ATSA Logical Reasoning: Word Problems Explained

The ATSA logical reasoning section, which candidates often call "the word problems" or "deductive reasoning," gives you a set of rules, such as who can be scheduled when or which object sits north of which, and asks what must be true, what could be true, or what follows if one more condition is added. Several candidate accounts name it as the hardest part of the test for them. It rewards a consistent method much more than raw speed.

The FAA doesn't publish descriptions of individual ATSA subtests. What follows is based on candidate reports and is hedged; your test may differ. All example puzzles on this page are original.

What is publicly known

Pearson VUE's FAA registration page described the ATSA as an aptitude assessment "consisting of multiple subtests" measuring "decision-making ability, spatial awareness, multitasking, and working memory, among others," with a maximum test duration of 2 hours 49 minutes plus up to 30 minutes of break time. A 2021 DOT privacy impact assessment says the content comes from two commercial vendors, Airways International Limited and SHL (the OIG audit calls them Airways New Zealand and SHL/CEB), and that subtest scores are combined by a proprietary algorithm into a single ATSA score. The DOT Office of Inspector General's 2023 audit found the FAA selected 7 of 15 candidate assessments for the battery after validating them against the job performance of certified controllers and the grades of Academy students.

None of those documents lists a logic subtest by name or gives its weight. The FAA's own position, stated on the same Pearson VUE page, is that the ATSA is an aptitude test for which "no studying or preparation is necessary." Familiarity with the format still helps many people avoid wasting time working out what a question is asking.

What candidates report about the format

• Rule-based word problems. A 2017 test taker described items like "If runway A is South of B, runway C is in between A and D, runway D is north of A, which of the following statements MUST be true?" A December 2024 account described a scheduling set: several interview times, rules such as one person must go before another, someone can't be last, someone can't take a certain slot, and then questions like who can take 10 a.m.
• Roughly 18 to 20 questions in about 20 minutes. The 2017 account said 18 questions in 20 minutes. A 2024 account mentioned 20 questions. Treat both as recollections that may not match the current version.
• Time pressure is a common complaint. Candidates describe running out of time, rushing the final questions or feeling mentally tired by this point, since it's usually reported near the end of the session.

We couldn't confirm from FAA material or candidate accounts whether you can go back to earlier questions in this section, so it's sensible to practice finishing each question before moving on.

What the section likely measures

Controllers apply sets of rules to changing situations, like separation standards, sequencing and restrictions, and they have to see what those rules allow and what they rule out. A subtest built on deductive reasoning under time pressure fits the "decision-making ability" Pearson VUE listed. That's our inference. The FAA hasn't said which skill each subtest targets.

A fully solved example (original)

This puzzle was written for this guide to show the reasoning style. It isn't an ATSA item.

Five trainees, Avery, Blake, Cruz, Dana and Ellis, each get one simulator slot at 8, 9, 10, 11 or 12 o'clock. No two share a slot.

1. Avery is scheduled earlier than Blake.
2. Cruz is at 10 or 11.
3. Dana is immediately after Cruz.
4. Ellis is not at 8.

Step 1: Build the cases before reading any question

Start from the most restrictive rules. Rules 2 and 3 together make a two-slot block: either Cruz at 10 and Dana at 11 (Case 1), or Cruz at 11 and Dana at 12 (Case 2). Then place the others.

• Case 1 (Cruz 10, Dana 11): Avery, Blake and Ellis share 8, 9 and 12. Ellis can't be 8 and Avery must be before Blake, so the options are Avery 8, Blake 9, Ellis 12, or Avery 8, Ellis 9, Blake 12.
• Case 2 (Cruz 11, Dana 12): Avery, Blake and Ellis share 8, 9 and 10. The options are Avery 8, Blake 9, Ellis 10, or Avery 8, Ellis 9, Blake 10.

That's four valid schedules in total, and in every one of them Avery is at 8. You can see why: Ellis can't be at 8 (rule 4), Cruz and Dana can't be (rules 2 and 3), and Blake can't be because Avery has to be earlier. Finding a fixed position like that early makes every later question faster.

Question 1: If Blake is at 9, who is at 10?

A. Avery
B. Cruz
C. Ellis
D. It could be either Cruz or Ellis

Answer: D. With Blake at 9, both Case 1 (Cruz at 10, Ellis at 12) and Case 2 (Ellis at 10, Cruz at 11) still work. An "if" question adds a condition, and you only need to check the schedules that survive it.

