THE GOLD STANDARD OF JUSTICE · SPECIALIST STUDY
Treat identification as memory evidence collected under controlled conditions
An eyewitness identification is not a photograph retrieved from storage. Perception is limited at the event, memory changes with time and exposure, and the identification procedure itself can alter what the witness later remembers. A fair process therefore records the witness’s own description before suspect exposure, uses suitable fillers, gives a genuine suspect-may-be-absent instruction, limits administrator influence, captures the first confidence statement in the witness’s own words, preserves the full procedure and tests the identification against independent evidence.
This guide belongs to the Justice subject root, the Gold Standard of Education, and the How X Works hub. It begins with a concrete decision, then tests whether the method still holds when the context changes.
Choose a reading route
Preserve memory before the lineup · chapters 1, 2, 3
Design a fair identification procedure · chapters 4, 5
Work and document the case · chapters 6, 7, 8
Corroborate, disclose and review · chapters 9, 10, 11
Learn and transfer without overclaiming · chapters 12, 13
The whole problem at a glance
| Question | Evidence needed | Unsafe shortcut |
|---|---|---|
| What did the witness perceive? | Contemporaneous conditions and first description | Later certainty replaces first account |
| Was the lineup fair? | Target, fillers, instructions and administrator record | Suspect stands out but selection counts |
| What was the first response? | Exact choice, latency and confidence words | Courtroom confidence becomes baseline |
| What supports it? | Independent corroboration and disclosure | Identification proves the whole case |
Later confidence can be sincere while still reflecting feedback or repeated exposure.
CHAPTER 1 OF 13 · Preserve memory before the lineup
1. Treat memory as reconstructive evidence
Separate perception at the event, retention over time and retrieval during questioning. Record conditions that affected each stage instead of treating memory as a permanent recording. The working unit here is one witness, one witnessed event, one documented identification procedure and one evidential claim at a stated time. State whose sample, memory, decision or record is being examined, what time boundary applies and which claim is still outside the evidence.
Worked example: A witness had a brief, obstructed view under poor lighting. The first report preserves those limits rather than replacing them with a later detailed narrative. Put the starting information, decision, later observation and consequence on separate lines. That chronology stops a coherent story from replacing a reproducible account.
Evidence to retain: duration; distance; lighting; obstruction; stress; attention; distinctive features; delay; first report; interviewer prompts Attach dates, versions, provenance, custody and the decision use. Record unavailable evidence explicitly; absence is not a negative result and cannot safely be turned into reassurance.
Boundary and misconception: A sincere witness can be mistaken, and a detailed later account is not automatically more accurate than a sparse first one. Test a credible competing explanation and name what evidence would change the conclusion. The pass condition is a bounded, reviewable claim, not a promise that the method removes uncertainty.
Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.
Changed-context transfer: Improve lighting but add divided attention and explain why reliability cannot be reduced to one factor. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.
Chapter 1 check · what should survive an audit
- Name the decision, unit, time boundary and responsible authority.
- Retain the supporting evidence and the strongest rejected explanation.
- Record the next action, review date, stop rule and reopening condition.
CHAPTER 2 OF 13 · Preserve memory before the lineup
2. Capture the witness’s description before suspect exposure
Use open prompts and record the witness’s own words, uncertainty and omitted features before showing photographs, composites or media. The working unit here is one witness, one witnessed event, one documented identification procedure and one evidential claim at a stated time. State whose sample, memory, decision or record is being examined, what time boundary applies and which claim is still outside the evidence.
Worked example: The witness describes height, clothing and one facial feature without naming ethnicity. The record preserves that account before investigators disclose a suspect. Put the starting information, decision, later observation and consequence on separate lines. That chronology stops a coherent story from replacing a reproducible account.
Evidence to retain: audio or video; exact prompt; verbatim description; confidence and uncertainty; time; prior discussions; images seen; interpreter Attach dates, versions, provenance, custody and the decision use. Record unavailable evidence explicitly; absence is not a negative result and cannot safely be turned into reassurance.
Boundary and misconception: A description need not perfectly match to be useful, but investigators should not retrofit it after learning who the suspect is. Test a credible competing explanation and name what evidence would change the conclusion. The pass condition is a bounded, reviewable claim, not a promise that the method removes uncertainty.
Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.
Changed-context transfer: Add an interpreter and specify how source words, translation and uncertainty are preserved. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.
