The Illusion of Accuracy

How Body-Worn Camera Footage Can Distort Evidence

November 2017

Table of Contents


Vanita Gupta, President & CEO
The Leadership Conference on
Civil and Human Rights

Each year, more and more police departments across the United States send their officers into the field wearing body-worn cameras.1 Many believed that by providing first-hand evidence of interactions between officers and the public, these cameras could enhance transparency, improve accountability, and foster greater public trust in local law enforcement. Yet, the promise of these cameras is not guaranteed; in fact, in many jurisdictions, it is not being realized. Without carefully crafted policy safeguards in place, there is a real risk that body-worn cameras could be used in ways that threaten civil and constitutional rights and intensify the disproportionate surveillance of communities of color.

To help ensure that these devices become tools of accountability, rather than instruments of injustice, a broad coalition of civil rights, privacy, and media rights organizations came together in 2015 to develop Civil Rights Principles on Body Worn Cameras.2 Shortly thereafter, The Leadership Conference on Civil and Human Rights and Upturn published the first iteration of the national Police Body-Worn Camera Policy Scorecard,3 an online resource that evaluates whether the body-worn camera policies of police departments are conforming to these principles and adequately protecting civil rights and civil liberties.

Our research has revealed that many police departments are failing to adopt adequate safeguards to ensure that constitutional rights are protected. In particular, we have discovered that year after year, the vast majority of the nation’s leading police departments with body-worn camera programs allow unrestricted footage review – meaning, officers are permitted to review footage from body-worn cameras whenever they’d like, including before writing their incident reports or making statements.

This report seeks to illuminate the ways that unrestricted footage review places civil rights at risk and undermines the goals of transparency and accountability. We urge police departments to instead adopt what we call “clean reporting,” a simple two-step process where an initial report is recorded based only on an officer’s independent recollection of an event and then a second, supplemental report can be added to a case file to address any clarifications after footage is reviewed. We make the case that in the interests of consistency, fairness, transparency and accountability, clean reporting should be adopted as a standard practice for all police departments with body-worn camera programs.

Ultimately, we must recognize that these cameras are just a tool, not a substitute for broader reforms needed to address police misconduct, build trust between police and communities of color, and ultimately fix our broken justice system. It will take much more work over a sustained period of time to ensure that our system of justice does not target or disproportionately impact low-income communities and communities of color. In the meantime, we will continue fighting for justice and working to build an America as good as its ideals.

Vanita Gupta, President & CEO
The Leadership Conference on Civil and Human Rights

  1. In fact, a nationwide survey in 2016 revealed that 95% of major police departments either had started, or had planned to start, BWC programs. In addition, Axon announced earlier this year that they would provide free body worn cameras to every police officer in the U.S.

  2. Principles available at

  3. Scorecard available at


In August 2014, five Marion County, Florida deputies chased down a suspected drug dealer named Derrick Price.4 One of the deputies had a body-worn camera, which captured the moments just before Mr. Price’s arrest. The footage appeared to show the deputies in an intense physical struggle: the camera was shaking heavily, and the deputies could be heard screaming at Mr. Price, “Stop resisting! Stop resisting!”

Marion County body-worn camera footage

The footage appeared to confirm what the deputies later wrote in their incident reports, that Mr. Price was indeed resisting arrest. One deputy wrote that after numerous verbal commands, “the defendant still would not comply,” a claim that seemed plausible enough, even though it’s not directly obvious from the video.5 Another deputy wrote that as the deputies were approaching, Mr. Price “turned his back toward officers at which time his hands were no longer visible”6 — this moment wasn’t captured by the body-worn camera because it happened just before the deputy turned his camera on. So while the footage did not show everything, it appeared to reinforce the deputies’ version of events.

Incident report of Marion County Deputy Adam R Crawford

Except their version of events was an illusion that heavily distorted the truth.

A surveillance camera from a nearby building told an entirely different story. That footage showed Mr. Price attempting to surrender voluntarily. He threw his hands up in the air and laid down on the pavement as the deputies were closing in. Once the deputies arrived, the surveillance footage shows them brutally beating Mr. Price, repeatedly striking his head and his torso.

Marion County surveillance footage

Without the existence of the surveillance footage, it’s likely that the deputies would have gotten away with the heinous beating of Mr. Price. But because other footage happened to exist, the public was able to learn what really happened. Four of the deputies later pled guilty to federal civil rights charges, including one deputy for obstruction of justice for falsifying his police report.7

The Derrick Price incident underscores why the circumstances under which officers are allowed to review body-worn camera footage and other video evidence must be critically examined. As police departments around the country continue to adopt body-worn cameras, it will soon be routine for police interactions to end up on video. But whether videos will help the public more accurately understand how a particular event truly unfolded — or whether, as in the Derrick Price incident, videos will be used simply to reinforce officer accounts — will depend on whether departmental policies and procedures limit when on-scene officers are allowed to watch footage.

Today, most major police departments that use body-worn cameras allow officers unrestricted footage review. Officers may review body-worn camera footage any time they wish, including before and during the process of writing their initial incident reports. Even in cases where officers’ actions are most closely scrutinized, including after a controversial use-of-force incident, officers are often permitted to review footage before giving investigators an initial interview.

