Video, Perception, and Hindsight

Video can be some of the strongest evidence available after a police critical incident. It can also create an illusion that the reviewer sees everything that mattered.

Body-worn cameras, surveillance systems, dash cameras, cell phones, and other recordings have transformed how police encounters are reconstructed.

Video can establish timing, movement, location, commands, weapon position, and contradictions between later accounts and recorded events.

But a camera is not a human observer.

It has a fixed field of view, a particular location, technical limitations, and no understanding of what it records. It may capture things a participant did not perceive and miss things that were obvious to someone standing nearby.

Review also occurs with advantages unavailable during the event: replay, slow motion, frame-by-frame inspection, multiple camera angles, transcripts, enhanced audio, and knowledge of the outcome.

The useful question is therefore not:

“Does the video show what happened?”

It is:

What does this recording reliably establish, what does it leave unresolved, and how does it relate to what participants could perceive in real time?

Where it fits in the incident lifecycle

Arrival & Approach

Body-worn camera may begin before officers reach the subject, preserving dispatch audio, officer discussion, approach routes, and the first view of the scene.

Encounter & Intervention

Video may capture commands, movement, weapon posture, use of force, distances, civilians, barriers, and changes in threat.

It can be particularly valuable for reconstructing sequence.

Post-Incident & Analysis

Video often becomes central to investigation, administrative review, litigation, media reporting, academic analysis, and public understanding.

Those later uses should distinguish the recording itself from interpretations built around it.

What to know

1. Video is evidence, not an all-seeing witness

A camera records what falls within its technical field.

It may not capture:

• events behind the officer;
• peripheral movement;
• depth accurately;
• objects blocked by bodies;
• activity outside the frame;
• subtle visual cues;
• tactile information;
• what another officer saw; or
• what occurred before activation.

A recording can therefore be both highly probative and incomplete.

2. Camera perspective is not human perspective

A body-worn camera is usually mounted on the torso, head, shoulder, or another fixed location.

Its perspective may differ from the officer’s eyes.

The camera may:

• point toward an object the officer is not looking at;
• miss something the officer sees by turning their head;
• exaggerate or compress apparent distance;
• move differently from the wearer’s visual focus; or
• be blocked during physical movement.

Reviewers should avoid assuming that visible on camera means perceived by the officer, or that not visible on camera means not visible to the officer.

3. What is outside the frame can still matter

An officer may react to:

• another person;
• a sound;
• movement outside the frame;
• an officer’s warning;
• a civilian’s location;
• a vehicle;
• a doorway; or
• an accessible weapon that the camera does not show.

The absence of an event from a recording is not automatically evidence that it did not occur.

Where possible, multiple sources should be combined.

4. Recorded audio has limits

Microphones can capture speech that a person under stress did not hear or process.

They can also fail to capture:

• distance effects;
• competing noise;
• overlapping voices;
• hearing impairment;
• radio traffic through an earpiece;
• auditory exclusion; or
• whether words were intelligible to a particular person.

A command being audible on playback does not conclusively establish that every participant heard or understood it.

5. Slow motion changes perception

Slow-motion and frame-by-frame review can be useful for determining sequence.

They can also make a rapidly developing event appear slower and more deliberate than it was.

A movement that occurred within a fraction of a second can be examined for minutes after the fact.

For timing questions, reviewers should understand both:

• what the recording reveals when examined carefully; and
• how quickly the event unfolded at normal speed.

6. Repeated viewing creates hindsight

A reviewer can watch the same event multiple times.

After several viewings, it becomes difficult not to anticipate:

• when the weapon will move;
• when the officer will fire;
• where the person will fall;
• whether another person will enter the frame; or
• how the encounter will end.

Participants did not have that knowledge.

Repeated viewing improves reconstruction but can also make uncertain moments appear predictable.

7. Video can contradict an account

The limitations of video do not make it weak evidence.

A sufficiently clear recording may establish that:

• a person was stationary rather than advancing;
• a weapon was or was not raised;
• commands occurred in a particular sequence;
• an officer was in a different location than described;
• significant time passed between uses of force; or
• a later factual assertion cannot be reconciled with the recording.

The appropriate response is not to privilege either video or testimony automatically.

It is to determine what each source can reliably establish.

8. Perception and factual reconstruction are different questions

Post-incident analysis often asks two related questions:

What physically happened?

and:

What could a participant reasonably perceive while it was happening?

Video may be excellent evidence for the first question.

The second may require additional information about:

• participant position;
• field of view;
• lighting;
• attention;
• timing;
• prior information;
• stress;
• obstruction; and
• events outside the camera frame.

Those questions should not be collapsed into one.

What the evidence can tell us

Video can help establish:

• sequence;
• elapsed time;
• subject movement;
• officer movement;
• weapon posture;
• commands;
• warnings;
• use-of-force timing;
• pauses between interventions;
• civilian locations;
• scene geometry;
• physical barriers; and
• inconsistencies between recorded events and later accounts.

Confidence increases when multiple independent recordings show the same event from different positions.

What the evidence does not establish

The fact that something appears on video does not, by itself, establish that:

• every participant perceived it;
• apparent distance is exact;
• audio was heard or understood;
• movement was intentional;
• the camera captured the complete scene;
• the officer’s attention was directed where the camera pointed;
• an event outside the frame did not occur; or
• a reviewer watching repeatedly has the same informational perspective as a participant acting once in real time.

Likewise, camera limitations do not mean clear recorded evidence should be disregarded.

Questions to keep in mind

Where is the camera?
Body, vehicle, building, bystander, or another location?

What does it not show?
Look for important people, objects, and areas outside the frame.

Is the apparent distance reliable?
Lens and perspective can affect depth.

What is the playback speed?
Watch important sequences at normal speed as well as frame by frame.

What could the participant actually see?
Camera direction and gaze direction may differ.

What information did the participant already have?
Prior threats or dispatch information can affect interpretation of later movement.

Is there another recording?
Multiple angles can resolve ambiguities that one camera cannot.

Does the video contradict a factual claim?
Identify the specific contradiction rather than treating video as globally superior or inferior to testimony.

The takeaway

Video can dramatically improve factual reconstruction.

But it should not be treated as a perfect substitute for human perception or as though a reviewer watching repeatedly occupies the same position as a participant acting once.

Use video to clarify the record without confusing the camera’s perspective with the participant’s perspective.