6 micro-decisions per IMAT question: a 100-minute decision framework

IMAT time management is not a single number you memorise on the morning of the test. It is a chain of small decisions made in a 100-minute window, with 60 questions spread across four sections of unequal difficulty. The first decision is whether to attempt a question at all; the second is how long to stay before flagging; the third is when to return; the fourth is what to sacrifice when a single hard problem eats seven minutes. Candidates who treat pacing as one global budget of 100 seconds per question usually run out of time on General Knowledge or finish Section 1 with stamina debt that bleeds into Section 4. The workable mental model is a per-section decision tree, anchored to realistic minute budgets and a strict triage threshold for each item family. The framework below is built around the IMAT exam format, with the 60-question / 100-minute envelope treated as four sub-budgets rather than one. Each sub-budget absorbs the section's native demand, a small contingency reserve, and a hard stop that triggers the next question. Most candidates reading this will lose marks not because they ran out of knowledge, but because the minute budget on the harder sections was never calibrated to the actual question mix they saw on test day.

Why the 100-minute envelope is misleading as a single budget

On paper, 60 questions in 100 minutes gives 100 seconds per question, and many candidates who prepare with that headline figure alone end the exam with three or four unanswered items in the last section. The arithmetic is correct, but the assumption that the four sections behave as a single queue of equivalent items is wrong. Section 1, Logical Reasoning and General Knowledge, contains ten mixed items that often include a long reading passage on culture, current affairs, or a quote, plus a small set of pure logic items that look short but contain a hidden two-step inference. Section 2, Biology, contains roughly 18 items, several of which demand a labelled diagram read against a passage. Section 3, Chemistry, contains roughly 12 items where stoichiometry and equilibrium calculations routinely take three to four minutes each. Section 4, Physics and Mathematics, contains roughly 8 + 4 items where a single mechanics problem can swallow five minutes if the candidate starts with the wrong variable.

When you treat the envelope as one budget, two failure modes show up. The first is over-investment in Section 1, where a candidate lingers on a hard quote-interpretation item because they are still fresh and treat the start of the test as low-stakes warm-up. The second is under-investment in Section 4, where the same candidate skips the multi-step physics item at minute 88 because the visible clock has triggered panic. The remedy is to plan four sub-budgets in advance, and to write them on the cover page during the first 30 seconds of the test. For most candidates I have worked with, a workable split is roughly 12 minutes for Section 1, 30 minutes for Section 2, 24 minutes for Section 3, and 28 minutes for Section 4, with a 6-minute contingency reserve that is held back from the start and only released after the first pass through each section is complete. That reserve is the single most important number on the page, because it is the only buffer that lets you re-enter a hard question without rewinding the whole schedule.

The sub-budget split also changes how you think about triage. A 100-second global budget implies that every question deserves the same investment, which is mathematically true but psychologically dishonest. In practice, a 45-second general knowledge item and a 4-minute equilibrium calculation are not the same investment, and forcing them into the same time box produces a queue of half-finished items at the 80-minute mark. The fix is to teach yourself a per-section hard stop, then trust it. A hard stop is a clock reading past which you flag the question and move on, regardless of how close you are to the answer. The hard stop is the most useful single device in the whole framework, because it removes the largest hidden cost of pacing, which is in-decision time after the productive work is done.

A four-section minute map and where the reserves should sit

The minute map below is the working schedule I would commit to if I were writing the cover page of an IMAT booklet. It assumes a 100-minute envelope, four sections, and a single 6-minute contingency block that is held back from the start. Read it as a contract with yourself, not as a forecast of how the test will feel in the moment.

Two numbers in the table are worth defending. The first is the 110-second hard stop for Biology, which is 10 seconds above the global average. Biology items look short, but a sizeable share of them come bundled with a diagram, a passage, or a non-obvious vocabulary term, and the difference between a 90-second cap and a 110-second cap is the difference between finishing Section 2 with three flags versus finishing it with seven. The second is the 180-second hard stop for the four Mathematics items in Section 4. These are the most expensive items on the test by expected minutes-per-mark, and the most tempting to chase. A 180-second cap is a deliberate trade: you accept that one Math item will be flagged and returned to, in exchange for never being trapped inside a quadratic at minute 94.