This report explains why police departments must carefully limit officers’ review of body-worn camera footage. Policies that permit unrestricted review of footage reduce the accuracy of officer reports, and undermine the independent evidentiary and investigative value of police reports. Many departments recognize that body-worn cameras will capture details that are different than what officers remember experiencing. Decades of psychology research has shown that watching video replays can easily change people’s memories, often subconsciously.

Because watching body-worn camera footage can alter an officer’s memory of an event, doing so will likely taint what officers write in their reports. This practice will make it more difficult for investigators, internal affairs, and courts to accurately assess what occurred and whether an officer’s actions were reasonable given what he or she perceived at the time. Unrestricted review policies undermine longstanding principles of procedural justice, violate the law enforcement community’s best practices for preserving eyewitness evidence, and threaten to erode community trust. These policies are antithetical to the primary purpose of body-worn cameras: to enhance police transparency and accountability.

But this does not have to be the case. Departments can ensure that officer reports are as accurate and independent as possible by requiring clean reporting. Clean reporting is a simple and sensible two-step process.

Step 1: Officers write an initial incident report before reviewing any footage.
Step 2: Once an initial report is filed, officers may then review any available footage and file a supplemental report.

The first step preserves a report’s independent evidentiary value, by requiring officers to explain — from their own perspective and memory — what they saw and what they did.

The second step gives officers the opportunity to add details from the footage, provide context, and explain any differences between their initial reports and what the footage shows. Taken together, the initial and supplemental reports represent an officer’s most detailed and accurate account of an incident.

We are not the first to suggest this solution.8 Indeed, some major departments have already implemented clean reporting for certain serious use-of-force incidents. Rather, we are insisting that departments adopt this practice far more broadly than it is today. There is no reason why clean reporting could not be adopted in every police department, for every incident with body-worn camera footage. If the purpose of body-worn cameras is to increase transparency and accountability, then clean reporting must become a standard practice.

  1. Auston Miller, “Ex-Marion deputy indicted in police beating; four others already pleaded guilty,” Ocala Star Banner (Jan. 27, 2016),

  2. Incident report of Adam R Crawford (Aug. 11, 2014), available at

  3. Incident report of Cody P Hoppel (Aug. 11, 2014), available at

  4. “Former Marion County Deputy Charged With Excessive Use Of Force,” Department of Justice, U.S. Attorney’s Office Middle District of Florida (Jan. 27, 2016),

  5. See, e.g. “Body-Worn Cameras In New York City: An Assessment of NYPD’s Pilot Program and Recommendations to Promote Accountability,” New York City Department of Investigation Office of the Inspector General for the NYPD (OIG-NYPD) (July 2015), available at “Where a police officer is the subject of or witness to a misconduct investigation either by CCRB, IAB, a District Attorney, or OIG-NYPD, the officer should be precluded from reviewing any recordings as soon as the investigating entity informs NYPD it has commenced an investigation. The officer should only then be permitted to review the video after providing a formal statement and completing any incident-related paperwork. … [I]n all instances, officers should be permitted to review video after an initial report and an official statement have been made, and file supplemental reports in order to document details of an incident recalled through footage review and maintain a complete and accurate record of events surrounding an incident. In this regard, consideration of any mitigating factors leading to differences between BWC footage and officer testimony should be considered before officers are disciplined for these discrepancies, and officers should not be sanctioned without clear evidence they intentionally and materially obscured the truth.” 

How Officers Use Body-Worn Camera Footage

Body-worn cameras have given officers a powerful new ability: they can now review videos to supplement their own memories of past events. Camera vendors have made it easy for officers to review footage out in the field, soon after an incident occurs. Some cameras have video displays built directly on the device, while others can sync with in-car video systems or officer smartphones with dedicated viewing applications.9 Each of these technologies give officers immediate access to video evidence and the opportunity to review footage before they write their incident reports.

Officer reports are often evidentiary records

Police officers regularly document their actions and observations while on shift. They often keep informal notes about where they go, who they interact with, and what they see. After many types of interactions, departments require officers to file formal reports to “document the full range of criminal and administrative incidents that warrant the attention of police and correctional officers.”10 After specific types of incidents, such as those involving use-of-force, officers are commonly required to fill out additional reports or worksheets.11 Officers are typically required to file incident reports by the end of their shifts.

Officer reports may ultimately be relied upon in a variety of contexts for years after they are written, by “supervisors, police and correctional administrators, prosecutors, defense counsels, judges, jurors, [and] wardens[.]”12

Officer reports traditionally serve a dual purpose: They are a core part of how police departments investigate and process cases, but just as importantly, they are often vital pieces of evidence in later court proceedings or internal affairs investigations.

As an investigative tool, officer reports can assist law enforcement in determining what additional investigative steps or resources are needed. Officer reports are often included in applications for search warrants to justify the gathering of further information and evidence. As new information comes to light, officers routinely file supplemental reports that add to or correct information in the initial report, while preserving the content of the original report.13 Ultimately, these reports are added to a case file that may include numerous reports, interviews, and evidence that collectively contribute to the police department’s record of an incident.

