How to study for the LSAT, according to how learning actually works

Most LSAT study advice is folklore passed between people who did well, which is a poor evidence base — the survivors are not a random sample. There is, however, a substantial research literature on how people acquire difficult skills, and a fair amount of it contradicts what LSAT prep normally looks like.

None of what follows is a promise about a score. Study method is one input among many, and no honest source can tell you what a particular routine will produce for a particular person. What the research supports is narrower and still useful: some ways of spending an hour leave more behind than others.

Testing yourself beats reviewing, by a lot

The most robust finding in the field is that retrieving information from memory strengthens it far more than re-reading it. Reviewing an explanation you already understand feels productive because it is fluent — nothing is difficult, so everything seems learned. Attempting a question cold feels worse and works better.

In practice: do the question before reading anything about how to do it. Get it wrong. Then read the explanation, which now lands on a question you have an actual stake in rather than a hypothetical one. The order matters more than the material.

Spacing beats massing

The same total study time produces more durable learning distributed across days than concentrated into one sitting. Four sessions of forty minutes across a week outperforms one long Saturday, reliably, in study after study — even though the long Saturday feels like more serious work.

The mechanism is uncomfortable but useful to know: some forgetting between sessions is not waste. Retrieving something you have partly lost is exactly the effortful act that consolidates it. A schedule that never lets you forget anything never lets you practice the recovery.

Mixing question types beats blocking them

This is the finding that most directly contradicts standard LSAT practice. The usual approach is a block of thirty Assumption questions, then a block of thirty Flaw. Research on interleaving consistently finds that mixing types during practice produces worse performance during the session and better performance on a later test.

The reason is specific to what a mixed set requires. When every question is an Assumption question, you never practice deciding what kind of question you are looking at — you practice executing a method you were handed. On the real exam nobody hands it to you. Blocked practice quietly omits the step that the test actually charges you for.

The practical compromise most of the literature supports: block briefly when a type is genuinely new to you, then mix. Blocking is for acquisition, mixing is for retention and transfer, and the mistake is staying in the first phase because it feels better.

Worked examples first, then fade the support

For a skill you do not yet have, studying a fully worked solution is more efficient than struggling through unaided problems — the struggle consumes attention that is not yet available for learning the method. But the advantage reverses as competence grows: past a point, continuing to study worked examples is less effective than solving.

This argues for a specific shape to a study plan. Full walkthroughs early in a question type. Then partial ones — the first step given, the rest yours. Then unaided practice with the explanation available afterward. Staying on full walkthroughs after you no longer need them is a real cost, not merely a neutral one.

Explaining why the wrong answers are wrong

On a five-choice question, checking that you picked the right answer uses about a fifth of the available information. The far more valuable review question is why each of the other four is wrong, and specifically what it was designed to catch — a reversal, an overstatement, a true-but-irrelevant fact, an answer to a question that was not asked.

Distractors are not random. They are written to catch identifiable errors, and the same errors recur. A student who can name why they were tempted has learned something that transfers to every future question built on that temptation. A student who only knows they got it wrong has learned one item.

What this adds up to

• Attempt before you read. The failed attempt is what makes the explanation stick.
• Shorter sessions, more days. Distributed practice beats concentrated practice at equal total time.
• Mix question types once a type is no longer new. Sorting the question is part of the skill.
• Fade the scaffolding deliberately — full walkthrough, then partial, then unaided.
• Review the four wrong answers, not just the one right one. Name what each was built to catch.
• Take timed sections under conditions resembling the real thing, but not daily. They measure; they do not teach much on their own.

One honest caveat about all of the above. This research is drawn largely from studies of other skills and other populations, and applying it to LSAT preparation is a reasonable inference rather than a measured result. It is a better foundation than folklore. It is not a guarantee, and anyone offering you one about a standardized test is not being straight with you.