From “Learning to Read” to “Reading to Learn”: How Does “Reading Comprehension” Work?
If you don’t know what the writer assumed you knew, you can’t very well learn what the writer intended you to learn.
When you were young, you spent years learning to read. With all the tricky rules and exceptions in English pronunciation, it took great effort to translate words on the page into sounds in your head.
It isn’t until about fourth grade that almost all English speakers become proficient readers of their own language — but some become better readers than others. From this point forward, reading skill is no longer a matter of how well you can sound out English words in your head. It’s more about reading comprehension: basically, how well can you learn from what you read?
The gap between “skilled readers” and “struggling readers” grows wider over time, a phenomenon known as the “Matthew effect,” referencing Matthew 25:29: “For to everyone who has will more be given, and he will have abundance; but from him who has not, even what he has will be taken away.”
Those who spent their young years reading more will benefit more from reading, as they’re better able to understand what they read, which makes reading a more satisfying experience, one that they’ll seek out more often— the rich get richer. But those who struggle to read will learn less, leaving them less prepared for future reading and less enthusiastic about reading more, so they’ll fall even further behind their skilled reader peers— the poor get poorer.
There’s no shortcut to improvement. Building your reading comprehension skill requires lots of reading. This is because you need lots of knowledge, both of words and of the world.
Vocabulary
It’s not a big deal to stumble upon a word you don’t know. Your brain will try its best to plug in a meaning that fits in context. That’s usually how a new word is learned — not by reading a formal definition in a dictionary, but by seeing how it functions in the real world.
A single exposure isn’t enough. Suppose you read, “Scientists developing string theory say that string can exist in a straight line, but it can also bend, oscillate, or break apart.” That sentence wouldn’t tell you what oscillate means. It’s not trying to teach you that; the writer used the word assuming you already know what it means, and she isn’t going to slow down to explain it. But now you’re aware of this formal term that has something to do with motion and shape.
If you saw the word again, in the Investopedia subheadline “Stocks oscillate between being overbought and oversold,” you’d figure that it probably means to waver, to fluctuate, to sway, to wigwag like a string. Then, if you read that a politician’s views on a matter have oscillated over time, your idea would be confirmed. You can more confidently conclude what the sentence means than you could if you’d never seen the word.
That’s how words are mentally stored — a collection of associations from the different contexts where you encountered them.
They’re not stored like a dictionary, with a fixed meaning you can recall. If I asked you to define “to spread,” you might struggle for a moment to explain what it means, even though you know you know it. Something like “to make something cover more surface area” would make sense, but why wouldn’t it make sense to say you “spread” water on a flowerbed, or “spread” a folded paper when you unfold it? You just know the word doesn’t work that way.
To understand what a word’s supposed to mean in a sentence, you should be intimately familiar with how the word is used. If not, you risk misunderstanding what the author meant in that sentence. If you don’t understand that sentence, you risk misunderstanding the whole idea the author was really trying to tell you.
The more unfamiliar words in a passage, the more this tax on comprehension adds up. It becomes a slow, interrupted, unreliable process that feels more like solving a puzzle than receiving new information.
Background Knowledge
Of course reading comprehension suffers if you can’t correctly identify the meaning of enough words. But it also suffers if you don’t know enough about the subject. This was tough for me to grasp — reading passages on tests can be about anything from ecosystems to artistic movements to the lives of 19th century pioneers, things even the highest scorers may not know that much about. Where does “prior knowledge” come in?
Just as authors don’t explain vocabulary words they assume the reader knows, they leave out information they assume the reader understands. Unstated references to long-term memory shape how you read everything, even text that seems simple on the surface. Consider this viral poem:
It could’ve been written more like this:
I woke myself up
Because my household couldn’t afford little things like an alarm clock
Dug in the dirty clothes basket to get my school uniform
Because we didn’t have enough quarters to run a load of laundry in our apartment building lately, to the point that all my school clothes are now dirty
Brushed my hair and teeth in the dark
Because my mother couldn’t pay the light bill on time so now all the lights are off
Even got my baby sister ready (even though I’m just a kid!)
‘Cause my mama had to work — she works as many hours as she can to provide for us
Got us both to school on time (a real achievement, because we could’ve easily shown up late)
To eat us a good breakfast that we couldn’t have gotten otherwise at home, since we depend on provided school meals for our nutrition
Then when I got to class the teacher fussed
Because I ain’t got a pencil? Clearly I have way bigger things on my plate. Struggling kids do way more than we give them credit for.
But communication would be awfully long and boring if we had to explain everything like that. If you have enough background knowledge, you’ll make the right inferences automatically. (If you don’t, the poem just sounds like a pointless list of complaints!) This subconscious process, linking ideas through quick inferences from prior knowledge, is how “Shut up! I’m trying to read!” makes sense and “Shut up! I’m trying to eat a sandwich!” doesn’t, even when you understand every word either way.
That’s not to say a reader needs to share all the author’s assumed prior knowledge to understand a passage. It’s possible to learn new things in passing, even if the author didn’t intend to teach you so — but only so much at a time.
A novel that takes place in 21st century London won’t explicitly explain every location, every custom, every dialectical word the author expects English readers to recognize, but an American reader who generally understands what’s going on can decipher bits in context along the way. It’s much harder to learn from reading, say, a 19th century Charles Dickens novel. A modern American reader is missing so much context that he may need to turn to SparkNotes to even understand what’s going on. (Meanwhile, a 21st century author writing about that same time and place for a modern audience would be much easier to read. The author knows when to explain aspects of this setting that Dickens’ audience wouldn’t need explained.)
In other words, when most of the “surely you knew” information is already in your long-term memory, you have more mental processing space for the “here’s what’s new” information. When it’s not, you’re juggling more new information in your working memory than you’re supposed to, creating a more difficult reading experience than the author intended.