The contingency reserve is the part most candidates leave out. Held back from the start, it has three uses, in this priority order: re-entry on a flagged Section 3 or 4 item where the hard-stop pass produced a strong second-read hypothesis; a recovery pass on Section 1 if the first pass went badly; and a final 60-second sweep to mark a guess on any item still blank. The reserve is not for thinking time on a fresh question. By the time the reserve is being deployed, the decision tree is already running, and every minute in the reserve is being spent on items that have a real probability of being correct. A reserve that is used up by minute 75 is a sign the minute map above was not honoured in Sections 1 and 2.

One tactical note on the reserve that is worth saying out loud. The reserve is a single block, not four small ones. Splitting the 6 minutes into four 90-second cushions and attaching one to each section feels safer, but in practice it produces four small allowances that are spent early on items that are not actually decision-relevant. Holding it back converts the reserve from a comfort blanket into a triage tool. Use it on the question with the highest expected value of an extra minute, not on the question that is currently in front of you.

Question-type triage: which items deserve your full minutes

Triage in IMAT is not guessing. It is a pre-committed rule about which items get the full minute budget, which get a fast read-and-eliminate pass, and which get a flag-and-return. The rule is built from three observable properties of each item: its expected solve time, its dependency on a diagram or passage, and the cost of being wrong. The cost of being wrong is constant across the test, because there is no negative marking, but the expected solve time and the dependency on stimulus material vary sharply between sections and even between items within a section.

• Section 1, fast-eliminate tier: pure logic items with a single quantifier shift, a one-step syllogism, or a direct factual recall item. These should be answered in 30 to 45 seconds, and the discipline is to not reread them.
• Section 1, full-minute tier: quote-interpretation items, items that combine a passage with a non-obvious inference, and items where two answer choices survive elimination. The full-minute tier is where the 75-second hard stop does its work.
• Section 2, fast-eliminate tier: one-word identification items on cell biology, Mendelian ratios, and standard organ-system functions. These are the workhorses of the section, and they should be answered in 40 to 60 seconds.
• Section 2, full-minute tier: items bundled with a labelled diagram, items where the answer depends on a passage, and items that mix two subdomains such as immunology and histology. The 110-second hard stop applies.
• Section 3, full-minute tier by default: stoichiometry, equilibrium, acid-base, and redox calculations. There is no fast-eliminate tier in Chemistry that is reliable across papers, and candidates who try to build one usually misread a passage item as a recall item.
• Section 4, tier A (Physics): mechanics and electromagnetism items with a single diagram. Target 130 to 150 seconds.
• Section 4, tier B (Mathematics): algebra, functions, probability, and geometry. Target 150 to 180 seconds, with the 180-second cap applied to a single item per paper.

The fast-eliminate tier in Section 1 is where most candidates can recover the largest block of minutes, because General Knowledge items in particular are designed to reward the candidate who knows the fact and penalises the one who reads the stem twice. The cost of the fast-eliminate tier is over-confidence, which is why the hard stop is paired with it. If a fast-eliminate item is not closed at 45 seconds, it has revealed itself as a full-minute item, and the rule has to be applied honestly.

The first-pass contract: read once, decide once, mark once

The first pass through a section is where the minute map is either honoured or destroyed. The contract for the first pass is simple to state and hard to execute: read the stem once, decide once, mark once. Three violations of this contract are common, and each one has a measurable cost in minutes. The first violation is the second reading of the stem. Candidates who second-read the stem are usually looking for reassurance, not information, and the second read consumes 20 to 30 seconds without changing the answer. The second violation is the in-stem pause, where the candidate stops midway through reading to consult memory, and the third violation is the post-decision hover, where the candidate returns to the answer they just marked and reconsiders.

The mechanical remedy is the read-once discipline. Read the stem from first word to last word, and only after the last word do you look at the answer choices. The five options in IMAT are not a feature of the stem, and they should not be read while the stem is being read. The second mechanical remedy is the mark-once rule. Once the answer sheet is filled, the item is closed. If a hard stop is reached and the answer is not yet final, the candidate marks the best guess and flags the item for re-entry. Marking the best guess is not optional, because the cost of leaving a question blank is higher than the cost of marking a wrong answer on a test with no negative marking.

Read-once and mark-once together protect the minute map by removing the two largest hidden time sinks: in-decision time after the productive work is done, and post-decision time spent revisiting a closed item. They also produce a clean psychological state at the end of the first pass, which is when the contingency reserve is most useful. Candidates who have violated the contract in Sections 1 and 2 typically arrive at Section 4 with no reserve and a flag list of nine items, which is not a recoverable state. The contract is not a productivity trick, it is the precondition for triage to function.

