No Products in the Cart
Back To School 2026 SPECIAL DISCOUNT CODE: WEBACK26 for 20% OFF
If you are preparing for the AP Psychology 2026 exam in the United States, the single most important thing to understand is this: the exam sitting in front of you on Tuesday, May 12, 2026 (12:00 p.m. local time) is not the exam your older siblings or favorite YouTube study channel took. The 2025 redesign shrunk the multiple-choice section from 100 questions to 75, retired the old free-response format, and installed two brand-new question types — the Article Analysis Question (AAQ) and the Evidence-Based Question (EBQ). Our close reading of the 2025 and 2026 papers shows exactly how the new exam behaves in practice:
This guide walks through real AP Psychology past papers from 2022–2026, quotes actual questions with their answers, and extracts the patterns that matter for your score. If you are hunting for authentic AP Psychology practice questions rather than imitations, you are in the right place.
| Section | Content | Time | Score weight |
|---|---|---|---|
| Section I — Multiple Choice | 75 questions, 4 answer choices each, scenario-driven | 90 minutes | 66.7% |
| Section II — Free Response | 1 Article Analysis Question (AAQ, 6 parts A–F) + 1 Evidence-Based Question (EBQ, 3 parts A–C) | 70 minutes | 33.3% |
The first two questions of the 2026 U.S. paper tell you everything about the redesign's philosophy. Here they are, verbatim, with the answers released alongside the paper.
Michaela is a researcher in a lab where she spends most of her days working quietly by herself. She shows little interest in speaking to others in the lab. When people praise her for her work, she seems uninterested. When her manager asks her about her career aspirations, she says she is fine just doing her work. She does not have any close relationships because she does not experience any pleasure from them.
Which personality disorder would best explain Michaela's symptoms?
A. Schizoid B. Histrionic C. Dependent D. Borderline
Answer: A — Schizoid
Expert analysis: This is the redesign in miniature. Nothing in the stem asks you to define schizoid personality disorder; instead you must diagnose from behavior — detachment, indifference to praise, no pleasure from relationships — and rule out three look-alike disorders. Histrionic (attention-seeking) and Dependent (clinginess) are its behavioral opposites, and Borderline involves instability rather than flat detachment. Question 1 being clinical psychology tells you the Mental and Physical Health unit is not a bolt-on; it leads the paper.
A local restaurant hires only individuals under 40 years old to work at their establishment. Which of the following concepts is best illustrated in this example?
A. Discrimination B. Self-serving bias C. Prejudice D. Out-group homogeneity bias
Answer: A — Discrimination
Expert analysis: This item tests the single most-confused distinction in the Social Psychology unit: prejudice is an attitude, discrimination is a behavior. Because the restaurant is acting — hiring only under-40s — the correct answer is discrimination. Students who memorize definitions without practicing scenario discrimination (pun intended) lose exactly this kind of point. Note the stem family: "Which of the following concepts is best illustrated in this example?" You will meet it dozens of times across the 75 questions.
We compared the 2026 U.S. paper against the 2026 International (Asia) and Late versions, the 2025 U.S. and International papers, and the pre-redesign 2022/2024 legacy papers. The table below uses only real year-and-question evidence from those papers.
| Year | Question | Topic | Difficulty | Pattern observed |
|---|---|---|---|---|
| 2026 U.S. | MCQ Q1 | Personality disorders (schizoid) | Medium | Clinical scenario opens the exam; diagnose-from-behavior |
| 2026 U.S. | MCQ Q2 | Discrimination vs. prejudice | Easy–Medium | "Best illustrated" scenario stem; social psych up front |
| 2026 U.S. | MCQ Q3 | Why a study is not an experiment (random assignment) | Medium | Methods reasoning over a real data table |
| 2026 U.S. | AAQ (items 77–84) | Cell-phone absence & anxiety; FOMO (Mannion & Nolan, 2020) | Hard | Real journal article; Part A = identify the research method |
| 2026 U.S. | EBQ (item 76) | Testing effect; retrieval of recently learned information | Medium | 3 sources; claim + evidence + reasoning |
| 2025 U.S. | MCQ Q1436 | Fluid intelligence & aging | Easy | Everyday-aging scenario; same application-first style |
| 2025 U.S. | MCQ Q1437 | Confirmation bias in relationships | Medium | "Best applies" stem family repeats year over year |
| 2025 U.S. | AAQ (items 1511–1516) | Misinformation & memory; mock-crime video (Peña, Klemfuss, Loftus & Mindthoff, 2017) | Hard | Same six-part AAQ skeleton; real stats table |
| 2025 U.S. | EBQ (items 1517–1519) | Social facilitation — whether the presence of others improves performance | Medium | Same three-part argument skeleton as 2026 |
| 2026 Asia | AAQ | MZ/DZ twin study of pain & opioid medication (Angst et al., 2012) | Hard | Within-subjects, double-blind design literacy required |
| 2026 Late | AAQ | Gratitude letters & happiness (Toepfer & Walker, 2009) | Medium–Hard | Longitudinal survey data; same Part A method question |
| 2022 / 2024 | Entire paper | Legacy format: 100 MCQ + 2 FRQs (concept application, research design) | — | Pre-redesign; useful for content drill, not for format rehearsal |
Read down the "Pattern observed" column and the conclusion assembles itself: the same stem families, the same AAQ/EBQ skeletons, and the same content priorities (clinical, social, research methods, cognition) recur in every administration we hold. The 2025 U.S. paper even shares a question-pool numbering range (items 1436–1519) with the same underlying item bank style used across versions — the content pools are parallel, and the skills are identical.