So when a student performs well on “reading comprehension” questions for a topic that’s unfamiliar, it’s thanks to sufficient prior knowledge for the new information to make sense.
If a passage is about, say, the War of 1812, good test-takers don’t all think, “I already knew all this!” But for a test-taker who doesn’t have a solid mental map of early America and concepts such as trade, territory, battles, strategy, and siege, it’s hard to make the right inferences. More mental energy is dedicated to guessing what a sentence is referring to than correctly considering the implications of what it means. The cost of each insufficiently understood sentence adds up, and soon enough, most of what the author was trying to say is lost on the reader.
Altogether, “reading comprehension” seems difficult to improve. Just read a lot? Stumble across as much vocabulary and information as you can? If it’s all based on the contents of a reader’s long-term memory, how could mastering a “reading comprehension” passage help learners with a whole different passage, of a whole different topic, with different vocabulary?
Are there any reading comprehension skills that are transferable? Fortunately, yes, to an extent.
“Reading Comprehension” Skills
The subconscious strategy of a skilled reader is not the same as what’s going on in a struggling reader’s head. Skilled readers understand that every sentence contributes to a main idea, each with its own role: introduce a concept, flesh out a concept, point out a potential counterargument, address a potential concern, add some needed nuance. By linking meaning across sentences, they can trace the structure of the full argument, and by identifying the true argument, they can draw better conclusions about what the author’s saying.
Struggling readers, on the other hand, view each sentence as an independent unit. They don’t see the overarching main idea. They don’t see whether the author is disagreeing or agreeing with a particular view. Together, the long series of sentences sounds like the author “just saying stuff.”
So if readers don’t readily apply reading strategies, it can help to explicitly teach them, according to a rich body of research. It’s not that readers will fill out graphic organizers, or ask themselves, What’s the main idea? What are the supporting claims?, for the rest of their reading lives. Practicing these strategies just helps them see that a passage is structured to serve a purpose, rather than “just saying stuff,” and that sentences must be interpreted in connection to other sentences.
One thing that distinguishes the effective programs from the unsuccessful ones studied is when readers are taught. These strategies must be learned after readers have mastered “reading” in the sense of decoding written words. It’s hard to think of the big picture when you’re taking it one word at a time; it’s hard to think beyond the page when all your mental processing space is spent identifying what words are even on the page. Considering that fluent reading in native English speakers isn’t solid until about fourth grade, it’s no wonder that the studied reading comprehension interventions in third grade or sooner never seem to work.
And what doesn’t matter much is how long readers are taught these strategies. Multiple meta-analyses show that of all the reading interventions found effective, the size of the gains is about the same; ones involving up to fifty sessions have the same effect as ones as short as ten sessions. Once a reader is aware of it, the proper reading mindset clearly doesn’t take long to learn. If a reader’s comprehension is still poor after practicing it, the solution is not more practice of the same thing.
That’s because “reading comprehension” strategies only do so much. Applying them is just a more helpful way to approach reading, just a way to build a more coherent mental model of the text —but you can’t “coherently” organize what you can’t understand in the first place. I scored perfect on the reading section of the SAT, but such “reading” aptitude wouldn’t help me decipher a dissertation in theoretical physics.
Since most reading comprehension depends on domain-specific knowledge, it’s counterproductive to spend so much instruction time on transferable strategies. Yet in schools today, that’s the main approach to raise “reading” standardized test scores. What could they be doing instead?
Clearly, the more you read, the more you eventually can read in the future. But not all “reading” is equal. A steady diet of action-packed graphic novels won’t very well make you a better reader, nor will struggling through works of Dickens and Shakespeare that you can barely decipher.
That’s the case for knowledge-rich education.
People wonder why it’s education’s job to deliver facts — don’t we forget most of them? Can’t we just consult Google if we really need to know something? Understanding the relationship between reading comprehension and memory, we’d see the flaws in that approach.
Facts don’t pass through the brain like pictures stored in a phone, installed and deleted within limited capacity. I could learn a wealth of information later in life, and I’m still never going to forget what “vulnerable” means, or what a neuron is, or who Kanye West is. Once a fact secures itself in a mental map, you’re changed for good.
So the key is to build the right mental maps early on — the knowledge of science, history, the arts, the world, and words that grown-up writing will expect readers to know.
This “cultural capital” helps you understand more texts that would otherwise seem inaccessible, teaching you more, drip by drip, about words and the world… which helps you read more texts that teach you, drip by drip, about words and the world… which helps you read more texts that teach you, drip by drip, about words and the world…to the point that you could be reading the same text as someone else and be able to comprehend it better than the other person with the same words on the screen.
It’s thanks to your long-term memory. “Memory” isn’t just something you dig into to spit out an answer for a test. You can’t read something and not be reminded of relevant things you remember, lighting up to tell you whether something is familiar or unfamiliar, mundane or surprising, concordant or contradictory. Long-term memory is always active when you process something new, as we just saw with the “‘Cause I Ain’t Got a Pencil” poem.
Not all students have received the right knowledge by the time reading tests expect. With too many gaps in a mental model of a passage, the test questions just feel like a guessing game. The infamous “fourth-grade slump,” in which students from disadvantaged backgrounds start lagging behind their peers in reading, is only noticed starting in fourth grade. Turns out, they were lagging behind their peers in accumulated knowledge all along, and the more knowledge-rich reading tests in fourth grade onward are where the gaps truly start to show.
When put like this, it should sound obvious now: To learn what the author assumed you’d learn, you should know what the author assumed you’d know. Of course it doesn’t make much sense to speak of “reading comprehension” as a transferable skill. Now that you’ve learned to read, you can’t comprehend all kinds of passages equally. What you can “read to learn” depends on what you know.