Officer reports also play a crucial evidentiary role.14 Police reports — along with physical and testimonial evidence — can help factfinders (such as internal affairs investigators or juries) determine what occurred and who was at fault. The narrative portion of an officer’s initial incident report should also help these factfinders understand the officer’s independent perspective on and recollection of what occurred. An officer’s perception of an event as it transpired can be crucial to assess, for example, whether a use-of-force was reasonable, which, according to the Supreme Court, “must be judged from the perspective of a reasonable officer on the scene . . . in light of the facts and circumstances confronting them.”15

But the adoption of body-worn cameras appears to be shifting the purpose of officer reporting away from this crucial evidentiary role.

Departments routinely allow officers to view footage while writing reports

The vast majority of the nation’s leading police departments with body-worn camera programs currently allow unrestricted footage review. In 33 of the nation’s 40 largest departments that have adopted camera programs, officers are allowed to review footage at any time, including before writing their incident reports.16 Some major departments, like the Los Angeles Police Department, even require officers to review footage during the report-writing process.17 Only a handful of departments even require their officers to clearly indicate on they used video while preparing their reports.

LAPD’s policy requires officers to review footage when writing reports

Police executives often defend the practice of unrestricted footage review by claiming that it makes incident reports more “accurate.”18 But as we explain below, this accuracy is an illusion, because reviewing footage — consciously or not — can distort what officers claim to remember and what they write in their reports. Reports aided by footage review are ultimately less accurate, because they no longer capture an officer’s own independent recollection of events.

San Bernadino PD’s policy recognizes that body-worn cameras capture different information that what officers perceive

A few departments, however, have recognized the importance of preserving an officer’s independent recollection, and have placed limits on when officers may review footage after serious use-of-force incidents. The Oakland Police Department, for example, prohibits its officers from viewing any footage “prior to being interviewed by the appropriate investigative unit” after serious uses of force, fatal or near fatal pursuits, and in-custody deaths.

Oakland PD prohibits officers from viewing footage after a “Level 1” use of force — until they have submitted an initial report

Departments in Atlanta, Baltimore, and San Francisco similarly require officers to complete an initial report or interview after incidents where their actions may be under close scrutiny.

But while these departments’ viewing restrictions are a step in the right direction, they are neither widespread nor do they go far enough. These departments’ policies are a small handful of exceptions in an otherwise rampant landscape of unrestricted footage review. And no major police department in the United States has yet adopted a comprehensive policy that preserves the independence of officer recollections after all incidents.

  1. For example, Axon’s body-worn cameras connect with the vendor’s mobile application, allowing the wearer to “ stream, tag, and replay videos from [their] phone.” Axon Body 2,

  2. See Bruce Berg, Gregory Gibbs, and Michael Miller, Report Writing for Police and Correctional Officers, 10 (2013). 

  3. E.g. NYPD’s Patrol Guide requires any officer involved in a use-of-force incident to submit a Threat, Resistance, or Injury (T.R.I.) Incident Worksheet. PG 221-03, available at In Houston, an officer involved in a use-of-force incident must submit an original incident report (or a supplement incident report if applicable) and complete a Use of Force form that includes the name of all employees on the scene and the on-duty supervisor, the specific reasons for using force (including detailed descriptions of all actions taken by the officer and the person against whom force was used), the injury sustained by that person, and “any other pertinent information.” Houston Police Department GO-600-17, available at

  4. See Bruce Berg, Gregory Gibbs, and Michael Miller, Report Writing for Police and Correctional Officers, 2-4 (2013) (“these reports document criminal or, in some cases, rule-violating incidents; or they may be used to maintain a record of investigative tasks as they are accomplished during that segment of the event; or they accurately record what transpired for certain other pertinent players in the game who, while not present, will need to know the facts of the incident (e.g., investigators, administrators, and prosecuting and defense attorneys) . . . there are a number of less obvious reasons to maintain accurate, well-written records. Among these are efforts on the part of agencies to support or defeat allegations of wrongdoing, or to provide necessary evidence of proper policy and procedures having been used to avoid liability lawsuits”). 

  5. See Bruce Berg, Gregory Gibbs, and Michael Miller, Report Writing for Police and Correctional Officers, 10 (2013). (“Supplemental reports typically contain areas to provide basic information that links them to the original incident or complaint report.”). 

  6. Reports are often the subject of discovery rulings prior to trial, and investigators and patrol officers frequently use reports to refresh their memories before testifying in court. See, e.g., Criminal Practice Institute Manual, Public Defender Service for the District of Columbia, §2.8.K – Police Reports (2015). 

  7. See Graham v. Connor, 490 U.S. 386, 396-97 (1989). 

  8. See BWC Scorecard, The Leadership Conference and Upturn (Nov. 14, 2017), E.g., Aurora Police Department, Dallas Police Department, Oklahoma City Police Department, Salt Lake City Police Department, San Antonio Police Department, San Diego Police Department, Tampa Police Department. 

