May 15, 2025 • 58 mins
What is in the This Week in Science Podcast? This Week: Science, Knowledge, & Power, Forever Chemicals, AI Exaggeration, Internal Inference, Prey Trapping, Labile Language, Librarians Saving Democracy, and Much More Science! Become a Patron! Check out the full unedited episode of our podcast on YouTube or Twitch. Remember that you can find TWIS in […] The post 14 May, 2025 – Episode 1015 – Librarians Will Save Democracy appeared first on This Week in Science - The Kickass Science Podcast.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:00):
This is Twists. This Week in Science, episode number ten fifteen,
recorded on Wednesday, May fourteenth, twenty twenty five. Librarians will
save Democracy, I really think so. Hey everyone, I'm doctor
Keee and tonight on the show, I'm going to fill
your head with chatty apes, ai, exaggeration and inference about
(00:25):
everything pretty much. But first, thanks to our amazing Patreon
sponsors for their generous support of Twists. You can become
a part of the Patreon community at patreon dot com.
Slash This Week in Science disclaimer disclaimer disclaimer. What do
(00:45):
you do when the system is broken? When you watch
your contributions set to the side, a token emotion that
others are going through around you, so many words unspoken.
Do you stay where you are and hope that the
fight is worth the capitulations keeping you up at night?
Do you speak out when others are silent, potentially to
become a sign of the violence they say that people
like you are getting up to Do you step down
(01:08):
when you do? Are you silent underground? Or do you
use your moment as a beacon for others, For sisters
and brothers who know you did what you could, that
you stood for science and knowledge? And working for the
good of society, just like we do here on This
Week in Science Coming up next.
Speaker 2 (01:33):
I've got the kind of mind I can't get enough.
I want to learn everything. I want to fill it
all up with new discoveries. It happen every day of
the week. There's only one place to go to find
a knowledge I seek. I want to know what lots
happen this week in science.
Speaker 1 (01:56):
Laws, good science, good science. To everyone, and welcome to
this week's episode of This Week in Science. It's me
doctor Kiki, and I'm here on my own this evening
without conversation partners Blair and Justin have other things there
(02:20):
up to tonight, and so I'm here to talk with you.
I'm so glad to be here, just honored that I
get to spend part of my week every week talking
about science with all of you. Oh yeah, so what
do we have on the show tonight? I have stories
about science in Washington Forever, chemicals, exaggerating AI, chatty apes,
(02:44):
pray manipulation, and astronomical inference. And of course everyone in
the chat room. I'm so glad you're here, and please
ask questions, chat it up, let's talk about stuff. I
am here to talk with you, and as we jump
into the show here tonight, I do want to remind
you all that subscribing to This Week in Science on
(03:04):
your favorite podcast platform is a great way to get
the edited version of the show every single week when
we do the show. You can also find us broadcasting
live Wednesday's eight pm Pacific time ISH on YouTube, Facebook,
and Twitch. You can also find new show notes and
episodes if you go to our website. There's also our
(03:27):
Patreon link and our Zazzle link if you want to
do those things at twists dot org twys dot org.
Are you ready for the science? Yes, I hope you're
ready for the science because I am not keeping silent. No,
We're going to talk about stuff tonight. First first, first
(03:50):
on my list as we dive into the science tonight
is a story about what's happening in Washington and what
has been going on this past week. The big story
this last week is Landra Nelson as resigning from the
National Science Board and the Library of Congress Scholars Council.
(04:14):
A letter was written in timemagazine, Time dot com Why
I'm resigning from positions at the National Science Foundation and
Library of Congress Aldre Nelson has acted as a member
Acting Director of the Office of Science and Technology Policy
until twenty twenty three. She has also been from twenty
(04:37):
twenty four on the National Science Board Library of Congress
Scholars Council. She's been doing these things four years. She
has been working within the institutions of our government and
believes in them and has been doing what she can
to make sure that science and all people in society
are being heard and represented in involved in the process
(05:02):
that is science, that science is responsive to the people
in our country. So anyway, she's been doing that. But
this last week and the May thirteenth is when this
letter came out. Alandra. Doctor Alandra Nelson wrote in her
notice that over the past few years, in these diverse
(05:26):
bodies that are guided by the legislative branch, not the judicial,
the head of the Library of Congress who was just
just fired this last week, the head librarian of the
Library of Congress. In the last week, along with Alandra
(05:47):
Nelson's resigning from the Library of Congress Scholars Council and
the National Science Foundation's Board, doctor Alandra Nelson has She
wrote that she was honored to be appointed. She's been
working for these things for oversight of the National Science
(06:07):
Foundation's mission, also oversight of the Library of Congress and
its mission, both of which are to a certain degree
controlled by Congress. The Library of Congress is part of
a constitutional mandate to preserve American documentation, political documentation, everything
(06:32):
that might be needed for members of Congress to do research,
to be able to learn and govern in an accurate
and responsive way. There is a children's part of the
library that's involved, but it's supposed to hold all of
the texts that are produced within the United States. It's
a massive, massive undertaking to organize this library. And this
(06:57):
last week Trump got rid of the act the head
of the Library of Congress and has put in an
acting Librarian of Congress who was one of his lawyers
while he was when he was found guilty of thirty
(07:17):
four fellony accounts of falsifying business records to conceal a
payment to an adult film star. According to NPR, this
person is Todd Blanche, the Deputy Attorney General of the
United States, and this person has been appointed. There were
a couple of other people who were appointed as additionally,
as the Librarian of Congress was kicked out, also acting
(07:42):
the Register of Copyrights and Director of the Copyright Office
were additionally kicked out. And thank you librarians for listening
to your lawyers and standing up for the legislative branch
Congress and the rights that are constitutionally given to Congress
(08:02):
to preserve knowledge. And they kept appointees that had been
appointed as acting Librarian and Acting Register and Director of
the Copyrights Office. They did not allow them into offices,
and so they are waiting to find out what Congress
tells them to do, because they are thankfully working for Congress,
(08:28):
not the President. Even the President does have the power
to be able to appoint these people to lead the
Library of Congress and also the Copyrights Office, those appointments
have to be confirmed by Congress and they haven't been.
So anyway, I just want to say thank you to
(08:51):
the librarians who are have been standing strong and are
are standing up for the constitution and separation of powers
within our government. Alandra Nelson in her letter and how
she finished this letter, she says, my resignations are both
(09:21):
an exit that amplifies are both an exit leaving and
a voice speaking up. They're not mutually exclusive. My resignations
are both an exit that amplifies the voices of others.
By departing these advisory roles, I am to speak more
clearly in my own language about what they have become
(09:41):
and what they ought to be. This is not an
abandonment of loyalty to these institutions' missions, but rather its
highest expression. The aim of the resignation, she writes, is
to break free of powers that seek to limit knowledge
and silence voice, to signal that certain bound three lines
have been crossed, to insist that advisory roles must expand
(10:05):
knowledge and be more than appendages to predetermined decisions. So
thank you, doctor Alandra Nelson for standing up for what
you believe in and taking this moment to give voice
to what is happening at our government center and how
it's impacting important organizations institutions like the National Science Foundation,
(10:30):
the Department of Energy, the Library of Congress, and how
that will impact information and knowledge and hence power within
our country moving forward. Additionally, this last week, the Environmental
Protection Agency announced just actually this last week, note today,
(10:51):
Oh my gosh. They announced today Wednesday that it's delaying
the timeline for water utilities to comply with getting rid
of forever chemical goals. So p fast, p fast, p
did a little bit. All those, all of them, pfh XA, PFNA, HPFO, PFBS,
(11:13):
there's a whole bunch. In twenty twenty or the earlier
Trump administration, the EPA under Trump, Lee Zelden, the EPA administrator,
said they were already working to protect Americans from p
fos and drinking water. And this started during that first
(11:33):
that first administration, and Leezelden said it's going to continue
under his leadership. However, this last week they said we're
gonna put a delay on it until twenty thirty one
for p PFOA and PFOS, and it's restinding and reconsidering
(11:53):
the limits per E, PFHXXS, PFNA, HPFO, pf BS. These
are regulations that were created during the Biden administration. In
April twenty twenty four, the Biden administration put limits on
these chemicals and the regulation requires communities and the companies
(12:19):
and infrastructure local infrastructure that provides water to either find
different water that's clean or to make sure that there
are filtration systems in place to remove these chemicals. We
know these chemicals are showing up more and more in
people and that they are, but the biologically accumulating, so
(12:42):
that when you drink water, if you eat a fish
who swim in that water, all this stuff is like
it's going to be in you for a really long time.
So anyway, Lee Zelden's like, no, this is in front
of Congress today, doesn't mean that the limits get weaker.
When I go through a process, we follow the law.
At the end of it, the final levels might be
(13:03):
a lower number, not a higher number. Yeah, they might
be Okay, I don't know. But what they're doing is
potentially getting rid of regulation of four compounds that were
previously regulated and extending compliance deadline for two of the
really big ones, PFOA and p FOSS. Now what happens
(13:26):
when you extend the deadline for compliance, Well, nobody complies
until they have to, and then they try and push
it off a little bit longer. So the EPA has
analyzed the costs to companies and it's about a one
point five billion dollars a year for companies to comply
(13:46):
with this regulation. It's a lot, it is. But pfas,
these per fluoro accolades, these forever chemicals cause cancer, heart attacks, strokes,
birth complications, other diseases that are afflicting communities, especially poor communities,
(14:08):
rural communities across our country. And by delaying the removal
of these things, we will continue to bio accumulate. We
will continue to see these compounds in our water, and
we will continue to see health impacts and deaths as
a result that can be specifically tied to these compounds
in our water. Of course, we want to make sure
(14:32):
that the companies who create these chemicals are also regulated.
