On another subject, I tried something different in my most recent post on Accumulating Glitches and I’d love to hear what you think — what works for you, what doesn’t, how I could do better.  The post is about ants which practice agriculture and what they might think of the way we farm…

Hauskaa Juhannusta to those of you in Finland!  If anyone will be at the WCSJ 2013 meeting in Helsinki next week, let me know and we can try to meet up!

Double-blind: what it is and why it matters

Double blind studies are important because humans can be biased, often without knowing it. Scientists aren’t immune to this kind of thing.  A wonderful paper published last year showed an unconscious bias (by men and women) against women when evaluating candidates for a lab manager position. This isn’t a question of honesty or integrity; it’s about things like confirmation bias and implicit associations.

Unconscious biases can shape our behaviour in lots of ways, and sometimes that can interfere with the outcome of an experiment.  Clever Hans, the horse who could count, is a famous example of how we can unconsciously shape the outcome of an experiment.  Hans amazed crowds in early 20th century Berlin with his ability to count, spell, remember names and other amazing feats.  Hans couldn’t actually do any of these things, of course, but it took some effort to demonstrate that.  There wasn’t any trickery involved.  Hans’ keeper hadn’t trained him to respond to any special cues; anybody could ask a question and Hans would tap out an answer with his hoof.  A professor eventually showed that Hans could only answer a question if the person asking it knew the answer.  It turned out that Hans was picking up on the unconscious responses of the person answering the question — say, a slight shift in gaze when he’d tapped out the right number — and using those as cues.  If the person didn’t know the answer, he couldn’t accidentally give it away and Hans would have no cues to bias his behaviour.

A “double blind” study is one that takes just such an approach to eliminate (or minimize) the risk of unconscious biases affecting the results.  If you want to find out whether someone can tell the difference between Coke and Coke Zero, it’s obvious that they shouldn’t know which glass is which when they taste them.  A double blind study goes further by ensuring that the person giving them the glasses also doesn’t know what’s in them.  That way, there’s no risk that they’ll accidentally give away which glass has Coke Zero because of an unconscious difference in their behaviour — say, how they handle the glasses or how intently they watch the person while they’re drinking.  To make this work, you need a third person who randomly labels the glasses A and B and is the only person who knows which is which.  The actual experiment and data gathering should be done completely independently of this knowledge and only combined afterwards.  In some cases, people even chose to analyse the data without knowing which is which and only change the labels back after everything is done.

Sample size and statistical significance

It’s always important to make sure that a study is based on enough data to support the results.  What does that actually mean?  Basically, it means that there should be enough observations to convince us that the outcome isn’t just due to dumb luck.  If I flip a coin once and it comes up heads, that’s hardly going to convince you that the coin is weighted and always lands heads-up.  You probably still wouldn’t be convinced if I pulled it off three times in a row or maybe even five times.  But ten consecutive flips, all heads? Twenty?  That can’t just be dumb luck, right?  Something must be going on.

That’s really all there is to it.  Say you want to test the effect of a new fertilizer on plant growth.  Individual plants are always going to grow differently, so you couldn’t just use two plants, one with fertilizer and one without.  Any differences you find might have nothing to do with the fertilizer — they might be because of they way the plants were positioned or inherent differences between them or anything else.  You need to use more plants — a larger “sample size” — to know whether any effects are due to the fertilizer.  How many plants you need depends on how strong the effect of the fertilizer is — the stronger the effect, the fewer plants it will take to be convinced.

Scientific research should always be done with a large enough sample to make it unlikely that the results are due to chance alone and proper statistics should be used to confirm the findings.  In science, such results are called “significant”, which is different from the word’s normal meaning (something like “a great deal”); significant results are just results that are unlikely to occur randomly.  Look out for phrases like “statistically significant” or “significantly more/less/greater/fewer/differences” when reading about research.  If you don’t see them, be cautious — especially in the face of extraordinary claims.  An example of recent research that epically failed to do this is the dreadful paper from last year falsely claiming there was a link between GMO food and tumour incidence.  (Incidentally, the observations in that paper also weren’t double-blind, which also raised a warning flag.)

Correlation or causation?

