Este blog es solamente de entrevistas. This blog is only of interviews.
The effects of teacher quality on early reading
Published on April 22, 2010 By askain In Pure Technology

Host – Robert Frederick

Hello and welcome to the Science Magazine Podcast for April 23rd, 2010. I'm Robert Frederick. This week: the effects of teacher quality on early reading; a chronology of Asian monsoons and megadrought; and a new resource for ecology research. All this and more, plus a wrap-up of some of the latest science news—including a story about how good dogs live longer—from our online daily news site, ScienceNOW.

Support for the Science Magazine Podcast is provided by AAAS: the American Association for the Advancement of Science. Advancing Science, Engineering, and Innovation throughout the World for the Benefit of All People. AAAS—the Science Society—at www.aaas.org.

Host – Robert Frederick

According to past research, children who fail to read proficiently are more likely to be held back in school, drop out, and even commit crimes. But there’s been some doubt as to whether a good teacher makes any difference in a child’s reading proficiency, or if reading achievement is rather influenced by the child’s genetics, family life, and the larger school environment. In a paper in this week’s Science, Jeanette Taylor and colleagues confirm that genes do indeed make a difference in children’s oral reading fluency, but that teacher quality does too. In particular, the team reports that “when children receive more effective instruction, they will tend to develop at their optimal trajectory.” In addition, “poor teaching impedes the ability of children to reach their potential.” I spoke with Taylor from her office at Florida State University.

Interviewee – Jeanette Taylor

We found that genes certainly do have an effect on early reading skills, but so does teacher quality, and that teacher quality has the chance to maximize potential of children’s reading gains.

Interviewer – Robert Frederick

What test were you and your team using to assess teacher quality?

Interviewee – Jeanette Taylor

We calculated an index based on the same measure that we used for the twins, and it’s a standardized test called Oral Reading Fluency, and it’s just a timed measure of how many words children can read within a passage, so it’s got a context to it – it’s not just words on a list. And we used those scores from the classmates of the twins to assess the quality

or effectiveness of the twins’ teachers. So it wasn’t based on the twins’ scores themselves, but on their classmates’ scores, and it was essentially a change, or residual index, such that it was, over time, over the course of the school year, how much gain in reading words per minute was achieved by the class given the initial level. So, it’s over and above what would be expected based on the class’ initial level. And we felt that that was a reasonable estimate of teacher effectiveness, because the scores ranged from, you know, very low scores – 11 words per minute (wpm), 8 wpm – that was the total gain of the class over the school year, which I wouldn’t say is very effective – to, you know, 120, 130, 140 wpm – that was the gain over time in the children’s performance on this test.

Interviewer – Robert Frederick

Did you and your team have enough participants, enough twins, to statistically control for things like socioeconomic status, parental education levels, gender of the student, and teachers?

Interviewee – Jeanette Taylor

Yeah, right. We didn’t control for those things, other than, I guess, we took advantage of the fact that the sample is diverse economically, so that we had a range. And that’s – we needed to take advantage of that, because in a restricted range – if, let’s say we only had a sample from upper-middle class counties in Florida, where the income is quite high and you’d expect that to be correlated with high parental education, there may not be the kind of effects that we saw, because you might expect that teaching and schools are uniformly good because they have a lot of resources and, you know, parental involvement. Well, we had the full range, so we didn’t want to control for that, or parse all that out, we needed to actually take advantage of that. And so, one of the nice things about this study is that the measured environmental variable is teacher effectiveness, but not based on anything that we got from the twins or their parents, so it wouldn’t be as confounded with genetics as other environmental measures could be.

Interviewer – Robert Frederick

What, if any, differences, then, were there in the Oral Reading Fluency scores when the twins were in the same class with the same teacher?

Interviewee – Jeanette Taylor

Yeah, if you looked at twins who had the same teacher versus those who didn’t, there was more similarity amongst the twins. So, that was an analysis that’s not reported in the paper, but it’s one thing that we looked at, which would get at the potential for teacher to serve as a shared environmental influence. And what we saw was that there was greater similarities, suggesting this shared environment, or common environmental influence, but we noticed that the data – we have data at 3–4 times per year on these same measures for these twins, and the differences that we saw, you could see those from the beginning, at least in the first and second grade, and we didn’t know if that was attributable, in part, to the confound of tracking, where students are placed – either by their teacher, or by the school, or by the parents request to have the kids in the same classroom, have a certain teacher, have a, you know, that kind of thing – and we felt like that that was probably not the most effective way to look at our data was to compare kids with the same classroom

versus not. And, also, in part, because there was a study that just came out – I think it was last year, from a large study that goes on in the United Kingdom on reading – that showed that even identical twins in the same classroom have different experiences – that they show significant differences in their experience of the classroom, and then sometimes in their achievement and things like that. So, we preferred to go with this different approach where we looked at the interaction. So, teacher quality – does it have a direct influence as an environmental variable on the sources of variance in reading ability? So these genetic, shared environmental, non-shared environmental sources of variability – are those influenced by this measured environmental variable of teacher quality? So that’s what we ended up thinking was probably better suited to our data.

