Big question: What must an explanation let this person do?
Research lock: 2026-08-28

Why this week matters

Explanation is not a single output. A developer debugging a model, a regulator evaluating compliance, a clinician calibrating reliance, and an applicant contesting a denial need different evidence. A plausible heatmap can be unfaithful; a faithful global model description can be useless to an affected person. Start with audience and action, then select and test the explanation.

Deeper teaching spine

1. Separate four ideas

  • Transparency: what exists, who made it, what data/process/version was used, and what happened.
  • Interpretability: how a person can understand the model or relationship itself.
  • Explainability: reasons or evidence accompanying an output or process.
  • Contestability: information and procedure that let someone challenge and change a consequential result.

These support one another but are not synonyms.

2. Use NIST's four principles as a test

NIST proposes explanation, meaningfulness to the intended consumer, explanation accuracy, and knowledge limits. Apply all four to every technique. An explanation that looks intuitive but does not reflect the model fails accuracy; a mathematically faithful explanation that the recipient cannot act on fails meaningfulness.

3. Prefer interpretable models where stakes justify it

Rudin argues that high-stakes decisions should often use inherently interpretable models rather than explaining black boxes after the fact. Treat “accuracy requires complexity” as an empirical claim requiring comparison, not a law. Students should demand a benchmark between an interpretable candidate and a black box under the actual task and constraints.

4. Understand what local attributions say

LIME approximates local behaviour around an input; SHAP assigns additive feature contributions relative to a background expectation. Results depend on perturbations, feature dependence, background data, model access, and implementation. They are explanations of a model prediction under assumptions—not causal reasons in the world.

5. Test explanations adversarially

Adebayo et al. showed that some saliency methods can remain visually similar after model parameters or labels are randomized. Slack et al. demonstrated adversarial models that can fool perturbation-based LIME and SHAP explanations. Explanation systems therefore need validation: sensitivity, stability, faithfulness, robustness, human comprehension, and decision utility.

6. Counterfactuals need feasibility and recourse

“If income were $5,000 higher, approval would change” may be understandable but not actionable, causal, stable, or fair. A recourse design should consider immutable or protected attributes, feature dependencies, cost, time, uncertainty, repeated use, and whether the institution can change its own policy rather than shifting the burden to the person.

7. Mechanistic interpretability changes the object

Anthropic's circuit-tracing work uses attribution graphs to propose computational structures inside language models. This may generate testable hypotheses about model behaviour. It does not provide a complete causal account, guarantee coverage, or directly satisfy an affected person's need for recourse. Compare it with output-level, procedural, and institutional explanations.

Case-study dossier

Wolf versus husky: a useful local clue

The classic LIME example suggested a classifier relied on snowy backgrounds. Ask what this establishes: a local surrogate found evidence consistent with a spurious feature. Ask what it does not: global behaviour, causality in the world, fairness, or field reliability. Design a follow-up test set that manipulates snow and animal independently.

Saliency sanity checks: visual appeal is not faithfulness

Show original and randomized-model saliency maps without labels. If students cannot distinguish them, ask whether the explanation reflects learned parameters. This is a vivid example of evaluating the explainer as a model in its own right.

Credit denial: explanation versus recourse

The U.S. CFPB has stated that creditors using complex algorithms still must provide specific, accurate adverse-action reasons. Students compare a feature-attribution chart, a reason code, and a feasible counterfactual. Which satisfies legal notice, understanding, and ability to contest?

Explanation selection matrix

AudiencePurposeCandidate artifactCore validation
DeveloperDebug spurious behaviourcounterfactual test suite, slice errors, local/global attributionchanges when model/data changes
Regulator/auditorAssess process and compliancedocumentation, logs, subgroup results, decision rulestraceable and reproducible
Domain professionalCalibrate relianceconfidence, similar cases, limits, actionable factorsimproves decisions without over-trust
Affected personUnderstand and contestspecific reasons, evidence, feasible recourse, appealunderstandable and can change outcome
ExecutiveAccept residual risksystem boundary, scenarios, control evidencecalibrated aggregation, no hidden tails

Visual evidence plan

VisualCapture targetTeaching useGuardrail
NIST four principlesNISTIR 8312Reveal one principle at a time and test an explanation.Principles are a framework, not a certification.
LIME wolf/huskyLIME paperCrop the example and design a falsification test.A local surrogate is not a causal proof.
SHAP attribution diagramNeurIPS paperShow base value → feature contributions → prediction.State the background and feature-dependence assumptions.
Saliency randomizationNeurIPS 2018 paper PDFCompare explanations before/after parameter randomization.A failed necessary test does not rank every alternative.
Circuit tracingAnthropic methods paperContrast internal computational hypotheses with user-facing explanation.Treat graphs as partial, method-dependent evidence.
Adverse-action requirementCFPB Circular 2022-03Crop the requirement for specific and accurate reasons.Explain U.S. credit-law scope.

Reading and citation ledger

  1. NIST, Four Principles of Explainable Artificial Intelligence.
  2. Rudin, “Stop explaining black box machine learning models for high stakes decisions”.
  3. Ribeiro, Singh & Guestrin, LIME.
  4. Lundberg & Lee, SHAP.
  5. Wachter, Mittelstadt & Russell, counterfactual explanations.
  6. Adebayo et al., “Sanity Checks for Saliency Maps”.
  7. Slack et al., “Fooling LIME and SHAP”.
  8. Anthropic, Circuit Tracing methods.
  9. CFPB Circular 2022-03.

Watch list

  • Never call an attribution a cause without a causal design.
  • Evaluate the explanation method on the actual model and use case.
  • Ask what action the explanation enables and whether the institution offers real recourse.