I mean, it's like plastic in the rivers. The best
way to get rid of plastic in the rivers, in
the ocean other places is by DA da DA making
less plastic. You don't put as much plastic in there, right,
But once it's in there, what do you do While
we're trying to filter it. We're trying to get the
plastic out. These forever chemicals they're not as visible as
(14:55):
those plastics. Microplastics are pretty bad too. But if we
can't only stop or regulate the production, oh, we really are,
and that's good. But you know, a million dollars here,
a few million dollars there, the companies are still breaking
in profits and they're still going to do what they
can to get around regulations. So what can we do?
(15:17):
Localities need to put in filtration or cleaner water sources
to protect people who live there. If you're paying taxes
an area, shouldn't your water? Shouldn't you be shouldn't shouldn't
be clean? I don't know. So anyway, the Safe Water
Drinking Act actually has a measure that I didn't know about.
It's called the Anti Backsliding Provision, and PR says it's
(15:43):
this provision deems it illegal to weaken a drinking water
rule once it is set. So these rules have been set,
and it seems as though the EPA itself is trying
to weaken them. Although you know, the companies that clean,
they're like, thank yeah, this is great. I love it.
Oh my gosh, hello Misha from Slab City. Yes, the
(16:08):
chemicals coming from the salt and sea. Absolutely, we've got
so many pepsi cola. You want more danger? I think
people are working on that for everyone. Okay, moving on
from these these Washingtonian stories, let's talk for a second
(16:34):
about Oh yay, here's danger large language model chatbots. Ah.
Oh my goodness. So we've talked about these a bunch
on the show, and the issue that has come up
in a study this week out of where is this
(16:54):
study from you? U nl E N yes, ooh clicked
universe in the Netherlands. Yes, Lars van der Kouche. Wait
wait note those the press officer, I am sorry you.
Peters at Utrecht Utrecht University and Benjamin chin Yee Western
(17:15):
University and University of Cambridge analyzed five thousand large language
model generated summaries of scientific papers. In seventy three percent
of the cases you want to guess accurate or inaccurate,
(17:36):
more accurate or less accurate. I'm giving you a chance,
more or less accurate. Dun, dun, dun. I haven't seen
any comments yet. I'm sure there's a little bit of
a delay here. Seventy three percent of these cases, the
(17:59):
model's summaryed them inaccurately. Over a year, they collected forty
nine hundred summaries from journals like Nature, Science, the Lancet.
And then their question was how accurate are the summaries
that Claude and chat GPT four and you know, Gemini
(18:19):
and others. What are these large language models doing and
how are they summarizing things? They tried to create accurate
summaries the chatbots, after the summaries were provided that were inaccurate,
were probed again to produce more accuracy. They were specifically
(18:42):
asked to avoid inaccuracies, and then the conclusions were exaggerated
even more often. When they were asked to be more accurate,
they doubled down on over generalization and exaggeration nearly twice.
As like, the researchers say, so, if you have policy makers,
(19:04):
for instance, who are like, oh, no, no, what this study
really says, and they're not asking the researchers, they're asking
chat GPT or Claude or whatever, rock, I don't know,
they're potentially over three quarters of the time, around three
quarters of the time, And if they're asking for like
really precise summaries, they're going to get less precise summaries.
(19:29):
They're going to be getting less and less accuracy and
more and more overstatement and broad claims. The researchers say
that previous studies, you know, suggest that these lms do
the broader claims because they're trained with texts that have
conclusions that generally have like broad claims or a broad
claim in the abstract or for example, he says, human
(19:55):
experts also tend to draw more general conclusions from Western
samples to all people. For example, the generalizations weren't necessarily
weren't problematic in the original articles that they found, but
the summaries were. One of the things that they suggest
(20:17):
here is that the reason that these AI generated science
summaries are not doing a great job of accurately presenting
the science that they're supposed to be summarizing is that
the lms are told to be sensational. They're basically told
(20:39):
to they want a positive response, they want people to
like what they say, and if you have a more
generalized broad statement, it doesn't really say much necessarily, and
if people go, oh yeah, all right, then the AI
(21:01):
has done what it was trained to do, what it
was programmed to do. The alignment, even though it's inaccurate,
the alignment of the model to its algorithm and to
the rules that it has been given in order to
train itself, it's a downward spiral. So the researchers actually
(21:24):
suggest that if you would like to use chatbots to
summarize any scientific news, then outs out there they'd say
Claude had the highest generalization accuracy. And additionally, if you
enforce indirect past tense reporting that helps in your prompts.
(21:45):
If you're a programmer, set the chatbots to lower temperature,
which is a parameter that reduces the chatbots creativity. So
don't let them be creative, just have them be accurate.
We don't want the We don't want the creativity. I
(22:06):
just find this very interesting and lower Laura d what
you're saying here, so they're human then doubling down in
their inaccuracies, right, Oh my gosh. Yeah, shocking, not so shocking.
I think the more we can all talk about the
issue with these lllms, these chatbots, and the algorithms and
(22:29):
the training, all this stuff that goes into the black
box of the model to create these versions of whatever
we're using, that if we if we go into it
knowing that it wants to make us happy, that it
wants us to give a positive response, because a positive
(22:50):
response means that we might engage more. Right, So it's
like scrolling on social media. The algorithms there are are
primed for engagement for you to scroll, scroll, scroll, maybe
click on an ad, maybe click on a thing oh
something something something ooh a baby, ah. You know, the
algorithms are personalized in a way that drives longer time
(23:15):
and more more engagement on their site. And that's what
the algorithms are trying to do in social media. Same
for these llms, their job is not to give you
accurate information even though you ask for it, which is
really weird, right, they are predicting, they're predicting things. It's
(23:40):
they're not. Yeah, they're the worst intern They're really really bad.
But anyway, I think people need to be aware of
this and we all need to be talking about this
so that we can use these tools well as opposed
to have them use us Dan Dan da will not
be good totally speaking of tools, haha, Oh, Aaron anathema.
(24:06):
That's a really good statement. This sounds like a case
of if you make people think that they're thinking, they'll
love you, but if you make them really think, they'll
turn on you in a heartbeat. Mm hmm, exactly. Okay,
just out of JPL this week. Researchers published in the
journal Nature on May fourteenth, and additionally in the journal
(24:30):
Nature Astronomy on April twenty third, two separate studies related
to the interiors of moons are moon and Well it's
not a moon moon, it's the asteroid Vesta. I don't
know what is it a Solar system moon. I know
it's orbiting out there. The asteroid Vesta and the Moon
(24:52):
are very different. However, the Moon is known to have
more of a core, some volcanic activity. There's a lot
of shifting and layers going on in the interior of
the Moon. We've been looking at it for a while.
Vesta as an asteroid, we don't really we haven't really
known about known a lot about. And people have thought
that Vesta is layered in the sense that you have
(25:16):
the heavier minerals and elements that go to the core
to create a core, and then you'd have like a
mantle and a crust. So these differentiated layers are typical
of bodies that have enough gravitational and mass that it
all get They all get sucked in anyway, both of
(25:37):
the Moon and Vesta. They are impacted by gravitational forces
from other solar objects. Right, So, the Moon orbits around
the Earth, and as it orbits around the Earth, the
tidal forces and gravitational forces of the Earth itself go
(26:00):
hey Moon ah, and the Moon goes ah ah, and
it gets squished and squosed, squished and squosed, and the
researchers finally took a method that has been used to
look at some other some other moons and the Solar System,
and they used the gravity recovery and interior laboratory data.
(26:21):
This was a mission. The spacecraft I love the names
of them, Ebb and Flow, orbited the Moon from December
thirty first, twenty eleven to December seventeenth, twenty twelve. So
Ebb and Flow orbit around the Moon. I don't know
if you're watching listening to the podcast right now, I
(26:42):
have my hands going in circles that make absolutely no
sense for an orbit around the Moon. But the methodology
that was used is line of sight to the Earth,
so that the Earth is tracking the location of those
craft in the orbit around the solar object, which speaking
(27:04):
of Solar system, with this moon, so the Moon is
going around the Earth, we're looking at the craft that
are orbiting the Moon going around the Moon, which is
going around the Earth. And as the Moon is going
around the Earth, it's being squished and squosed by the
Earth's tidal forces, and Ebb and Flow we're also impacted
(27:28):
in a little squished and squos by the Moon's gravitational forces.
And so there are small movements based on gravity that
are the result of large pockets of mass, so mineral deposits,
lots of rock, etc. Versus other parts that maybe are
(27:48):
a less dense area of the Moon. And so over time,
very many years actually of these two craft orbiting the
Moon and of all the little wobbles, it's like a
Doppler shift, the little wobbles that come from the gravitational
(28:09):
forces of the Moon itself as it is being impacted
by the gravity of the Earth. It allows us to
infer the mass of the Moon. It allows us to
create a gravitational map of the Moon. And so the
researchers in effect, with this new study that was just published,
(28:33):
they have, according to this press release that was written,
oh it's the person who wrote it is the name
is not here. That's too bad. They have the most accurate,
most detailed lunar gravitational map yet. And so this could
allow future missions to calculate better places to land, so
(28:56):
location and time on the Moon, and to able to
maybe find places where there's hollows where we could put
you know, human colonies, or maybe that there is a
massive deposit of something. But it's it gave us information
(29:18):
the inside of the Moon, reinforcing that the Moon itself
has the near side more squistion squosed than the far side,
because the near side of the Moon is always kind
of being impacted by the tidal forces of the Earth,
and so the volcanic activity is really coming from all that.
(29:40):
It's like the face of the Moon is the Earth's
squeeze toy, and it feels it. It's been. It's been.