Interviewer – Robert Frederick

To what degree, then, are teachers having an impact on student reading achievement?

Interviewee – Jeanette Taylor

Yeah. You know, data from non-twins that are out there in the education literature – and one of the nice things about this paper is that it’s a true collaboration among folks that come from different perspectives, so we’ve got me, the behavioral geneticist, and we’ve got education researchers in the group, and folks focused on reading research, in particular – and we know that the literature on non-twins shows a, not, I mean, admittedly, it doesn’t appear to be a huge effect in terms of controlling for other things, but that there’s an effect of teacher quality on reading gains and on reading achievement, and what we tried to do was say, “Does that influence the genetic piece that we know other studies have also shown, and so, can we see that effect?” And certainly the effect was there. So, the moderation is there – that’s what our data seemed to show, that there’s this interaction such that at high, high levels of teacher quality, yet at the extreme tail of the distribution, you’re going to see variability in children’s reading being accounted for much more by genetics – so, it’s going to highlight those genetic differences and allow kids to probably reach those potentials, whatever they are. But at the lowest end of teacher quality, you’re going to see those genetic differences are much smaller in terms of how much they’re associated with the differences between kids and their ability to read. So, most people are going to be in that middle range, right? They’re going to be in that middle range, and you’re going to get this genetic influence, and our data are just helpful, I think, in helping us modify what we know about genetic influence on reading, in that it is influenced by this teacher quality variable. Our MZ analyses—or identical twins analyses—where we were able to look at a number of pairs who didn’t match up in terms of their reading ability – one of the members of the identical twin pair was above the group mean for the whole sample and the other twin, the co-twin who’s genetically identical, was below the mean. So they were discordant in their reading ability performance. And when we looked at teacher quality, there was a significant difference there, too, of, you know, close to 10 points difference in teacher quality on average, so that the member of the pair performing better had the better quality teacher.

Interviewer – Robert Frederick

Was that effect greater for when each twin was in his or her own class compared to when each twin was in the same class with the same teacher?

Interviewee – Jeanette Taylor

For the identical twin analyses where they were what we called discordant, or they didn’t have similar scores – one was above the mean of the group and the other one was below, so a higher performer and a lower performer – we only compared those kids who had different teachers, because if they had the same teacher, they would have had the exact same teacher score, or quality score, and so there wouldn’t have been a difference for that. So we only looked at the kids with different teachers who were discordant. It is possible, it was possible in the database that there could have been identical twins who also were discordant within a classroom – so they had the same teacher, but, nonetheless, weren’t performing at the same level – I don’t think we looked for those, but that’s possible, and again, that goes back to that issue of – it doesn’t mean that identical twins will even experience the same teacher the exact same way. But three are – it is possible that in our big sample that there were a few pairs that shared a teacher but were nonetheless discrepant. And that would be an interesting, separate question of, “Why are they different?” But, again, we have only the data from their achievement scores.

Interviewer – Robert Frederick

Any insights, then, from this study on what the higher quality teachers, as measured by this study, are doing in the classroom, so that other teachers might employ the same teaching strategies?

Interviewee – Jeanette Taylor

Yeah, that’s a good question, and, you know, I’d hate to not give an answer, but our study – we really only looked at the gains of the kids in class. So, you know, one of the things that we’ve talked about, or my co-authors and I, in terms of finding this greater genetic variability at this high end of teacher quality measure, is that, you know, it could point to the idea that there are some things that good teachers do alike. So, I think it highlights that we definitely should be looking for those kind of factors, and, perhaps, there’d be a point where we could get a little more nuanced in our investigation and learn a little bit more about the kinds of methods that are used by the teachers that, at least their kids are making big gains, according to our index. I hate to not give an answer, but we are certainly limited in terms of what we have, since we really only had scores from the kids in the class – twins and non-twins.