  9. Los Angeles Police Department Department Manual Section 3/579.15, Body Worn Video Procedures (Apr. 23, 2015), available at

  10. Former D.C. Metropolitan Police Department Chief Cathy Lanier “advocated for police officers to be able to view the body-worn cameras before writing an initial report because it is a basic tool that can help to improve report accuracy, officer decision making, and police-community relations.” Perry Stein, “Should officers be able to review body camera footage before writing police reports?, ” The Washington Post (Dec. 17, 2015),

The Distorting Effects of Unrestricted Footage Review

Unrestricted footage review creates an illusion of accuracy because it produces a false impression about how much officers actually remember about an incident. It makes officers’ memories appear to be more accurate, and thus more credible, than the memories of other eyewitnesses — which can distort how an independent factfinder, like a judge or a jury, might understand how an incident truly unfolded. In the worst cases, because of the inherent limits of body-worn cameras, unrestricted footage review allows officers to square their version of events to the footage, and potentially create false beliefs about what actually happened.

Unrestricted footage review can taint what officers remember

Video footage can influence how officers remember a particular event unfolding, and in turn, change what officers ultimately write in their incident reports.

A rich body of cognitive science research shows how the introduction of post-event information — such as reviewing video footage — can alter people’s memories of events.19 In fact, people can be led to believe they experienced entire events that did not occur,20 and are sometimes so confident in their “pseudo-memories” that they go on to describe them in substantial detail.21 Researchers have also found that “the post-event distortion of memory is permanent and occurs without conscious awareness.”22 As the authors of one study explained, “[p]eople often misattribute feelings of fluency for familiarity when thinking about an event. Familiarity, in turn, is mistaken for a signal that the event must have happened in the past.”23

This body of research underscores a key point: when officers review footage, they’re not only priming or refreshing their own memories — there’s also a real risk that their memories will actually change. Whether consciously or not, officers may fail to describe all of the details of an incident, perhaps even omitting details that may turn out to be relevant later that occurred off-camera or before the camera was turned on. In addition, officers may learn new facts by watching the footage, and report them as though they had perceived these facts at the time of the incident. Given that even slight changes in memory can impact how an officer ultimately describes his or her own experience within an incident report, departments must protect the integrity of these reports by requiring that officers document their untainted perception before watching any footage.

Unrestricted footage review can skew what officers write

Despite common misperceptions, camera footage is not inherently more objective or more reliable than other evidence. It shows one perspective of an incident — a perspective that is not identical to the officer’s perspective. Footage can be limited by the camera’s angle, field of view, and other factors. A lively encounter could result in blurry footage, or an officer who draws a firearm could inadvertently block the camera’s line of sight. Footage on its own will quite often be inconclusive and subject to interpretation.24

Sometimes, cameras will capture more than officers actually witness, even if the cameras are worn on officers’ bodies. For example, an officer wearing a body camera may turn his or her head away at a critical moment and miss something that the camera captured.25 Cameras can also have enhanced technological capabilities, such as night vision, which capture more detail than officers can themselves perceive.26

Conversely, camera footage can be incomplete, in some cases capturing less than what officers actually witnessed. Officers may turn their cameras on too late, or turn cameras off too early, meaning that key moments may not be captured.27 The camera’s angle, distance, lighting, and other factors may also mean that the camera does not clearly capture what actually occurred. For example, chest-mounted cameras are often obscured when an officer raises a weapon.28 It’s also entirely possible that an officer will react to an event not captured on video at all, even if his or her body camera was properly recording.

These unavoidable differences between an officer’s own perspective and the camera’s perspective makes it possible — and perhaps likely — that watching footage will influence what officers would otherwise write in their incident reports. With pressure to match their stories to the video, officers might report new facts that they learned from the video that they had not noticed at the time of the incident. And it’s easy to imagine officers simply providing a written narration of what’s shown in the footage, at the expense of providing a full and independent account of their actual experience and state of mind at the time.

Some major departments recognize these risks, even as they choose not to address them. For example, the Philadelphia Police Department, in its body-worn camera policy, warns that:29

Philadelphia PD points out that body-worn camera footage is not a substitute for officers perceptions

The San Diego Police Department’s policy includes similar caveats:30

San Diego PD recognizes that body-worn camera video can be incomplete

As these two departments recognize, video footage presents only one limited perspective.31 It’s a perspective that doesn’t always align with the experience of the officer. In the end, it’s not realistic to expect officers to remember everything that happened, especially during high-stress encounters, so it’s also not realistic to expect that officers’ initial reports will perfectly match the videos.

By allowing officers the opportunity to watch footage before writing their reports, departments actually place undue pressure on their officers to conform their reports to what the video appears to show, rather than what the officers actually saw and experienced. This is why officers should be permitted to watch video and provide explanatory context — but only after they have filed initial reports that capture their own perspective.

Unrestricted footage review can undermine the evidentiary value of police reports

When factfinders seek the truth, they look to compare and contrast independent pieces of evidence, in an effort to explain or rule out competing hypotheses about what took place. But unrestricted footage review creates an unnecessary dependency between the camera footage and an officer’s report, narrowing the amount of independent evidence available in a case. When officers can watch body footage before writing their initial reports, the independent evidentiary value of those reports is lost.

