hey, welcome to 12tone! let's try an experiment.
I'll play you a song, and you try to figure out what it is.
ready? here we go.
(bang) did that seem too short?
ok, let's try another one.
(bang) got it? well, how about this? (bang) ok, let's do one more for fun.
(bang) in case you missed any, those were Hit Me Baby One More Time by Britney Spears,
Every Breath You Take by The Police, Michael Jackson's Thriller, and Nirvana's Smells Like
Teen Spirit.
so how'd you do?
statistically speaking, it's likely you got at least one, and you may have even got all
four.
but each clip was less than half a second long: how were you able to put together so
much information in such a short time?
well, now we're getting into brain stuff and I'm just a music theorist, so I'm gonna have
to call for backup from my neuroscientist friend, Alie Astrocyte from Neurotransmissions!
Hey Cory!
Thanks for having me.
To be able to recognize short clips like these, we've got to pull from our long-term memory.
Scientists split long-term memories into a couple of different categories - declarative,
or explicit, and nondeclarative, or implicit.
Our memories of facts and events - and art and music - are declarative memories.
Certain brain regions are really important for letting you store lyrics like when you
memorized Hit Me Baby One More Time back in 1998 - and for remembering exactly what those
opening riffs sounded like when you heard them just now.
Scientists believe that our memories are processed through the hippocampus.
The hippocampus is a seahorse-shaped region located deep inside the brain.
The importance of this region in memory formation became clear after a man named Henry Molaison
had his hippocampus removed to cure his epilepsy.
After he woke up from the surgery, doctors discovered that he could no longer form new
memories!
The memories themselves are believed to be stored in the connections between neurons.
Even though we don't really grow many new neurons as we age, our brain cells are plastic
- that means they're adaptable and can change.
That's how our brains are able to connect different kinds of information together - by
wiring together different neurons and brain regions.
these specific results are from a 2010 paper written by music psychologist Carol Krumhansl.
the full paper is linked in the description, but the main point was to better understand
the relationship between memory and music.
that is, when we commit a song to memory, how is that information connected? in order
to explore that, Professor Krumhansl collected a list of 28 popular songs and took tiny slices
of them to play for her experimental subjects.
and I mean tiny: 400 milliseconds, to be exact.
when she tested the clips with the music identification software Shazam it was unable to even recognize
them, but her subjects were able to identify them about 25% of the time.
One interesting result was a strong correlation between confidence and accuracy: that is,
when a test subject knew the song, they also knew that they knew it.
After giving an answer for each song snippet, the subjects would give a confidence rating
from 1 to 6, with 1 being "no idea" and 6 being "I definitely know both the title and
the artist".
Despite having shades of grey on the scale (six, to be exact), there were clear clusters
at the extremes: roughly 70% of answers were marked with either a 1 or a 6.
And that confidence was justified: about 95% of 6s had the correct answer and 97% of correct
answers were marked with a 6.
she also asked them to identify a couple other properties of the song, including decade of
release, style, and emotional feel.
the last two are hard to measure objectively, so the results were compared to the subject's
own judgments when listening to longer clips from the same song afterwards.
but the most interesting one to me is the release date, because of a well-known phenomenon
known as the reminiscence bump.
The reminiscence bump is a phenomenon where we are more easily able to recall memories
from certain ages.
In this case, from about ten years old to about thirty.
The theory says that we remember things during this time because there's just a lot of
stuff going on.
Everything is new and interesting, we figure out who we are, and we go through a lot of
developmental, educational, and life milestones during this time.
Since things typically tend to stabilize after that, less bandwidth goes towards creating
new memories.
But what's interesting about the reminiscence bump is that ALL memories seem to be stronger
- even things that aren't so important.
because our memories of that period are so much stronger, it stands to reason that we
would be better at recognizing songs from that era, but that's not what Krumhansl's
results showed.
her subjects were college students born around 1990, and while the most recognized clip was
a Britney Spears song that came out in '98, second place went to Respect by Aretha Franklin,
from 1967 and, overall, there was no significant correlation between release date and recognition.
of course, it's not like children are never exposed to older music, but the point is we
don't seem to be better at recognizing songs from our own generation.
interestingly, though, even when participants failed to recognize the song, they were still
pretty good at guessing when it came out, which means that the differences in tone and
timbre between musical eras is very deeply ingrained in our minds.
but release date isn't the only thing that might've affected recognition.
for instance, the chorus is generally considered the most memorable part of the song: we often
call it the hook, and it's the part most likely to get stuck in your head.
but is it actually easier to recognize? to test this, Krumhansl included two clips from
each song: one from the chorus and one from somewhere else. and while chorus clips did
do slightly better on average, she concluded that the difference wasn't statistically significant.
but there was one factor that was incredibly significant.
any guesses? it's actually really simple: do you hear the name of the song?
clips that included either the full name or part of it were over twice as likely to be
recognized, which makes a lot of sense.
compare this (bang) to this.
(bang) it's a lot easier to place that second one, isn't it?
But science isn't science unless it's repeatable, so Krumhansl tried another variation on the
experiment.
Before, she used 400 millisecond clips.
But in the second experiment, she used only 300 milliseconds, and the results are predictable:
people got a lot worse.
In fact, they only managed to identify the song 11% of the time.
That said, most of the other observed patterns held.
People mostly either confidently identified both song and artist or had no idea at all.
And that last part is important: it was very rare for participants in either study to identify
just one of those factors, implying that a song's name and its artist are very closely
linked in our brains and we have difficulty accessing one without finding the other as
well.
These results make sense, given what we understand about how our brains link related information.
Even beyond this study, researching the effects of music on the brain is a whole subfield
of cognitive neuroscience.
As far as we can tell, music is a uniquely human development - so understanding how we
create and understand music might help us understand what makes our brains so special.
The brain regions dedicated to processing sound are larger in humans than in other species,
and we developed specialized brain regions to process language.
But music is a way of expressing ourselves that doesn't require words.
We know that music can have a physical effect on the brain - a small study in 2010 showed
that subjects who felt "chills" while listening to instrumental music showed an
increase in dopamine signaling in the striatum.
This means that listening to music isn't just psychologically pleasurable - it actually
increases the signaling of the brain's pleasure and reward centers.
There's even evidence that music therapy may be helpful for patients with mood disorders
like depression, and rhythmic movement therapy accompanied by music can help stroke and Parkinson's
patients recover function.
So there's a lot of good evidence that music is important to us, and /good/ for us - but
we've still got a long way to go before we understand where music came from, or why
it's so important.
so that's something you may not have known about your brain, but here's something you
may not have known about mine: I'm autistic.
I talked with Alie about what that means, as well as what scientists know about the
condition, over on her channel, Neurotransmissions.
I learned a lot, and I've put a link to that video in the description, which you should
really check out.
anyway, thanks for watching! if you want to help make these videos possible, please consider
supporting 12tone on Patreon or checking out our store.
you can also join our mailing list to find out about new episodes, like, share, comment,
subscribe, and keep on rockin'.
Không có nhận xét nào:
Đăng nhận xét