Interviewer – Robert Frederick

Jeanette Taylor, thank you very much.

Interviewee – Jeanette Taylor

Yes, thank you so much.

Host – Robert Frederick

Jeanette Taylor of Florida State University is lead author of a paper on how teacher quality moderates the genetic effects on early reading. Read the paper and a special section on literacy in this week’s Science.

Host – Robert Frederick

Also in this week’s Science: an unscientific poll about unconventional journals. We’d like your thoughts about whether the benefits of a non-peer-reviewed journal, such as Medical Hypotheses, outweigh the risks. Here’s what the poll is about: according to the Medical Hypotheses journal’s website, the journal “will consider speculative and non- mainstream scientific ideas provided they are coherently expressed.” But the journal has drawn fire for publishing papers that some say are detrimental to healthcare efforts. For example, one paper published by the journal denied the link between HIV and AIDS. In this week’s Science, find Letters that argue for and against the unconventional journal’s position on publishing speculative ideas, and our poll to ask you what you think. Find our unscientific poll about unconventional journals on our website, www.sciencemag.org, and click on "Take our Readers' Poll."

 

Host – Robert Frederick

Drought is a normal part of the climate for almost every part of the world, but when droughts go on year after year, they can greatly upset human settlements and even put an end to entire civilizations. For example, historical records show that past megadroughts in Asia coincide with the end of China’s Ming Dynasty. And in a paper in this week’s Science, Ed Cook and colleagues report that the end of the Khmer civilization in what is now Cambodia also coincided with a megadrought. The research team reconstructed the chronology of hundreds of years of Asian monsoon droughts based on a network of tree ring data, providing a long-term context for monsoon variability. I spoke with Cook from his office at the Lamont-Doherty Earth Observatory of Columbia University in New York.

Interviewee – Ed Cook

What we’ve done is provide the first spatial reconstruction of the Asian Monsoon Complex, the rather complicated varieties of the Asian monsoon that strike India, China, and Southeast Asia, and down even into Australasia, and we’ve done that from a network of long tree ring chronologies.

Interviewer – Robert Frederick

How does your team’s model, based on tree ring chronologies, compare to models based on ice cores, or corals, or other proxy data?

Interviewee – Ed Cook

What makes our product quite different is that we can provide an actual spatial representation of Asian monsoon variability. Records like ice cores and speleothems, for example, are extremely useful.

Interviewer – Robert Frederick

These are cave deposits, right?

Interviewee – Ed Cook

Yes, that’s correct. Yes, speleothems are cave deposits, and those records are extremely useful, but they tend to be at fixed locations with not enough locations being available, typically, to provide a spatial representation of monsoon variability. The tree ring network that we’ve developed, 327 chronologies in total, basically cover most of the region affected by the Asian monsoon system, and consequently, our tree ring network can be used to reconstruct patterns of drought variability over the region back in time.

Interviewer – Robert Frederick

How did you and your team calibrate or test your team’s model?

Interviewee – Ed Cook

What we do is, we first assemble the network of tree ring chronologies, of course, and then we also obtain through various sources the gridded instrumental data that we use for calibrating the tree rings in terms of a particular climate variable – in this case summer monsoon drought variability. The instrumental climate data we used was the Palmer Drought Severity Index for the June, July, August summer monsoon season, and it’s a 534 grid point array of data points spaced 2.5 degrees by 2.5 degrees across the Asian monsoon region. And we can calibrate this grid of drought indices or use this grid to calibrate the tree rings, such that at each grid point we have a tree ring reconstruction of drought, and this provides us with the spatial details back in time.

Interviewer – Robert Frederick

And how far back does that Palmer Index go?

Interviewee – Ed Cook

Well, the instrumental data goes back, with quality, back to about 1951, and then it degrades in quality back in time further, but it does go back as far as 1870, the particular data set for instrumental records. The tree rings, however, can take us back hundreds of years into the past, and, in fact, what we present in our Science paper is a set of drought reconstructions that go back to 1300 – approximately 700 years in the past.

Interviewer – Robert Frederick

What, if any, insights have you and your team gleaned from your chronology of Asian monsoons in looking at the time period of known historical droughts?

Interviewee – Ed Cook

Well, what we’ve done is to try to do some preliminary evaluations of monsoon failures and megadroughts in our reconstructions, and how they might compare to what is in the 20th century record of recorded drought variability from rain gauges and such. And we find that when one goes back in time, back as far as 700 years, there are some clear periods of very unusual drought—monsoon failure—that is certainly as significant, or more significant, in strength compared to anything that’s indicated in the instrumental records for the 20th century.

