HSW-0071 · How Studying Works
“I have already spent two hours on these notes.”
“I have used this method all year.”
“I bought the course, so I should finish it.”
“I have done fifty questions this way. I cannot change now.”
Each sentence sounds responsible. Sometimes persistence really is the right decision. But each sentence also contains a trap: it uses a cost that has already been paid to decide what should happen next.
This is the study version of the sunk-cost problem.
A sunk cost is time, effort, money or identity investment that cannot be recovered by continuing. The decision in front of the learner is not whether the past investment was worthwhile. That decision is gone. The decision is whether the next unit of time is better spent continuing, modifying, pausing or stopping.
This article preserves nearby canonical owners. How Studying Works | Knowledge Investment owns why durable capability can compound. How to Plan Properly | Know When the Plan Must Change owns plan revision. How to Grow Up Properly | Do Not Confuse Hard Work With Useful Work owns the broader distinction between effort and value. HSW-0071 owns the decision error that occurs when past study expenditure is allowed to govern future allocation even after the expected return has changed.
Past effort deserves respect, not control
There is a moral instinct behind sunk-cost thinking: do not waste what you have already invested.
That instinct is often useful. It encourages commitment, patience and follow-through. But a past investment cannot be made less wasted by adding more waste to it.
The correct question is not “How much have I already put in?” It is “Given what I now know, where should the next hour go?”
Persistence and sunk-cost persistence are not the same thing
Students need persistence. Difficult learning often requires repeated attempts, delayed payoff and uncomfortable practice.
So “stop when it is hard” is terrible advice.
The distinction is about future evidence.
- Productive persistence: continue because the next attempt still has a reasonable expected learning return.
- Sunk-cost persistence: continue mainly because stopping would make the previous effort feel wasted.
The first is forward-looking. The second is backward-looking.
Current research shows why persistence needs calibration
Persistence is strongly associated with learning success, but recent research also warns against treating more persistence as automatically better. A 2026 study of adolescents’ STEM achievement found persistence to be an important predictor while also reporting non-linear and interactive effects: under some conditions, high persistence could become less adaptive when curiosity was low. See Be curious and persistent: joint roles of curiosity and persistence in adolescents’ STEM achievements.
Decision research supplies the other half of the picture. A 2025 experimental paper on escalation of commitment found that a devil’s-advocate approach could reduce the tendency to keep investing in failing projects: Mitigating escalation of commitment through error management climate and the devil’s advocate approach. The educational transfer is not that studying is a business project. It is that decision-makers benefit from a procedure that deliberately challenges continued investment.
The classic open-access meta-analytic review of the sunk-cost effect also found clear evidence that prior unrecoverable investments can bias later decisions: On the sunk-cost effect in economic decision-making: a meta-analytic review.
Where sunk costs appear in studying
The note-making trap
A learner spends weeks producing beautiful notes. Evidence then shows that rereading them is not producing retrieval or transfer. The learner keeps making more because abandoning the method would make the previous notes feel pointless.
The course trap
A paid course is badly matched to the learner’s syllabus or level. Because money has been spent, the learner gives it priority over better free or school-provided material.
The question trap
A student has already spent twelve minutes on a four-mark question and refuses to move on because “I have come this far.” The next three marks elsewhere are sacrificed to protect the twelve minutes already lost.
The tutor or tool trap
A method, platform or support arrangement no longer fits, but changing feels like admitting the previous choice was wrong.
The identity trap
“I am the kind of student who studies this way.” At that point the investment is no longer only time. It is identity. Change feels like self-contradiction.
Use marginal value, not total history
Economics offers a useful decision frame: compare the expected benefit of the next unit of investment with its opportunity cost.
For studying, ask:
- What is the next 30 minutes expected to improve?
- What evidence says this method is still changing capability?
- What else could the same 30 minutes do?
- If I had not already invested anything, would I choose this method today?
That last question is especially powerful. It temporarily removes history from the decision.
The restart test
Imagine your current notes, app, course, revision routine or question strategy disappeared tonight. Tomorrow, knowing what you know now, would you rebuild it the same way?
