Quantitative transfer refers to the ability to take mathematical or numerical reasoning learned in one setting and successfully apply it in a different, sometimes very different, context. Research consistently shows that people can achieve this kind of transfer, but it rarely happens automatically. The gap between learning a quantitative skill and deploying it somewhere new is one of the most studied problems in educational psychology, and the findings point to specific teaching strategies and cognitive conditions that make transfer far more likely. How those conditions work, and why the brain sometimes resists bridging the gap on its own, turns out to be more interesting than the simple question of whether transfer is possible.
What Makes Quantitative Transfer So Difficult
The classic problem with transfer is that people tend to anchor what they learn to the surface features of the context where they learned it. A student who can solve a word problem about trains may struggle with a structurally identical problem about water flowing into tanks, because the two look different on the surface even though the underlying math is the same. This is not a failure of mathematical ability; it is a feature of how human memory works. We encode context along with content, so retrieving the right strategy in a new situation requires recognizing the deep structure beneath unfamiliar packaging.
Researchers have traditionally defined transfer as applying knowledge from one situation to another, but a more recent framework flips the perspective. Rather than asking whether a learner can reproduce a specific procedure in a new setting, “actor-oriented transfer” looks at how learners personally construct relationships of similarity between activities. Under this lens, transfer is not a yes-or-no event that either matches the expert’s expectation or doesn’t. It is about what connections the learner actually sees, which may be productive even when they are unexpected.1ScienceDirect. Quantitative reasoning in a reconceived view of transfer This shift matters because it reframes the teaching challenge: instead of drilling identical procedures and hoping students reproduce them elsewhere, instruction can focus on helping learners notice structural similarities across contexts.
Schema Training Before Practice
One of the clearest findings in the transfer literature is that teaching learners to recognize the underlying structure of a problem type, what researchers call a “schema,” substantially improves their ability to handle unfamiliar versions of that problem. The timing of this training matters. A study on mathematical problem solving found that participants who received schema training at the outset, before they practiced solving problems, outperformed those who received the same schema training after practice. The advantage was especially pronounced on far transfer problems, the kind that look quite different from the training examples.2PubMed. The effects of early schema acquisition on mathematical problem solving
Schema-based transfer instruction, or SBTI, takes this idea further by explicitly teaching students about the surface features that make problems appear novel even when they share a known solution structure. In a randomized study with third graders, teachers were assigned to a control condition, a standard SBTI condition, or an expanded SBTI condition that incorporated more challenging transfer features. On a measure designed to approximate real-life problem solving, the expanded SBTI group outperformed the standard SBTI group, which in turn outperformed the control group.3American Educational Research Journal. Expanding Schema-Based Transfer Instruction to Help Third Graders Solve Real-Life Mathematical Problems The stepwise improvement suggests that the more deliberately instruction addresses what makes problems look unfamiliar, the better students become at seeing through surface differences to the quantitative structure underneath.
A separate randomized controlled study with second graders found that schema-broadening instruction improved word-problem performance and, critically, helped students represent problems with algebraic equations. That last detail is worth pausing on: children who were not yet formally studying algebra began spontaneously using algebraic representations because the schema training promoted that kind of structural thinking.4PubMed Central. The Effects of Schema-Broadening Instruction on Second Graders’ Word-Problem Performance and Their Ability to Represent Word Problems with Algebraic Equations
Concreteness Fading
A recurring tension in math education is whether to start with concrete examples (hands-on objects, real-world scenarios) or abstract representations (symbols, variables, general rules). The evidence on quantitative transfer suggests that the answer is both, in that order. A technique called “concreteness fading” begins instruction with concrete materials and then gradually strips away the contextual details until the learner is working with purely abstract representations.
In a study with undergraduates learning a new mathematical concept, those in a concreteness fading condition showed the best transfer performance. Students who worked only with abstract representations from the start did somewhat better than those who only used concrete materials, but the advantage was not robust. The fading group outperformed both.5Learning and Instruction. “Concreteness fading” promotes transfer of mathematical knowledge A parallel set of three experiments with children replicated the pattern: in all three, children in the concreteness fading condition exhibited better transfer than children in other conditions.6Learning and Instruction. Benefits of “concreteness fading” for children’s mathematics understanding
The logic is intuitive once you see it. Concrete materials give learners an entry point, something to mentally grab onto. But if instruction stays concrete, students bind the concept to that specific physical context and struggle to apply it elsewhere. Fading the concreteness forces learners to extract the quantitative structure on their own, which is exactly the skill that transfer demands. Starting abstract, on the other hand, can leave learners without enough grounding to form a meaningful representation in the first place.