Interviewer – Robert Frederick

Does your chronology show any patterns or conditions that might have led to the mega- droughts, or monsoon failure?

Interviewee – Ed Cook

Yeah, the best we can tell right now is that there’s a fairly tight coupling, as it were, between what happens in the tropical Pacific Ocean – what the tropical Pacific sea surface temperatures are doing – and what happens to the monsoon in various regions of Asia. It’s basically related to the El Niño-Southern Oscillation system that is well-known for controlling hydroclimatic variability over large parts of the globe. The summer monsoon in Asia, particularly in India and Southeast Asia, is quite sensitive to sea surface temperatures in the tropical Pacific, and that seems to be driving a lot of the variability we think we see in terms of changes in wetness and dryness or strength of the monsoon.

Interviewer – Robert Frederick

Beyond insights into historical droughts, what, if any, new insights does your team’s tree ring based chronology provide as to the conditions or patterns that led to other weather conditions that aren’t in the historical record?

Interviewee – Ed Cook

Well, we’re still just evaluating the reconstructions themselves – we haven’t gotten too deeply into it – but the real significance of this product, the Monsoon Asia Drought Atlas, relates to how it can be used as a tool for modeling studies to better understand the actual mechanisms or processes that lead to the monsoons varying the way they do, or have in the past, and also now, and, hopefully, even in the future. The drought atlas, the Monsoon Asia Drought Atlas, provides the spatial patterns of variability that modelers need to really understand the mechanisms and processes that are responsible for the reconstructed variability. And I think that’s going to be the big payoff, is just coming up with improved models that can be used to understand and even, perhaps, to predict monsoon variability in the future.

Interviewer – Robert Frederick

What would you say to a skeptic who might say that our mucking with the climate means analyzing past data won’t help in predicting the future?

Interviewee – Ed Cook

Well, there’s an element of truth to that in the sense that, you know, we could, for example, be entering into a kind of a no-analog state of climate variability due to unusual greenhouse gas forcing while, on the other hand, the underlying processes of climate variability are still the same. And so, to better understand and to model variability, even in the future, requires an understanding of the processes and the dynamics that go into monsoon variability, and that can be understood from the past, even though conditions today might be different. And models can take into account that changing state, in terms of forcing, as the models do, you know, routinely in understanding what’s happening today and perhaps what will happen in the future. So, I don’t think it’s a limitation, per se, that the conditions in the past are not quite the same as they are today with respect to

the forcings involved.

Interviewer – Robert Frederick

Ed Cook, thank you very much.

Interviewee – Ed Cook

You bet.

Host – Robert Frederick

Ed Cook of the Lamont-Doherty Earth Observatory of Columbia University is lead author of a paper on Asian monsoon failure and megadrought during the last millennium.

Music

Host – Robert Frederick

For decades, the astronomy and geosciences communities have worked with funding agencies to build costly telescopes and enormous ships, all without devouring their annual budgets. Now, the ecology community in the United States is preparing to do the same thing by constructing a network of observational sites to track long-term ecological change on a continental scale. But as Science’s Elizabeth Pennisi reports in this week’s issue, success of the over $400 million project will depend on whether ecologists can change how they do science.

Interviewee – Elizabeth Pennisi

So, this story is about the National Ecological Observatory Network, which is a new program established by the National Science Foundation to monitor the environment across the whole United States. It’s a program that’s been in the making for a decade now, has undergone various permutations in terms of what it should look like, and what it should do, and how it should operate, but it’s getting to the point now where they’re almost ready to actually start building. And it will be a network of 20 main sites and about 40 secondary sites that will have the same kind of environmental monitoring with the same kind of instruments on the same schedule to gather data that you can compare across all the sites.

Interviewer – Robert Frederick

How are the sites distributed?

Interviewee – Elizabeth Pennisi

The sites are distributed across the United States. They went through a process where they evaluated all sorts of climate and vegetation and other kinds of data—information about what the environments were in different parts of the country. In fact, they had a kilometer-by-kilometer analysis where they would sort of build a matrix of all these variables and figure out what areas are most like each other and sort of cluster them into what they call “domains.” And they came up with 20 domains across the country that represent different ecosystems, more or less, across the whole United States, Hawaii, Puerto Rico, and Alaska.

Interviewer – Robert Frederick

So, how is this National Ecological Observatory Network different from previous ecological monitoring projects?

