Shrink the first move below the resistance.

Because the math of procrastination changes. Piers Steel's Temporal Motivation Theory frames procrastination as a function of task aversiveness and delay vs expected value: when an exam is six weeks away, the perceived delay is huge and the brain discounts the task to near zero, so anything else wins. During finals week the delay collapses and the value spikes, but task aversiveness also spikes, because you are now staring at 8 to 12 lectures you have not opened and the perceived effort to start is enormous. The early-semester version is a delay problem; the finals-week version is an aversiveness problem. Most procrastination advice (Pomodoros, time-blocking, quiet rooms) does not address aversiveness, it addresses delay. That is why those tactics stop working in finals week even when they worked in October.

Both. ADHD makes the task-initiation drag larger and the dopamine of switching to your phone more rewarding, so the effect is bigger and earlier. Neurotypical students also procrastinate during finals week, just later in the cycle and usually with more self-loathing about it. A 2007 meta-analysis by Steel put the lifetime prevalence of academic procrastination at around 80 to 95% among college students. You are not broken; you are inside the normal distribution. What changes the outcome is not whether you procrastinate but how small you can make the first move when you finally sit down.

Pomodoro is a session-pacing tool, not a session-starting tool. It assumes you are already at the desk doing the work and you need help with rhythm and break timing. The actual failure mode of finals-week procrastination is upstream of that: it is the 47 minutes between deciding to study and opening the first slide. Pomodoro does nothing about those 47 minutes. Worse, the Pomodoro framing (work in 25-minute blocks) sets the activation bar at 25 minutes, which is plenty large to keep procrastinating against. The thing that works is committing to 5 minutes of an active task, then negotiating from there. If 5 minutes is still too much, drop to one card.

Two systems compete. The reflective system knows the exam is Friday and you should study. The reactive system runs on immediate reward gradients and notices that the slide deck is gray and effortful while Instagram is colorful and zero-cost. The reactive system wins because the rewards are immediate. The standard intervention is to add discipline to the reflective side; this rarely works under finals-week stress because cognitive load is already maxed out. The intervention that does work is reducing the cost gradient on the studying side, until the reactive system stops finding studying aversive enough to avoid. A 5-minute drill with visible progress on a tree fights the reactive system on its own terms. A 4-hour reread block does not.

It is not enough on day one. It is plenty on day five. The Cramming Procrastinator pattern Studyly built around looks like this: open the app for the first time on Sunday, do five minutes on one deck because that is all you have in you, get a tree to stage 2, close it. Monday do ten minutes because the tree from Sunday is still there. By Thursday you are doing 45 minutes a sitting because each sitting now produces visible movement on six trees, and the dashboard looks like a forest you are not willing to abandon. The mechanic is not 'five minutes is enough.' It is 'five minutes today produces five minutes plus momentum tomorrow.' The version where you wait for a 4-hour block produces zero minutes every day until exam morning.

Different page, same product. The night-before and three-night protocol live in our cramming guides: see the all-nighter and retrieval-vs-rereading writeups for the 6-hour and 18-hour versions. The short answer at the 3-day mark is: dump every PDF you have not opened into one drop, let them convert in parallel (about 60 seconds per 90-slide deck for around 200 cards), and start drilling MCQ cold on the deck for the earliest exam. The miss list from that first drill is your actual study plan, not the syllabus.

It is childish in the sense that mobile games are childish, and mobile games are the most successful behavior-shaping technology of the last 20 years. The variable that breaks finals-week study sessions is not the academic difficulty of the material, it is the absence of visible progress. Highlighting a textbook for an hour produces a slightly more colorful textbook. Drilling 30 cards in Studyly grows the tree on that deck from stage 2 to stage 3, and you can see the difference. The pure-pedagogy literature ignores this because it is not a pedagogy variable. The behavior-design literature does not ignore it. We borrowed from the second pile.

It is built around the same loop, with one critical structural difference. Forest and other meditation apps grow a tree if you do not touch your phone for 25 minutes. Studyly grows a tree because you completed retrieval drills against your own slides. The reward is contingent on the work, not on phone abstinence. The point is identical (visible progress sustains the session) and the substrate is different (real practice questions on your own lecture material, scored against a published rubric, with auto-rephrasing on revisit so you cannot pattern-match the second pass).

Same intervention, smaller. If 5 minutes feels like too much, the right starting move is one card. Open one deck, answer one MCQ, close it. The neurochemistry of anxiety-driven avoidance is that the imagined task (3 hours of unfamiliar material under time pressure) triggers the same threat response as a real danger and the brain freezes. A single card cannot trigger that response because the perceived effort is too small to be threatening. Doing one card breaks the freeze. Doing one card a day for the first two days of finals week is genuinely better than waiting for a 4-hour block that never materializes.

Two structural differences and one behavioral one. Structural: Studyly enforces a four-criterion rubric (factual correctness, stem clarity, distractor plausibility, question-type coverage) before each card ships, scoring 81.3 on a held-out three-document eval where Turbolearn scored 57.8 on the same documents and rubric; ChatGPT enforces no rubric and distractors are often one-obvious-right. Studyly also auto-rephrases the stem on every revisit so you cannot memorize the wording, which is what kills static-deck cramming. Behavioral: opening Studyly and uploading is one act with a visible outcome (a tree); opening ChatGPT means writing a prompt, then writing another prompt, then copy-pasting outputs, then losing them in a tab. The activation cost is higher and the visible progress is lower, which means the procrastinator never quite starts.