What Scientists Found About Audiobooks — Transcript
Full transcript
- 0:00Which is better, reading a book or
- 0:02listening to an audiobook? Over 40% of
- 0:05Americans think that audiobooks don't
- 0:08count as real reading.
- 0:10And I'll be honest, I would usually
- 0:11agree. When someone tells me they
- 0:13finished [music] 50 books this year, but
- 0:15they just listened to them, it does feel
- 0:18a little bit like cheating. But it turns
- 0:19out I might be wrong, at least
- 0:22partially. Scientists have now mapped
- 0:24exactly what happens inside our heads
- 0:27when we listen to a story versus when we
- 0:29read one on a physical page. And the
- 0:31neuroscience tells a more complicated
- 0:34and even slightly contradictory [music]
- 0:36story. But the science does give an
- 0:38answer and it might just change the way
- 0:40that you consume [music] your content.
- 0:52So, check this out. Researchers at UC
- 0:54Berkeley had nine people listen to hours
- 0:56[music] of narrative stories inside an
- 0:59fMRI scanner. Every word was mapped to
- 1:03brain activity. Now, this is what your
- 1:05brain looks like when you're listening
- 1:07to a story. Every color represents a
- 1:10different category of meaning. Then, the
- 1:13same people came back and they read the
- 1:16exact same stories. Again, in the same
- 1:19scanner with the same words, the content
- 1:21didn't [music] change. And here's the
- 1:23same brain reading. Notice anything?
- 1:26Yeah, they're basically identical. All
- 1:29the red color is the brain processing
- 1:31meaning the exact [music] same way, no
- 1:33matter how the words arrived. The
- 1:36temporal cortex, the parietal cortex,
- 1:38and the prefrontal cortex are all
- 1:40responding pretty much identically. Your
- 1:44brain doesn't really care how the words
- 1:46arrived. It's already extracting meaning
- 1:48the same way. So, reading [music] and
- 1:50listening are far more similar than they
- 1:53are different.
- 1:54But both are a lot more complex than we
- 1:57give them credit for. Now, spoken
- 1:59language goes back tens of [music]
- 2:01thousands of years. Written language
- 2:04maybe 6,000. The printing press 600, and
- 2:08audiobooks about 90 years. [music] Your
- 2:11brain didn't build a separate new system
- 2:14for reading. It borrowed one it already
- 2:16had [music] for listening. So, if the
- 2:18brain wiring is the same, does anyone
- 2:22actually learn better one way than the
- 2:24other? Well, let's take a look.
- 2:31Let's get into the evidence because
- 2:32there is evidence showing that learning
- 2:35via audio is just not so good. Back in
- 2:392010, textbook publishers were pitching
- 2:42audio versions of their books [music] to
- 2:44professors. Now, Professor David Daniel
- 2:47wanted evidence that they actually
- 2:49worked before forcing students to pay
- 2:51for them. Seems [music]
- 2:52reasonable. So, he ran the experiment.
- 2:56He randomly divided his psychology
- 2:58students into two groups. One group read
- 3:01the textbook chapter. The other one
- 3:03listened to the audio version. It was
- 3:05the same content and they had 25 minutes
- 3:08each. After that, they had to take a
- 3:10high-stakes [music] quiz. So, what did
- 3:13the data show? Like how much worse were
- 3:15they performing? Was it just a little
- 3:16bit or was it like, "Okay, this is
- 3:18detrimental."
- 3:19>> I figured there was a difference between
- 3:20an A and a D, like 28% different. It was
- 3:23not pretty. Right?
- 3:25>> Both groups had the same content and the
- 3:27same amount of time, but the listeners
- 3:29scored [music] an average of 28%
- 3:32lower on the quiz. Now, the final number
- 3:35is telling, but so is what happened
- 3:37before the quiz.
- 3:39>> But before I told them I divided them
- 3:40up, I asked them, "Which group do you
- 3:42want [music] to be in?" Almost all of
- 3:43them wanted to be in the audio version.
- 3:45Most people, if there's an option
- 3:46between something easier and something
- 3:48harder, they're going to take the easier
- 3:49version to get the the result.
- 3:50>> So, they went in confident, then came
- 3:54the quiz.