Interviewee – Elizabeth Pennisi

This project, whose acronym is NEON, is meant to be a 30-year project, because they want to have long-term data, so you can actually follow the environment long enough to see if there are any trends or any changes. It’s not the first time that NSF has gone into long-term ecological research. The National Science Foundation has had another long- term ecological research project called the Long-Term Ecological Research Network that has been going on for 30 years, but which was done on a piece-by-piece basis. So, in other words, a researcher in Colorado would make a proposal to NSF saying, “I want to establish a long-term research project in this mountain range,” and NSF would approve it, and he or she would do the experiments that they wanted to do, and do the kind of monitoring that they wanted to do. And maybe someone in Virginia would do another long-term ecological research place, but would do a different set of experiments and do a different set of measurements in a different way. And so, there’s no way to compare what you’d find in Colorado with what you find in Virginia. And NEON is meant to get around that, in other words, because everybody is doing the same thing, the same measurements, so you can compare.

Interviewer – Robert Frederick

What kind of monitoring will all these stations be doing?

Interviewee – Elizabeth Pennisi

Well, there’s a variety of things. They’ll be monitoring of, sort of weather and climate parameters; they’ll be monitoring carbon dioxide flux. Then there’s also a biological component where they’ll be catching insects and trapping mice and digging up microbes and sequencing the microbes to see what’s there. There’ll be vegetation surveys. They’re also planning to do airplane flyovers where they’ll take pictures with cameras, and they have these sensitive instruments that look at the chemistry of the vegetation, so you can tell basically how green the vegetation is, how much productivity there is. So there’s a pretty extensive list of stuff that they’ll be studying.

Interviewer – Robert Frederick

So, in setting up this project, did researchers have a model for creating an observatory network on a national scale, or is this something new, internationally even?

Interviewee – Elizabeth Pennisi

This is something new. I don’t know if anybody’s done anything internationally, but for the United States, this is something brand new.

Interviewer – Robert Frederick

With something so new, what, if anything, are project organizers saying it will take for NEON, this National Ecological Observatory Network, to succeed?

Interviewee – Elizabeth Pennisi

Well, NEON is different for ecologists, because ecologists are used to working on their own, gathering their own data, sort of doing their own thing. And NEON is different in that NEON and NEON personnel will be actually gathering the data and providing a lot of different data for any researcher to use. So, the ecologist doesn’t have the same control over the data that he or she might normally have. So it requires a sort of a change in mindset from ecologists: number one, to sort of accept data that they didn’t gather; number two, to think in terms of national scale or regional scale instead of thinking in terms of the particular study site where you were working. So, it’s a little bit of a risk, because nobody really knows exactly how to use these data and nobody knows whether ecologists will be lining up to take advantage of it.

Interviewer – Robert Frederick

What’s been the reaction of the ecologists you spoke with to this large-scale environmental science project?

Interviewee – Elizabeth Pennisi

The reaction was mixed. I mean, a lot of people are very enthusiastic about it – they think this is what’s needed to sort of make the next step in ecology and to sort of make the next leap in understanding global climate change. Some people were a little bit wary of the amount of money that it’s going to cost. It’s an over $400 million project, which is huge from the ecological research point-of-view. And some people think that observing is not enough—they would rather see global climate change experiments going on, and the number of global climate change experiments that will go on at these different sites is limited.

Interviewer – Robert Frederick

So, a real change, then, for people who are even considering going into ecology as to what to expect?

Interviewee – Elizabeth Pennisi

Right. This is new-age ecology in the way that genomics is new-age genetics, and the idea is that this is something for young scientists to think about devoting their careers to, thinking in terms of how you take advantage of all these data, how you think about things on a national scale. It will be an exciting opportunity for young people.

Interviewer – Robert Frederick

Elizabeth Pennisi, thank you very much.

Interviewee – Elizabeth Pennisi

Well, thank you very much.

Host – Robert Frederick

Science’s Elizabeth Pennisi reports on the National Ecological Observatory Network and the new standard for tracking long-term ecological change across the United States.

Music

Host – Robert Frederick

As reported in The New York Times last week, the number of women dying in childbirth has dropped dramatically from about 526,000 in 1980 to about 343,000 in 2008. The source of the figures comes from a paper put out by The Lancet. Here with more about it is Science Policy Forum editor Brad Wible.

Policy Forum Editor – Brad Wible

The new figures are surprising because the approximate number of maternal deaths has appeared to hold steady for decades, leading to a widespread perception of little progress in the field. The World Health Organization’s latest figures still put maternal deaths at about half a million per year.

