7.3 Encoding Vehicles, Signage, Numbers & Scene Layout
Key Takeaways
- Alphanumeric strings such as plate numbers and addresses are best encoded in fixed chunks rather than as continuous runs of characters.
- Vehicles are encoded reliably with a fixed attribute order — colour, body style, make, distinguishing marks, plate — because a fixed order prevents omission under time pressure.
- Encode the relative position of objects, since many questions ask what was near, left of, or behind something rather than what was simply present.
- Read every character of any string in the image, because near-miss distractors typically alter the middle characters rather than the beginning or end.
- Text in the image — signage, clock faces, house numbers, prices — is high-yield material that candidates routinely skip while studying objects.
7.3 Encoding Vehicles, Signage, Numbers & Scene Layout
Sections 7.1 and 7.2 covered the study protocol and the encoding of people. This section covers everything else in the frame — vehicles, text, numbers, and where things sit relative to one another. These are the details that questions most often turn on, and the ones candidates most often fail to encode because they have no system for them.
A reminder of scope before we start. The examination does not test job knowledge, so you are not expected to know New York plate series, which counties use which prefixes, vehicle model years, or how a collision is classified. What is measured is whether you can look at an image for two minutes and afterwards state accurately what was in it. Everything below is a memory technique, not domain knowledge.
1. Alphanumeric Strings: Chunk in a Fixed Pattern
Plate numbers, house numbers, phone numbers, unit numbers and prices are pure recall with no meaning to hang on, which makes them the most fragile items in the frame. Two techniques carry almost all the benefit.
Chunk consistently. A seven-character string read as seven separate characters exceeds working memory. Read as two or three chunks, it does not.
| Raw string | Poor encoding | Good encoding |
|---|---|---|
| KJH4820 | K-J-H-4-8-2-0 (seven chunks) | KJH / 48 / 20 (three chunks) |
| 5512 Ridge Road | five-five-one-two | 55 / 12 on Ridge |
| 03:15 on a wall clock | three fifteen | quarter past three |
Use the same chunk pattern every time — letters as a group, then digits in pairs — so that recall has a shape to follow.
Attach a hook to each chunk. A meaningless chunk is far more durable when tied to something: KJH becomes initials of a name you know, 48 becomes a familiar age or address, 20 becomes a round number. The hook does not need to be clever; it needs to exist.
Read every character, including the middle. Distractor options in this examination are commonly built by rearranging or swapping the middle characters while leaving the beginning and end intact — so that KJH 4820 competes against KJH 4028 and KJH 4802. A candidate who encoded "KJH four-something-twenty" cannot separate them. This is the same discipline required by the flexibility of closure items in section 5.3, and practising it serves both blocks.
2. Vehicles: Encode in a Fixed Attribute Order
When a vehicle appears in a photograph, run a fixed sequence rather than noticing whatever stands out. A fixed order guarantees that nothing is skipped and gives your recall a checklist to walk during the questions.
| Order | Attribute | What to fix in memory | Common recall failure |
|---|---|---|---|
| 1 | Colour | Base colour, plus any second colour on roof, hood or panels | Merging similar shades — silver and grey, navy and black |
| 2 | Body style | Sedan, coupe, hatchback, SUV, pickup, van; door count | Calling a crossover an SUV, or missing a two-door |
| 3 | Make or badge | Manufacturer if the badge or grille is legible; otherwise skip it rather than guess | Inventing a make from a general impression |
| 4 | Distinguishing marks | Damage, mismatched panel, roof rack, decals, ladder, missing hubcap | Encoding the vehicle generically and losing the one unique feature |
| 5 | Plate and position | Plate characters if legible, plus where the vehicle sits in the frame | Recalling the plate but not which vehicle carried it |
Attribute 4 deserves emphasis. In a photograph containing several vehicles, questions frequently hinge on which one had the ladder on the roof or the dented rear quarter. Generic encoding — "three cars, one white" — answers nothing.