"I]f a case goes to trial, jurors are more likely to be swayed hearing what they consider to be two corroborating lines of evidence, when in fact they are hearing the video footage twice — once on screen and once through the eyes of the witness who watched it. . . . If it is important to know exactly what happened, then viewing the video footage will always be more accurate than the account of an officer, in which case the officer does not need to see the footage. If it is important to know an officer’s perception of an event, then it is important to preserve his memory untainted by viewing the video footage. In neither case is viewing the video footage recommended."
- Kathy Pezdek, The Marshall Project

The need to preserve the independence of evidence is especially acute in investigations of use-of-force incidents, allegations of officer misconduct, and other situations where an officer’s actions are under scrutiny. Critical to such investigations is an officer’s state of mind during the time of the incident.32 Yet unrestricted footage review allows officers to potentially conform their entire report to match what’s shown in the video, which greatly reduces the evidentiary value of that report. Investigators lose — forever — the ability the access an independent account of an officer’s state of mind.

Unrestricted footage review also contradicts investigative best practices. Throughout the criminal justice process, precautions are taken to shield witnesses from being exposed to sources of information that might influence their independent recollections of what occurred. Examples of this principle are evident throughout the investigative process, from the crime scene to the courtroom. Suspect lineups are often administered one witness at a time to avoid having witnesses influence one another.33 The D.C. Metropolitan Police Department, for example, prohibits group interviews of subjects, members, or witnesses of use-of-force incidents, and sequesters officers under investigation until after they are interviewed by investigators.34 Similarly, autopsy reports are often withheld from the public to avoid tainting witness testimony.35 And once cases go to trial, witnesses are generally prevented from listening to each other’s testimony to prevent later testimony from being intentionally or subconsciously skewed.36

Law enforcement executives, prosecutors, and local leaders have even used similar logic to withhold body camera video from being publicly released. For example, after the police shooting of Sylville Smith in Milwaukee, Wisconsin Attorney General Brad Schimel argued that releasing body camera footage “would compromise the integrity of the investigation” and that he could not “have witness statements colored or tainted by what they are seeing from other sources.”37 Los Angeles Police Chief Charlie Beck similarly justified withholding video evidence in a police involved shooting by arguing that it “shouldn’t be out there to taint the jury pool, nor should it be out there to taint witnesses.”38

By that same logic, unrestricted footage review taints officer statements and risks compromising the integrity of investigations. Officers are often key witnesses to an event. The same rationale that justifies policies and practices for withholding footage from potential witnesses and jurors should therefore also apply to police officers in their viewing of body camera footage.

Unrestricted footage review can unduly inflate officer credibility

In the most basic evidentiary sense, police officers are simply witnesses to alleged criminal activity and to the resulting actions of anyone on the scene. Yet unrestricted footage review gives officers the opportunity to augment their initial incident reports with information that would not otherwise be available to them from their own memory. This makes officers’ reports artificially consistent with video footage and appear to be unnaturally comprehensive and credible, particularly compared to reports of other witnesses to events. Police officers already enjoy a high degree of credibility in criminal cases,39 and artificial consistency between officer testimony and video evidence could cause factfinders to unjustifiably see an officer as even more credible than other witnesses overall, which could have distorting effects on an entire proceeding.40

The consequences of this undue credibility are stark. If officers are inclined to distort the truth, letting them watch footage of an incident before writing an initial report will give them the opportunity to misrepresent what happened more effectively. These officers could tailor their reports to conform to what was captured on video, and omit important information that the footage failed to capture. But if the same officers have to document their behavior without the ability to go back and watch the footage first, it makes it more difficult to get away with lies and other false narratives.

When unrestricted review is permitted, any discrepancies between the two records are less likely to be uncovered, let alone scrutinized — preempting and obfuscating what might be important lines of investigative inquiry. If an officer’s account of an event and the associated body camera footage remain independent, discrepancies can be noted and evaluated by factfinders, leading to more accurate and thorough understanding of the incident.

Unrestricted footage review can erode procedural justice and trust

Building and maintaining community trust is one of the cornerstones of successful law enforcement.41 While body cameras are frequently sold to communities as a way to improve this trust, allowing police officers to watch body camera footage before writing reports could undermine the purpose of these tools, leading many in the community to double down on opinions that the justice system is inherently unfair.

Community members develop opinions about an interaction with law enforcement based not only on the outcome of the encounter, but also the fairness of the process for addressing their concerns. In fact, research has shown that the process adopted by police departments is often more important than the outcome of the encounter in shaping a community member’s experience of an interaction.42 Allowing police officers to view body-worn camera footage before writing their reports violates community confidence, undermining two pillars of procedural justice: fairness in the process and transparency in actions.43

First, unrestricted viewing is an unfair process from the perspective of the community. While officers are allowed to consult video footage before writing their statements, other witnesses are not afforded this opportunity and must rely on their own memories. For example, if an individual witnessed an armed robbery at a bank, the police would not show that person footage from the bank’s security camera before taking a statement from that witness. Civilians and witnesses who directly interact with the police are similarly not allowed to view relevant body-worn camera footage prior to making initial statements, while officers often are. This inconsistency between the way officers and civilians are treated erodes the public’s perception of procedural fairness in policing.

