Verbal Reasoning on the Psychometric test

What it is testing

Three things at once: how much language you have, how carefully you read, and whether you can tell what follows from a text and what merely sounds as though it does. The last is where most of the difficulty lives — a plausible sentence the passage does not actually support is the standard wrong answer.

The question types

Analogies

A pair of words, and four pairs to choose from. The task is to name the relation in the first pair precisely enough to test it against the others — “a tool and what it is used on” discriminates; “they go together” does not. Vocabulary matters, but a precise relation beats a large vocabulary on most items.

Logic and inference

Short passages followed by a question about what can be concluded, what would weaken an argument, or which assumption it rests on. Everything needed is in the passage; outside knowledge is usually the trap rather than the route.

Reading comprehension

A longer text — often academic, often on a subject you have never studied — followed by several questions. The unfamiliarity is deliberate and is not a disadvantage: the answers are all in the text.

Where marks are usually lost

• Answering from knowledge instead of from the passage. If you happen to know the topic, that is a risk here rather than an advantage.
• Nearly-right alternatives. An option that is true but does not answer the question asked is the most common wrong choice.
• Absolutes. Always, only, never — a strong claim needs strong support in the text.
• Reading the passage twice. The section is about 20 minutes long; a second full read is rarely affordable.

How Vizo practises it

A verbal item cannot be checked by arithmetic the way a quantitative one can, so it goes through a different gate: a written rationale for the correct answer, a stated reason each distractor is wrong, and rejection of any item whose alternatives are not mutually exclusive. Every passage and every question is original and written for Vizo in Hebrew or Arabic — the two languages the December 2026 sitting is offered in.

Wrong answers are classified by the kind of mistake rather than only marked wrong, and return for review on a spacing schedule. See the writing task for the other component of this domain score, and Quantitative Reasoning for the other half of the general score.