It's been squished and squashed and heat up a lot,
heat it up a lot. Anyway, Vesta, which is way
out there, and the Dawn spacecraft went and hung out
and they were able to get data through the Deep
Space Network radio radiometric system to get imaging data from
(30:07):
the Dawn spacecraft when it visited Vesta for almost a year,
and they found that it doesn't have layers in the
way that other solar objects do, and the same kind
of thing line of site vest is going around this asteroid.
And because of the deformities in this weird chunk of
(30:28):
rock out in space, there's some places with lots of iron,
other places where there's not a lot, and the deformities
of the whole asteroid were then kind of shifting the
orbit of Dawn. And then because we were looking at
Dawn like here, there it is, we're checking the Doppler
(30:49):
wobble and what's it doing. And we got such accurate
data that they were able to determine that Vesta has
what they think is a very homogeneous interior that either
there was maybe it hasn't been around long enough to
actually get distinct layers, maybe it doesn't have enough iron
in it for a core to be created. Really, they
(31:13):
think that what it suggests is the that it was
formed because of a massive, massive, I don't know, massive
collision in space somewhere, but Vesta is impacted by Jupiter
and Mars, and so it's kind of there's some really
(31:35):
interesting stuff. Maybe Vesta, you know, Vesta wobbles as it spins,
and so the inertia, the moment of inertia is what
this study, this research is looking at. And you can't
just measure the moment of inertia of a planetary body,
(31:56):
you can't do it. So they're inferring it based on
the wobble of the craft that are orbiting these things.
So it's pretty cool. It's a new methodology and it's
been used Juno and the Galileo spacecraft did Jovian satellites.
iOS gravity was also measured, and so we've been able
(32:19):
to determine these things through the years. But this is
some of the most accurate. And it's been like a
decade in the works for this research looking at this
data and applying this technique to this data set. So
it's pretty cool. I know, Pepsi cola. Space science is
so cool. Yeah, and yes, Nisha a wacky waving inflatable
(32:43):
flailing arm tube, lady tonight. Yes, that's the way it's working.
Oh my goodness. I don't know. Maybe someday we can
all use, you know, some little orbiting thing to figure
out our moments of inertia. No, thank you, I don't
think I want that. Moving on, I had two studies
(33:07):
this week related to trapping of prey, and the first
one is one of Blair's favorites, the assassin bug and
the assassin bug. This study looking at this particular species
of assassin bud bug Paha ben paha paha ben Kakia
(33:31):
pilliceps Paha ban Kakia pilliceps. This assassin bug, it's a
scary looking bug. Oh my goodness, I'm gonna share a
picture of this link of this thing here. This particular
assassin bug really really really likes their prey species of
(33:52):
a certain species of stingless bee. I mean number one,
the bee doesn't sting, so that makes it easier. And additionally,
the bees have an arrangement where they put a resin
around the outside entrances to their hives, and normally invaders
to the hive come in predators, and they get stuck
(34:15):
in the resin, and then the bees can come and
even though they don't sting, they can still potentially like
push a predator out or get rid of it in
some way. Is that how this assassin bug uses the resin?
Does it just allow itself to get stuck in the resin?
Oh no, no, no, no, no, no, it does not.
(34:37):
This assassin bug takes the resin, breaks it off of
the outside of this the entrance to the bee hive,
and then rubs it on its legs and then sticks
its legs out into the entrance near the entrance, and
the researchers don't know if the assassin bug ug is
(35:01):
if or not the assassin if the bees are approaching
because of the smell of the resin having been been disturbed,
so the resin's normally sitting there. Maybe because it was disturbed,
that is a signal to the bees to come out
and to see what the mother is right or is
(35:23):
there something else going on because these assassin bugs are
not getting stuck in it. The researchers say that they
are using that they are using the resin as a tool.
And so this tool they use not as if they
(35:44):
are choosing to use it and they know what they're
doing with it and they're changing the way that they
use it. They always use it the same way. The
researchers say, it's like an instinct that these assassin bugs.
It's like they're born with the knowledge that are going
to go to this prey species hive and they are
going to rub resin on their legs and then eat
(36:08):
the stingless bees that they're going to catch. Use it
to catch the stingless piece. Yes, it's a sticky bandit's
exactly Paul Disney, Yeah, and uh Misha exactly trapping their
security alarms off for a snack. So that's potentially what
is exact exactly what's going on here. But the fact
(36:30):
that they manipulate the resin is why the researchers are
suggesting that this is tool use, but they aren't saying
that it is like a cognitive strategy. They're saying it's
an instinct that they instinctively use the tool. And the
reason that the different, the reason they're calling it a
(36:51):
tool is because they're not just putting their legs near
the resin and sitting near the resin waiting for the bees.
They're picking it up, they're breaking it apart, and they're
they're putting it on their arms or their legs. I mean,
I guess it's an instinct, so it's all legs. Yeah,
but hey, that's how they catch their yummy, stingless bee. Praise.
(37:14):
That's what they want and this is how they get it.
The researchers did find that the there was a difference
in which legs they used. They they found the they
had more prey if they were able to use their
(37:35):
four and their mid legs as opposed to no resin
or their back legs, which makes sense the front and
mid legs are closer to the mouth, right closer to
the mouth parts. The fact that it's an instinct, though,
is something that I find very interesting, and the idea
(37:57):
that tool use can be instinctual. Is a new idea
for me, Like, we don't know how to use pencils, right,
Nobody says this is how you do it. This is
what you go. Use this pencil and you'll communicate. That's
not that's not a thing. This telephone. These are tools, right,
(38:21):
Our hands are tools, but things that we put on
our bodies. And I think I think there's a real
area for inquiry here about the idea of tool use
as or a specific tool tool use, not general tool use,
but a specific tool that's instinctual. And how does that
(38:45):
instinct not just fear of snakes, not just fear of spiders.
How does the instinct to rub your legs in resin
and wait at the mouth of a beehive? How does
that get into your little insect brainstem. I think it's
(39:07):
very think it's very fastening. AnyWho, totally interesting stuff. So
let's talk about flamingos. Now, this is our other tool
wielding Well, no, they're not tool they're beak wielding animals.
And the interesting thing that was just discovered about flamingos
(39:27):
that came out of UC Berkeley this last week and
was published in the proceedings of the National Academy of Sciences.
Researchers from UC Berkeley, Georgia Institute of Technology, Kennessas State University,
and the Nashville Zoo. They worked with Chilean flamingos at
the Nashville Zoo. The researchers what The lead researcher on
(39:52):
this study, Victor or Taker Jimenez at UC Berkeley. He
specializes in biomechanics and was at the zoo with his
family and it was looking at the flamingo beaks and said,
they're supposed to just be filter feeders, which means that
they shouldn't be doing lots of motions or active. They're
(40:13):
supposed to be passive filter feeders, which means stick your
beak in the water and the water goes through your
beak and you passively get food. That's what everyone thinks,
serene single legs standing flamingos with their head in the water,
just like drinking straws. Not even anyway, it's more active
(40:34):
even than drinking straws. So apparently through his studies he
was able to determine that the l shaped beak of
the flamingo allows them to basically have a little efficient straw.
That's in the water. They clatter, chatter. They chatter their
beaks something like ten times a second, twelve times a second.
(40:58):
I can't even do it something like that anyway, it's
a really fast chattering of the beak, and the chattering
of the beak creates a section and a turbulence in
the water that enables them to actively filter feet. Even
beyond this, he looked at their feet and they have
(41:19):
webbed feet. Flamingos don't just have long bird toes. They
have webbing between their toes. But it's not a firm webbing.
It's very floppy. And he did studies to find out
about the difference between floppy flamingo feet and ferm flamingo feet,
(41:42):
and what he was able to find is that floppy
flamingo feet are better for mixing up the bottom of
the water and pulling organisms like brine shrimp out of
the sediments to get into the water column where they're
chattering beak could be used to enable the flamingos to
(42:06):
uh basically be you know, be pink drinking straws in
the water. Yes, so these flamingos, these Chilean flamingos, they
are not passive filter feeders. Oh no, no, you don't
see anything other than maybe a slight ripple on the
top of the water. But what they are doing is
(42:26):
actually very active, and they are specifically adapted to be
able to as efficiently as possible get their food source
into their mouths as they can. Go flamingos, Go, go flamingos.
(42:46):
I think they're so pretty. They are very smelly. If
you've been to a zoo where flamingos are, they're kind
of stinky birds. But I have a new I have
a new appreciation for the lay in flamingos because of
this study. The fact that they're there going. The chattering
(43:09):
their beaks and the shape of their beak is very
specific to how they feed, which I think is really cool.
I never really thought of the way that their beaks
are formed and how that might contribute to the flow
(43:34):
of food and water. But it's kind of like, I
guess if you have your finger on the end of
a straw. The chattering is like if you have your
finger on the end of a straw and you move
up and down and down against the end of the word,
and you pull water up the column because of water tension.
So I'm imagining that that is something to do with it,
but I might be wrong. Oh, here's another here's another
(43:58):
video I'm going to show really quickly, the brine shrimp.
Here comes a floppy flamingo foot Ooh it creates four texes,
no bortices. There we go. So much fun, so much,
(44:19):
so much fun.
Speaker 3 (44:20):
Oh.