The new study cited several reasons for the positive findings, including lower pregnancy rates in some countries, better nutrition and access to health care, more education for women, and an increase in the availability of trained professionals to assist with births.

As for the drastic difference between the new stats and previous estimates, Lancet editor Richard Horton says his journal’s study was based on more and better data as well as more sophisticated statistical methods. He also says he was pressured to delay his journal’s findings while public health officials readied to ask for maternal health aid at upcoming United Nations meetings. “This is one of those instances,” he told The New York Times, “when science and advocacy can conflict.”

Host – Robert Frederick

That was Policy Forum editor Brad Wible with a policy update from Science and the AAAS Center for Science, Technology, and Congress.

Music

Host – Robert Frederick

Finally today, David Grimm, editor of Science’s online daily news site, ScienceNOW, is here with a wrap-up of some of the latest science news, including a story about how good dogs live longer. Their disposition has an effect, Dave?

Interviewee – David Grimm

It turns out it does, Rob. This story is all about what factors make certain species live longer than others – you know, why does an elephant live for years and mice only live for months. And it’s a hard thing to test, because a lot of this may have happened via natural selection, you know, there may have been certain traits that were selected for that maybe conferred a certain advantage to an animal – maybe making it more aggressive, but being more aggressive made it a better hunter and maybe it was able to procure food better, but also maybe that same trait also shortened its life span, because it was so aggressive maybe that its heart beat really fast, and that really increased its metabolism and caused it

to run out its life span a lot faster than maybe another animal would.

Interviewer – Robert Frederick

So, natural selection versus people selection?

Interviewee – David Grimm

Well, right. This is a hard theory to test because we can’t just take a bunch of animals and naturally select them ourselves over millions of years and figure out what happens. But dogs have been a nice natural experiment because we have been, as you say, self- selecting these animals for over 10,000 years for specific traits, like, you know, some dogs have been bred to be good hunting dogs, and some dogs have just been bred for looking nice on a couch. And so, what a researcher did in this study was to look at various breeds of dogs and see if there was any correlation between personality and life span. And that’s actually what he found – he found that obedient dogs, like German Shepherds and Bichon Frises, they actually live very long for their size, whereas hard-to- train dogs, like Beagles and Pomeranians, generally died earlier than similarly sized breeds. And what’s more, the researcher also found that peaceful dogs, like Newfoundlands, which are these really large dogs, and Labradors, tend to burn less energy per kilogram than really aggressive dogs do, like Fox Terriers and Great Danes. And so, what the researcher thinks is, you know, this is all a correlation, but what he saw is that these dogs with the sort of more peaceful, more obedient personalities, tended to be the types of dogs that lived longer for their size range than dogs on the opposite end of the spectrum.

Interviewer – Robert Frederick

Are researchers willing to extend this hypothesis, or extend this correlation, to other species, like us?

Interviewee – David Grimm

Well, it’s a good question, Rob. There’s nothing in this study to indicate that this has any application to humans, whether more peaceful, or, I don’t know how you would define an obedient human, but definitely more easy-going humans might live longer, although we do know that stress can be an important factor for shortening life span, so it’s possible that there are some implications for humans, as well.

Interviewer – Robert Frederick

Okay. So, what other stories have you brought with you this week?

Interviewee – David Grimm

Well, Rob, from personality in dogs to intelligence in birds, and specifically crows, this next story is all about a type of crow known as the New Caledonian crow, which is actually known as a “feathered ape,” because it has been demonstrated to be very smart in a lot of previous experiments.

Interviewer – Robert Frederick

What kind of smartness are we talking about here?

Interviewee – David Grimm

Well, believe it or not, there have been some studies that have shown that these crows can actually understand some rudimentary physics, and that they can actually recognize themselves and other humans. We actually a while ago in a podcast...

Interviewer – Robert Frederick

Oh, right, I remember, they were attacking people who were wearing Dick Chaney masks.