Chapter 2 check · what should survive an audit
- Name the decision, unit, time boundary and responsible authority.
- Retain the supporting evidence and the strongest rejected explanation.
- Record the next action, review date, stop rule and reopening condition.
CHAPTER 3 OF 13 · Preserve memory before the lineup
3. Build an exposure timeline before designing the lineup
Identify news, social media, police images, co-witness conversations and earlier procedures that may make a face familiar for reasons unrelated to the event. The working unit here is one witness, one witnessed event, one documented identification procedure and one evidential claim at a stated time. State whose sample, memory, decision or record is being examined, what time boundary applies and which claim is still outside the evidence.
Worked example: The witness saw the suspect’s arrest photo online. The lineup proceeds only with the exposure documented and the resulting limitation disclosed. Put the starting information, decision, later observation and consequence on separate lines. That chronology stops a coherent story from replacing a reproducible account.
Evidence to retain: device or media exposure; dates; image content; co-witness contact; prior show-ups or lineups; investigator disclosures; witness account Attach dates, versions, provenance, custody and the decision use. Record unavailable evidence explicitly; absence is not a negative result and cannot safely be turned into reassurance.
Boundary and misconception: Familiarity is not source memory; recognising a face does not prove where the witness encountered it. Test a credible competing explanation and name what evidence would change the conclusion. The pass condition is a bounded, reviewable claim, not a promise that the method removes uncertainty.
Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.
Changed-context transfer: Move the online exposure after the first lineup and distinguish contamination of later confidence from the first selection. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.
Chapter 3 check · what should survive an audit
- Name the decision, unit, time boundary and responsible authority.
- Retain the supporting evidence and the strongest rejected explanation.
- Record the next action, review date, stop rule and reopening condition.
CHAPTER 4 OF 13 · Design a fair identification procedure
4. Choose the lineup target and avoid multiple-suspect pressure
Use one suspect per procedure and fillers selected from the witness description, not merely people who resemble the suspect. The working unit here is one witness, one witnessed event, one documented identification procedure and one evidential claim at a stated time. State whose sample, memory, decision or record is being examined, what time boundary applies and which claim is still outside the evidence.
Worked example: Investigators have two possible suspects. Separate lineups prevent the witness from treating the presence of both as a cue that one must be the offender. Put the starting information, decision, later observation and consequence on separate lines. That chronology stops a coherent story from replacing a reproducible account.
Evidence to retain: suspect basis; witness description; lineup composition; prior procedures; order and spacing; administrator plan Attach dates, versions, provenance, custody and the decision use. Record unavailable evidence explicitly; absence is not a negative result and cannot safely be turned into reassurance.
Boundary and misconception: A lineup should test recognition, not ask which police-selected person looks most plausible. Test a credible competing explanation and name what evidence would change the conclusion. The pass condition is a bounded, reviewable claim, not a promise that the method removes uncertainty.
Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.
Changed-context transfer: Add a second witness with a different description and build a separate lineup rather than recycling the first. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.
Chapter 4 check · what should survive an audit
- Name the decision, unit, time boundary and responsible authority.
- Retain the supporting evidence and the strongest rejected explanation.
- Record the next action, review date, stop rule and reopening condition.
CHAPTER 5 OF 13 · Design a fair identification procedure
5. Select fillers that make the suspect not stand out
Choose enough fillers who fit the description and do not share only a feature unique to the suspect. Review the array as a whole and document replacements. The working unit here is one witness, one witnessed event, one documented identification procedure and one evidential claim at a stated time. State whose sample, memory, decision or record is being examined, what time boundary applies and which claim is still outside the evidence.
Worked example: The suspect has a visible facial mark mentioned by the witness. Comparable placement is handled across images so the mark does not reveal the police choice. Put the starting information, decision, later observation and consequence on separate lines. That chronology stops a coherent story from replacing a reproducible account.
Evidence to retain: first description; filler source; image age and quality; distinctive-feature handling; similarity review; final array; rejected fillers Attach dates, versions, provenance, custody and the decision use. Record unavailable evidence explicitly; absence is not a negative result and cannot safely be turned into reassurance.
Boundary and misconception: Fillers need not be clones, but the suspect should not be the only person who matches the witness’s description or presentation style. Test a credible competing explanation and name what evidence would change the conclusion. The pass condition is a bounded, reviewable claim, not a promise that the method removes uncertainty.
Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.
Changed-context transfer: Change the witness description from general to highly distinctive and explain the increased filler challenge. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.
Chapter 5 check · what should survive an audit
- Name the decision, unit, time boundary and responsible authority.
- Retain the supporting evidence and the strongest rejected explanation.
- Record the next action, review date, stop rule and reopening condition.
Worked lineup review
| Observation | Risk | Next evidence | Protection |
|---|---|---|---|
| Witness saw social-media image | Pre-procedure exposure | Exposure timeline and first description | Do not present lineup as uncontaminated |
| Suspect is only person with feature | Relative judgment | Rebuild fillers or document flaw | No suggestion that one must be chosen |
| Immediate confidence is tentative | Normal uncertainty | Record exact words and video | Do not inflate after confirmation |
| Two witnesses discussed case | Memory conformity | Separate accounts and contact history | Treat independence as unresolved |
A safeguard protects the evidence only when its use and exceptions are recorded.
CHAPTER 6 OF 13 · Work and document the case
6. Use blind or blinded administration and neutral instructions
Prevent the administrator from knowing or signalling the suspect where feasible, and tell the witness the offender may or may not be present and the investigation will continue either way. The working unit here is one witness, one witnessed event, one documented identification procedure and one evidential claim at a stated time. State whose sample, memory, decision or record is being examined, what time boundary applies and which claim is still outside the evidence.
Worked example: A computer-controlled procedure delivers standard instructions and records the administrator’s limited role. The witness is not told that police caught the right person. Put the starting information, decision, later observation and consequence on separate lines. That chronology stops a coherent story from replacing a reproducible account.
Evidence to retain: administrator knowledge; blinding method; instruction script; witness confirmation; deviations; technical logs; observer roles Attach dates, versions, provenance, custody and the decision use. Record unavailable evidence explicitly; absence is not a negative result and cannot safely be turned into reassurance.
Boundary and misconception: Even subtle tone, repetition or body language can influence a witness; good intentions do not substitute for control. Test a credible competing explanation and name what evidence would change the conclusion. The pass condition is a bounded, reviewable claim, not a promise that the method removes uncertainty.
Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.
Changed-context transfer: Run the procedure in a small agency without a second officer and design a documented blinded alternative. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.
Chapter 6 check · what should survive an audit
- Name the decision, unit, time boundary and responsible authority.
- Retain the supporting evidence and the strongest rejected explanation.
- Record the next action, review date, stop rule and reopening condition.
CHAPTER 7 OF 13 · Work and document the case
7. Choose sequential or simultaneous presentation under a stated policy
Follow the jurisdiction’s validated procedure and training, making order, randomisation, stopping rules and prior exposure visible. Do not claim one format is universally superior in every setting. The working unit here is one witness, one witnessed event, one documented identification procedure and one evidential claim at a stated time. State whose sample, memory, decision or record is being examined, what time boundary applies and which claim is still outside the evidence.
Worked example: A sequential lineup uses a pre-generated random order and records each decision before the next image. The policy and deviation log travel with the evidence. Put the starting information, decision, later observation and consequence on separate lines. That chronology stops a coherent story from replacing a reproducible account.
Evidence to retain: policy version; presentation mode; randomisation; viewing controls; decision after each item; stopping rule; repeat-view requests Attach dates, versions, provenance, custody and the decision use. Record unavailable evidence explicitly; absence is not a negative result and cannot safely be turned into reassurance.
Boundary and misconception: Changing format without controlling instructions, fillers and feedback can make outcome differences impossible to interpret. Test a credible competing explanation and name what evidence would change the conclusion. The pass condition is a bounded, reviewable claim, not a promise that the method removes uncertainty.
Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.
Changed-context transfer: Keep the same images but switch presentation mode and identify which comparison is now confounded. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.
Chapter 7 check · what should survive an audit
- Name the decision, unit, time boundary and responsible authority.
- Retain the supporting evidence and the strongest rejected explanation.
- Record the next action, review date, stop rule and reopening condition.
CHAPTER 8 OF 13 · Work and document the case
8. Work a complete identification case
Join event conditions, first description, exposure timeline, fair lineup, first selection and confidence, later feedback and corroboration in one chronology. The working unit here is one witness, one witnessed event, one documented identification procedure and one evidential claim at a stated time. State whose sample, memory, decision or record is being examined, what time boundary applies and which claim is still outside the evidence.