Common pitfalls and how to avoid them

The minute map and the triage tiers above fail in predictable ways. The list below is the short version of the most common failures, and for each one a tactical counter that fits inside the 100-minute envelope. These are drawn from the same mistake class that shows up across most candidates I have worked with, and they are the failures that are most likely to cost a candidate a full point on a 90th-percentile paper.

• The Section 1 rabbit hole. A long quote-interpretation item looks interesting and the candidate commits five minutes. Counter: pre-commit the 75-second hard stop, and if the item is still open at the stop, mark the most defensible of the two surviving answer choices and flag it.
• The Section 2 diagram misread. The candidate counts the wrong feature on a labelled diagram and solves a clean problem against the wrong stimulus. Counter: any item with a diagram gets a 10-second diagram audit at the start, before the stem is read.
• The Section 3 calculation drift. The candidate starts a stoichiometry problem, reaches an intermediate, second-guesses the intermediate, and restarts. Counter: the read-once rule, plus a hard stop at 130 seconds, with a flag rather than a restart.
• The Section 4 chasing tail. The candidate falls in love with a single Physics item and chases it for six minutes. Counter: the 150-second cap on Physics items, applied without exception. The cap is more important than the answer.
• The hidden Section 1 tax. General Knowledge items in Section 1 consume more minutes than candidates plan for, because they reward breadth, not depth. Counter: budget 12 minutes for Section 1, not 10, and accept that two items will be best-guess flagged.
• The empty reserve. The reserve is spent on a Section 1 item that was not actually a triage candidate. Counter: the reserve is a single block, used only after the first pass, and only on items where a second read has a high expected value.
• The post-decision hover. The candidate returns to a marked item and re-marks it. Counter: the mark-once rule, and the recognition that the second mark is almost always worse than the first, because the candidate has lost the context of the first read.

Two of these pitfalls deserve a second mention because they are the ones that consistently cost candidates a full point on a 90th-percentile paper. The first is the Section 4 chasing tail, which is the most expensive single mistake on the test, because it burns minutes that are needed for the next item and for the reserve. The second is the empty reserve, which is the most common structural mistake, because the reserve is not visible until it is missing. A candidate who reaches minute 88 with no reserve and three flagged Section 4 items has lost the test in Section 1, not in Section 4, even though the panic shows up in Section 4.

Re-entry protocol: how to spend the contingency reserve

The contingency reserve is deployed in a single pass after the first read of every section is complete. The re-entry pass is not a re-read of the whole test. It is a targeted pass on the flagged list, with a strict order of priority. The order is built from the expected value of an extra minute, not from the order in which the items appeared on the test.

1. First priority: a Section 3 or 4 item where the first pass produced a strong hypothesis that was not yet confirmed. These items are the highest expected value, because the productive work is already done and the extra minute converts a guess into a defended answer.
2. Second priority: a Section 2 item bundled with a diagram where the first pass skipped the diagram audit. A 10-second diagram audit on a flagged item is one of the highest-yield uses of reserve time.
3. Third priority: a Section 1 item where two answer choices survived elimination. The mark-once rule means the first mark is already on the answer sheet, and a re-read can break the tie.
4. Fourth priority: a Section 4 Mathematics item where the first pass was abandoned at the 180-second cap. These are the hardest items to recover, and they are last on the priority list for that reason.
5. Fifth priority: any item still blank. With roughly 30 to 60 seconds left in the reserve, a final sweep marks a guess on any item that has not been answered. There is no penalty for a wrong answer, and an unanswered item is the only outcome worse than a guess.

The re-entry protocol has a feature worth flagging explicitly. It is not symmetric across sections. Section 1 items get a smaller share of the reserve than their item count would suggest, because Section 1 items have the lowest expected value of an extra minute. A Section 1 item that survived the first pass is, by construction, the kind of item where extra time has already been spent. A Section 3 or 4 item that survived the first pass is, by construction, the kind of item where the productive work was done in the first read and the second read converts the work into a mark. The asymmetry is not a preference, it is a reflection of where the productive minutes are.

Stamina and section transitions

Stamina is the silent variable in the 100-minute envelope. Sections 1 and 2 are where stamina is built or destroyed, and Sections 3 and 4 are where it is spent. The transition from Section 2 to Section 3 is the highest-leverage transition on the test, because the question type changes from recognition-heavy Biology to calculation-heavy Chemistry, and the cognitive demand changes with it. Candidates who arrive at Section 3 still solving in Section 2 mode tend to under-read the Chemistry stem and over-rely on recall, and the cost shows up in the stoichiometry items.