It’s been said a thousand times, but correlation isn’t the same thing as causation.  The fact that two things happened to the same group of people or at the same time (or one after another) doesn’t mean that one causes the other.  It’s not just that it might be a coincidence — there are statistical tests to check that.  Even if two things always happen together and only happen together, there’s no guarantee that one caused the other.  There might be another factor underlying them that causes both to happen.  Nearly everyone who regularly drank water has died and nearly everyone who is dead regularly drank water.  Of course, it would be absurd to suggest that drinking water causes death.  Rather, the fact that we’re alive means that we have to drink water regularly and that we will eventually die.  This example is intentionally absurd, but the same logic applies when a study shows that eating X or doing Y is linked with higher rates of cancer or heart disease.  The two observations might co-occur, but that doesn’t necessarily mean that one causes the other.

It can be hard to show that one thing causes another, particularly since causality can be quite a thorny philosophical issue.  The ideal way would be to rewind the Universe and redo the observation with just one thing changed to see if the outcome is different.  Of course, that’s impossible to actually do, but scientists try to get as close as possible.  The idea is to repeat an observation many times and try to hold as many things constant as possible, so the only factor that changes is the one you think is causing an effect.  Since it’s impossible to make sure everything in the Universe is exactly the same, it’s always possible that some hidden cause will get missed, but these sorts of experiments are still more convincing than a simple correlation would be.

Keep an eye out while you’re reading and see whether there are convincing experiments showing that X caused Y or if there’s just a correlation between the two.  Correlations are important — they can be useful markers of risk, for example, or even be the first step in getting researchers to do experiments and show a causal link.  By itself, though, correlation simply isn’t good evidence of causation, so don’t let anyone get away with pretending it is.

Isn’t that what science writers should be doing?

Yes, science writers have a responsibility to evaluate research and present it to the public with an explanation of the various short-comings and so forth.  It’s far better, though, to have a critical and thoughtful public that can evaluate the information they get — whether from science writers, pundits, media figures, or politicians.  Ultimately, you are the only gatekeeper of your mind and the person responsible for what you do or don’t believe.  Science writers should do their job, but all of us should learn to read and listen actively and critically, especially in the face of extraordinary claims.

The main thing is to read and listen critically and to always ask questions and challenge ideas.  There are some subtleties I’ve glossed over and lots of other things to look out for, but I’ll stop there.  What about you? How do you decide whether or not to believe what you’re reading?

Figures like MC Escher’s Waterfall and the Penrose triangle are frequently described as impossible. The images exist, so clearly the pictures themselves aren’t “impossible”. The impossibility of these structures comes from our minds attempt to interpret them as two-dimensional projections of three-dimensional structures; that is, we understand the images but think that the objects they depict couldn’t exist. The three-dimensional structures appear to bend back on themselves in ways that violate our intuition of space.

To explain this discrepancy, we need to understand how our visual system constructs a three-dimensional world from what we see. We form images of the world when light strikes the retina, a sheet of light-sensitive cells at the back of our eyes; this creates a flat picture from which we have to reconstruct the world around us. In principle, the image we see could correspond be generated by an infinite number of different worlds. For example, a straight line somewhere in our sight might represent a straight line out in the world, but it might also be a flat surface viewed edge-on — say, a piece of paper seen from the side. In fact, a the paper wouldn’t even have to be rectangular; a circle (or paper of any other shape) viewed edge-on would still look like a straight line. Our visual system uses a set of simple and powerful rules to make sense of the images it gets and to construct a single, coherent world view from them. One such rule is the rule of “generic views”, which simply means that we prefer to interpret what we see as though there were nothing special about our vantage point. If a straight line in your sight was actually a sheet of paper, then a slight shift of view would reveal the paper; in other words, you would have to be quite lucky to have such a perspective of a piece of paper. We tend to process images as though we’re not in such a privileged position, which is why we see straight lines simply as lines and why the Necker cube in the middle looks like a 3-d cube while the flanking images look more like flat surfaces:

The images on the left and right could be cubes, but you would have to be looking at them from a particular angle in order for the parts to overlap so conveniently. Instead, we prefer to see them as flat shapes. The rule of geometric views, together with rules about things like occlusion, the apparent distance between objects, or the interpretation of 2-d curves in 3-d, enables our visual system to construct a three-dimensional interpretation of the flat images cast upon our retina. When applied to images like Escher’s Waterfall or the Penrose Triangle, these rules generate objects that violate our expectations of the physical world. In fact, it’s perfectly possible to build a real, physical object that looks like the Penrose Triangle when viewed from the correct angle, but the rule of generic views prevents us from constructing the correct shape when we see it.

Figure-ground illusions, such as the Rubin vase or the young/old woman at the top of the post, appear ambiguous because they present equally two valid ways of subdividing the image. Our visual system uses concave curves (areas where a figure curves inwards) to break an image into parts which it then tries to combine into a recognizable object. In these illusions, two different sets of complementary concave curves divide the image into equally recognizable objects; as a result, we switch between different interpretations: a vase or a pair of faces; a young woman or an old one.

