5.3 Cross-Scenario Counts, Order and Background
Key Takeaways
- Memory questions can aggregate details across several videos, so each scene needs a simple comparable tag.
- Order questions are easier when scenarios are grouped into beginning, middle, and end before exact comparisons are made.
- Background recall should focus on large, readily visible anchors connected to a character or event.
Why cross-scenario recall is different
A direct question asks for one fact from one scene. A cross-scenario question asks you to search the mental set of all 15 videos. The official examples include:
- How many scenarios involved a character wearing a short-sleeved shirt?
- Which of several described scenarios appeared earliest?
- Which scenario included a particular event?
These questions reward consistent encoding. If each scene is remembered using a different set of features, comparison becomes difficult. Use a repeatable tag: role, issue, visible feature, background anchor, sequence position.
Count categories, not microscopic attributes
For cross-scene counts, notice broad binary categories while viewing:
- Short sleeves / long sleeves.
- Indoors / outdoors.
- Supervisor / coworker / member of public.
- One speaker / multiple speakers.
- Vehicle present / absent.
- Large doorway, portrait, detector, or barrier visible / absent.
The official sample count concerns short sleeves, a readily observable category. It does not require distinguishing nearly identical color tones or reading distant serial numbers. Use categories that remain stable in memory.
When answering, reconstruct the relevant scenes one at a time and keep a mental tally. Do not estimate from a vague impression such as "several." If one scene is uncertain, compare the answer options; widely separated counts may allow a confident choice without perfect recall.
Build a rough timeline
Fifteen scenarios are easier to order when grouped:
- Beginning: the first few distinctive scenes.
- Middle: a cluster around the halfway point.
- End: the final few scenes before Memory Skills begins.
Attach one strong anchor to each cluster. The official samples place the protester scenario before the day-off request, fatigue discussion, and checkpoint delay. A question asks which described event appeared earliest; recognizing the protester as the first sample is enough without assigning every scene an exact number.
During practice, briefly label a video as early, middle, or late, then refine exact order between two candidates when needed. Story chaining can help: "protest, overtime, fatigue, aide." The chain preserves order through meaning rather than arbitrary numbers.
Reconstruct the background from the person outward
An official sample asks what was located behind a balding person in a dark blue shirt. The explanation identifies a brown door, while also noting a picture of the White House. This is a large-background-anchor task.
To encode a scene:
- Focus first on the speaker and issue.
- Expand outward to one object immediately behind or beside the speaker.
- Note one larger setting cue.
- Preserve the relation: "brown door behind aide, picture also behind."
A relation is more useful than an inventory. Remembering "door" without where it appeared may cause source confusion. Remembering "brown door behind the aide" binds object, location, and character.
Event-to-scenario matching
A question may ask which scenario contained a particular event. Store each scene's distinctive verb:
- protester refused to move;
- supervisor requested day-off work;
- coworker described fatigue;
- aide complained about delay.
When the prompt supplies an event, retrieve the verb and then the associated character and setting. This method also separates scenes that share a security location but involve different interactions.
A retrieval sequence for each item
- Identify whether the item is count, order, background, or event matching.
- Name the relevant scene tags without looking at choices if possible.
- Read every choice and reject details imported from other scenes.
- For counts, tally scene by scene.
- For order, compare timeline clusters, then exact sequence.
- For background, visualize the person and expand behind or beside them.
- Select and advance; 12 questions share eight minutes.
Practice set design
Use five to eight short clips per drill. After all clips, ask at least one question of each type:
- Which person said a specified statement?
- How many scenes shared an obvious feature?
- Which scene appeared earliest?
- What large object was behind a named person?
- Which statement changes one fact?
- Which scene included a specified action?
Increase the number of clips gradually toward 15. Keep the details readily observable and avoid turning practice into trivia. Review whether the miss came from encoding, source confusion, order, or retrieval.
Protect judgment performance
Memory is Section 3, but its source material belongs to Section 2. Do not become so preoccupied with background that you miss the scenario's core problem and rate actions poorly. The most efficient details often serve both tasks: speaker identity, request, reason, action, and setting. Use the permitted replay when a central detail was missed, then return to independent SJT ratings.
The official design rewards integrated observation: understand what is happening, notice what is readily visible, and retain how the scenes relate across time.
What is the best way to prepare for an official-style question asking which scenario appeared earliest?