid = Not checked
| in_outline = Yes
| in_category_outline = Yes
| templates_complete = Done
| formatting_standard = Meets standard
| symmetry_check = Not applicable
| self_report_flagged = No
| terminology_consistent = Yes
| standing_check = Self-assessed only
| drift_report_status = None open
| depends_on = Slave Owner Game; Observable Behavior Rule; Slave Owner Game/False Positives
}}
}}

Revision as of 09:49, 17 July 2026

CYCLE Calibration position

Status — Active Development

This page is a conceptual instrument under Permanent Beta. It declares a real calibration position, not a finished product waiting to ship. Checking continues; an edit is only required when evidence demands it. Stage: Seed to Fruit.

Feedback welcome — especially clarity, failure modes, and calibration gaps. Use discussion or Contribute.




Sovereign-Games-OG-Image.jpg

Meta

Slave Owner Game/Diagnostic Inversion Test

Type Core Diagnostic Game
Functional Layer
Application Layer Multi-Layer
Category Core Diagnostic
Version 0.4
Maturity Development
Last Calibration 2026-07-29
Status Permanent Beta
Description Tests whether this diagnostic is being applied symmetrically — the same evidence standard, regardless of who is being diagnosed. Tribalism is the most visible failure mode, but not the only one. Passing this test is necessary but not sufficient for a valid diagnosis.

Core Principles

  • Reality gets final vote
  • See the Game. Refuse the Game. Build Better.
  • Permanent Beta

Navigation

Related


Canonical Question: Would you reach the same conclusion, using the same evidence, if the identity of the person or group being diagnosed were swapped for someone or something you feel differently about — and was that swap itself a fair one?

Diagnostic Inversion Test

The **Diagnostic Inversion Test** is a symmetry check. It asks whether a diagnosis is being applied using the same standard of evidence, regardless of who is being diagnosed. Tribalism (treating your own group more leniently than an out-group for identical behavior) is the most common and visible failure of this symmetry, which is why it is often used as the primary example. However, it is only one instance of a broader problem.

    • Passing this test is necessary, but not sufficient.** A symmetric conclusion is not automatically correct. This test only rules out one specific failure mode (diagnostic asymmetry). The actual mechanism test (see Observable Behavior Rule and False Positives) must still be run and passed.

The Test

Take the specific claim you are about to make and ask:

    • Would I reach the same conclusion, using the same evidence and the same standard, if I substituted someone or something I feel differently about?**

If the conclusion changes based on identity rather than evidence, the diagnosis is not calibrated.

Failure Modes

Diagnostic asymmetry can appear in several forms. Tribalism is the most visible, but the others are equally real and often harder to detect because they do not obviously involve group identity.

  • **Tribal Bias** — Applying a looser standard to one’s own in-group and a stricter standard to an out-group for behavior that is otherwise identical.
  • **Familiarity Bias** — Applying a looser standard to someone known personally and a stricter standard to a stranger, even within the same group or category.
  • **Recency/Sympathy Bias** — Applying a looser standard to someone currently sympathized with (for example, because they were recently harmed), even when the behavior being evaluated is unrelated.
  • **State-Dependent Bias** — Applying a stricter standard when angry, hurt, or personally invested, and a looser standard when calm — producing different verdicts on the same evidence depending on the evaluator’s emotional state.
  • **Status Bias** — Applying a looser standard to someone with greater social, institutional, or professional power (assuming their position signals legitimacy), or conversely, applying a stricter standard specifically because someone holds power.

None of these require an out-group. A person can apply this diagnostic asymmetrically to two members of their own family, political side, or friend group based purely on current favor or emotional state.

Meta-Bias Warning: Gaming the Test

This test can itself be gamed. A person can satisfy its surface requirements while defeating its purpose by choosing a substitution deliberately designed to be easy to distinguish from the real case.

    • Example:** Substituting “a stranger you’ve never met” for “your close friend” is usually a weak test, because differences in available evidence can explain asymmetry without requiring bias.

A genuine substitution holds power differential, context, and available evidence roughly constant, and only varies the identity or relationship being checked.

    • If a person cannot clearly articulate *why* their chosen substitution is fair *before* running the test (not after the result is known), the test has likely been gamed rather than genuinely applied.**

Substitution Fairness Criteria

A valid substitution should hold roughly constant:

  • **Power differential** — Comparable relative power between the parties.
  • **Context** — Comparable relationship type or setting.
  • **Stakes and evidence available** — Comparable amount and quality of evidence.

If a substitution fails these criteria, the test has not actually been run — it has only been performed.

How to Run the Test in Practice

1. State the specific claim and the specific evidence supporting it. 2. Identify who is being diagnosed. 3. Choose a substitution that satisfies the Substitution Fairness Criteria above. Be able to explain *why* it is fair **before** running the test. 4. Substitute while holding the evidence constant. Ask honestly: Does the conclusion change? 5. If the conclusion changes, identify which failure mode is producing the difference. 6. If the conclusion holds under a fair substitution, the diagnosis has passed this test. It has **not** thereby been proven correct — only shown to not be failing this particular check.

Relationship to Other Pages

This test supports several other pages that depend on consistent application:

Development Breadcrumb

The five failure modes listed above are a first-pass taxonomy. Future calibration should test: - Whether these categories are genuinely distinct or overlapping. - Whether additional failure modes exist. - Whether the Substitution Fairness Criteria are sufficient to prevent gaming, or whether gaming can still occur within technically “fair” substitutions.

Calibration Dependencies

See the Game. Refuse the Game. Build Better.



Page Transparency & Calibration

This page does not have its own dedicated Calibration Log or Decision Records — both are tracked at the Main Page level.

This page is under continuous calibration in line with the Permanent Beta principle.



Calibration References

This page is calibrated against the following core standards and reference materials:



Calibration Dependent

Pages that list this page as a load-bearing dependency:

Page Priority Instrument Grade Last Updated Cycle Status Drift Status
Slave Owner Game/How This Framework Can Be Abused Supporting Development 2026-07-20 Current Breadcrumb-Open

If this page is edited substantively, review the list above per the Ripple Review rule — see Calibration Dependencies: Standards and Process#Rule: Core-Priority Changes Trigger Mandatory Ripple Review.


Calibration Dependencies

Pages this page relies on as load-bearing dependencies: Slave Owner Game • Observable Behavior Rule • Slave Owner Game/False Positives • Slave Owner Game/Legitimate Authority If incorrect, edit the `depends_on` field in Admin Page Status — do not edit this section directly, it is auto-generated.



Page Reference

Title Slave Owner Game/Diagnostic Inversion Test
URL https://www.thesovereigngames.com/wiki/Slave_Owner_Game/Diagnostic_Inversion_Test
Description Tests whether this diagnostic is being applied symmetrically — the same evidence standard, regardless of who is being diagnosed. Tribalism is the most visible failure mode, but not the only one. Passing this test is necessary but not sufficient for a valid diagnosis.
Category Core Diagnostic



Retrieved from "https://www.thesovereigngames.com/index.php?title=Slave_Owner_Game/Diagnostic_Inversion_Test&oldid=2844"