Worked example: A witness with a moderate viewing opportunity gives a specific description, sees no suspect image before a blinded lineup, selects one person and says “I think that is him, but I am not certain.” The exact words remain the baseline. Put the starting information, decision, later observation and consequence on separate lines. That chronology stops a coherent story from replacing a reproducible account.
Evidence to retain: event record; first account; exposure log; lineup and fillers; instructions; video; selection; latency; confidence; feedback; corroboration Attach dates, versions, provenance, custody and the decision use. Record unavailable evidence explicitly; absence is not a negative result and cannot safely be turned into reassurance.
Boundary and misconception: The fictional case shows procedure, not guilt. Identification addresses identity evidence and does not establish every element of an offence. Test a credible competing explanation and name what evidence would change the conclusion. The pass condition is a bounded, reviewable claim, not a promise that the method removes uncertainty.
Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.
Changed-context transfer: Add pre-lineup social-media exposure and rewrite the evidential claim without erasing the original response. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.
Chapter 8 check · what should survive an audit
- Name the decision, unit, time boundary and responsible authority.
- Retain the supporting evidence and the strongest rejected explanation.
- Record the next action, review date, stop rule and reopening condition.
CHAPTER 9 OF 13 · Corroborate, disclose and review
9. Record the first response and confidence in the witness’s own words
Capture the choice, non-choice or rejection, decision time, certainty words and behaviour before any confirming feedback. Preserve recordings rather than reducing them to a checked box. The working unit here is one witness, one witnessed event, one documented identification procedure and one evidential claim at a stated time. State whose sample, memory, decision or record is being examined, what time boundary applies and which claim is still outside the evidence.
Worked example: Immediately after selection the witness says “about sixty percent.” The officer records the words without translating them into “positive identification.” Put the starting information, decision, later observation and consequence on separate lines. That chronology stops a coherent story from replacing a reproducible account.
Evidence to retain: complete video; exact selection; confidence statement; latency; requests to view again; administrator response; time stamp Attach dates, versions, provenance, custody and the decision use. Record unavailable evidence explicitly; absence is not a negative result and cannot safely be turned into reassurance.
Boundary and misconception: Later confidence can grow after feedback, repetition or learning case facts; it should not replace the first uncontaminated statement. Test a credible competing explanation and name what evidence would change the conclusion. The pass condition is a bounded, reviewable claim, not a promise that the method removes uncertainty.
Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.
Changed-context transfer: Delay the confidence question until after an officer says “good job” and explain why the baseline is lost. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.
Chapter 9 check · what should survive an audit
- Name the decision, unit, time boundary and responsible authority.
- Retain the supporting evidence and the strongest rejected explanation.
- Record the next action, review date, stop rule and reopening condition.
Identification evidence gate
| Gate | Pass evidence | Remain open when |
|---|---|---|
| Perception | Viewing conditions and first account preserved | Exposure or description missing |
| Procedure | Fair fillers, instructions and blind control | Administrator influence or standout unresolved |
| Response | First choice and confidence recorded | Only later certainty available |
| Corroboration | Independent evidence and disclosure complete | Identification is carrying every proposition |
The gate organises review; admissibility and weight remain matters for the governing law and decision-maker.
CHAPTER 10 OF 13 · Corroborate, disclose and review
10. Prevent and disclose feedback and post-identification contamination
Avoid confirming comments, repeated exposure, suggestive briefings and co-witness discussion. When exposure occurs, document it instead of pretending neutrality survived. The working unit here is one witness, one witnessed event, one documented identification procedure and one evidential claim at a stated time. State whose sample, memory, decision or record is being examined, what time boundary applies and which claim is still outside the evidence.
Worked example: After selection, a witness learns that another witness chose the same person. The later interview records this exposure and does not treat increased certainty as independent confirmation. Put the starting information, decision, later observation and consequence on separate lines. That chronology stops a coherent story from replacing a reproducible account.
Evidence to retain: all contacts; exact feedback; media; repeated images; co-witness communication; preparation sessions; later statements; confidence changes Attach dates, versions, provenance, custody and the decision use. Record unavailable evidence explicitly; absence is not a negative result and cannot safely be turned into reassurance.