- 3:54>> Well, the first question on the quiz was
- 3:56which group would you have rather been
- 3:58in? And the people who wanted to be who
- 4:00were in the audio condition almost
- 4:02universally wish they were in the other
- 4:03condition. They knew they didn't know.
- 4:05Now, they didn't know they didn't know
- 4:07until they had to
- 4:08be forced with recalling it in front of
- 4:11a quiz. If I had
- 4:12I asked them while they were learning
- 4:13it, they said, "Oh, I'm fine. This is
- 4:14great." But when when they started to
- 4:15think, "Do I know this stuff?" Uh-oh, I
- 4:17don't.
- 4:19And they didn't.
- 4:20>> They felt like they were learning, but
- 4:22they [music] weren't. That's kind of
- 4:24scary. How often do you finish a podcast
- 4:26feeling smart and realize an hour later
- 4:29you couldn't explain a single thing you
- 4:31[music] just heard? This happens to me
- 4:33pretty often.
- 4:34But why does it happen? Why would
- 4:36students perform worse with audio? There
- 4:38are a couple of reasons.
- 4:40First, reading on paper engages our
- 4:43decoding [music] pathway. Your eyes scan
- 4:46the letters, recognizing and
- 4:47distinguishing [music] shapes. Then in
- 4:49the brain, grapheme-to-phoneme mapping
- 4:52converts symbols [music] to sounds, and
- 4:55working memory blends these into
- 4:57meaningful
- 4:58>> [music]
- 4:58>> units. Second, audiobooks strip out
- 5:01spatial cues.
- 5:03When you read, your brain builds a map.
- 5:05You remember an idea was at the top of
- 5:08the left page or a definition was
- 5:10halfway through the chapter.
- 5:12Audio doesn't quite have that.
- 5:14Third, your eyes regress constantly
- 5:17while reading. You go back, recheck,
- 5:20re-read. [music]
- 5:21Technically, you can rewind audio, but
- 5:24come on, be honest. [music] Almost
- 5:26nobody does. And other studies back this
- 5:28up. Compared to reading aloud and
- 5:30reading silently to yourself, [music]
- 5:32listening to audio leads to the highest
- 5:35rate of mind wandering.
- 5:37>> [music]
- 5:37>> And the same study found that the audio
- 5:39group had the poorest memory performance
- 5:42and the least [music]
- 5:43interest in the material.
- 5:44And lastly, there are no natural places
- 5:47to stop.
- 5:48>> There were no pages that were natural
- 5:50places to pause, to build the knowledge.
- 5:53This kept going and going and going.
- 5:55>> So, that's the case against [music]
- 5:57audio, pretty damning. But, hold on
- 6:00because the next study tells a different
- 6:03story.
- 6:05>> [music]
- 6:08>> Beth Rogowsky is a professor at
- 6:10Bloomsburg University of Pennsylvania.
- 6:12[music]
- 6:12She was an English teacher for 14 years
- 6:15and wondered what role different
- 6:17learning mediums have on her students.
- 6:20So, she asked, "Does modality matter?"
- 6:23And then she went ahead to test it. She
- 6:25invited 91 college-educated adults and
- 6:28randomly assigned them to
- 6:30>> [music]
- 6:30>> three groups. Group one listened to
- 6:33sections of Unbroken,
- 6:34>> [music]
- 6:35>> a book about World War II. Group two
- 6:38read the exact same sections, but on a
- 6:40Kindle. [music] And group three did
- 6:42both, listening and reading at the same
- 6:44time. After that, everyone took the
- 6:47exact [music] same 48-question quiz.
- 6:50The quiz had basic comprehension
- 6:52questions.
- 6:53>> [music]
- 6:53>> For example, what branch of the military
- 6:55did the castaways represent? The
- 6:58participants took the exact same quiz 2
- 7:00weeks [music] later.
- 7:01>> We found no difference in the three
- 7:03groups.
- 7:04>> Here are the results. The Y axis is the
- 7:07[music] quiz score. The X axis has the
- 7:09audio, text, and dual modality groups.
- 7:13No significant differences, not
- 7:15immediately and not 2 weeks later, and
- 7:17not between men and women. And Beth was
- 7:21genuinely surprised.
- 7:22>> I was surprised because I thought that
- 7:26there would be some effects. Um I was
- 7:29thinking that with the dual modality,
- 7:34you know, a double dose will do better.
- 7:37And
- 7:38yeah, not at all.