Speaker 1 (44:20):
We can also see more tornado like vortices when the
L shaped beak of the flamingo is retracted. They actually
three D printed a flamingo beak to see exactly how
it changed the water column as it was removed from
the water. So this L shaped flamingo beak as it's
(44:44):
pulled upwards retraction induces tornado like vortices. There's so much
going on there. It's so exciting. I really I think
I probably find it way too exciting. Okay, I have
one more study that I brought for this evening, and
(45:06):
then we can chat a little bit more if you
would like. Kevin reared and yeah, the plastic flamingos don't
smell so much exactly. I know. So how much did
chimpanzees talk? I mean, I'm not talking about Coco the gorilla,
talking about chimpanzee language. How much is chimpanzee language, How
(45:30):
versatile is it? How much does it change or is
it fixed? Apparently, researchers for a very long time thought
that chimpanzees had fixed call vocalizations. And even though studies
with bonobos are more closely related primate relatives that it
(45:56):
looks as though they may have more complex language structure,
chimpanzees hadn't been looked at in this particular way. And
we've seen through previous studies that there's a vocal repertoire
of twelve main single call types. So these are single
vocal units, and there's some variants around that for particular
call types, and so there are criterion and then they
(46:22):
have to be able to put these twelve calls into
vocal sequences and then use them across different sequences. And
so the question is what's going on there? Are they
just like big apple friend, I'm just I'm not going
to food, oh hawk. I mean, I have no idea,
(46:44):
I don't know what their calls are. But these researchers said, Okay,
there's got to be more going on, and let's take
a look at what's happening. So they found that there
are compositional call combinations that these that chimpanzees utter bigrams
(47:07):
or two call vocal sequences, and they say that they
may combine information about self identity and either food like
a pant hoot plus a grunt or subordination a pant
hoot and panted grunt, or information about danger and recruitment
who bark sequence. I love all these. So there's question
(47:32):
about events specifific specificity these bigrams, how much combination can
be used, and so how broad is the language syntax
that chimpanzees use. There isn't a lot of work that
has really dug into this previously. And so these researchers
looked at four comminatorial mechanisms corresponding to four different scenarios.
(47:57):
And so they checked out mechan that were involved here
and they were like, oh, we're going to show you
a picture, and then what do you think about this picture?
And what are you gonna what are you gonna do
in this situation? Go chimpanzees whoo. And as they went
through these studies, they found the chimpanzees, the chimpanzees are
(48:17):
like fifty of them in the study that they were using,
that there actually was a fairly versatile combination past sequence
or different for different units in different situations, which I
don't know about you. I don't think it's that. I
don't think it's that weird, honestly, But the researchers that
(48:41):
are saying, hey, this is really awesome because it suggests
that the uh, the syntax or grammar aspect, the syntax
aspect of human speech of Homo sapien, Homo habilists, you know,
going way way way back, maybe even started earlier than
(49:05):
our genus itself, that maybe the basis for language had
its start much much earlier in primate evolution. And there's
a part of me that goes, duh, of course it did.
It just we're the first, Oh my gosh, just because
(49:27):
we've been able we know what our words are, and
we haven't been able to figure out what chimpanzees are
saying to each other. We're appreciative of bird song and
we use it as a model for human speech and
the complexity and dynamics of bird song. We look at it,
it's like, oh, oh, it's amazing, But we don't think
(49:47):
whoa convergent evolution and that there's more being said, there's
more contextual information being shared than we've figured out anyway. Yeah,
it's exciting, so everyone be excited. Your language, my language, language,
(50:12):
the language that we use to communicate wherever we are
in the world. The basis for it came before our brains,
came before us, and it probably came before a common
ancestor that we share with a chimpanzee. So exciting. I know,
(50:34):
I love it. But that's what I did. I put
these stories together into a show for everyone tonight, and
I just talked and talked and talked, and I can't
believe it's already been an hour. My goodness, how do
we do that? Everybody? Did everything? Anything happen in your
lives this week? Anything sciencey that you're interested in.
Speaker 2 (50:59):
Yeah?
Speaker 1 (50:59):
I thought they were fun stories. Thanks for being here
and being here for me sharing them with you, because
I mean, honestly, that's what I really love to do,
is try and share these stories with you. I find
things and I go, that's really cool, and then I
want to find out, you know, if I can find
(51:21):
a neat angle to them, why they're interesting to me,
why they might be interesting to you, what other questions
can we ask about them? Thanks? Paul Disney. Has been
a fun it's been a fun hour. Yeah, thank you
so much for being here, all of you. Thank you. Thanks,
Pepsi Cola. I am gonna finish the show then, oh
(51:44):
my gosh, it's so weird not to have my co
hosts here. They're off. They'll be back again next week.
But all of you, thank you for being here. And
I hope you enjoyed the show or what it was
with me talking about science for an hour. I hope
you liked it. I really do want to thank as
always the people who help make the show possible, my
(52:06):
co hosts who aren't here, all of you in the
chat room who are here listening so that I don't
feel like I'm just talking at a camera and not
actually interacting with you. I appreciate you being here so much,
and those of you who help keep the chatroom really good,
a good, healthy, respectful place to be. I appreciate you
being a part of that effort. Let's all be good
(52:26):
to each other. Fata, thank you for your help with
show notes and social media. And I did try to
look into our social media stuff and it's a meta
side of things that I'm having issues with. I got
to work that out. Gord arn Ler, thank you for
being here, Identity for Thank you for recording the show,
(52:46):
and Rachel, thank you for editing the show and as always,
thank you, Thank you. Thank you to our twists or
Patreon sponsors, Thank you too. Robert Norland, Robert W. Farley,
Lauren Gifford, Dana Lewis, Even Mandel, Alan Viola, Erin Anathema,
(53:08):
Arthur Keppler, What to see you Tonight, Aaron, Arthur Kepler,
Craig Potts, Mary Kurtz, Teresa Smith, Richard Badgebob Coles, Kent North, Coote,
George Chorus, bra Bella's Arm, John Rattaswami, Chris Bosniak, Big Art,
Chef's Stad, Donald Stiles AKA Don Stylo, Ali Coffin, Schue,
Bruce eah For far Old Boomer eight four to seven
p I G. Stephen Alberon, Darryll My, Jack Andrews Swanson
(53:28):
fretis one of four, Sky Luke, Paul Ronovich, Kevin Brden, Noodles, Jack,
Brian Carrington, davidy young Blood, Shawn Clarence Slam, John McKee,
Greg Riley, Mark Hessenflow, Steve Lison and ak A Z mckenn, Hayes,
Howard tandis rapping Richard Brendan Minish, Johnny Gridley, grim mide Gee,
Burton Lottimore, Flying Out, Christopher Dryer, Artium, Greg Briggs, John Outwood,
Ridy Garcia de Wolkinson, Rodney Lewis, Paul Phillips, Shang, Kurt
(53:51):
Laars and Craig Land and Sue Astor, Jason Olds, David Nahbor,
Eric Knapp, Lawn makes Eo, Adam Mischkhon, Kevin Perchan, Bob Calder,
Margie Paul dy Disney, Patrick beccararo, and Johnny's Day. Thank
you everyone. Your support on Patreon right now is very,
(54:11):
very appreciated. Thanks to everyone who's able to be here
for the show. Thank you for those of you who
are able to support the show through Patreon and PayPal.
I know there are others of you out there who
are supporting us on PayPal, and I do appreciate that
method of support as well. I really can't do the
show without all of you. If you are interested in
(54:32):
heading over in supporting Twists as a Patreon spot sponsor,
I can't speak tonight. Head over to Twist dot org,
click on the Patreon link and choose your level of support.
We also have our Zazzle store where you can go
and find cool Twist merchandise. Click on that Zazzle link.
Parts of the purchase do go to help. This Week
(54:54):
in Science Oh well, next week's show. Oh my my gosh,
we'll be back. I'll be back, and I do believe
my co hosts will be back, Justin and Blair. I'll
be back. We will be speaking with each other and
you about science news. Wednesday apm Pacific time, broadcasting live
(55:16):
from Twitch, YouTube, and Facebook. Look for this Week in
Science hit subscribe and notifications. If you want to watch
us live every week, but if you want to listen
to us as a podcast, make sure you look for
this Week in Science wherever podcasts are found and subscribe.
Make sure you share with a friend as well. That's
if you enjoyed the show. I mean, even didn't enjoy
(55:37):
the show, you should just subscribe your friends anyway, and
then they could be like, ah, why'd you do that? No.
For more information on anything you've heard here, today's show, notes,
links to stories, etc. Go to Twist dot org, where
you can also sign up for a newsletter. I will
send one. I promise it's going to be great when
I do. We love your feedback as well. So if
(56:00):
you have a topic you want us to cover, address
suggestion for an interview, you know, just a comment, let
us know on one of our social media accounts, or
usually twists Twist Science or this week in Science out
there wherever we are, or you can just send us
an email. And if you do, though, make sure you
put Twists into the subject line so that your email
(56:21):
does not get sucked into the L shaped beak of
a Chilean flamingo because I am not going near there
because it's smells fishy. We really look forward to discussing
science with you again next week, and we hope that
we will see you again. If you learned anything from
(56:43):
the show, remember it's all in your head.
Speaker 4 (56:51):
Just tweaking science. This week in science, This week in science,
This week in science. It's the end of the world.
So I'm setting up shop. Got my batter unfurl it
says the scientist is in. I'm going to sell my advice,
(57:11):
show them how to stop the robots with a simple device.
Speaker 5 (57:15):
I'll reverse all the warming with a wave of my hand,
and it'll cost you is a couple of grass. Because
this week science is coming your way.
Speaker 4 (57:27):
So everybody listens to what I say.
Speaker 5 (57:31):
I use the scientific method for all that it's worth,
and I'll broadcast my opinion all over the.
Speaker 1 (57:40):
Well.
Speaker 5 (57:40):
It's this week in science, This week in science, This
week in science is iscire Science.
Speaker 3 (57:48):
This week in science, This weekend Science, This week in science,
This week is science. This week in science, This week
in science, This week in science, This week in side
Speaker 1 (58:08):
Mm HM
This is Twists. This Week in Science, episode number ten fifteen,
recorded on Wednesday, May fourteenth, twenty twenty five. Librarians will
save Democracy, I really think so. Hey everyone, I'm doctor
Keee and tonight on the show, I'm going to fill
your head with chatty apes, ai, exaggeration and inference about
(00:25):
everything pretty much. But first, thanks to our amazing Patreon
sponsors for their generous support of Twists. You can become
a part of the Patreon community at patreon dot com.