Boundary and misconception: A witness may honestly remember increased certainty without remembering the source of that increase. Test a credible competing explanation and name what evidence would change the conclusion. The pass condition is a bounded, reviewable claim, not a promise that the method removes uncertainty.
Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.
Changed-context transfer: Move the case into repeated pretrial hearings and create a record of every suspect exposure. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.
Chapter 10 check · what should survive an audit
- Name the decision, unit, time boundary and responsible authority.
- Retain the supporting evidence and the strongest rejected explanation.
- Record the next action, review date, stop rule and reopening condition.
CHAPTER 11 OF 13 · Corroborate, disclose and review
11. Protect independence across multiple witnesses
Separate interviews and procedures, prevent shared lineup orders where policy requires variation, and trace information flow between witnesses. The working unit here is one witness, one witnessed event, one documented identification procedure and one evidential claim at a stated time. State whose sample, memory, decision or record is being examined, what time boundary applies and which claim is still outside the evidence.
Worked example: Two witnesses arrive together but are interviewed separately, receive different randomised arrays and are asked about any conversation before the procedure. Put the starting information, decision, later observation and consequence on separate lines. That chronology stops a coherent story from replacing a reproducible account.
Evidence to retain: arrival and separation; contact history; independent first accounts; array order; administrator; procedure times; later contact Attach dates, versions, provenance, custody and the decision use. Record unavailable evidence explicitly; absence is not a negative result and cannot safely be turned into reassurance.
Boundary and misconception: Two matching identifications are not independent corroboration if both were shaped by the same image, rumour or conversation. Test a credible competing explanation and name what evidence would change the conclusion. The pass condition is a bounded, reviewable claim, not a promise that the method removes uncertainty.
Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.
Changed-context transfer: Discover a group chat after both lineups and reassess which evidence remains independent. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.
Chapter 11 check · what should survive an audit
- Name the decision, unit, time boundary and responsible authority.
- Retain the supporting evidence and the strongest rejected explanation.
- Record the next action, review date, stop rule and reopening condition.
CHAPTER 12 OF 13 · Learn and transfer without overclaiming
12. Use show-ups only within their narrow, documented rationale
When an immediate one-person identification is permitted, record necessity, timing, instructions, presentation and alternatives, and disclose its suggestive nature. The working unit here is one witness, one witnessed event, one documented identification procedure and one evidential claim at a stated time. State whose sample, memory, decision or record is being examined, what time boundary applies and which claim is still outside the evidence.
Worked example: Police detain a person near the scene minutes after an offence. The witness is told the person may not be involved, and the full procedure is recorded under local policy. Put the starting information, decision, later observation and consequence on separate lines. That chronology stops a coherent story from replacing a reproducible account.
Evidence to retain: legal basis; urgency; time and distance; description before show-up; instructions; custody presentation; exact response; available alternatives Attach dates, versions, provenance, custody and the decision use. Record unavailable evidence explicitly; absence is not a negative result and cannot safely be turned into reassurance.
Boundary and misconception: Speed can preserve memory but the one-person structure is inherently suggestive; a show-up should not be normalised as an easy substitute for a lineup. Test a credible competing explanation and name what evidence would change the conclusion. The pass condition is a bounded, reviewable claim, not a promise that the method removes uncertainty.
Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.
Changed-context transfer: Extend the delay and remove the emergency; identify when a lineup becomes the safer available procedure. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.
Chapter 12 check · what should survive an audit
- Name the decision, unit, time boundary and responsible authority.
- Retain the supporting evidence and the strongest rejected explanation.
- Record the next action, review date, stop rule and reopening condition.
CHAPTER 13 OF 13 · Learn and transfer without overclaiming
13. Corroborate, disclose and review without circularity
Test identification against independent physical, digital, alibi and timeline evidence. Disclose procedure materials, exposure and deviations so the opposing party and court can examine reliability. The working unit here is one witness, one witnessed event, one documented identification procedure and one evidential claim at a stated time. State whose sample, memory, decision or record is being examined, what time boundary applies and which claim is still outside the evidence.
Worked example: Location data conflicts with the identification. Investigators preserve both and test timing and device attribution rather than using the identification to dismiss the conflict. Put the starting information, decision, later observation and consequence on separate lines. That chronology stops a coherent story from replacing a reproducible account.
