HSW-0091 · How Studying Works
Two learners can arrive at the same chapter with the same amount of study time and still experience completely different difficulty.
One sees the new idea as a small extension of something familiar. The other sees a wall of new symbols, terms and decisions. The difference is not always intelligence, motivation or effort. Sometimes it is the route that came before.
This is learning path dependence: the idea that what becomes easy, hard, visible or even thinkable later can depend on the sequence of earlier learning.
The concept is familiar in economics, technology and institutional history. Systems do not begin each day from zero. Earlier choices shape later costs and available options. Learning behaves the same way. A learner’s current state is partly the accumulated consequence of previous explanations, representations, habits, shortcuts, successes and gaps.
This article does not replace eduKateSG’s Learning Adjacency, which asks what the next learnable thing is beside current knowledge, or How Learning Dependencies Work, which owns prerequisite structure. Path dependence asks a wider systems question: how does the route already travelled alter the cost and character of the route ahead?
Learning is not only a destination problem
Schools often describe learning in destination language: master fractions, understand photosynthesis, write an argument, solve simultaneous equations.
But the route matters.
A student who learned fractions through strong part-whole, ratio and number-line models may meet algebraic fractions with multiple conceptual anchors. Another student who survived fractions mainly through memorised procedures may reach the same syllabus point with far fewer routes into the new idea.
The same next lesson can be a small step for one learner and a reconstruction project for another.
Prior knowledge changes the cost of new learning
This is not merely an analogy. Current educational research continues to show how strongly prior knowledge shapes instruction. A 2025 study in Educational Psychology Review examined whether instruction-first or problem-solving-first works differently depending on learners’ prior knowledge. A 2025 experimental study in Smart Learning Environments likewise investigated personalised learning based on prior knowledge. See the Educational Psychology Review study and the Smart Learning Environments study.
The practical point is simple: prior knowledge does not merely add to new knowledge. It changes how new knowledge is processed.
Four ways a learning path leaves a footprint
1. It changes what the learner notices. Experts see structure that novices do not because earlier learning has built categories and contrasts.
2. It changes working-memory demand. A familiar sub-skill can be compressed into one usable unit. An unfamiliar one has to be rebuilt step by step.
3. It changes which errors are likely. Old shortcuts and misconceptions become default moves.
4. It changes future optionality. Some foundations open several later routes; others support only a narrow local task.
Mathematics: sequence can create or remove future friction
Mathematics is full of path dependence because later ideas repeatedly reuse earlier representations.
If negative numbers are learned as arbitrary sign rules, algebra may later feel like a collection of exceptions. If equality is learned relationally, equations have a stable underlying logic. If graphs are learned as pictures rather than representations of relationships, later coordinate work becomes harder to generalise.
The route matters because each new topic inherits not just facts, but habits of interpretation.
English: reading habits shape writing options
English learning also leaves a route behind. A student who has spent years reading for gist only may struggle when later tasks require precise evidence tracking. A writer trained mainly to decorate sentences may find it hard to build causal coherence. A learner who has developed strong paragraph architecture can use vocabulary more effectively because the words have somewhere purposeful to go.
The later writing problem is therefore partly a history problem. What kind of reading, noticing and composing did the learner repeatedly do before this task arrived?
Science: models are stepping stones, not disposable pictures
Science education frequently uses simplified models before more sophisticated ones. That is sensible. The danger appears when the learner is not told what the first model can and cannot explain.
A useful early model should support later refinement. A brittle model can create resistance because the learner must first unlearn what was treated as absolute.
Recent research on conceptual change keeps this problem visible. A 2025 study in Evolution: Education and Outreach used concept mapping to trace students’ changing conceptions of evolution, showing how learning can be understood as movement among organised concept states rather than simple accumulation. See the study.
Path dependence does not mean destiny
This distinction matters.
A learner who took a weak route is not trapped forever. Path dependence means that the starting state of the next decision has been changed. It does not mean the route cannot be repaired.
Good teaching can create a new path by rebuilding a missing representation, replacing a misleading shortcut, or connecting a familiar idea to a stronger model.
The earlier route affects the repair cost, not the learner’s worth.
The financial analogy: option value
Some educational foundations are valuable not because they deliver immediate marks, but because they preserve future choices.
A broad mathematical foundation can keep advanced routes open. Strong reading can support science, humanities and professional learning. Clear writing can improve performance wherever ideas have to be communicated.
This is close to Learning Optionality. The difference is that optionality focuses on the value of keeping doors open; path dependence explains how earlier routes either preserve or narrow those doors.
The systems route: standards are sequences, not lists
Curriculum design often looks like a list of content statements. In operation, it is a sequencing problem.
If a national curriculum places concepts in an order that assumes unavailable prerequisites, teachers must compensate. If transitions between primary, secondary and post-secondary education do not align, students experience hidden resets. If assessment rewards shortcuts earlier, later teachers inherit the habits those shortcuts created.
This is why learning progressions matter. eduKateSG’s How Learning Progressions Work owns the mapping of intermediate states. Path dependence adds the historical insight: moving through those states in different ways can leave different residue.
The training route: sequence changes transfer
Adult training often compresses learning because time is expensive. That makes sequence even more important.
If workers are taught a software sequence before they understand the operational decision the software supports, they may become interface-dependent. If they first understand the decision logic, they can often adapt when the software changes.
The same capability can therefore be taught through a narrow route or a transferable route.
How to audit your own learning path
- Name the current difficulty. What feels expensive now?
- Trace backwards two or three layers. What earlier knowledge does this task assume?
- Look for a representation gap. Did you learn only one form of the idea?
- Look for a shortcut that became a habit. What rule works locally but fails more generally?
- Identify the strongest existing anchor. What do you already understand that can become the new bridge?
- Rebuild one dependency at a time. Do not restart the entire subject unless evidence requires it.
- Retest the new route in a later task. The repair is successful when the next topic becomes cheaper.
Teachers can design paths, not just lessons
A strong lesson can still sit inside a weak sequence. That is why teaching quality has a temporal dimension.
- Activate the specific prior knowledge the new lesson needs.
- Make the bridge explicit instead of assuming students see it.
- Use representations that can survive later sophistication.
- Warn learners when a model is intentionally simplified.
- Revisit earlier ideas when later topics reveal their real value.
- Preserve enough variation that the learner does not confuse one route with the concept itself.
The world route
Path dependence is everywhere outside school. Technologies inherit standards. Cities inherit roads. Institutions inherit rules. Companies inherit processes. People inherit habits.
Education is one of the places where we can learn to see this clearly. Every current capability has a history. Every future capability will inherit something from what we practise today.
Good studying does not only ask, “Can I reach the next answer?” It asks, “What kind of route am I building for the learner I will be next year?”