Preference Reversal Probe¶
Experimental probe — instantiates Revealed Preference Validation Against Indifference Curves
Deliberately re-presents the same options in an altered frame or order to see whether the chooser's ranking flips — separating a stable preference from a framing artifact.
Passive data can show a chooser prefers A to B, but it cannot tell you whether that preference would survive a different presentation. Preference Reversal Probe finds out by intervening: it re-offers the same underlying options in a changed frame — reordered, relabeled, differently anchored, or as a slightly altered counterfactual bundle — and watches whether the choice flips. Its defining move is the deliberate manipulation of presentation while holding the real options fixed, so that any reversal is evidence the preference was constructed by the frame rather than held independently of it. Where the archetype's other consistency checks read contradictions off existing records, this one creates the comparison it needs by running a controlled variation.
Example¶
A SaaS company suspects its "most popular" tier choice is an artifact of how the pricing page is arranged. Rather than only auditing the existing page, it runs a probe: the same three plans are shown to matched cohorts of new visitors, varying only the presentation — in one arm the mid tier is anchored beside a much pricier "enterprise" plan; in another it is shown next to a bare-bones cheap plan; in a third the order is reversed. The underlying plans and prices never change.
The mid tier's selection share swings sharply between arms — chosen far more when flanked by an expensive anchor than when it sits at the top of the list. That swing is a preference reversal: the same person, facing the same real options, ranks them differently depending on the frame. The probe's conclusion is not "customers prefer the mid tier" but "the observed preference for the mid tier is partly manufactured by anchoring, and is not a stable trade-off the company can safely build revenue projections on." Only an intervention that varied the frame while fixing the options could establish that.
How it works¶
- Fix the options, vary the frame. Hold the real, underlying alternatives constant and change only presentation — order, anchor, label, default salience — across matched arms or repeated occasions.
- Introduce counterfactual variants. Where useful, add a decoy or a reframed version of an option to test whether it drags the ranking.
- Detect the flip. Compare rankings across frames and register a reversal when the same chooser (or matched group) orders the same options differently.
- Attribute the swing to the frame. Because the options were held fixed, a reversal is charged to presentation, not to any change in what was on offer.
Tuning parameters¶
- Frame manipulation strength — how aggressively presentation is varied. Strong manipulations reveal fragility clearly but may create a reversal no natural setting would.
- Counterfactual distance — how far the probe's altered options sit from the real ones; near variants are realistic, far ones test robustness but strain external validity.
- Within- vs between-subject design — re-offering to the same chooser detects individual reversals but risks memory and consistency pressure; matched groups avoid that but add noise.
- Reversal threshold — how large a ranking swing counts as a genuine reversal versus sampling noise.
- Number of frames — more frames map the preference's sensitivity surface but multiply cost and comparisons.
When it helps, and when it misleads¶
Its strength is causal: because it manipulates the frame while holding options fixed, a reversal is strong evidence that a "preference" is constructed rather than stable — something no amount of passive observation can establish, since in the wild frame and choice are confounded.[n1] It is the archetype's tool for stress-testing whether an inferred preference is real enough to build on.
Its failure mode is that the probe can manufacture the very instability it measures: an artificial, extreme, or repeated presentation may induce reversals that no natural choice setting ever would, over-diagnosing fragility. It also generalizes only as far as its frames resemble reality. The classic misuse is running a contrived manipulation and concluding the underlying preference is worthless, when the real lesson is only that this frame moves it. The guarding discipline is to keep manipulations within the range of frames the chooser actually encounters and to report the reversal as sensitivity to a specific, named frame rather than as global incoherence.
How it implements the components¶
Preference Reversal Probe realizes the archetype's interventional-robustness test:
counterfactual_option_probe— its core act: presenting reframed or counterfactual option sets to see whether the choice changes.consistency_test_suite— it applies the reversal test across frames, registering a flip as a framing-induced inconsistency.
It does not passively catalog the real choice environment's defaults and layout (choice_architecture_audit, anomaly_explanation_boundary) — that is Choice Architecture Confound Audit, which inspects the frame that exists rather than imposing new ones.
Related¶
- Instantiates: Revealed Preference Validation Against Indifference Curves — it stress-tests whether an inferred preference survives a change of frame.
- Consumes: Choice Bundle Normalization supplies the option representation the probe re-presents in altered frames.
- Sibling mechanisms: Budget Set Reconstruction · Choice Bundle Normalization · Marginal Substitution Estimator · Indifference Region Visualization · Dominance Violation Scan · Revealed Preference Consistency Matrix · Choice Architecture Confound Audit · Stated vs Revealed Gap Report · Ethical Preference Inference Review
Editorial Notes¶
Form Classification¶
Form family: Experiment, Test & Rehearsal
Rationale: Preference Reversal Probe operates as an active test, trial, simulation, drill, or rehearsal that generates evidence through a deliberate attempt or perturbation because it deliberately re-presents the same options in an altered frame or order to see whether the chooser's ranking flips — separating a stable preference from a framing artifact.
Independent corroboration: The frozen evidence defines Preference Reversal Probe as 'Deliberately re-presents the same options in an altered frame or order to see whether the chooser's ranking flips — separating a stable preference from a framing artifact', so its operative form is Experiment, Test & Rehearsal.
Review outcome: Independent reviewer agreement; high confidence.
Origin Attribution¶
Primary origin: Behavioral Economics
Origin pattern: Cross-disciplinary synthesis
Present-day reach: Multi-domain
Rationale: Preference reversals under changed framing or elicitation are a canonical behavioral-economics finding.
Related originating lineages:
- Psychology — Psychology independently developed experimental framing and order-effect paradigms.
Review resolution: Both blind reviewers agree that behavioral economics is the primary origin. Reconciliation resolves origin mode disagreement. Formative alternate lineages are retained as psychology; later breadth of use is recorded separately as domain_reach=multi_domain, while origin_mode=cross_disciplinary_synthesis describes the relationship among origin lineages.
Review outcome: Reconciled after independent review; high confidence.
Notes¶
[n1] A preference reversal is a systematic flip in which option a person favors depending on how the choice is elicited or framed; Sarah Lichtenstein and Paul Slovic documented the phenomenon experimentally in the early 1970s, and it remains the canonical demonstration that revealed rankings can be constructed by presentation rather than read off stable underlying values. ↩