Attribute 5 contains a trap worth naming. When two or more vehicles appear, bind each detail to its vehicle at the moment you encode it. "White pickup, ladder, plate ending 20, far left" is a bound record; "there was a ladder and a plate ending 20" is not, and a question asking which vehicle bore the plate will defeat it.
3. Text in the Image Is High-Yield
Candidates study objects and skip words. Question writers do the opposite, because text is unambiguous and easy to write items about. Deliberately sweep the frame for:
- Signage — shop names, street signs, posted notices, opening hours
- Numbers — house numbers, prices, aisle numbers, room numbers, exit numbers
- Clocks — analogue or digital, and whether they show a plausible time
- Vehicle markings — company names, phone numbers, unit numbers
- Labels on objects — brands on boxes, text on a bag or a shirt
Text is also easier to hold than visual detail, because it converts directly into language. Reading a sign to yourself once encodes it more reliably than looking at an object for the same duration.
4. Encode Relationships, Not Just Contents
A large share of memorization questions ask about position: what was to the left of the counter, what sat between the two doorways, which direction the footprints pointed, what stood behind the person in the foreground.
An inventory-only encoding — a mental list of what was present — cannot answer these. Build relationships as you scan:
- Choose one anchor per photograph. The largest fixed object works best: a counter, a doorway, a parked vehicle, a signpost. Anchors do not move and are easy to recall.
- Locate everything else relative to the anchor. "Toolbox on the floor, left of the counter." "Blue van behind the signpost." "Clock above the rear door."
- Note orientation as well as position. Which way a person faced, which way a vehicle pointed, which direction footprints or an arrow indicated. Orientation questions are common and orientation is rarely encoded by accident.
- Count deliberately. How many people, how many vehicles, how many boxes. Counts are frequently asked and almost never remembered unless counted on purpose.
The anchor technique also aids recall order. During the questions, mentally return to the anchor and work outward along the same path you used while studying — a structured retrieval route recovers more than free recall does.
5. A Two-Minute Working Routine
Applied to a single photograph, everything above fits a repeatable routine:
| Time | Action |
|---|---|
| 0:00–0:15 | Take in the scene type and choose your anchor |
| 0:15–1:00 | Structured scan (quadrants or raster), encoding people first, then vehicles in fixed attribute order |
| 1:00–1:25 | Text sweep — every sign, number, clock and label in the frame |
| 1:25–1:45 | Relationships and counts, anchored to your chosen object |
| 1:45–2:00 | Rapid confirmation pass on the two or three details you feel least sure of |
Then move to the next photograph, whether or not you feel finished. Three photographs share six minutes, and a candidate who spends three minutes on the first has two minutes for the remaining two.
6. Drill
Extend the publisher's practice method with these variations, using photographs you have never seen:
- Plate drill. Find a photograph containing a legible plate or house number. Study two minutes, then write the string from memory. Have someone generate three near-miss variants that differ only in the middle characters and see whether you can still identify the original.
- Multi-vehicle binding drill. Use an image with three or more vehicles. Afterwards, write a bound record for each: colour, body, marks, position. Check how many attributes migrated to the wrong vehicle.
- Relationship recall. After studying, answer only positional questions: what was left of X, what was behind Y, which way did Z face. This is the weakest area for most candidates and improves quickly once practised deliberately.
During a memorization study period a candidate sees a photograph containing a white pickup with a ladder on the roof rack, a grey sedan with front-end damage, and a blue van. Afterwards he is asked which vehicle had visible damage and answers "the pickup." What encoding error most likely produced this?
A candidate encodes a licence plate in a photograph as "starts with RTB, ends in 7." The answer options are RTB 4917, RTB 4197, RTB 4971 and RTB 9417. What is the problem with his encoding, and what would have prevented it?
A candidate has two minutes to study a photograph of a shop interior. He spends the full two minutes looking carefully at the objects on the shelves and the two people at the counter. Which category of question is he most likely to have prepared himself poorly for?