Many illusions are based on contextual clues confusing our visual system. For example, the two slabs in the image below appear to be different colours but can be seen to be identical by covering up the middle portion of the image (where they meet). Although the two slabs are the same shade of grey, they have opposite gradients around the point where they meet; the top slab gradually gets darker near the intersection while the bottom slab becomes lighter. Our visual system processes these gradients as clues about lighting and interprets them to mean that the top slab is being lit while the lower one is in shadow. Since the two slabs are the same colour and cast the same light onto our eyes, our visual system interprets the one that seems to be in shadow as though it were a lighter shade — which it would have to be in order to look the same colour even though it’s in shadow. If you cover the intersection with your finger or a strip of paper, the misleading gradients disappear and you can see that the two slabs are the same colour.

Likewise, the Shepard tables use perspective and other contextual clues (such as the legs) to fool our visual system into thinking that two tabletops are different shapes, although measurement will confirm that they are in fact identical.

In completion illusions, such as the Kanizsa triangle, our visual system constructs apparent objects that don’t really exist. Most people viewing this image will see a whiter-than-white triangle, often with distinct edges, although the only things in the picture are the lines and incomplete circles. Researchers have explored a variety of possible explanations for why we perceive illusory contours forming shapes in these images. The underlying figure doesn’t have to be symmetric in order for the illusion to work, as can be seen in the figure on the right:

Factors like the amount of contrast (i.e., the ratio of black to white) also didn’t provide a satisfactory explanation. Junctions between light and dark or different shades of grey are used by our visual system as clues about where objects meet and how they’re lit. Completion illusions work because we often interpret certain kinds of junctions (those shaped like and L or a T) as evidence of one object occluding another. In the Kanizsa triangle, the (approximately) L-shaped junctions in the Pacman-like circles and the junctions interrupting the lines convince our visual system that a white triangle is floating above the other shapes, blocking portions of them from our sight.

Illusions work because they represent cases where the rules used by our visual system generate ambiguous or unexpected interpretations of the world around us. Given that the a natural question arises: why would we evolve a set of rules that fails to accurately depict the world around us? The answer to that question can be found in my most recent post on Accumulating Glitches, where I present the case that natural selection has not equipped us with a truthful visual system.

Refs
Hoffman, Donald Visual Intelligence. New York: W.W. Norton & Company (1998).
Adelson, E.H. Lightness Perception and Lightness Illusions. In The New Cognitive Neurosciences, 2nd ed., M. Gazzaniga, ed. Cambridge, MA: MIT Press, (2000).
Eagleman, D. (2001). TIMELINE: Visual illusions and neurobiology Nature Reviews Neuroscience, 2 (12), 920-926 DOI: 10.1038/35104092

In English, “science” refers to a specific kind of knowledge and practice.  Physics certainly qualifies, along with biology and chemistry.  Economics and psychology probably count as “science” to many people, though they’re usually distinguished by being called ‘social sciences’.  On the other hand, many disciplines, ranging from history to literary analysis, would never be called science; instead, they fall under the heading “humanities”.

This distinction may seem perfectly natural to native English speakers, but it’s not actually universal.  Several languages, including a few closely related to English, don’t admit the same categories.  The Swedish word for science, for example, is vetenskap, but its meaning is actually somewhat more general; the same is true of the Dutch wetenschap and the German Wissenschaft.  These words all refer to the systematic and serious pursuit of knowledge and the search for general principles through rigorous study, regardless of the field of inquiry.  Vetenskap encompasses the whole gamut of disciplines that English divides into natural sciences, social sciences and humanities.

This isn’t just a peculiar quirk of a handful of Germanic languages.  The Arabic word علم (roughly pronounced ilm (actually `lm)) and the Finnish tiede are similar; both are translations of “science” but actually mean systematic knowledge more broadly.  One can speak of litteraturvetenskap in Swedish (literally “literature-science”) or uskontotiede in Finnish (literally “science of religion”) where in English we might say “comparative literature” and “religious studies”.  When I first saw the world litteraturvetenskap, it took me a moment to properly understand it; I initially went with the literal translation and tried to figure out what “literary science” might be.  Betraying my quantitative bias, I thought it had something to do with analysing word frequencies or usage patterns.