Slash This Week in Science disclaimer disclaimer disclaimer. What do
(00:45):
you do when the system is broken? When you watch
your contributions set to the side, a token emotion that
others are going through around you, so many words unspoken.
Do you stay where you are and hope that the
fight is worth the capitulations keeping you up at night?
Do you speak out when others are silent, potentially to
become a sign of the violence they say that people
like you are getting up to Do you step down
(01:08):
when you do? Are you silent underground? Or do you
use your moment as a beacon for others, For sisters
and brothers who know you did what you could, that
you stood for science and knowledge? And working for the
good of society, just like we do here on This
Week in Science Coming up next.
Speaker 2 (01:33):
I've got the kind of mind I can't get enough.
I want to learn everything. I want to fill it
all up with new discoveries. It happen every day of
the week. There's only one place to go to find
a knowledge I seek. I want to know what lots
happen this week in science.
Speaker 1 (01:56):
Laws, good science, good science. To everyone, and welcome to
this week's episode of This Week in Science. It's me
doctor Kiki, and I'm here on my own this evening
without conversation partners Blair and Justin have other things there
(02:20):
up to tonight, and so I'm here to talk with you.
I'm so glad to be here, just honored that I
get to spend part of my week every week talking
about science with all of you. Oh yeah, so what
do we have on the show tonight? I have stories
about science in Washington Forever, chemicals, exaggerating AI, chatty apes,
(02:44):
pray manipulation, and astronomical inference. And of course everyone in
the chat room. I'm so glad you're here, and please
ask questions, chat it up, let's talk about stuff. I
am here to talk with you, and as we jump
into the show here tonight, I do want to remind
you all that subscribing to This Week in Science on
(03:04):
your favorite podcast platform is a great way to get
the edited version of the show every single week when
we do the show. You can also find us broadcasting
live Wednesday's eight pm Pacific time ISH on YouTube, Facebook,
and Twitch. You can also find new show notes and
episodes if you go to our website. There's also our
(03:27):
Patreon link and our Zazzle link if you want to
do those things at twists dot org twys dot org.
Are you ready for the science? Yes, I hope you're
ready for the science because I am not keeping silent. No,
We're going to talk about stuff tonight. First first, first
(03:50):
on my list as we dive into the science tonight
is a story about what's happening in Washington and what
has been going on this past week. The big story
this last week is Landra Nelson as resigning from the
National Science Board and the Library of Congress Scholars Council.
(04:14):
A letter was written in timemagazine, Time dot com Why
I'm resigning from positions at the National Science Foundation and
Library of Congress Aldre Nelson has acted as a member
Acting Director of the Office of Science and Technology Policy
until twenty twenty three. She has also been from twenty
(04:37):
twenty four on the National Science Board Library of Congress
Scholars Council. She's been doing these things four years. She
has been working within the institutions of our government and
believes in them and has been doing what she can
to make sure that science and all people in society
are being heard and represented in involved in the process
(05:02):
that is science, that science is responsive to the people
in our country. So anyway, she's been doing that. But
this last week and the May thirteenth is when this
letter came out. Alandra. Doctor Alandra Nelson wrote in her
notice that over the past few years, in these diverse
(05:26):
bodies that are guided by the legislative branch, not the judicial,
the head of the Library of Congress who was just
just fired this last week, the head librarian of the
Library of Congress. In the last week, along with Alandra
(05:47):
Nelson's resigning from the Library of Congress Scholars Council and
the National Science Foundation's Board, doctor Alandra Nelson has She
wrote that she was honored to be appointed. She's been
working for these things for oversight of the National Science
(06:07):
Foundation's mission, also oversight of the Library of Congress and
its mission, both of which are to a certain degree
controlled by Congress. The Library of Congress is part of
a constitutional mandate to preserve American documentation, political documentation, everything
(06:32):
that might be needed for members of Congress to do research,
to be able to learn and govern in an accurate
and responsive way. There is a children's part of the
library that's involved, but it's supposed to hold all of
the texts that are produced within the United States. It's
a massive, massive undertaking to organize this library. And this
(06:57):
last week Trump got rid of the act the head
of the Library of Congress and has put in an
acting Librarian of Congress who was one of his lawyers
while he was when he was found guilty of thirty
(07:17):
four fellony accounts of falsifying business records to conceal a
payment to an adult film star. According to NPR, this
person is Todd Blanche, the Deputy Attorney General of the
United States, and this person has been appointed. There were
a couple of other people who were appointed as additionally,
as the Librarian of Congress was kicked out, also acting
(07:42):
the Register of Copyrights and Director of the Copyright Office
were additionally kicked out. And thank you librarians for listening
to your lawyers and standing up for the legislative branch
Congress and the rights that are constitutionally given to Congress
(08:02):
to preserve knowledge. And they kept appointees that had been
appointed as acting Librarian and Acting Register and Director of
the Copyrights Office. They did not allow them into offices,
and so they are waiting to find out what Congress
tells them to do, because they are thankfully working for Congress,
(08:28):
not the President. Even the President does have the power
to be able to appoint these people to lead the
Library of Congress and also the Copyrights Office, those appointments
have to be confirmed by Congress and they haven't been.
So anyway, I just want to say thank you to
(08:51):
the librarians who are have been standing strong and are
are standing up for the constitution and separation of powers
within our government. Alandra Nelson in her letter and how
she finished this letter, she says, my resignations are both
(09:21):
an exit that amplifies are both an exit leaving and
a voice speaking up. They're not mutually exclusive. My resignations
are both an exit that amplifies the voices of others.
By departing these advisory roles, I am to speak more
clearly in my own language about what they have become
(09:41):
and what they ought to be. This is not an
abandonment of loyalty to these institutions' missions, but rather its
highest expression. The aim of the resignation, she writes, is
to break free of powers that seek to limit knowledge
and silence voice, to signal that certain bound three lines
have been crossed, to insist that advisory roles must expand
(10:05):
knowledge and be more than appendages to predetermined decisions. So
thank you, doctor Alandra Nelson for standing up for what
you believe in and taking this moment to give voice
to what is happening at our government center and how
it's impacting important organizations institutions like the National Science Foundation,
(10:30):
the Department of Energy, the Library of Congress, and how
that will impact information and knowledge and hence power within
our country moving forward. Additionally, this last week, the Environmental
Protection Agency announced just actually this last week, note today,
(10:51):
Oh my gosh. They announced today Wednesday that it's delaying
the timeline for water utilities to comply with getting rid
of forever chemical goals. So p fast, p fast, p
did a little bit. All those, all of them, pfh XA, PFNA, HPFO, PFBS,
(11:13):
there's a whole bunch. In twenty twenty or the earlier
Trump administration, the EPA under Trump, Lee Zelden, the EPA administrator,
said they were already working to protect Americans from p
fos and drinking water. And this started during that first
(11:33):
that first administration, and Leezelden said it's going to continue
under his leadership. However, this last week they said we're
gonna put a delay on it until twenty thirty one
for p PFOA and PFOS, and it's restinding and reconsidering
(11:53):
the limits per E, PFHXXS, PFNA, HPFO, pf BS. These
are regulations that were created during the Biden administration. In
April twenty twenty four, the Biden administration put limits on
these chemicals and the regulation requires communities and the companies
(12:19):
and infrastructure local infrastructure that provides water to either find
different water that's clean or to make sure that there
are filtration systems in place to remove these chemicals. We
know these chemicals are showing up more and more in
people and that they are, but the biologically accumulating, so
(12:42):
that when you drink water, if you eat a fish
who swim in that water, all this stuff is like
it's going to be in you for a really long time.
So anyway, Lee Zelden's like, no, this is in front
of Congress today, doesn't mean that the limits get weaker.
When I go through a process, we follow the law.
At the end of it, the final levels might be
(13:03):
a lower number, not a higher number. Yeah, they might
be Okay, I don't know. But what they're doing is
potentially getting rid of regulation of four compounds that were
previously regulated and extending compliance deadline for two of the
really big ones, PFOA and p FOSS. Now what happens
(13:26):
when you extend the deadline for compliance, Well, nobody complies
until they have to, and then they try and push
it off a little bit longer. So the EPA has
analyzed the costs to companies and it's about a one
point five billion dollars a year for companies to comply
(13:46):
with this regulation. It's a lot, it is. But pfas,
these per fluoro accolades, these forever chemicals cause cancer, heart attacks, strokes,
birth complications, other diseases that are afflicting communities, especially poor communities,
(14:08):
rural communities across our country. And by delaying the removal
of these things, we will continue to bio accumulate. We
will continue to see these compounds in our water, and
we will continue to see health impacts and deaths as
a result that can be specifically tied to these compounds
in our water. Of course, we want to make sure
(14:32):
that the companies who create these chemicals are also regulated.
I mean, it's like plastic in the rivers. The best
way to get rid of plastic in the rivers, in
the ocean other places is by DA da DA making
less plastic. You don't put as much plastic in there, right,
But once it's in there, what do you do While
we're trying to filter it. We're trying to get the
plastic out. These forever chemicals they're not as visible as
(14:55):
those plastics. Microplastics are pretty bad too. But if we
can't only stop or regulate the production, oh, we really are,
and that's good. But you know, a million dollars here,
a few million dollars there, the companies are still breaking
in profits and they're still going to do what they
can to get around regulations. So what can we do?