The practical advice is short. At the Section 2 to Section 3 transition, take three to five seconds to mentally reframe the work. Section 3 is calculation-heavy, the hard stop is 130 seconds, and the read-once rule is the most important device in the section. At the Section 3 to Section 4 transition, reframe again. Section 4 is the section where the 150-second and 180-second caps are applied without exception, and the temptation to chase is highest. The transition is not a rest, and it is not a moment to reread the minute map. It is a 3-second reset that protects the next 28 minutes from the previous section's habits.

How the minute map interacts with IMAT scoring

The IMAT scoring system is built around a single ranked score, with no penalty for wrong answers, which is the structural feature that makes the minute map above safe to use. A wrong answer and an unanswered item produce the same contribution to the final score, but a flagged item that is re-entered in the reserve has a strictly higher expected contribution than an item that is left blank at the end of the test. The minute map is designed to convert the reserve into marks on the items where the expected value of an extra minute is highest, and to leave the items where the expected value is lowest in the best-guess state produced by the first pass.

This is also the reason the hard stops are calibrated to the section rather than to a global 100-second average. The global average treats every item as having the same expected solve time, which the scoring system does not reward. The scoring system rewards marks, and marks are concentrated in the items where the candidate has a real probability of being correct given an extra minute. The hard stops are calibrated to surface those items, and the reserve is calibrated to spend the extra minutes on them.

Adapting the minute map to your own paper

The minute map above is a starting point, not a prescription. The right way to use it is to commit it to memory, run it on a full timed IMAT paper, and then adjust the section budgets by the differences you observe. A candidate whose Section 1 first pass consistently takes 16 minutes should not lower the Section 2 budget to compensate. They should accept that Section 1 is genuinely a 16-minute section for them, and rebalance the reserve and the Section 4 budget to match. The minute map is a personal tool, and the only version of it that matters is the one that has been calibrated against your own first-pass times.

Two diagnostic questions are worth running after every timed paper. First, which two items consumed the most minutes on the first pass, and were they Section 3 or 4 items? If the answer is yes, the hard stops are working. If the answer is no, the hard stops are being violated. Second, how many minutes of reserve were actually deployed, and on which items? If the reserve was deployed on Section 1 items, the re-entry protocol is being violated. If the reserve was unused, the minute map is too generous and the surplus should be redistributed. The diagnostics are how the minute map becomes a personal instrument, and they are the part of the preparation that cannot be outsourced to a textbook.

Putting it together: a 100-minute walkthrough

Imagine the first 100 minutes of test day, with the minute map on the cover page. The first 30 seconds are spent writing the four sub-budgets and the 6-minute reserve on the booklet cover. Minutes 0 to 12 are Section 1, in read-once, mark-once mode, with the 75-second hard stop applied to every item. The result is a flagged list of one or two Section 1 items, and roughly 10 minutes used. Minutes 12 to 42 are Section 2, in fast-eliminate mode for the recognition items and full-minute mode for the diagram and passage items, with the 110-second hard stop. The result is a flagged list of three to five Section 2 items, and 30 minutes used. Minutes 42 to 66 are Section 3, in full-minute mode by default, with the 130-second hard stop. The result is a flagged list of two to four Section 3 items, and 24 minutes used. Minutes 66 to 94 are Section 4, with the 150-second Physics cap and the 180-second Mathematics cap, and 28 minutes used. Minutes 94 to 100 are the contingency reserve, deployed in the priority order above, with the final 60 seconds reserved for the blank-sweep.

This is the 100-minute walkthrough that the minute map is designed to produce, and the only way to know whether it fits your own test-day behaviour is to run it on a full timed paper. The walkthrough is not a script. It is a structure, and the structure's job is to keep the productive work in productive minutes and the triage decisions in their correct priority order. Most candidates reading this will recognise at least one of the pitfalls above as a habit they already have, and the minute map is the device that makes the habit visible before it costs a point on test day.

TestPrep İstanbul's IMAT preparation programme builds the minute map into the diagnostic phase, so the per-section budgets are calibrated against the candidate's own first-pass times rather than borrowed from a generic template. The framework above is the starting point, and the personal calibration is what makes it work on a real paper.

Frequently asked questions