There isn’t really an etymological basis for the distinction that English makes. The word science comes to us from Latin via Old French; the root is the Latin verb scire, which simply means “to know”.  (I do find it fascinating that this may, in turn, come from an older Proto-Indo-European root sci- meaning “to divide/cut/separate” — who knew reductionism was embedded in the very word itself!)  It’s a relatively modern distinction (dating back to the mid-1800s) and we still sometimes use the word in the broader sense, but the modern usage certainly tends to reinforce our awareness of the differences between these fields of knowledge rather than highlighting their similarities.  To my mind, this somehow mirrors the modern conception of a decoupling between science and philosophy, a view which has always struck me as somewhere between narrow-minded and deeply flawed.  I find that I far prefer a more general approach to defining our pursuit of knowledge, whatever form that might take.

• Do you think there’s a good reason to distinguish between natural sciences, social sciences and humanities?
• How do other languages you know deal with the words “science” and “knowledge”?
• Do you think there’s such a thing as “literary science”?  Should Shakespeare scholars be called “literary scientists”?

(This post was inspired by HBG Casimir’s essay “When does jam become marmalade?” in A Random Walk in Science.)

It turns out that these viral stowaways aren’t just surprisingly common; they’re also strikingly important.  Once incorporated into the host’s genome, the viral genes can become tamed over time, eventually acquiring new functions.  Last year, Carl Zimmer wrote about viral genes that have played a critical role in the evolution of mammals.  Syncytin is a gene that’s essential for the placenta to properly connect with the uterus.  Synctin turned out to be a viral gene that infected our ancestors, but the story doesn’t end there.  Mice also have synctin genes which came from viruses, but they’re different from ours; the same is true of rabbits and of carnivorans.  In other words, on at least six different occasions ancestral mammals were infected by a virus that got incorporated into their genome and helped them make a placenta.

More recently, Carl has also written about the amazing discovery that viral genes are also important in deciding whether or not cells are totipotent — that is whether they’re able to produce any other kind of cell or only cells like them.  Genes are switched on and off by promoters, and researchers had discovered that the promoters of a whole suite of genes that are active specifically in totipotent cells were originally from viruses.

Now, a team of scientists have found that the same is true of primates in general.  DNA is packed quite tightly to fit into cells, so it has to be partially unfolded to makes genes accessible and allow them to be activated.  The researchers used data from the ENCODE project to study a map of accessible parts of the human genome in different cell types.  One of the things that they found was that virally-derived genes were much more common in these open regions than they were in the genome overall; in other words, the open regions were enriched in viral genes.  When the researchers categorized these regions based on which other species we share them with, they discovered that most of the virally-derived open regions in our genome appeared during the evolution of primates.

This figure from the paper shows the open regions of our genome lined up with a tree according to which species have similar regions. The colours in the pie charts show the proportion of different genetic elements in open regions of our DNA. The purple fraction (labelled “LTR/ERV”) is the viral genes that have stowed away in our genome — and in the genomes of other primates.

Many of these old virus genes turned out to be promoters and other stretches of DNA that regulate gene activity.  Most of the open regions also turned out to be specific to just a few cell types, suggesting that virally-derived genes might not just be important in making cells totipotent, but also in determining which genes are active in different types of cells.  The researchers claim that their results show that infection by viruses which got incorporated into the genome “considerably transformed the transcriptional landscape during primate evolution”.

But that’s not the limit of what viruses have been credited with.  The most exciting contribution they may have made is also the most contentious.  Unlike bacteria, our cells store their DNA inside a nucleus, and some scientists believe that the nucleus may have originally been a virus which infected an ancient bacterium.  Instead of copying itself and destroying the bacterium, it just took control of the cell’s machinery and, eventually, usurped the original genome.  Patrick Forterre even believes that viruses may have been instrumental in the ancient switch from RNA to DNA.  These are controversial suggestions and the jury’s still out, but one thing is certain: viruses have played a major role in shaping the evolution of life on Earth.

Ref
Jacques, P., Jeyakani, J., & Bourque, G. (2013). The Majority of Primate-Specific Regulatory Sequences Are Derived from Transposable Elements PLoS Genetics, 9 (5) DOI: 10.1371/journal.pgen.1003504

That’s all from me this time!

Faced with the rich diversity of living beings around us, humans have proven unable to resist the temptation to try to organize and categorize them. We have a natural tendency to classify things, a habit that’s deeply rooted in our cognition and use of language. Our brain excels at recognizing patterns (and thus finding meaning where it doesn’t exist), an ability that allows us to interact with the world using names — like “chair” — that we might be hard-pressed to properly explain. In fact, it’s surprisingly difficult to define even a seemingly straightforward word like “chair” in a way that would let us recognize everything that should be included (from office chairs and recliners to stools and wheelchairs) but nothing that shouldn’t (like tables, tree stumps, or other things we might decide to sit on).