- 7:40>> So wait, Professor Daniel and
- 7:42researchers have found massive gaps, but
- 7:45Beth and colleagues found basically
- 7:47nothing. Why?
- 7:50I'll answer that soon, but there's one
- 7:52thing I have to address. When I first
- 7:53read Beth's study, I noticed something
- 7:55in the acknowledgements that, as a
- 7:57journalist, made me raise an eyebrow.
- 8:00The study was partially funded by
- 8:02Audible. Mhm. The audio company. So, I
- 8:05asked her about it.
- 8:06>> They were completely hands-off with us.
- 8:10We set up the study and wrote the
- 8:14results, and they were completely
- 8:15hands-off. We chose the book, the
- 8:17material, and the results are the
- 8:19results, which doesn't really say much
- 8:23for any modality. Maybe Audible would
- 8:25have liked it to have turned out
- 8:27differently, but it is what it is.
- 8:29>> Okay, but there is another issue with
- 8:31her study that I think it's worth
- 8:33knowing. The reading group wasn't
- 8:35reading paper books. They were reading
- 8:37books on Kindles. And screen reading
- 8:40tends to score lower than ink on paper.
- 8:43We have a whole video on this, so go
- 8:45check it out. So, if this study had
- 8:47pitted traditional print [music] against
- 8:50audio, print might have had a benefit.
- 8:54The researchers acknowledge this
- 8:55themselves. They also caution that
- 8:58Unbroken is a narrative non-fiction and
- 9:00may not represent Dan's technical
- 9:03writing,
- 9:04>> [music]
- 9:04>> which brings us back to the
- 9:06contradiction.
- 9:12So, we have two studies that appear to
- 9:14disagree. David Daniel found a 28% gap.
- 9:18Beth Rogowsky found no gap at all. Both
- 9:21are rigorous studies. Both are
- 9:23peer-reviewed. What's going on? Well,
- 9:26the answer is in what the participants
- 9:28were trying [music]
- 9:29to do. David's students were studying
- 9:31textbook material for a college
- 9:33psychology quiz. Beth's participants
- 9:35were reading a narrative for a
- 9:38Daniel Willingham, a cognitive
- 9:39psychologist at the University of
- 9:41Virginia, who studies how the brain
- 9:43learns, outlined this problem for me.
- 9:46>> Most people experience narratives as
- 9:49easier to follow in an audiobook format.
- 9:52You get a lot of support from your own
- 9:55knowledge of the conventions of a
- 9:58narrative. When you're reading something
- 10:01like Aristotle, information-dense
- 10:04texts are organized hierarchically. As
- 10:07I'm on page, say six, I'm really,
- 10:11ideally, supposed to be relating that to
- 10:14what I read on page two. That's very
- 10:17difficult to do. It's easier with print.
- 10:21It's easy to flip back.
- 10:22>> That's the key. Narrative content flows
- 10:25forward. Textbook content is layered and
- 10:28hierarchical.
- 10:30You have to hold pieces in your head and
- 10:32connect them. The thing is that audio
- 10:34doesn't really give you that
- 10:35flexibility.
- 10:37>> People who love print have a way of
- 10:39reading that is more contemplative.
- 10:44You like to really marinate in the text.
- 10:48You like to set the pace, and there are
- 10:50moments you like to sort of pause and
- 10:53really think about the idea that you've
- 10:56just read. And this is one of the key
- 10:59differences between reading print
- 11:00[music] and read and listen to an
- 11:02audiobook. With print, you set the pace.
- 11:04With an audiobook, the narrator is
- 11:06setting the pace. And so you don't have
- 11:09those opportunities to sort [music] of
- 11:11like, "No, I want to enjoy that sentence
- 11:14that I just read. That was beautiful."
- 11:15Or I want to reflect on that idea, cuz
- 11:18the damn narrator is off, you know,
- 11:20they're still they're still talking.
- 11:22>> So the contradiction isn't really a
- 11:25contradiction. Both findings are true.
- 11:28The format should match the purpose.
- 11:31Okay, but what about both at once? What
- 11:34if we combine them, reading and
- 11:36listening at the same time? Well, I used
- 11:39to do this in grad school all the time.