Question 2: Which of the following must be true?

A. Dana is at 12.
B. Ellis is at 9.
C. Avery is at 8.
D. Blake is earlier than Cruz.

Answer: C. It holds in all four schedules. A fails in Case 1, and B fails in the schedule with Blake at 9. D fails when Blake is at 12 and Cruz is at 10. To reject a must-be-true option, one counterexample is enough.

Question 3: Which of the following could be true?

A. Ellis is at 11.
B. Blake is at 11.
C. Ellis is at 12.
D. Cruz is at 12.

Answer: C. Avery 8, Blake 9, Cruz 10, Dana 11, Ellis 12 satisfies every rule. The 11 o'clock slot always belongs to Cruz or Dana, which eliminates A and B, and D breaks rule 2. To accept a could-be-true option, one valid example is enough.

A quick ordering item

Tower P is north of Tower Q. Tower R is south of Tower Q. Tower S is north of Tower R. Which must be true: (A) S is north of Q, (B) P is north of R, or (C) S is south of P? B is the only certain one, because P is north of Q and Q is north of R. S only has to be north of R, so it could sit between R and Q, or north of P, which rules out both A and C.

A method you can repeat under time pressure

1. Write the rules in shorthand. "A < B," "C = 10/11," "CD block," "E ≠ 8." Shorthand is faster to scan than sentences and makes a forgotten rule easy to spot. If you're given something to write on at the test center, use it. If not, practice holding a simple slot diagram in your head.
2. Start with the most restrictive rule. Blocks ("immediately after"), fixed positions and two-option rules cut the possibilities fastest.
3. Split into cases when a rule has only two or three options. Two small diagrams are quicker and safer than one diagram full of question marks.
4. Look for forced positions. Ask which slots each person can't take. A deduction like "Avery must be 8" answers several questions at once.
5. Match the test to the question. Must be true means it holds in every case. Could be true means it holds in at least one. Cannot be true means it holds in none. "If" means add the condition and discard the cases that break it.
6. Reuse your work. Questions in the same set share the rules, so your case diagram stays valid for all of them.

Mistakes that cost the most points

• Reading "before" as "immediately before." "Avery is earlier than Blake" allows a gap, while "Dana is immediately after Cruz" doesn't.
• Carrying a condition over from an "if" question. "If Blake is at 9" applies only to that question.
• Treating could as must. Finding one schedule where an option works proves it could be true, not that it must be.
• Re-solving from scratch for every question. This is where most of the time goes.
• Getting stuck. If a question isn't resolving after a reasonable try, pick the best-supported option and move on. Candidates who ran short of time describe rushing the last few questions, and a stuck question can cost you several you could have answered.

How to practice

This section responds well to practice because the method becomes automatic. Work sets of scheduling and ordering puzzles against a timer, then review every miss and name the error: a misread rule, a missed case, or a could/must mix-up. Once untimed accuracy is solid, tighten the clock. It also helps to practice after a mentally tiring task, since candidates usually meet this section late in the session.

You can try a short logical reasoning demo with original puzzles. Full timed sets and mock runs are in the Complete plan. See pricing. Our puzzles are written to practice the reported format. They aren't copies of real ATSA items, and no amount of practice guarantees a particular band. The reading comprehension guide covers the same must-be-true discipline applied to prose, and how to prepare for the ATSA puts all seven sections into one plan.

Frequently asked questions

What kinds of logic questions are on the ATSA?

Candidates mostly describe rule-based word problems: scheduling (who takes which time slot) and ordering or position puzzles (which item is north or south of another), with questions about what must, could or cannot be true. The FAA doesn't publish a list of question types.

How long is the ATSA logical reasoning section?

The FAA doesn't publish per-section timings. Candidate accounts describe roughly 18 to 20 questions in about 20 minutes, which works out to about a minute per question.

Do I need formal logic training for this section?

No. The questions are written in plain language. What helps is a consistent method for turning rules into a quick diagram and checking options against every possible case.

Is the logic section weighted more heavily than the others?

That isn't public. The DOT's privacy impact assessment says subtest scores go through a proprietary algorithm to produce a single ATSA score, and candidates only see their band, not section scores.