Despite these difficulties, we’ve been classifying organisms throughout the history of human thought, from Aristotle’s division between plants and animals to modern scientific nomenclature. The modern classification system is based on grouping organisms into units called ‘species’; species, in turn, group together into a larger units called genus, family, order, and so on through the nested hierarchy of life. What make a species, though? Why should a particular group of organisms be thought of as a unit and given a distinct name? How do we decide which organisms make up a species?

Read the rest over at Accumulating Glitches…

At the moment, sequencing an entire genome is still expensive enough to be out of most people’s reach.  Instead, most personal genomics companies offer genotyping services.  Unlike sequencing, where all three billion letters of your DNA are read, genotyping involves looking at 500,000 or a million locations in your genome.  The locations are chosen because they’re close to genes known to play a role in certain characteristics or diseases or because variation at that location is statistically linked to a disease or trait.  By reading a few letters of DNA at these locations, scientists can find out which variant a person has.

For example, a person with “GG” at a particular location might have a higher chance of having blue eyes, while someone with “AA” at the same location would likely have brown eyes.  It’s not really that simple, though.  Eye color, like many other traits, is based on lots of genes which interact to determine the actual color.  It’s important to realize that genotyping doesn’t give you definite answers — you just get the odds.  Knowing how to interpret those odds is important, too.  You might be twice as likely as average to develop a certain disease, but if the average person has a 4% chance, your odds would still be pretty good.  You’d have an 8% chance of getting the disease, meaning that 92 of 100 people like you wouldn’t actually get sick.  Conditions like diabetes can also be strongly influenced by lifestyle choices.  Knowing you’re at risk can help you make better decisions, as long as you understand how to evaluate the numbers.

Despite the caveats, this sort information can be quite important; it can even change your life.  The actress Angelina Jolie recently decided to have a double mastectomy after a genetic test confirmed that she had a high chance of getting breast cancer.  She discovered that she was carrying a mutant form of the gene BRCA1 which normally repairs DNA.  The mutation made it more likely that she would develop breast or ovarian cancer, giving her odds of 87% and 50%, respectively.  Based on her family history, she decided to undergo the surgery as a preventative measure.  Following the operation, the likelihood of her developing breast cancer dropped to under 5%.

You’re not just learning about yourself, however.  You’re also learning about your relatives, since they share some of your genome.  Parents and children share 50% of their genes, as do full siblings.  If you’ve got a mutated version of a particular gene, there’s a 1/8 chance that your cousin has it, too.  Finding out about your own genome can mean inadvertently invading other people’s privacy.  Even if you don’t tell your relatives what you’ve learned, you might change your behaviour in ways which would tip them off.  Do you have the right to make that decision for them?  At what point do our individual rights give way to the rights of others and communal considerations?

I think this is pretty exciting technology and I’d love to get genotyped or, better yet, have my whole genome sequenced.  The question I’d like to pose is: Would you like to get genotyped or have your genome sequenced?  I’m not really asking about the issues of insurance, etc, here, but we can discuss them if people want.  I’m more curious about things like:

• Are you interested in finding out more about where your ancestors came from and what percentage of your DNA comes from Neanderthals?
• Would you just want the fun bits and none of the scary/important health-related stuff? Or maybe you’re only interested in the serious stuff?
• What about incurable conditions like Parkinson’s or Huntington’s?  Would you want to know if you’re more likely to develop them?  Would you change your life if you were?
• Given that you’re finding out for your whole family, would you ask for their consent? Do you think they should have a right to veto your decision?
• What about your partner — would they get a say?  Do you have an obligation to tell them about what you might pass on to your children?

• Do you think we’ll ever achieve the same efficiency as ants — e.g., to be able to find the shortest path back without actually doing the math?

It might come as a surprise, but we already do this sort of thing all the time. We may not all be very good at path finding, but most mature humans are able to catch an object that’s thrown towards them. We do it reflexively and don’t usually think about the computational effort it would take to explicitly calculate the path it will take. Based on just a few visual cues (and some experience), we account for things like the aerodynamics of the object, the force of the pass, the angle of the throw, and so forth. It’s actually pretty amazing once you think about it.