- 11:41Audio at 1.5, eyes on the page. It feels
- 11:44like a superpower, and I thought it was
- 11:45quite helpful. But, a meta-analysis of
- 11:4830 studies with nearly 2,000
- 11:50participants tested exactly this, and
- 11:54the overall effect was trivially
- 11:55positive. A bump barely clearing
- 11:58statistical noise. So, basically, it
- 12:01doesn't seem to work. Beth Stol modality
- 12:03group confirmed it. She expected a
- 12:05double dose to help, but it didn't. More
- 12:08is not always better. So, if you enjoy
- 12:12it, keep doing it. It'll help you stay
- 12:14engaged. Just don't expect it to make
- 12:16you a super learner.
- 12:22Now, here's something important I have
- 12:24to flag, because all three researchers
- 12:26brought it up independently. And it's a
- 12:28thing that worries them most about how
- 12:30their findings will be used.
- 12:32>> If we apply this to children, so I came
- 12:35from the K-12 environment. I taught
- 12:39middle level English for a long time.
- 12:41Students need to be able to read. And
- 12:45when we allow them to just listen to the
- 12:48audio, we are denying them practice in
- 12:52reading. We don't want this to be
- 12:54misinterpreted as, "Okay, let's just
- 12:58have students listen to audiobooks and
- 13:00not tackle reading." Creating a literate
- 13:03society is important.
- 13:05>> None of these studies were about kids
- 13:06[music] learning to read. They were
- 13:08about adults who already can. The
- 13:11finding is about comprehension, [music]
- 13:13not about building the skill of reading
- 13:14itself.
- 13:16These are two completely different
- 13:17things. So, is listening to a book
- 13:20really the same as reading one? Here's
- 13:22the honest answer. For [music] general
- 13:24comprehension, reading and listening are
- 13:26functionally equivalent, the same.
- 13:29The brain processes meaning the same
- 13:31way. [music]
- 13:31>> If you're doing a book club or just want
- 13:35to relax and you're driving [music] or
- 13:38commuting to work, listening to an
- 13:40audiobook is perfectly fine. It's you're
- 13:43going to get the same
- 13:45comprehension out [music] of it.
- 13:47>> However, for deep learning of complex
- 13:49material, print holds a small but
- 13:52consistent advantage with self-pacing
- 13:54and spatial memory [music] cues.
- 13:56And crucially, the attentional demands
- 13:59of reading reduce mind-wandering.
- 14:02>> [music]
- 14:02>> So, use audiobooks to expand the breadth
- 14:05of what you consume during time that
- 14:07would otherwise be [music] dead. All the
- 14:10researchers I spoke with said the same
- 14:12thing about their own habits. Beth
- 14:14listens in the car and [music] while
- 14:15folding laundry, but reads text when
- 14:18she's preparing for a class.
- 14:19>> I don't start with [music] print. I
- 14:21start and end with print. I leave
- 14:23audiobooks solely for [music]
- 14:26stuff that is, you know, pretty light
- 14:28stuff.
- 14:29>> Here's the single most important [music]
- 14:31finding. The variable that matters the
- 14:33most is not the format. It's attention.
- 14:36An audiobook listened to with full
- 14:39concentration will beat reading a print
- 14:41book while distracted [music] and vice
- 14:44versa. And the same applies to other
- 14:46audio mediums like podcasts.
- 14:48>> If you listen to a podcast and you want
- 14:49to remember it, stop it and say, "Do
- 14:52[music] I remember what I just what I
- 14:53just heard?" Check yourself
- 14:56before you wreck yourself.
- 14:58>> So, the medium is not the message. Your
- 15:01focus is. If you found this interesting,
- 15:04drop a comment. And which do you think
- 15:06is more effective for you,
- 15:09listening to or reading a book, and why?
- 15:11Share this video with friends and
- 15:13family. They might learn a thing or two.
- 15:16And if you want to see more content like
- 15:18this one exploring the science of human
- 15:20cognition and learning, make sure to
- 15:23subscribe.
- 15:25Thank you for watching or
- 15:27listening.
- 15:33>> [music]
About this transcript
This page contains the full transcript of What Scientists Found About Audiobooks by The Upgrade with Makai Elías Calles, generated from the public captions YouTube serves with the video. The transcript has 2,420 words across 401 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.
What you can do with it
Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to choose a caption track available on the video.
Free YouTube transcript tool
YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, caption language selection, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.