The Innocence Project has pushed a fairly consistent package of reforms aimed at making eyewitness identifications more accurate and less suggestive, both in police procedures and in court.

Core Lineup Procedure Reforms

  • Double-blind administration
    Lineups should be run “double-blind,” meaning the officer administering the lineup does not know who the suspect is, and the witness also does not know who (if anyone) in the lineup is the suspect.

This is meant to prevent even subtle cues—tone of voice, body language, where a photo is placed—from nudging the witness toward a particular person.

  • Sequential (one-at-a-time) presentation
    Instead of showing all photos or people at once (simultaneous), witnesses should see them one at a time and make a yes/no decision on each before seeing the next.

This reduces “relative judgment,” where the witness picks the person who looks most like their memory compared to others, even if none is actually the perpetrator.

  • Clear pre-lineup instructions to the witness
    The witness must be told that:

    • The perpetrator may or may not be present in the lineup.
    • The administrator does not know who the suspect is.
    • The investigation will continue whether or not the witness makes an identification.

These instructions are designed to relieve pressure to “pick someone” and reduce false IDs.

Lineup Composition and Documentation

  • Fair, non-suggestive lineup composition
    • Only one suspect per lineup.
* Fillers (non-suspects) should be chosen based on the witness’s **original description** , not based on how much they look like the suspect.

* The suspect must not stand out (for example, not the only person of a certain race, age range, or facial hair; no distinctive clothing or background).

* Witnesses should not see multiple lineups containing the same suspect, which can artificially boost a sense of familiarity.

  • Pre-lineup interview and evidence-based suspicion
    Before any lineup is conducted, officers should:

    • Take a careful statement of the witness’s description and level of certainty.
    • Ensure there is genuine, evidence-based reason to treat a person as a suspect before putting them in a lineup.

This helps anchor fillers to an independent description and avoid using lineups as “fishing expeditions.”

  • Immediate confidence statement
    Right after the identification, the officer should document—in the witness’s own words—how confident they are, before any feedback, discussion, or later information can inflate that confidence.
  • Full recording of the procedure
    The Innocence Project and allied groups urge video or at least audio recording of the entire identification procedure so courts can later review whether it was conducted fairly and non-suggestively.

Limits on Highly Suggestive Practices

  • Limiting “showups”
    A “showup” is when police present a single suspect (for example, on a sidewalk or in a squad car) and ask, “Is this the person?”
    The Innocence Project has supported recommendations to sharply limit showups or subject them to strict safeguards because they are highly suggestive and prone to error.

  • Cautions and training for officers
    Model policies include:

    • Guidance for dispatchers and first responding officers on not feeding details to witnesses.
    • Non-leading interview techniques and consistent procedures when handling multiple witnesses to avoid cross-contamination of memories.

Courtroom and System-Level Proposals

Beyond police procedures, the Innocence Project and regional Innocence Projects have also backed reforms inside the courtroom and broader justice system:

  • Allowing expert testimony on eyewitness reliability
    Courts should permit psychologists and other experts to explain, in plain terms, why memory is fallible, how stress, weapon focus, cross-racial identification, and suggestion can distort it, and why high confidence at trial is not always a sign of accuracy.
  • Better jury instructions
    Judges should give modern, research-informed instructions helping jurors evaluate eyewitness evidence (for example, addressing cross-racial IDs, poor viewing conditions, delay, and suggestive procedures).
  • Requiring corroboration in eyewitness-only cases
    Some Innocence Projects have advocated that convictions should not rest solely on uncorroborated eyewitness identification, especially where procedures were weak or risk factors for misidentification were high.
  • Statewide policies and model statutes
    The national Innocence Project has worked with legislatures and law- enforcement agencies to promote statewide standards that incorporate “sequential double-blind” lineups, detailed written policies, training, and data collection on identifications.

Summary (TL;DR)

In practical terms, the Innocence Project’s proposals to improve eyewitness ID accuracy cluster around:

  1. How lineups are run : double-blind, sequential, clear instructions that the culprit may not be present.
  1. How lineups are built and documented : fair fillers based on the witness’s description, one suspect per lineup, immediate confidence statements, full recording.
  1. What courts and lawmakers do : allow expert testimony, strengthen jury instructions, demand corroboration, and adopt statewide “sequential double-blind” policies.

All of these are aimed at reducing the risk of misidentification, which has been a major driver of documented wrongful convictions uncovered through DNA testing.

Information gathered from public forums or data available on the internet and portrayed here.