Another example is recognizing voices or faces. Humans are wired to find faces in their surroundings, even in common, everyday objects.  Not only that, but we can instantly pick out a friend’s face in a crowd or recognize their voice against the background buzz of a party.  That sounds pretty trivial unless, like Steve Royster or KH, you can’t do it for some reason.  Recognizing faces and voices are both computationally challenging tasks which are essentially effortless for a healthy human being. In fact, we can still do these things pretty accurately even if the voice is a bit different (e.g., because of illness or a bad phone connection) or the face is distorted.  We can even pick out, recognize and understand a voice when it’s mixed in with a couple of guitars, a keyboard, and some drums.  We’re also quite good at reading subtle clues in a person’s face, voice, or body language to gauge how they’re feeling, a trick which animators and advertisers take advantage of to bring their characters to life.

(Images by workbyknight)

The last example I’ll give is our use of language. Producing words takes an impressive co-ordination of many, many muscles, but our ability to understand what we hear is also pretty amazing. We get a continuous stream of sound as an input and manage to instantly break it down into recognizable chunks which carry meaning and construct images in our mind.  Most of us don’t have to think about how incredible these abilities are, but people with aphasia can have difficulty coming up with and producing words or even understanding a language they know.

These are all examples of complicated computational tasks which are so easy for humans that we often don’t realize how hard they are or how efficiently we solve them. To my mind, this is probably very similar to how ants “calculate” the shortest path home from a food source or how other animals solve other complex problems. It’s interesting that several of the examples of human efficiency have to do with social interaction — facial & vocal recognition, language, etc. That probably reflects the kinds of forces that have been shaping our cognitive development during evolutionary history.

Recognizing a face or using language might not seem quite the same as optimizing a path, but they are the sorts of things that humans excel at. Could we train the human brain to excel at other tasks instead? I don’t know. The brain is amazingly flexible, so we probably could. Some people who’ve lost their sight have learned to navigate by sound (i.e., to echo-locate like bats or dolphins). It’s possible that a specific sort of programme could train humans to excel at unusual tasks like path-finding, but I would imagine it would have to be in response to a strong need or a conscious and planned manipulation.

• Do you think humans can ever achieve that sort of coordination or discipline?

Again, language is an excellent example of the sort of impressive unconscious coordination humans are capable of. Not only do we all learn a language, but we constantly reinvent it together through use. New words, phrases and styles aren’t usually the result of a decision by a central authority; they tend to emerge organically from the spontaneous behaviour of large groups of humans. A similar argument could be made for art. Nobody planned the transition from impressionist music through blues and jazz to rock; it just happened as a result of thousands of independent decisions and urges in many different musicians. It was “in the air” — in other words, it was an unconscious coordination of their behaviour.

I guess the question might be whether we could harness that sort of unconscious coordination to improve the way we work or produce things. My feeling is that this is something that would be
challenging with most traditional labour roles and structures.  The cases where you do see this kind of behaviour are things like art, language, culture,…perhaps science, too. It might also be reflected somewhat in the Open Source movement. These are all fields in which the participants are invested, motivated, independent, and communicative. The coordination isn’t imposed or even constructed into the system; it simply emerges from the independent behaviour of individuals with similar interests and goals.

That’s all from me for now!  Over to you:

• Can you think of any other examples of complex tasks we do without learning or thinking about them?
• What about examples of people retraining part of their brain for a different task?
• How much do you think we can rewire our brains?
• Is there anything you’d like to train the human brain to do reflexively?
• Do you want to learn to echo-locate? (I do!)

The wasp Hymenoepimecis argyraphaga parasitizes the spider Plesiometa argyra, somehow manipulating it into building a web for the wasp larva.  The female wasp lays her egg on a spider, where it hatches into a larva that sucks the spider’s hemolymph (the equivalent of blood in insects, spiders, and other arthropods).  Eventually, the larva makes the spider stop building its usual orb-shaped web and build a different shaped one instead, a remarkable feat of manipulation which we still don’t understand.  The larva then sucks the spider dry and builds a cocoon for itself, using the spider’s web to hang itself in the air where it’s protected it from predators on the ground.

Glyptapanteles wasps lay their eggs inside caterpillars which serve as a food source for the larvae once they hatch.  After they’ve munched on the caterpillar’s insides, the larvae emerge and pupate; while the wasps are metamorphosing, the caterpillar defends the pupae against predators by swinging its head violently.  This behaviour isn’t seen in unparasitized caterpillars, but scientists don’t know for sure how the wasp manipulates the caterpillar, though it’s thought that it involves modifying the central nervous system in some way that we still don’t understand.