Interviewee – David Grimm

Exactly. And so, we know these crows are smart. The question is “How smart are they?” And one classic experiment that’s been done with these crows is to show that they can use tools. Researchers give them a stick and they use the stick to maybe probe down a hole and get a worm out of it. And actually, that is not so unique to crows, in fact, other animals, even ants, can use tools. But what was interesting with crows is that researchers showed that if they gave crows a really small stick, say, a stick that couldn’t get all the way down a hole to get a worm, the crows could use that small stick to get a larger stick, and then they would use that larger stick to get the worm. And researchers had claimed from this study, which was conducted a couple years ago, that not only does this show the crows are smart, because they know how to use tools, but that they actually are capable of this higher cognition – of this ability to sort of plan ahead. They’re not just saying, “Hey, I need a tool to get that worm,” they’re saying, “Hey, I need this tool to get that tool, and that tool’s going to help me get a worm” – this really complex flow of logic. Researchers say, you know, we’ve really only seen that in primates.

Interviewer – Robert Frederick

But other scientists call that into doubt?

Interviewee – David Grimm

Exactly. A lot of other scientists had said, some skeptics of the study had said, “You guys are really sort of over-interpreting this result.” And what their criticism was is that when the crows were given this smaller stick, they weren’t really thinking ahead that, “Hey, this smaller stick’s going to help me get the larger stick, which is going to help me get the worm” – they just associated this smaller stick with the eventual food reward. So they saw the small stick as something good that helped them get the larger stick, and that helped them get the worm.

Interviewer – Robert Frederick

So they saw the small stick as a good thing, the larger stick as a good thing – whatever the case, just go get the good thing.

Interviewee – David Grimm

Exactly. The skeptics said this is all just an association – you know, one good thing gets me another good thing – and, you know, there is some intelligence there, it doesn’t really represent a complex sort of planning or being able to follow this flow of logic that

researchers have associated with humans and other primates.

Interviewer – Robert Frederick

So, what about this new study?

Interviewee – David Grimm

Well, this new study is all about an attempt to get around that problem. What researchers did was they said, “Let’s get crows to really hate this small stick. You know if these skeptics are saying that these crows are just using the small stick because they like it or they associate it with food, let’s get them to really dislike it – to view it as what they call an “unattractive object” – and see if they can still use it in the same way.” So, what they did was they took a few crows and they gave them the small stick, and they also presented the crows with something to use the stick on. They would, you know, put a treat in a hole or behind a cage, and the crows could never get this treat because the stick was too small. So they would try to use the stick and they would give up after a few attempts, and, in fact, they would actually have behaviors that indicated they really disliked this stick – after a few attempts, they would sort of throw the stick away or toss it down. They really came to dislike this stick, exactly as researchers hoped they would. And then, what researchers did was, they repeated this classic experiment where they gave these crows, again, the small stick, but this time the crows could use the small stick to get a larger stick to get the reward. But they made things even more unattractive for the small stick. So, already they knew the crows didn’t like this small stick any more, but now they tied this small stick to a rope on a branch, so not only did the crows have to get the stick that they didn’t really like, they really had to go through a lot of machinations to get this stick. And what the researchers found was that, even though the crows didn’t like the small stick and the small stick was really hard to get, when the crows realized that they could use the small stick to get a larger stick to get their reward, they used it. And there’s actually a really cool video in this story which I highly encourage everybody to take a look at. This is a video of a female crow, and what’s really special about her is – a lot of these other crows I just talked about, they had to make a few attempts with the small stick and the larger stick before they really got the hang of knowing, “Okay, the small stick is going to get me the larger stick, which is going to get me the reward” – but you’ll see in this video, there’s a female crow that basically looked at the whole setup. She saw the small stick hanging from the tree; she saw the large stick on a table; and she saw the food reward hidden behind something that she could only get with the large stick. And without any trial and error, after sort of staring at the setup for a couple of minutes, she goes ahead, she goes straight to that small stick, grabs it, goes to the big stick, grabs that, and uses the big stick to get the reward. So she worked out this whole thing in her mind, didn’t have to fail at all and got everything on the first try. You can see all this happening in the video.

Interviewer – Robert Frederick

So, does this convince researchers that that’s what’s going on inside the crow’s brain?

Interviewee – David Grimm

Well, you know, it’s hard to say whether this is going to convince all the skeptics. At

least one outside expert thought this was a very powerful experiment. The researchers do say, though, that it does really address this criticism of the previous study that, you know, the crows were just using these small sticks because they were making these positive associations with them, and it really does get us further along the lines of being able to claim that maybe animals that we don’t think are as intelligent as humans or other primates, like birds, they actually might be capable of this very advanced cognition.

Interviewer – Robert Frederick

Okay. So, last story. What’s this last one about?

Interviewee – David Grimm

Well, Rob, I feel like we’re working our way down the food chain here a little bit, but this last story is about fish, and specifically mercury levels in tuna. And this study is claiming that the tuna that you get in sushi at restaurants actually has a lot more mercury in it than the tuna-filled sushi that you would get at a supermarket.