Second, there is little transparency as to when police officers are actually consulting camera footage in practice. Departments rarely explicitly state which evidence officers are allowed to use in their reports, and the community is often unaware that officers can change their statements to match what is shown on the body-worn camera footage before filing, without any paper trail of those changes. Too few departments require officers to note in their reports when body camera footage was consulted.

Because it violates these two pillars of procedural fairness, unrestricted footage review contributes to the community’s distrust of local law enforcement. Responsible policies like the two-step, clean reporting process described below, on the other hand, encourages transparent and accountable outcomes needed to foster public trust.

  1. Douglas J. Narby, Brian L. Cutler, and Steven D. Penrod. “The effects of witness, target, and situational factors on eyewitness identifications.” (1996) (noting that production of verbal statements and use of mug shots and police sketches can influence witness memory); Wright, Daniel B., Gail Self, and Chris Justice. “Memory conformity: Exploring misinformation effects when presented by another person.” British journal of psychology 91.2 (2000): 189-202. See also Robert A. Nash and Kimberley A. Wade. “Innocent but proven guilty: Eliciting internalized false confessions using doctored‐video evidence.” Applied Cognitive Psychology 23.5 (2009): 624-637,which found that watching video footage of an event can change what people know and experienced to be the objective truth of an event. Experimenters accused participants of cheating on a computerized gambling task — something both the experimenters and participants knew had not happened. But when the experimenters showed some participants a digitally-manipulated video of themselves cheating, those participants were more likely to report believing they had actually cheated than others accused of the same thing but who only knew such a video existed. Participants who saw the video were also more likely to sign a written confession, without resistance; and Kimberley A. Wade, Sarah L. Green, and Robert A. Nash. “Can fabricated evidence induce false eyewitness testimony?” Applied Cognitive Psychology 24.7 (2010): 899-908, where study, participants completed the same computerized gambling task in pairs; later, they were falsely told that their partner had cheated. Even though experimenters emphasized that participants should not confirm the accusation unless they actually saw their partner cheat, participants who saw video “proving” the claim were three times more likely to sign a statement confirming they witnessed their partner cheating — and that they understood their signed statement would lead to disciplinary action against their partners. Some of the participants who signed the statement also described incriminating accounts of what they supposedly saw. 

  2. Elizabeth F. Loftus, David G. Miller, and Helen J. Burns. “Semantic integration of verbal information into a visual memory.” Journal of experimental psychology: Human learning and memory 4.1 (1978): 19; Daniel B. Wright and Elizabeth F. Loftus. “How misinformation alters memories.” Journal of Experimental Child Psychology 71.2 (1998): 155-164. 

  3. Elizabeth F. Loftus, and Jacqueline E. Pickrell. “The formation of false memories.” Psychiatric annals 25.12 (1995): 720-725. 

  4. Kathy Pezdek, “Should Cops Get to Review the Video Before They Report?,” The Marshall Project (Aug. 13, 2015),

  5. Id. 

  6. See generally Adam Benforado, Frames of Injustice: The Bias We Overlook, Indiana Law Journal 85 (2010). See also Robert N. Kraft, “The influence of camera angle on comprehension and retention of pictorial events,” Memory & Cognition 15.4 (1987): 291-307 (finding that “camera angle can profoundly affect the meaning viewers ascribe to pictorial events”); Percy H. Tannenbaum and James A. Fosdick, “The effect of lighting angle on the judgment of photographed subjects,” Educational Technology Research and Development 8.5 (1960): 253-262 (finding that lighting angle “can influence some of the connotative judgments of … photographed models”); Dan M. Kahan, David A. Hoffman, and Donald Braman, “Whose eyes are you going to believe? Scott v. Harris and the perils of cognitive illiberalism,” Harvard Law Review 122 (2009): 08-18 (finding viewers of they key video in Scott v. Harris had “sharp differences of opinion along cultural, ideological, and other lines” about what the video showed). See also Roseanna Sommers, “Will Putting Cameras on Police Reduce Polarization?” 125 Yale L. J. 1304 (2016) (finding that “despite the seeming objectivity of the camera, video footage remains susceptible to biased interpretation by observers such as grand jurors. . . . This holds true for factual or objective judgments, such as whether a weapon was present and whether physical force was used; evaluative judgments, such as whether the police officer treated the citizen fairly; and global judgments, such as whether the police officer deserves to be sanctioned for misconduct.”) 

  7. Some body cameras are affixed to the brim of an officer’s hat, or to an officer’s glasses. But these cameras have downsides, too: head-mounted cameras are more shaky, making it more difficult to capture clear and steady footage. 