Inventing Before Being Told
Another effective lever for transfer is the sequence of exploration and instruction. In a study comparing two approaches, one group of students was told a mathematical principle first and then practiced applying it. A second group was given contrasting cases and asked to invent their own explanations before receiving any formal instruction. The inventing group performed significantly better on a delayed transfer test.7American Psychological Association (Journal of Educational Psychology). Practicing Versus Inventing With Contrasting Cases: The Effects of Telling First on Learning and Transfer
The reason appears to be about depth of encoding. Students who tried to make sense of contrasting cases before being told the answer were forced to attend to the deep structure of the problems. They noticed which features mattered and which did not. Students who were told the principle first and then practiced it tended to reproduce the procedure without building the same structural awareness. The inventers were not necessarily better at recreating surface features, but they were significantly better at recreating the deep mathematical structure, and deep structure is what transfers.
This finding has a clear practical implication: if you want quantitative reasoning to stick and travel, give learners a chance to wrestle with the underlying structure before handing them the formula. The struggle is not a sign that instruction is failing. It is the mechanism through which transferable understanding gets built.
Worked Examples and Self-Explanation
Worked examples, where students study a fully solved problem step by step rather than trying to solve it from scratch, are another well-supported tool for promoting transfer. A quasi-experimental study with eighth graders found that an example-based learning model significantly reduced cognitive load throughout the problem-solving stages and produced better outcomes on both retention and near-transfer tests compared to a problem-solving model.8RANGE: Jurnal Pendidikan Matematika. Effect of Example-Based Learning Model on Micro Level Cognitive Load and Knowledge Transfer The link between cognitive load and transfer makes sense: when a learner is overwhelmed by the demands of figuring out a problem, they have little mental bandwidth left to extract the general principle that would help them recognize a similar structure later.
Worked examples become even more powerful when paired with self-explanation prompts, where students are asked to explain to themselves why each step in the example works. A study with fourth and fifth graders found that self-explanation attempts significantly improved scores on a “preparation for future learning” measure, though the benefit was strongest for students with at least average prior knowledge and for those who completed more practice worksheets.9PubMed. Preparing 4th and 5th graders to learn algebra with worked examples and self-explanation prompts The practical takeaway for teachers is that worked examples should not be a passive activity. Asking students to articulate why a step makes sense, even briefly, pushes them toward the kind of structural understanding that underlies transfer.
There is an important caveat here about prior knowledge. The self-explanation benefit was not uniform across all students; those with below-average prior knowledge did not see the same gains. This fits a broader pattern in the transfer literature: interventions that work well for learners who already have some foundation can be less effective for true beginners, who may lack the cognitive scaffolding to make productive self-explanations. The implication is not that worked examples are useless for struggling students, but that the surrounding supports may need to be different.
What the Brain Does During Transfer
Neuroimaging research has started to reveal what happens in the brain when someone successfully applies an abstract rule in a new context. A study using pattern classification analysis of brain activity found that the lateral prefrontal cortex encodes decision rules in a way that generalizes across contexts. Researchers trained a classifier to identify four different decision rules from brain activity during practiced tasks, then tested whether the same classifier could identify those rules during entirely novel tasks the participants had never seen before. It could, at accuracy significantly above chance.10PubMed Central. Rapid Transfer of Abstract Rules to Novel Contexts in Human Lateral Prefrontal Cortex
This is a stringent test: it requires that the neural pattern representing a rule during familiar tasks is essentially the same pattern that activates when the rule is applied in an unfamiliar context. The fact that the prefrontal cortex passes this test tells us something about how the brain supports quantitative transfer. It appears to maintain abstract representations that are at least partially independent of the specific context where a rule was first learned.
Complementary research on the hippocampus and prefrontal cortex has shown that these two brain regions format information at different levels of abstraction. In one study, the prefrontal cortex organized its population activity on a ring-shaped manifold that enabled high-accuracy classification of task variables even in novel conditions, essentially an abstract format. The hippocampus, by contrast, organized its activity in a way that more closely mirrored the physical spatial structure of the task environment. The prefrontal cortex could generalize across conditions; the hippocampus could not.11Cell Reports. Geometry of hippocampal-prefrontal neural manifolds during memory generalization This division of labor helps explain why transfer is possible but effortful: the prefrontal cortex builds abstract representations that can travel, while the hippocampus ties memories to specific contexts. Transfer likely requires the prefrontal system to override or supplement the context-bound representations anchored in the hippocampus.