Interviewer – Robert Frederick

Why the difference?

Interviewee – David Grimm

Well, the difference has to do with the type of tuna that’s actually used in these different types of sushi, and nobody had really done a comprehensive study of it before. You know, researchers knew that, you know, there’s a few tuna species that tend to end up in sushi. And why tuna itself tends to accumulate so much mercury in the first place is that the species at the top of the food chain in the ocean, the higher up you are on the food chain, the more animals you’ve eaten who have themselves eaten more animals, and if you’ve got mercury starting at the lowest levels, this really builds up a lot – especially if you’re talking about something like a bluefin tuna, which can weigh more than 500 kilograms, you’re talking about potentially a lot of mercury accumulating in these fish.

Interviewer – Robert Frederick

So, the restaurants serve tuna from these larger fish than the ones in the supermarket?

Interviewee – David Grimm

Exactly. And they tend to serve sushi that has tuna from species that accumulate more of this mercury. And, in fact, what the researchers did here, which really hadn’t been done before, is they employed something called “DNA bar coding”. And they essentially took samples of tuna – they took a 100 sushi samples from 54 restaurants and 15 supermarkets – and what DNA bar coding does is it basically, it allows you to look for specific genetic markers that would distinguish one species from the next – and what they found was they found that, you know, when you combine restaurant tuna and supermarket tuna, you’re talking about five species. And what the team found was that restaurant sushi tends to contain bigeye tuna or lean bluefin tuna, and these types of tuna had concentrations of mercury that were actually 4% above what the FDA recommended a limit is for mercury. And the reason we care about mercury is it’s been shown to cause severe neurological problems in humans, especially it can be a real problem for pregnant women and young

children. And the researchers found that the type of tuna that’s usually found in supermarkets is yellowfin tuna, which tends to be cheaper, and it’s a more plentiful species, but it contains less mercury than restaurant tuna, but it still exceeds the FDA limits for mercury. So, even with the supermarket tuna, you’re getting more mercury than you should be getting; you’re just not getting as much as you’re getting, typically, when you eat out at a restaurant.

Interviewer – Robert Frederick

So, what do researchers hope will come of this study?

Interviewee – David Grimm

Well, from the consumer angle, the researchers are saying, you know, if you eat a lot of sushi, make sure it’s not a lot of tuna sushi, you know, eat this tuna sushi in moderation because of the elevated mercury levels. And for the purveyors of the sushi, he’s saying they really should be more proactive about putting labels on the sushi they’re selling, so consumers know what sort of tuna’s being used and they can cross-reference that with what types of tuna tend to have more mercury in them.

Interviewer – Robert Frederick

Okay, well thanks, Dave.

Interviewee – David Grimm

Thanks, Rob.

Interviewer – Robert Frederick

So what other stories are you looking into for ScienceNOW, or on the policy blog, ScienceInsider?

Interviewee – David Grimm

Well, Rob, for ScienceNOW, we’ve got a story about how taking a nap may help you learn better; and continuing the neuroscience angle, a story about these brain-training games – maybe you’ve seen them on mobile devices – do they actually work, do they actually help improve cognition? And for ScienceInsider, Science’s policy blog, we’ve got a story about the Iceland volcano and how it’s impacting science. Specifically, it’s having a very large impact—negative impact—on scientific meetings, and what exactly that impact is. And also a story about the invasive lion fish, which is decimating fish in the Caribbean. The National Science Foundation has just announced a big grant to study and potentially help stop these fish. So be sure to check out all of these stories on the site.

Host – Robert Frederick

David Grimm is the editor of ScienceNOW, the online daily news site of Science. You can check out the latest science news plus all the stories from the science policy blog, ScienceInsider, at news.sciencemag.org.

Music

Promo

Support for the Science Magazine Podcast is provided by AAAS—the Science Society— at www.aaas.org.

Host – Robert Frederick

And that wraps up the April 23rd, 2010, Science Magazine Podcast. If you have any comments or suggestions for the show, please write us at sciencepodcast@aaas.org. The show is a production of Science Magazine. Jeffrey Cook composed the music, and I'm Robert Frederick. On behalf of Science Magazine and its publisher, AAAS—the Science Society—thanks for joining us.


Comments
No one has commented on this article. Be the first!
Meta
Views
» 156
Comments
» 0
Category
Sponsored Links