(15:17):
Localities need to put in filtration or cleaner water sources
to protect people who live there. If you're paying taxes
an area, shouldn't your water? Shouldn't you be shouldn't shouldn't
be clean? I don't know. So anyway, the Safe Water
Drinking Act actually has a measure that I didn't know about.
It's called the Anti Backsliding Provision, and PR says it's
(15:43):
this provision deems it illegal to weaken a drinking water
rule once it is set. So these rules have been set,
and it seems as though the EPA itself is trying
to weaken them. Although you know, the companies that clean,
they're like, thank yeah, this is great. I love it.
Oh my gosh, hello Misha from Slab City. Yes, the
(16:08):
chemicals coming from the salt and sea. Absolutely, we've got
so many pepsi cola. You want more danger? I think
people are working on that for everyone. Okay, moving on
from these these Washingtonian stories, let's talk for a second
(16:34):
about Oh yay, here's danger large language model chatbots. Ah.
Oh my goodness. So we've talked about these a bunch
on the show, and the issue that has come up
in a study this week out of where is this
(16:54):
study from you? U nl E N yes, ooh clicked
universe in the Netherlands. Yes, Lars van der Kouche. Wait
wait note those the press officer, I am sorry you.
Peters at Utrecht Utrecht University and Benjamin chin Yee Western
(17:15):
University and University of Cambridge analyzed five thousand large language
model generated summaries of scientific papers. In seventy three percent
of the cases you want to guess accurate or inaccurate,
(17:36):
more accurate or less accurate. I'm giving you a chance,
more or less accurate. Dun, dun, dun. I haven't seen
any comments yet. I'm sure there's a little bit of
a delay here. Seventy three percent of these cases, the
(17:59):
model's summaryed them inaccurately. Over a year, they collected forty
nine hundred summaries from journals like Nature, Science, the Lancet.
And then their question was how accurate are the summaries
that Claude and chat GPT four and you know, Gemini
(18:19):
and others. What are these large language models doing and
how are they summarizing things? They tried to create accurate
summaries the chatbots, after the summaries were provided that were inaccurate,
were probed again to produce more accuracy. They were specifically
(18:42):
asked to avoid inaccuracies, and then the conclusions were exaggerated
even more often. When they were asked to be more accurate,
they doubled down on over generalization and exaggeration nearly twice.
As like, the researchers say, so, if you have policy makers,
(19:04):
for instance, who are like, oh, no, no, what this study
really says, and they're not asking the researchers, they're asking
chat GPT or Claude or whatever, rock, I don't know,
they're potentially over three quarters of the time, around three
quarters of the time, And if they're asking for like
really precise summaries, they're going to get less precise summaries.
(19:29):
They're going to be getting less and less accuracy and
more and more overstatement and broad claims. The researchers say
that previous studies, you know, suggest that these lms do
the broader claims because they're trained with texts that have
conclusions that generally have like broad claims or a broad
claim in the abstract or for example, he says, human
(19:55):
experts also tend to draw more general conclusions from Western
samples to all people. For example, the generalizations weren't necessarily
weren't problematic in the original articles that they found, but
the summaries were. One of the things that they suggest
(20:17):
here is that the reason that these AI generated science
summaries are not doing a great job of accurately presenting
the science that they're supposed to be summarizing is that
the lms are told to be sensational. They're basically told
(20:39):
to they want a positive response, they want people to
like what they say, and if you have a more
generalized broad statement, it doesn't really say much necessarily, and
if people go, oh yeah, all right, then the AI
(21:01):
has done what it was trained to do, what it
was programmed to do. The alignment, even though it's inaccurate,
the alignment of the model to its algorithm and to
the rules that it has been given in order to
train itself, it's a downward spiral. So the researchers actually
(21:24):
suggest that if you would like to use chatbots to
summarize any scientific news, then outs out there they'd say
Claude had the highest generalization accuracy. And additionally, if you
enforce indirect past tense reporting that helps in your prompts.
(21:45):
If you're a programmer, set the chatbots to lower temperature,
which is a parameter that reduces the chatbots creativity. So
don't let them be creative, just have them be accurate.
We don't want the We don't want the creativity. I
(22:06):
just find this very interesting and lower Laura d what
you're saying here, so they're human then doubling down in
their inaccuracies, right, Oh my gosh. Yeah, shocking, not so shocking.
I think the more we can all talk about the
issue with these lllms, these chatbots, and the algorithms and
(22:29):
the training, all this stuff that goes into the black
box of the model to create these versions of whatever
we're using, that if we if we go into it
knowing that it wants to make us happy, that it
wants us to give a positive response, because a positive
(22:50):
response means that we might engage more. Right, So it's
like scrolling on social media. The algorithms there are are
primed for engagement for you to scroll, scroll, scroll, maybe
click on an ad, maybe click on a thing oh
something something something ooh a baby, ah. You know, the
algorithms are personalized in a way that drives longer time
(23:15):
and more more engagement on their site. And that's what
the algorithms are trying to do in social media. Same
for these llms, their job is not to give you
accurate information even though you ask for it, which is
really weird, right, they are predicting, they're predicting things. It's
(23:40):
they're not. Yeah, they're the worst intern They're really really bad.
But anyway, I think people need to be aware of
this and we all need to be talking about this
so that we can use these tools well as opposed
to have them use us Dan Dan da will not
be good totally speaking of tools, haha, Oh, Aaron anathema.
(24:06):
That's a really good statement. This sounds like a case
of if you make people think that they're thinking, they'll
love you, but if you make them really think, they'll
turn on you in a heartbeat. Mm hmm, exactly. Okay,
just out of JPL this week. Researchers published in the
journal Nature on May fourteenth, and additionally in the journal
(24:30):
Nature Astronomy on April twenty third, two separate studies related
to the interiors of moons are moon and Well it's
not a moon moon, it's the asteroid Vesta. I don't
know what is it a Solar system moon. I know
it's orbiting out there. The asteroid Vesta and the Moon
(24:52):
are very different. However, the Moon is known to have
more of a core, some volcanic activity. There's a lot
of shifting and layers going on in the interior of
the Moon. We've been looking at it for a while.
Vesta as an asteroid, we don't really we haven't really
known about known a lot about. And people have thought
that Vesta is layered in the sense that you have
(25:16):
the heavier minerals and elements that go to the core
to create a core, and then you'd have like a
mantle and a crust. So these differentiated layers are typical
of bodies that have enough gravitational and mass that it
all get They all get sucked in anyway, both of
(25:37):
the Moon and Vesta. They are impacted by gravitational forces
from other solar objects. Right, So, the Moon orbits around
the Earth, and as it orbits around the Earth, the
tidal forces and gravitational forces of the Earth itself go
(26:00):
hey Moon ah, and the Moon goes ah ah, and
it gets squished and squosed, squished and squosed, and the
researchers finally took a method that has been used to
look at some other some other moons and the Solar System,
and they used the gravity recovery and interior laboratory data.
(26:21):
This was a mission. The spacecraft I love the names
of them, Ebb and Flow, orbited the Moon from December
thirty first, twenty eleven to December seventeenth, twenty twelve. So
Ebb and Flow orbit around the Moon. I don't know
if you're watching listening to the podcast right now, I
(26:42):
have my hands going in circles that make absolutely no
sense for an orbit around the Moon. But the methodology
that was used is line of sight to the Earth,
so that the Earth is tracking the location of those
craft in the orbit around the solar object, which speaking
(27:04):
of Solar system, with this moon, so the Moon is
going around the Earth, we're looking at the craft that
are orbiting the Moon going around the Moon, which is
going around the Earth. And as the Moon is going
around the Earth, it's being squished and squosed by the
Earth's tidal forces, and Ebb and Flow we're also impacted
(27:28):
in a little squished and squos by the Moon's gravitational forces.
And so there are small movements based on gravity that
are the result of large pockets of mass, so mineral deposits,
lots of rock, etc. Versus other parts that maybe are
(27:48):
a less dense area of the Moon. And so over time,
very many years actually of these two craft orbiting the
Moon and of all the little wobbles, it's like a
Doppler shift, the little wobbles that come from the gravitational
(28:09):
forces of the Moon itself as it is being impacted
by the gravity of the Earth. It allows us to
infer the mass of the Moon. It allows us to
create a gravitational map of the Moon. And so the
researchers in effect, with this new study that was just published,
(28:33):
they have, according to this press release that was written,
oh it's the person who wrote it is the name
is not here. That's too bad. They have the most accurate,
most detailed lunar gravitational map yet. And so this could
allow future missions to calculate better places to land, so
(28:56):
location and time on the Moon, and to able to
maybe find places where there's hollows where we could put
you know, human colonies, or maybe that there is a
massive deposit of something. But it's it gave us information
(29:18):
the inside of the Moon, reinforcing that the Moon itself
has the near side more squistion squosed than the far side,
because the near side of the Moon is always kind
of being impacted by the tidal forces of the Earth,
and so the volcanic activity is really coming from all that.
(29:40):
It's like the face of the Moon is the Earth's
squeeze toy, and it feels it. It's been. It's been.
It's been squished and squashed and heat up a lot,
heat it up a lot. Anyway, Vesta, which is way
out there, and the Dawn spacecraft went and hung out
and they were able to get data through the Deep
Space Network radio radiometric system to get imaging data from
(30:07):
the Dawn spacecraft when it visited Vesta for almost a year,
and they found that it doesn't have layers in the
way that other solar objects do, and the same kind
of thing line of site vest is going around this asteroid.