The 2025 U.S. paper was the first redesigned administration, which makes its first two questions a preview of everything 2026 doubled down on. Quoted verbatim:
Jenna's grandmother is beginning to experience a decline in her fluid intelligence. Which of the following activities is most likely to be affected by this decline?
A. Memorizing a poem B. Writing a letter to a friend C. Playing a previously learned, complex piece of music on the piano D. Solving a new and complex puzzle quickly
Answer: D — Solving a new and complex puzzle quickly
Expert analysis: Fluid intelligence = novel problem-solving and speed; crystallized intelligence = accumulated knowledge and well-practiced skills. Options A–C are all crystallized or rehearsed activities, so the decline bites hardest at option D. This is the cognition equivalent of the 2026 discrimination item: one clean distinction, four plausible traps.
People often do not see their romantic partners' flaws because their partners present their most positive selves at the beginning of a relationship. People tend to continue to notice the pleasant qualities that are consistent with their first impression of their partners and overlook the unpleasant qualities.
Which of the following concepts best applies to this situation?
A. Self-serving bias B. Confirmation bias C. Actor/observer bias D. Fundamental attribution error
Answer: B — Confirmation bias
Expert analysis: Noticing what fits your first impression and discarding what contradicts it is textbook confirmation bias. The three distractors are all attribution-family terms — exactly the cluster students blur together. Put this next to 2026 U.S. Q2 and the family resemblance is unmistakable: a two-sentence social scenario, four bias/bias-adjacent options, one precise distinction.
Now the two free-response questions from the 2026 U.S. paper — first the EBQ, then the AAQ.
"Using the sources provided, develop and justify an argument about what strengthens a person's ability to retrieve recently learned information."
Structure: Part A — propose a specific, defensible claim; Part B — support it with evidence from one source and explain the link using a psychological concept; Part C — add evidence from a different source and justify it with a different concept.
Expert analysis: The psychological core here is the testing effect — effortful retrieval beats rereading. A strong Part A claim might read: "Actively recalling information, rather than simply rereading it, significantly improves later retention." (That is an expert-written sample claim, not an official scoring guideline.) Notice how student-life-friendly the topic is: 2025 U.S. asked whether the presence of others improves performance; 2026 U.S. asks how to strengthen retrieval. The EBQ is where your study-skills knowledge literally pays you back.
"In this study, the researchers examined whether the absence of cell phones contributes to anxiety in college students." — 128 students at a midsize private U.S. university were randomly assigned to a Phone Present group (phone on the table, instructed not to use it) or a Phone Absent group (phone locked in a box). Both groups' phones rang audibly two minutes into a ten-minute waiting period; anxiety was then measured with a 20-question survey using a 4-point Likert scale ("I am tense," "I am worried"). Participants in the Phone Absent group reported significantly higher mean anxiety scores, and the researchers interpreted the gap through the lens of "fear of missing out" (FOMO). The article is a real publication: Mannion, K. H., & Nolan, S. A. (2020). Cogent Psychology, 7(1), Article 1869378.
Part A: "Identify the research method used in the study."
Answer: Experiment
Expert analysis: Random assignment to conditions plus a manipulated variable (phone access) makes this an experiment — full stop. But the AAQ does not stop at Part A. The six-part skeleton then asks for the operational definition of anxiety (scores on the 20-question Likert survey), interpretation of the statistically significant difference in means, an ethical guideline in action (informed consent and the right to withdraw; debriefing about the deception), generalizability (a single private-university sample), and finally whether the findings support the FOMO claim. Every 2025–2026 AAQ we hold — misinformation and memory (2025 U.S.), selfie behaviors in Chinese teens (2025 International), the twin opioid study (2026 Asia), gratitude letters (2026 Late) — runs through this identical six-step gauntlet.
Predicted difficulty: moderate-to-slightly-above-average. The 75-question MCQ section is friendlier in length than the old 100-question grind (72 seconds per question instead of 42), but the per-question cognitive load is higher because nearly every item is a mini-scenario. The free-response section is where scores are made: 14 rubric points across the AAQ and EBQ, worth a third of your composite.
Three takeaways from the evidence above. First, the 2026 U.S. exam is fully knowable: the stem families, the AAQ's fixed six parts, and the EBQ's claim–evidence–reasoning skeleton repeat across every 2025–2026 administration we hold, from the U.S. main paper to the Asia and Late versions. Second, the highest-yield preparation is not rereading a textbook — it is working through real papers under real timing, because the exam's difficulty lives in scenario application, not in obscure terminology. Third, every quoted question above comes from an authentic paper; each one you have already seen is a question you cannot be surprised by. Walk into Bluebook on May 12 having out-trained the format, and a 5 stops being a hope and starts being a plan.
Every question quoted in this guide — and hundreds more from the 2022–2026 U.S., International, and Late administrations — is waiting for you in the full collection.
Recent Popular Subjects