If yes, continue with confidence.
If no, the gap between “what I would choose now” and “what I keep doing because I already started” is where sunk-cost thinking may be operating.
Mathematics: do not protect a long wrong method
In an exam, students sometimes continue a failing route because too much working has already been written.
The better decision may be to circle the useful intermediate result, restart from a cleaner representation, or move on and return later. The existing working is not a reason to keep feeding the wrong method.
Outside the exam, the same principle applies to revision. Fifty repetitions of a weak method do not create an obligation to perform the fifty-first.
English: drafts are investments, not sacred objects
A paragraph can contain beautiful sentences and still be structurally wrong for the task.
Writers often preserve weak sections because too much effort went into them. This is why editing can feel painful: deletion makes labour invisible.
But the reader does not award marks for invisible effort. The draft exists to serve the final response.
Keep what serves the argument. Salvage useful lines. Remove what no longer earns its place.
Science: abandon explanations that evidence no longer supports
Science has a deep cultural lesson for studying: investment in an explanation does not make the explanation true.
If new evidence contradicts the model, the answer is not to protect the model because it took a long time to learn. The model must be revised, bounded or replaced.
This is one reason scientific thinking matters beyond Science lessons. It teaches disciplined disloyalty to explanations when reality stops supporting them.
The systems route: preserve assets, not commitments
Changing a study method does not require throwing everything away.
Separate the asset from the commitment.
- Old notes may still be a useful reference even if note-making stops.
- A course may contain three excellent modules even if the rest is abandoned.
- A failed question may reveal a prerequisite gap even if the attempted route is discarded.
- A tutor relationship may have produced valuable diagnostics even if the teaching arrangement changes.
Good decision-making salvages value without becoming hostage to history.
The financial route: portfolio discipline
Investors rebalance because future expected returns and risks change. They do not keep every allocation forever simply because it was once chosen.
Study allocation should also respond to evidence. The learner who has become strong in one area and weak in another should move time. The previous distribution is not a contract.
This connects to Learning Portfolio Rebalancing. That article owns how allocation moves when evidence changes. HSW-0071 explains one psychological force that can prevent the move.
The teacher and parent problem: praise can accidentally lock the method
If adults repeatedly praise only effort quantity—hours spent, pages written, worksheets completed—students can become reluctant to stop low-value work because stopping looks like reduced diligence.
Praise should include intelligent reallocation:
- “You noticed that this method was no longer helping.”
- “You kept the useful part and changed the weak part.”
- “You stopped before the problem consumed the whole session.”
- “You used evidence rather than pride to decide.”
Use pre-committed stop rules
Sunk-cost bias is easier to resist when the stopping condition is defined before the emotional investment grows.
- If I cannot explain the method after three worked examples and five attempts, I diagnose the prerequisite.
- If a resource repeatedly conflicts with the authoritative syllabus, I stop using it as a primary source.
- If a four-mark exam question exceeds my time boundary, I mark it and move.
- If a revision method produces no improvement across two retests, I change the method before adding more hours.
These are not universal numbers. They are examples of decision rules that protect the future from the emotional weight of the past.
A five-question sunk-cost audit
- If I were starting today, would I choose this again?
- What evidence shows the next unit of effort is likely to help?
- What opportunity am I giving up by continuing?
- Am I protecting learning—or protecting the feeling that my previous effort was justified?
- What useful assets can I keep even if I stop the current route?
The world route
Adults face sunk-cost decisions in projects, careers, technologies, policies and organisations. The ability to stop feeding a failing route without denying the value of what was learned is a major form of judgment.
School can teach this gently. A student who can say, “That method taught me something, but the evidence says I should not spend the next hour on it,” is learning far more than revision technique.
Changing course does not erase the past investment. It prevents the past investment from owning the future.
Final rule
Respect the work already done. Learn from it. Salvage what still has value.
Then decide the next move using current evidence, future return and opportunity cost.
The hours behind you are history. The hour in front of you is still allocatable.
Previous in the numbered series: HSW-0070 · Study Exception Handling.