And because of the deformities in this weird chunk of
(30:28):
rock out in space, there's some places with lots of iron,
other places where there's not a lot, and the deformities
of the whole asteroid were then kind of shifting the
orbit of Dawn. And then because we were looking at
Dawn like here, there it is, we're checking the Doppler
(30:49):
wobble and what's it doing. And we got such accurate
data that they were able to determine that Vesta has
what they think is a very homogeneous interior that either
there was maybe it hasn't been around long enough to
actually get distinct layers, maybe it doesn't have enough iron
in it for a core to be created. Really, they
(31:13):
think that what it suggests is the that it was
formed because of a massive, massive, I don't know, massive
collision in space somewhere, but Vesta is impacted by Jupiter
and Mars, and so it's kind of there's some really
(31:35):
interesting stuff. Maybe Vesta, you know, Vesta wobbles as it spins,
and so the inertia, the moment of inertia is what
this study, this research is looking at. And you can't
just measure the moment of inertia of a planetary body,
(31:56):
you can't do it. So they're inferring it based on
the wobble of the craft that are orbiting these things.
So it's pretty cool. It's a new methodology and it's
been used Juno and the Galileo spacecraft did Jovian satellites.
iOS gravity was also measured, and so we've been able
(32:19):
to determine these things through the years. But this is
some of the most accurate. And it's been like a
decade in the works for this research looking at this
data and applying this technique to this data set. So
it's pretty cool. I know, Pepsi cola. Space science is
so cool. Yeah, and yes, Nisha a wacky waving inflatable
(32:43):
flailing arm tube, lady tonight. Yes, that's the way it's working.
Oh my goodness. I don't know. Maybe someday we can
all use, you know, some little orbiting thing to figure
out our moments of inertia. No, thank you, I don't
think I want that. Moving on, I had two studies
(33:07):
this week related to trapping of prey, and the first
one is one of Blair's favorites, the assassin bug and
the assassin bug. This study looking at this particular species
of assassin bud bug Paha ben paha paha ben Kakia
(33:31):
pilliceps Paha ban Kakia pilliceps. This assassin bug, it's a
scary looking bug. Oh my goodness, I'm gonna share a
picture of this link of this thing here. This particular
assassin bug really really really likes their prey species of
(33:52):
a certain species of stingless bee. I mean number one,
the bee doesn't sting, so that makes it easier. And additionally,
the bees have an arrangement where they put a resin
around the outside entrances to their hives, and normally invaders
to the hive come in predators, and they get stuck
(34:15):
in the resin, and then the bees can come and
even though they don't sting, they can still potentially like
push a predator out or get rid of it in
some way. Is that how this assassin bug uses the resin?
Does it just allow itself to get stuck in the resin?
Oh no, no, no, no, no, no, it does not.
(34:37):
This assassin bug takes the resin, breaks it off of
the outside of this the entrance to the bee hive,
and then rubs it on its legs and then sticks
its legs out into the entrance near the entrance, and
the researchers don't know if the assassin bug ug is
(35:01):
if or not the assassin if the bees are approaching
because of the smell of the resin having been been disturbed,
so the resin's normally sitting there. Maybe because it was disturbed,
that is a signal to the bees to come out
and to see what the mother is right or is
(35:23):
there something else going on because these assassin bugs are
not getting stuck in it. The researchers say that they
are using that they are using the resin as a tool.
And so this tool they use not as if they
(35:44):
are choosing to use it and they know what they're
doing with it and they're changing the way that they
use it. They always use it the same way. The
researchers say, it's like an instinct that these assassin bugs.
It's like they're born with the knowledge that are going
to go to this prey species hive and they are
going to rub resin on their legs and then eat
(36:08):
the stingless bees that they're going to catch. Use it
to catch the stingless piece. Yes, it's a sticky bandit's
exactly Paul Disney, Yeah, and uh Misha exactly trapping their
security alarms off for a snack. So that's potentially what
is exact exactly what's going on here. But the fact
(36:30):
that they manipulate the resin is why the researchers are
suggesting that this is tool use, but they aren't saying
that it is like a cognitive strategy. They're saying it's
an instinct that they instinctively use the tool. And the
reason that the different, the reason they're calling it a
(36:51):
tool is because they're not just putting their legs near
the resin and sitting near the resin waiting for the bees.
They're picking it up, they're breaking it apart, and they're
they're putting it on their arms or their legs. I mean,
I guess it's an instinct, so it's all legs. Yeah,
but hey, that's how they catch their yummy, stingless bee. Praise.
(37:14):
That's what they want and this is how they get it.
The researchers did find that the there was a difference
in which legs they used. They they found the they
had more prey if they were able to use their
(37:35):
four and their mid legs as opposed to no resin
or their back legs, which makes sense the front and
mid legs are closer to the mouth, right closer to
the mouth parts. The fact that it's an instinct, though,
is something that I find very interesting, and the idea
(37:57):
that tool use can be instinctual. Is a new idea
for me, Like, we don't know how to use pencils, right,
Nobody says this is how you do it. This is
what you go. Use this pencil and you'll communicate. That's
not that's not a thing. This telephone. These are tools, right,
(38:21):
Our hands are tools, but things that we put on
our bodies. And I think I think there's a real
area for inquiry here about the idea of tool use
as or a specific tool tool use, not general tool use,
but a specific tool that's instinctual. And how does that
(38:45):
instinct not just fear of snakes, not just fear of spiders.
How does the instinct to rub your legs in resin
and wait at the mouth of a beehive? How does
that get into your little insect brainstem. I think it's
(39:07):
very think it's very fastening. AnyWho, totally interesting stuff. So
let's talk about flamingos. Now, this is our other tool
wielding Well, no, they're not tool they're beak wielding animals.
And the interesting thing that was just discovered about flamingos
(39:27):
that came out of UC Berkeley this last week and
was published in the proceedings of the National Academy of Sciences.
Researchers from UC Berkeley, Georgia Institute of Technology, Kennessas State University,
and the Nashville Zoo. They worked with Chilean flamingos at
the Nashville Zoo. The researchers what The lead researcher on
(39:52):
this study, Victor or Taker Jimenez at UC Berkeley. He
specializes in biomechanics and was at the zoo with his
family and it was looking at the flamingo beaks and said,
they're supposed to just be filter feeders, which means that
they shouldn't be doing lots of motions or active. They're
(40:13):
supposed to be passive filter feeders, which means stick your
beak in the water and the water goes through your
beak and you passively get food. That's what everyone thinks,
serene single legs standing flamingos with their head in the water,
just like drinking straws. Not even anyway, it's more active
(40:34):
even than drinking straws. So apparently through his studies he
was able to determine that the l shaped beak of
the flamingo allows them to basically have a little efficient straw.
That's in the water. They clatter, chatter. They chatter their
beaks something like ten times a second, twelve times a second.
(40:58):
I can't even do it something like that anyway, it's
a really fast chattering of the beak, and the chattering
of the beak creates a section and a turbulence in
the water that enables them to actively filter feet. Even
beyond this, he looked at their feet and they have
(41:19):
webbed feet. Flamingos don't just have long bird toes. They
have webbing between their toes. But it's not a firm webbing.
It's very floppy. And he did studies to find out
about the difference between floppy flamingo feet and ferm flamingo feet,
(41:42):
and what he was able to find is that floppy
flamingo feet are better for mixing up the bottom of
the water and pulling organisms like brine shrimp out of
the sediments to get into the water column where they're
chattering beak could be used to enable the flamingos to
(42:06):
uh basically be you know, be pink drinking straws in
the water. Yes, so these flamingos, these Chilean flamingos, they
are not passive filter feeders. Oh no, no, you don't
see anything other than maybe a slight ripple on the
top of the water. But what they are doing is
(42:26):
actually very active, and they are specifically adapted to be
able to as efficiently as possible get their food source
into their mouths as they can. Go flamingos, Go, go flamingos.
(42:46):
I think they're so pretty. They are very smelly. If
you've been to a zoo where flamingos are, they're kind
of stinky birds. But I have a new I have
a new appreciation for the lay in flamingos because of
this study. The fact that they're there going. The chattering
(43:09):
their beaks and the shape of their beak is very
specific to how they feed, which I think is really cool.
I never really thought of the way that their beaks
are formed and how that might contribute to the flow
(43:34):
of food and water. But it's kind of like, I
guess if you have your finger on the end of
a straw. The chattering is like if you have your
finger on the end of a straw and you move
up and down and down against the end of the word,
and you pull water up the column because of water tension.
So I'm imagining that that is something to do with it,
but I might be wrong. Oh, here's another here's another
(43:58):
video I'm going to show really quickly, the brine shrimp.
Here comes a floppy flamingo foot Ooh it creates four texes,
no bortices. There we go. So much fun, so much,
(44:19):
so much fun.
Speaker 3 (44:20):
Oh.
Speaker 1 (44:20):
We can also see more tornado like vortices when the
L shaped beak of the flamingo is retracted. They actually
three D printed a flamingo beak to see exactly how
it changed the water column as it was removed from
the water. So this L shaped flamingo beak as it's
(44:44):
pulled upwards retraction induces tornado like vortices. There's so much
going on there. It's so exciting. I really I think
I probably find it way too exciting. Okay, I have
one more study that I brought for this evening, and
(45:06):
then we can chat a little bit more if you
would like. Kevin reared and yeah, the plastic flamingos don't
smell so much exactly. I know. So how much did
chimpanzees talk? I mean, I'm not talking about Coco the gorilla,
talking about chimpanzee language. How much is chimpanzee language, How
(45:30):
versatile is it? How much does it change or is
it fixed? Apparently, researchers for a very long time thought
that chimpanzees had fixed call vocalizations. And even though studies
with bonobos are more closely related primate relatives that it
(45:56):
looks as though they may have more complex language structure,
chimpanzees hadn't been looked at in this particular way. And
we've seen through previous studies that there's a vocal repertoire
of twelve main single call types. So these are single
vocal units, and there's some variants around that for particular
call types, and so there are criterion and then they
(46:22):
have to be able to put these twelve calls into
vocal sequences and then use them across different sequences. And
so the question is what's going on there? Are they
just like big apple friend, I'm just I'm not going
to food, oh hawk. I mean, I have no idea,
(46:44):
I don't know what their calls are. But these researchers said, Okay,
there's got to be more going on, and let's take
a look at what's happening. So they found that there
are compositional call combinations that these that chimpanzees utter bigrams
(47:07):
or two call vocal sequences, and they say that they
may combine information about self identity and either food like
a pant hoot plus a grunt or subordination a pant
hoot and panted grunt, or information about danger and recruitment
who bark sequence. I love all these. So there's question
(47:32):
about events specifific specificity these bigrams, how much combination can
be used, and so how broad is the language syntax
that chimpanzees use. There isn't a lot of work that
has really dug into this previously. And so these researchers
looked at four comminatorial mechanisms corresponding to four different scenarios.