  8. E.g. Camera vendor PatrolEyes offers body-worn camera models enhanced with night vision and high definition recording. “PatrolEyes HD Body Camera,”

  9. See, e.g., Craig Timberg, “A cop fires. A teen dies. Yet six police body cameras somehow miss what happens.,” Washington Post (Mar. 20, 2017), _available at

  10. Justin Davenport, “Cameras worn by Met’s armed police are ‘blocked whenever they aim their guns,’” Evening Standard (Oct. 19, 2015),

  11. Philadelphia Police Department Directive 4.21 §1.B, avaialble at

  12. San Diego Police Department Procedure, 1.49 - Administration, available at [](]( 

  13. However, despite these warnings, both Philadelphia PD and San Diego PD still both allow unrestricted footage review. 

  14. “The “reasonableness” of a particular use-of-force must be judged from the perspective of a reasonable officer on the scene, rather than with the 20/20 vision of hindsight.” Graham v. Connor, 490 U.S 386 (1989). 

  15. To “ensure fair and proper proceedings,” NYPD Detective Supervisors conducting police lineups to identify the perpetrator of a crime are instructed to interview each witness separately to “obtain and record [a] description of [the] suspect.” Policy further mandates that Detective Supervisors prevent witnesses from speaking to each other before, during, or after viewing a lineup, and to separate witnesses after each viewing. NYPD Patrol Guide, PG 208-24: Identification Lineups/Showups, Gould Publications, pp. 447-449 (2002 Edition). 

  16. Metropolitan Police General Order GO-RAR-901.08, “Use of Force Investigations,” as of 12/01/16. Available at

  17. Many police departments routinely prevent the release of video footage to the public out of concern that footage might taint eyewitness accounts of what occurred. Following the recent fatal police shooting of Keith Lamont Scott, Charlotte-Mecklenburg Police Chief Kerr Putney refused to release dash-cam footage to the public out of concern that doing so would “jeopardize the investigation.” See, e.g., Richard Winton, Sarah Parvini, and Corina Knoll, “The battle for footage after the El Cajon shooting: ‘The country is begging for a video’,” The Los Angeles Times (Sept. 20, 2016), After a fatal police shooting in El Cajon, CA, authorities refused to release bystander footage of the incident due to concern that the footage “could influence what witnesses say and make it more difficult to get to the truth.” See also “Charlotte police refuse to release shooting video as 3rd night of protests stay largely peaceful,” The Chicago Tribune (Sept. 23, 2016),

  18. The leading textbook on the rules of evidence provides a concise explanation for why this is necessary: “[s]eparation prevents improper influence during the trial by prohibiting witness-to-witness communication both inside and outside the courtroom.” [Kenneth S. Broun, George E. Dix, Edward J. Imwinkelried, D. H. Kaye, Robert P. Mosteller, E. F. Roberts, John W. Strong, and Eleanor Swift, McCormick on Evidence, Title 2, Ch. 5, §50, p. 114 (2006).] The text elaborates: “If a witness hears the testimony of others before she takes the stand, it is much easier for the witness to deliberately tailor her own story to that of other witnesses. Witnesses may also be influenced subconsciously. In either event, the cross-examiner will find it more difficult to expose fabrication, collusion, inconsistencies, or inaccuracies in the testimony of a witness who has already heard other witnesses testify. [Kenneth S. Broun, George E. Dix, Edward J. Imwinkelried, D. H. Kaye, Robert P. Mosteller, E. F. Roberts, John W. Strong, and Eleanor Swift, McCormick on Evidence, Title 2, Ch. 5, §50, p. 114 (2006).]” Separating witnesses from each other and from evidence has repeatedly been held up as a national best practice. The National Institute for Justice (the research, development, and evaluation agency of the Department of Justice) released a report in 1999 entitled “Eyewitness Evidence: A Guide for Law Enforcement” that recommended best practices for capturing witness testimony. The report describes the procedure for obtaining a witness statement at the crime scene, including encouraging “the witness to avoid contact with the media or exposure to media accounts concerning the incident,” and instructing witnesses “to avoid discussing details of the incident with other potential witnesses.” [Eyewitness Evidence: A Guide for Law Enforcement, National Institute of Justice (1999), p. 16,] In 2013, the National Forensic Science Technology Center released a similar guide, “Crime Scene Investigation: A Guide for Law Enforcement,” that reiterates the recommendation that witnesses be secured and separated at the crime scene in order to maintain integrity and minimize contamination before giving an initial statement. [Crime Scene Investigation: A Guide for Law Enforcement, National Forensic Science Technology Center (2013),] 

  19. Stephen Gossett, “Milwaukee Shooting Footage Will Not Be Released Ahead of Investigation,” Chigaoist (Aug. 22, 2016),

  20. Kate Mather and Richard Winton, “Officers in fatal Venice shooting wore privately owned body cameras,” Los Angeles Times (May 12, 2015),

  21. “‘In criminal cases officers are given a higher degree of credibility,’ said Tom Grover, a former Albuquerque Police Department sergeant who now works as a defense lawyer. ‘They are seen as having no stakes in the matter, of just doing their duty.’” Albert Samaha, “Blue Lies Matter”, BuzzFeed News (Jan. 17, 2017),

  22. See Barry Kamins, Practical Tips for New York Law Enforcement, NYPD Patrol Guide (2002 Edition) (“If one had to choose a single trait or characteristic that is most significant for law enforcement personnel, that trait would be credibility. . . . [T]he issue of credibility arises over and over again because, in every prosecution, an assistant district attorney must rely on the officer’s version of the facts . . . the officer’s recollection of the events is crucial to the People’s case. A small twist in the facts, a slight colorization of events, or a tiny omission of a detail can make all the difference in the world.”). 