From Simulations to Real Tasks
One practical question about quantitative transfer is whether skills learned in virtual or simulated environments carry over to the physical world. This matters because simulations are increasingly common in classrooms, and if the learning they produce is stuck in the virtual context, their value is limited. A study investigating young students’ learning from simulations found that they did transfer procedural knowledge to real equipment tasks, with some students also demonstrating basic conceptual transfer.12British Journal of Educational Technology. From simulations to real: Investigating young students’ learning and transfer from simulations to real tasks
The distinction between procedural and conceptual transfer is worth noting. Procedural transfer means students could perform the same steps on real equipment that they had practiced in the simulation. Conceptual transfer means they understood the underlying principles well enough to apply them in ways that went beyond just reproducing steps. The fact that procedural transfer was more robust than conceptual transfer fits the broader pattern: surface-level skills move more easily than deep understanding. But the presence of at least some conceptual transfer is encouraging, because it suggests that well-designed simulations can serve as a stepping stone toward genuine quantitative understanding, not just rote procedural memory.
Statistical Training and Everyday Reasoning
Perhaps the most ambitious claim in the quantitative transfer literature is that training in formal mathematical principles can change how people reason about everyday, non-academic situations. A classic pair of experiments tested this directly by teaching participants about the law of large numbers in brief laboratory training sessions. The result was striking: training increased both the frequency and the quality of statistical reasoning that participants applied to a wide variety of everyday problems, not just problems that resembled the training examples.13Cognitive Psychology. The effects of statistical training on thinking about everyday problems
This finding is notable because it pushes back against a pessimistic view of transfer that was dominant for decades. Early research often concluded that transfer was rare and fragile, that people learned things in one context and left them there. The statistical training studies showed that when the principle being taught is genuinely general, and when the training explicitly connects it to the kind of reasoning people encounter in daily life, transfer to everyday thinking is achievable. The key seems to be that certain quantitative principles, like the idea that small samples are unreliable, map naturally onto intuitions people already have. Training sharpens and formalizes those intuitions rather than trying to install something entirely foreign.
Computational Thinking as a Transfer Vehicle
A growing body of research examines whether learning computational thinking, the problem-solving approach central to programming and computer science, transfers quantitative and analytical skills to other subject areas. A systematic review with meta-analysis found a generally significant transfer effect from computational thinking skills to other domains.14Journal of Computer Assisted Learning. The transfer effects of computational thinking: A systematic review with meta‐analysis and qualitative synthesis
A more detailed meta-analysis pooling results from 37 studies and over 7,800 students found a moderate overall transfer effect, with moderate effects for both near and far transfer. The analysis identified several interesting patterns. Cognitive benefits (like improved problem-solving and mathematical reasoning) showed stronger transfer effects than noncognitive benefits (like motivation or attitudes toward learning). Educational level, sample size, instructional strategies, and intervention duration all moderated the strength of transfer, with educational level and sample size remaining significant moderators even for far-transfer effects.15International Journal of STEM Education. The transfer effect of computational thinking (CT)-STEM: a systematic literature review and meta-analysis
The fact that computational thinking shows moderate far transfer is worth emphasizing, because far transfer, applying skills in a domain that is substantially different from the training domain, is the hardest kind to achieve. The research here connects to a broader discussion in cognitive psychology about analogical reasoning and how learners map structures from one domain onto another.16ACM Transactions on Computing Education. From One Language to the Next: Applications of Analogical Transfer for Programming Education When students learn to decompose problems, recognize patterns, and think algorithmically in a programming context, those habits of mind can reshape how they approach quantitative problems in science, mathematics, and engineering. The transfer is not guaranteed, and it depends heavily on how instruction is designed, but the evidence that it can happen across meaningfully different domains is now reasonably solid.
How Spatial and Numerical Cognition Develop Together
Quantitative transfer has developmental roots that go deeper than classroom instruction. Research on early cognition suggests that spatial and numerical thinking are intertwined from a young age and develop along several parallel lines: changes in capacity and precision, differentiation of a generalized magnitude system into separable dimensions, formation of a discrete number system, mapping of whole numbers onto a continuous number line, and acquisition of abstract knowledge about relationships between these systems.17PubMed Central. Thinking about quantity: the intertwined development of spatial and numerical cognition
This developmental picture matters for understanding quantitative transfer because it reveals that the ability to move between different representations of quantity is not something grafted on by formal education. It is built into the architecture of how children come to understand the world. The shift from a generalized sense of “more” and “less” to a precise understanding of number, and then to the ability to see relationships between different quantitative systems, mirrors the progression that transfer demands: from context-bound intuition to abstract, portable understanding. Children who develop strong connections between spatial and numerical thinking early on may be better positioned for the kind of quantitative transfer that formal schooling later requires, though the research on this link is still developing.
This also suggests that efforts to improve quantitative transfer should not focus exclusively on teaching specific mathematical procedures. Building the broader cognitive infrastructure, the ability to see quantity in different forms and to move fluidly between representations, may be at least as important. Activities that connect spatial reasoning to numerical reasoning, or that ask children to translate between physical and symbolic representations of the same quantity, are likely doing transfer work even when they do not look like traditional math instruction.