(47:57):
And so they checked out mechan that were involved here
and they were like, oh, we're going to show you
a picture, and then what do you think about this picture?
And what are you gonna what are you gonna do
in this situation? Go chimpanzees whoo. And as they went
through these studies, they found the chimpanzees, the chimpanzees are
(48:17):
like fifty of them in the study that they were using,
that there actually was a fairly versatile combination past sequence
or different for different units in different situations, which I
don't know about you. I don't think it's that. I
don't think it's that weird, honestly, But the researchers that
(48:41):
are saying, hey, this is really awesome because it suggests
that the uh, the syntax or grammar aspect, the syntax
aspect of human speech of Homo sapien, Homo habilists, you know,
going way way way back, maybe even started earlier than
(49:05):
our genus itself, that maybe the basis for language had
its start much much earlier in primate evolution. And there's
a part of me that goes, duh, of course it did.
It just we're the first, Oh my gosh, just because
(49:27):
we've been able we know what our words are, and
we haven't been able to figure out what chimpanzees are
saying to each other. We're appreciative of bird song and
we use it as a model for human speech and
the complexity and dynamics of bird song. We look at it,
it's like, oh, oh, it's amazing, But we don't think
(49:47):
whoa convergent evolution and that there's more being said, there's
more contextual information being shared than we've figured out anyway. Yeah,
it's exciting, so everyone be excited. Your language, my language, language,
(50:12):
the language that we use to communicate wherever we are
in the world. The basis for it came before our brains,
came before us, and it probably came before a common
ancestor that we share with a chimpanzee. So exciting. I know,
(50:34):
I love it. But that's what I did. I put
these stories together into a show for everyone tonight, and
I just talked and talked and talked, and I can't
believe it's already been an hour. My goodness, how do
we do that? Everybody? Did everything? Anything happen in your
lives this week? Anything sciencey that you're interested in.
Speaker 2 (50:59):
Yeah?
Speaker 1 (50:59):
I thought they were fun stories. Thanks for being here
and being here for me sharing them with you, because
I mean, honestly, that's what I really love to do,
is try and share these stories with you. I find
things and I go, that's really cool, and then I
want to find out, you know, if I can find
(51:21):
a neat angle to them, why they're interesting to me,
why they might be interesting to you, what other questions
can we ask about them? Thanks? Paul Disney. Has been
a fun it's been a fun hour. Yeah, thank you
so much for being here, all of you. Thank you. Thanks,
Pepsi Cola. I am gonna finish the show then, oh
(51:44):
my gosh, it's so weird not to have my co
hosts here. They're off. They'll be back again next week.
But all of you, thank you for being here. And
I hope you enjoyed the show or what it was
with me talking about science for an hour. I hope
you liked it. I really do want to thank as
always the people who help make the show possible, my
(52:06):
co hosts who aren't here, all of you in the
chat room who are here listening so that I don't
feel like I'm just talking at a camera and not
actually interacting with you. I appreciate you being here so much,
and those of you who help keep the chatroom really good,
a good, healthy, respectful place to be. I appreciate you
being a part of that effort. Let's all be good
(52:26):
to each other. Fata, thank you for your help with
show notes and social media. And I did try to
look into our social media stuff and it's a meta
side of things that I'm having issues with. I got
to work that out. Gord arn Ler, thank you for
being here, Identity for Thank you for recording the show,
(52:46):
and Rachel, thank you for editing the show and as always,
thank you, Thank you. Thank you to our twists or
Patreon sponsors, Thank you too. Robert Norland, Robert W. Farley,
Lauren Gifford, Dana Lewis, Even Mandel, Alan Viola, Erin Anathema,
(53:08):
Arthur Keppler, What to see you Tonight, Aaron, Arthur Kepler,
Craig Potts, Mary Kurtz, Teresa Smith, Richard Badgebob Coles, Kent North, Coote,
George Chorus, bra Bella's Arm, John Rattaswami, Chris Bosniak, Big Art,
Chef's Stad, Donald Stiles AKA Don Stylo, Ali Coffin, Schue,
Bruce eah For far Old Boomer eight four to seven
p I G. Stephen Alberon, Darryll My, Jack Andrews Swanson
(53:28):
fretis one of four, Sky Luke, Paul Ronovich, Kevin Brden, Noodles, Jack,
Brian Carrington, davidy young Blood, Shawn Clarence Slam, John McKee,
Greg Riley, Mark Hessenflow, Steve Lison and ak A Z mckenn, Hayes,
Howard tandis rapping Richard Brendan Minish, Johnny Gridley, grim mide Gee,
Burton Lottimore, Flying Out, Christopher Dryer, Artium, Greg Briggs, John Outwood,
Ridy Garcia de Wolkinson, Rodney Lewis, Paul Phillips, Shang, Kurt
(53:51):
Laars and Craig Land and Sue Astor, Jason Olds, David Nahbor,
Eric Knapp, Lawn makes Eo, Adam Mischkhon, Kevin Perchan, Bob Calder,
Margie Paul dy Disney, Patrick beccararo, and Johnny's Day. Thank
you everyone. Your support on Patreon right now is very,
(54:11):
very appreciated. Thanks to everyone who's able to be here
for the show. Thank you for those of you who
are able to support the show through Patreon and PayPal.
I know there are others of you out there who
are supporting us on PayPal, and I do appreciate that
method of support as well. I really can't do the
show without all of you. If you are interested in
(54:32):
heading over in supporting Twists as a Patreon spot sponsor,
I can't speak tonight. Head over to Twist dot org,
click on the Patreon link and choose your level of support.
We also have our Zazzle store where you can go
and find cool Twist merchandise. Click on that Zazzle link.
Parts of the purchase do go to help. This Week
(54:54):
in Science Oh well, next week's show. Oh my my gosh,
we'll be back. I'll be back, and I do believe
my co hosts will be back, Justin and Blair. I'll
be back. We will be speaking with each other and
you about science news. Wednesday apm Pacific time, broadcasting live
(55:16):
from Twitch, YouTube, and Facebook. Look for this Week in
Science hit subscribe and notifications. If you want to watch
us live every week, but if you want to listen
to us as a podcast, make sure you look for
this Week in Science wherever podcasts are found and subscribe.
Make sure you share with a friend as well. That's
if you enjoyed the show. I mean, even didn't enjoy
(55:37):
the show, you should just subscribe your friends anyway, and
then they could be like, ah, why'd you do that? No.
For more information on anything you've heard here, today's show, notes,
links to stories, etc. Go to Twist dot org, where
you can also sign up for a newsletter. I will
send one. I promise it's going to be great when
I do. We love your feedback as well. So if
(56:00):
you have a topic you want us to cover, address
suggestion for an interview, you know, just a comment, let
us know on one of our social media accounts, or
usually twists Twist Science or this week in Science out
there wherever we are, or you can just send us
an email. And if you do, though, make sure you
put Twists into the subject line so that your email
(56:21):
does not get sucked into the L shaped beak of
a Chilean flamingo because I am not going near there
because it's smells fishy. We really look forward to discussing
science with you again next week, and we hope that
we will see you again. If you learned anything from
(56:43):
the show, remember it's all in your head.
Speaker 4 (56:51):
Just tweaking science. This week in science, This week in science,
This week in science. It's the end of the world.
So I'm setting up shop. Got my batter unfurl it
says the scientist is in. I'm going to sell my advice,
(57:11):
show them how to stop the robots with a simple device.
Speaker 5 (57:15):
I'll reverse all the warming with a wave of my hand,
and it'll cost you is a couple of grass. Because
this week science is coming your way.
Speaker 4 (57:27):
So everybody listens to what I say.
Speaker 5 (57:31):
I use the scientific method for all that it's worth,
and I'll broadcast my opinion all over the.
Speaker 1 (57:40):
Well.
Speaker 5 (57:40):
It's this week in science, This week in science, This
week in science is iscire Science.
Speaker 3 (57:48):
This week in science, This weekend Science, This week in science,
This week is science. This week in science, This week
in science, This week in science, This week in side
Speaker 1 (58:08):
Mm HM
Popular Podcasts
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Crime Junkie
Does hearing about a true crime case always leave you scouring the internet for the truth behind the story? Dive into your next mystery with Crime Junkie. Every Monday, join your host Ashley Flowers as she unravels all the details of infamous and underreported true crime cases with her best friend Brit Prawat. From cold cases to missing persons and heroes in our community who seek justice, Crime Junkie is your destination for theories and stories you won’t hear anywhere else. Whether you're a seasoned true crime enthusiast or new to the genre, you'll find yourself on the edge of your seat awaiting a new episode every Monday. If you can never get enough true crime... Congratulations, you’ve found your people. Follow to join a community of Crime Junkies! Crime Junkie is presented by audiochuck Media Company.
The Breakfast Club
The World's Most Dangerous Morning Show, The Breakfast Club, With DJ Envy, Jess Hilarious, And Charlamagne Tha God!