Evidence to retain: full procedure video; filler array; policy; deviations; first description; exposure log; confidence changes; independent evidence; disclosure record Attach dates, versions, provenance, custody and the decision use. Record unavailable evidence explicitly; absence is not a negative result and cannot safely be turned into reassurance.
Boundary and misconception: Evidence is not independent if it was generated from the identification and then used to validate that same identification. Test a credible competing explanation and name what evidence would change the conclusion. The pass condition is a bounded, reviewable claim, not a promise that the method removes uncertainty.
Evidence discipline: record the observation, version or time boundary that supports this chapter’s conclusion. Preserve the rejected alternative as well as the selected action. That makes a later review able to distinguish a changed world from an undocumented change of mind.
Changed-context transfer: Remove the digital evidence and identify which propositions remain supported only by memory evidence. The method transfers only if the decision question and measurement meaning survive the change. If either changes, rebuild the comparison rather than copying the earlier threshold or weight.
Chapter 13 check · what should survive an audit
- Name the decision, unit, time boundary and responsible authority.
- Retain the supporting evidence and the strongest rejected explanation.
- Record the next action, review date, stop rule and reopening condition.
Verification lab: work the decision twice
Design a blinded lineup from a fictional first description. Document filler selection, distinctive-feature treatment, instructions, randomisation, the exact response and the first confidence statement.
Repeat the procedure after the witness sees a suspect image online. Preserve the same lineup but rewrite the reliability claim and disclosure package.
Audit two witnesses for independence. Draw every information path among media, investigators and co-witnesses, then identify which apparent corroboration is circular.
Final review: distinguish recognition, source memory, confidence, procedure fairness and independent corroboration. If one courtroom identification is doing all five jobs, the reasoning is overclaimed.
Limits and misconceptions
Memory is not a recording
Perception, delay, retrieval and later exposure all shape the evidence.
A fair lineup tests recognition
Fillers, neutral instructions and blind administration reduce relative judgment and influence.
First confidence is time-sensitive
Capture it before feedback; later sincerity cannot reconstruct a lost baseline.
Identification still needs corroboration
Independent evidence and full disclosure keep one memory claim from deciding every proposition.
Questions readers ask
Why tell the witness the suspect may be absent?
- It reduces pressure to choose the closest-looking person when the offender is not in the lineup.
What makes a filler fair?
- A filler should fit the witness’s description so the suspect does not stand out for a distinctive feature or presentation.
Why use a blind administrator?
- An administrator who does not know the suspect cannot intentionally or unintentionally cue the witness.
Does high confidence prove accuracy?
- No. The timing and collection conditions matter, especially whether confidence was recorded before feedback.
Can witnesses discuss the event?
- Such discussion can contaminate memory and reduce independence; contacts should be prevented where possible and documented when they occur.
Is a show-up the same as a lineup?
- No. It presents one person and is inherently suggestive, so its necessity, timing and safeguards require careful review.
Practise, verify and continue
Create an identification evidence map with viewing conditions, first description, exposure timeline, lineup target, fillers, instructions, blinding, presentation mode, exact response, first confidence, feedback, witness independence and corroboration.
Apply the method to voice identification. Preserve first description, exposure control, fair alternatives, blind administration, exact confidence and corroboration, while rebuilding the perception and comparison limits for audio.
Practical evidence check
Before implementation, write a one-page evidence map with the decision, unit of analysis, version or time boundary, inputs, outcomes, uncertainty, authority and stop rule. Attach one worked case that passes and one that fails. A reviewer should be able to reproduce the reasoning without relying on the memory of the person who built the system.
Run one changed-context rehearsal before approval. Alter a mechanism rather than a cosmetic detail, then identify which calculation, safeguard or governance gate must change. Record any conclusion that does not transfer. This short rehearsal is often the fastest way to expose a hidden assumption before it becomes an operational dependency.
Related eduKate reading
Evidentiary justice, reliability and contestability
Procedural justice, voice, neutrality and trustworthiness
Wrongful-conviction review, evidence preservation and remedy
Authoritative sources and scope
National Institute of Justice overview of eyewitness identification
National Institute of Justice guidance on more reliable police lineups
National Academies report: Identifying the Culprit
These sources support general safeguards and evidence practices. They do not define the law of every jurisdiction, determine admissibility or decide an individual case. All worked cases are fictional. Actual investigations, interviews, identification procedures, disclosure and remedies require the governing law, competent counsel and the responsible authority or court.