The emerald cockroach wasp (Ampulex compresa) may be the best understood parasitoid wasp.  The adult female grabs its host (the cockroach Periplaneta americana) and stings it twice between the head and abdomen.  The first sting paralyzes the cockroach’s forelegs for a few minutes, giving the wasp time to place her second, more precisely targeted sting.  The wasp then goes off to find a burrow, leaving the cockroach unattended; the cockroach doesn’t take the opportunity to escape while the wasp is away, but instead stays put and grooms itself excessively.  When the wasp returns, she clips the cockroach’s antennae and gorges herself on the hemolymph oozing from the wounds.  She then grabs the stump of one antenna and leads the cockroach to the burrow she’s prepared.  The cockroach stays docile throughout the whole process, making no effort to resist or fight off its captor.  Once in the burrow, the wasp lays an egg and glues it to the leg of the immobile cockroach before entombing it alive.

(Even though someone repeatedly calls the wasp “he” in this video, it’s a female.)

By injecting wasps with radioactively labelled amino acids which would get incorporated into their venom, researchers were able to see to where the venom got delivered by the wasp’s sting.  They found that the stinger was able to penetrate the protection around the cockroach’s nervous system and deliver the venom to three specific bundles of nerves.  Some studies have even suggested that the stinger may be able to sense and recognize nervous tissue in the cockroach’s head, explaining how it can deliver the venom so precisely.  Venom is first delivered to a nerve bundle in the thorax, where it causes brief paralysis.  The following two injections, which reduce the cockroach’s motivation to walk away from threats, are directed to nerve bundles in its head.

By experimenting with stung cockroaches, scientists have learned that the venom affects the central nervous system by making it harder for the cockroach to respond to a stimulus.  For example, cockroaches normally walk away if they’re poked just once, but after being stung they need four consecutive pokes to start walking — and then they walk more slowly.  Researchers were also able to reproduce some of these effects by injecting one of the nerve bundles in the cockroach’s head with the drug procaine, confirming that it was responsible for the changes in the roach’s behaviour.  The effect of the wasp’s sting eventually wears off; the cockroach will recover in about five days if the wasp doesn’t lay an egg on it.  During that time, though, it remains still and doesn’t respond as well to stimuli.

Once the egg hatches, the larva cuts a hole into the roach’s leg and feeds on its hemolymph.  After around seven days of sucking the roach’s blood, the larva moves into it and feeds on its innards, slowly killing it.  The cockroach’s body isn’t just a resource for the wasp; it can also serve as a rich source of food for microorganisms.  The young wasp needs to protect itself from these competitors.  Recently, a team of scientists learned how they accomplish this. By watching through a hole in the roach (covered with a glass plate), they saw the larva secrete drops of clear liquid inside the body and spread them around (and made a video of it).  Chemical analysis showed that the liquid contained antimicrobial substances which stop a wide range of bacteria from growing.  Free from competition, the larva eats nearly everything inside the roach before building a cocoon for itself and metamorphoses into an adult.  Six weeks later, the fully grown adult bursts out of the cockroach’s body.

Parasitoid wasps excel at manipulating the neurochemistry of their hosts; studying their toxins can teach us a lot about the neuro-anatomy of their hosts and provide us with insights and drugs which can be useful for neuroscience.  As with everything else in biology, there’s still a great deal to learn.  We still don’t know what makes the cockroach groom itself after being stung by Ampulex, let alone how other wasps accomplish their amazing feats of manipulation.  Darwin may have taken parasitoid wasps as a mark against divine design, but to me they’re yet another fascinating example of the beauty of the natural world.  What about you?  Do you find the antics of parasitoid wasps fascinating or gross?  Are they beautiful or repulsive?

Refs
Libersat, F., & Gal, R. (2012). What can parasitoid wasps teach us about decision-making in insects? Journal of Experimental Biology, 216 (1), 47-55 DOI: 10.1242/jeb.073999
Herzner, G., Schlecht, A., Dollhofer, V., Parzefall, C., Harrar, K., Kreuzer, A., Pilsl, L., & Ruther, J. (2013). Larvae of the parasitoid wasp Ampulex compressa sanitize their host, the American cockroach, with a blend of antimicrobials Proceedings of the National Academy of Sciences, 110 (4), 1369-1374 DOI: 10.1073/pnas.1213384110

Autoimmune diseases, in which the immune system malfunctions and attacks the body’s own cells, tend to affect females more often than males.  Dr. Jayne Danska and her colleagues investigated the basis of this difference by studying a strain of mice with high rates of type 1 diabetes, which is an autoimmune disease.  Female mice are twice as likely to develop the disease as males, and the disease is also more common in animals living in clean conditions, in keeping with the hygiene hypothesis.

