Three tools, each valuable alone, each leaving a specific gap when used in isolation: a question bank builds knowledge without testing whether you can deploy it; a Tutor repairs misconceptions without checking they hold under pressure; a simulation tests performance without necessarily repairing what it reveals. Connected in one loop, each feeds the next, and the whole takes less time than running three disconnected routines.
What each tool does and does not do
Question bank: builds and tests knowledge through retrieval practice, reveals gaps at the level of individual concepts, and does not test whether that knowledge translates into performance under time pressure with a patient in front of you. Tutor: works through the reasoning behind a wrong answer rather than supplying the right one, repairs the underlying misconception, and does not itself test whether the repair survives a live encounter. Simulation: tests knowledge deployed as performance, reveals gaps in structure, communication, safety and pacing, and produces a transcript, but a weak performance not connected to remediation simply repeats.
The loop
Learn: question bank practice reveals a knowledge gap. Retrieve: Tutor works through the misconception underneath it. Perform: a simulation on related content tests whether the repaired knowledge deploys under pressure. Inspect: the transcript shows exactly what happened. Remediate: Tutor addresses whatever the performance revealed, which may be a communication or structural issue rather than the original knowledge gap. Repeat: a prescribed next case tests whether that correction held.
Why this takes less time, not more
The tools are not run as three separate routines; each session moves through the loop once. A short question block reveals a gap, a brief Tutor exchange repairs it, a single simulation tests it, a quick transcript review identifies the next target. Compared with running an hour of questions, an hour of separate tutoring and an hour of unconnected simulation, the connected loop spends less total time and every minute of it is directed by what the previous stage revealed.
A weekly rhythm
Three or four short weekday sessions: a question block, one Tutor exchange on the most significant miss, one simulation on related content, a brief transcript review, one target recorded for next time. One longer weekend session: a themed mini-circuit or full circuit in exam mode, reviewed for patterns across stations, with Tutor remediation on the most persistent one. Throughout: spaced retesting on previously repaired gaps, checking that corrections have become durable.
What happens without the loop
A candidate running the three tools separately, an hour of questions on Monday, a Tutor session whenever a gap is noticed, a simulation attempt disconnected from either, experiences each as a standalone task with its own setup cost and no memory of what the others revealed. A knowledge gap identified in Monday's questions may never reach a relevant simulation at all; a weak simulation performance may never trace back to the specific misconception underlying it. The loop's value is not any single tool working better in isolation, it is that the output of each stage becomes the input to the next automatically, removing the manual effort of connecting them yourself and ensuring nothing identified simply falls through the gap between tools.
Adapting the loop to limited time
On a genuinely constrained day, the full loop does not need every stage to add value: even a single question missed, twenty seconds spent naming why, and one simulated exchange testing the corrected concept captures most of the loop's benefit in under five minutes. The discipline that matters is not running every stage exhaustively every day, it is never treating a stage as complete without at least briefly connecting it forward to the next one, since an isolated question bank session that never reaches remediation or testing is exactly the pattern the loop exists to prevent.
Treat the loop as a habit to build gradually rather than a rigid protocol to follow perfectly from day one; even partial, imperfect use of the connection between these three tools outperforms running them as entirely separate, disconnected routines.
Frequently asked questions
Does every question-bank miss need a simulation?
No: the loop connects the tools, it does not require every stage for every gap; a factual miss may need only Tutor repair and a later retest, while a reasoning or performance gap is where simulation adds specific value.
What if the simulation reveals a different problem from the original knowledge gap?
That is the loop working as designed: knowledge and performance gaps are different, and simulation frequently reveals a communication or structural issue the question bank could never have surfaced, which then becomes the next remediation target.
Can this loop run on the app during a busy week?
Yes: the weekday sessions fit short windows through the app, with the longer circuit reserved for a desktop weekend session, and progress carries across both.