  23. Building Trust Between the Police and the Citizens They Serve: An Internal Affairs Promising Practices Guide for Local Law Enforcement, U.S. Department of Justice Office of Community Oriented Policing Services,

  24. Laura Kunard and Charlene Moe, Procedural Justice for Law Enforcement: An Overview. Washington, DC: Office of Community Oriented Policing Services (2015). 

  25. Id. 

Clean Reporting Preserves the Independence of Evidence

Adopting a clean reporting process and limiting when officer may review footage is the only fair way to address the concerns raised in this report, about memory distortion, evidence preservation, undue officer credibility, procedural justice, community trust, transparency, and accountability. Clean reporting is a simple and sensible approach that all witnesses, including officers, should be required to follow whenever body camera footage and other video evidence is available.

Clean reporting works as follows: initially, officers would write down only the details of an event as they remember it, providing a formal statement and completing any incident-related paperwork. After filing this initial report, officers would then be permitted to review relevant video and file a supplemental report, clearly noting when they watched footage and what information was added after watching it.

Several law enforcement agencies already use a similar approach in some cases, including the Atlanta and San Francisco Police Departments. The San Francisco Police Department’s policy comes close to clean reporting — but only in certain case like officer-involved-shootings and in-custody deaths:44

San Francisco’s policy outlines a two-step process for footage review following certain uses of force

The fact that these policies have already been adopted in a few agencies suggests that a two-step, clean reporting process would not be overly burdensome, particularly since officers across departments often already file supplemental reports as new information comes to light.

Although limited two-step policies represent steps in the right direction, clean reporting should be a standard process in all cases for two reasons. First, any officer interaction could reasonably result in a civilian misconduct complaint, and when adjudicating those complaints, it’s critical to have preserved an officer’s independent explanation of the interaction. A limited policy that only requires clean reporting after serious use-of-force incidents would mean that many complaint cases (say, for a traffic stop or a stop-and-frisk) would have distorted officer reports. Second, mandating a clean reporting process across all cases would reduce ambiguity for officers, since reporting practices would remain consistent regardless of the type of interaction recorded by the body camera. This would lead to more standardized administration of the reporting process within law enforcement agencies, and greater clarity in the courts about what evidence officers were exposed to before they wrote their reports.

Many police executives and line officers have resisted the clean reporting approach, expressing fear they will be penalized for discrepancies between the report and the video by juries and others who are prone to look for evidence of police misbehavior. We are not suggesting that officers should be penalized or disciplined for every discrepancy — just for those where there is clear evidence someone intentionally and materially obscured the truth.45 The two-step, clean reporting process we propose allows factfinders to explore and understand differences between an officer’s perspective and independent video — to determine what is a legitimate discrepancy and what is an intentional one — ultimately leading to a just and fair outcome for all parties.

Clean reporting would benefit everyone across the criminal justice system. It would improve judicial outcomes by preserving the independence of evidence in the service of fair and thorough fact-finding. It would reassure the public that body-worn cameras have utility as tools of accountability, rather than simply instruments for enhanced policing. And importantly, it would help officers themselves, whose actions could be judged based on their perceptions at the time, rather than being held to account for everything that’s shown in footage.

If body-worn cameras are help to drive consistency, fairness, and justice, it’s imperative for all police departments with camera systems to prohibit unrestricted footage review and adopt clean reporting as a standard practice.

  1. Oakland Police Department General Order I-15.1 (§IV.A.2), available at

  2. Body-Worn Cameras in NYC: An Assessment of NYPD’s Pilot Program and Recommendations to Promote Accountability, The Office of the Inspector General for the NYPD (July 2015),

About This Report

This report was written by Harlan Yu and Miranda Bogen at Upturn, with assistance from The Leadership Conference Education Fund’s Sakira Cook, Shin Inouye, Tyler Lewis, Patrick McNeil, Leslie Paluch, and Corrine Yu, who was an editor of the report.

We would like to acknowledge the research contributions of Alexander Atkins, Allison Johnson, Sanchi Khare, Derek Lanebeck, Zachary Martinez, Michael Nagelberg, Jennifer Porter, and Rebecca Williams from the Federal Legislation Clinic at Georgetown Law.

We would like to acknowledge the research contributions of Alexander Atkins, Allison Johnson, Sanchi Khare, Derek Lanebeck, Zachary Martinez, Katherine Moore, Michael Nagelberg, Michael Perez, Jennifer Porter, Sierra Walker, and Rebecca Williams from the Federal Legislation Clinic at Georgetown Law.

This report was supported by the Ford Foundation, the Open Society Foundations, and the Media Democracy Fund.


Return to top