When the researchers raised mice which had no gut bacteria, they found that the difference in susceptibility between the males and females disappeared.  The two sexes now developed autoimmune diabetes about as often as each other, at a rate in between that of normal males and females.  The team used genome sequencing technology to take a census of the gut  bacteria in normal male and female mice and found something startling.  Young male and female mice had similar kinds of gut bacteria, but around puberty the microbiome of the two sexes began to diverge; by the time they were adults, females and males had distinct populations of bacteria living in their gut.

To figure out whether these differences were linked with resistance to autoimmune diseases, the team gave doses of bacteria from adult males to young female mice.  As they grew up, these mice developed a new kind of microbiome, a community of gut bacteria that was different from that of both males and females.  They also proved to be resistant to type 1 diabetes, with the incidence rate dropping from 85% in normal females to 25% in the treated females.

Steroid hormones like testosterone are often used to treat autoimmune diseases, and one reason for the lower susceptibility of males might be the fact that they have higher levels of testosterone.  Earlier research had shown that castrated males were more likely to develop autoimmune diabetes and testosterone treatments could protect females, so the researchers also checked the levels of testosterone in the females after giving them bacteria from males.  Not only did they find that these mice had higher levels of testosterone than normal females, but they also showed that the improved resistance depended on the increase in testosterone.  When female mice were given a testosterone inhibitor along with the bacteria from male mice, the rate of diabetes returned to normal.

“It was completely unexpected to find that the sex of an animal determines aspects of their gut microbe composition, that these microbes affect sex hormone levels, and that the hormones in turn regulate an immune-mediated disease,” said Dr. Danska.  It’s a pretty startling result which may impact how we think about a variety of autoimmune diseases in humans, from multiple sclerosis to rheumatoid arthritis.  Like a lot of science, though, it generates nearly as many questions as it does answers, from the possible role of the micrbiome in pubescent development to how the gut bacteria actually regulate testosterone.  It also clearly underlines the fact that our microbiome isn’t really “something we have” but is a part and parcel of who we are, helping defining each of us, both as an individual and an ecosystem.

Ref
Markle JG, Frank DN, Mortin-Toth S, Robertson CE, Feazel LM, Rolle-Kampczyk U, von Bergen M, McCoy KD, Macpherson AJ, & Danska JS (2013). Sex differences in the gut microbiome drive hormone-dependent regulation of autoimmunity. Science (New York, N.Y.), 339 (6123), 1084-8 PMID: 23328391

The new blog will have a more didactic tone than Inspiring Science, perhaps something like my series on natural selection.  We’re planning a bunch of exciting posts, from explaining basic concepts to exploring fascinating evolutionary stories and the nuances of the evolutionary process.  In addition to the evolution blog, there will be blogs about geology, oceanography, environmental science, psychology, neuroscience, and physics.  I can’t say much more about it now, but stay tuned for more news!

Of course, I’m not going to stop writing on Inspiring Science.  I’ve nurtured this blog for the past 15 months and grown with it, so nothing’s going to change here.  Thank you to everyone who’s been reading, commenting, and encouraging me.  I started this blog to wade my way into the world of science writing; now, as I’m finishing my PhD and looking uncertainly towards the future, I’m really pleased to be taking the next step along that journey.

How would such a mechanism actually work in practice?  The team came up with several different designs where the last link could pull on the remainder of the falling chain  The most straightforward one is based on the idea of a tilted rod hitting the floor.  Since the rod is at an angle, one end will strike the floor first; the other end of the rod will pivot around the contact point with the ground, making it fall faster .  The free end speeds up for the same reason that you turn quickly if you grab onto something while running or skating: conservation of angular momentum.

When a rod falling at an angle hits the floor, the free end speeds up.

Anything attached to the free end of the rod (say, a rope connecting it to the rest of the chain) will get tugged downwards by the free end.  In this way, the impact with a surface actually creates a force which “pulls” in the rest of the chain.  It’s counter-intuitive and may even sound crazy, but it’s true.  The trio built chains based on their design — they look something like a wonky rope-ladder — and used a high-speed camera to film a pair of them falling side-by-side.  Just as they’d predicted, the chain that hit the table fell faster than the free-falling one (click the picture to see a video):

The chain on the left hits the table and so falls faster (click image for video) [credit: Ruina lab]

They also tested an “ordinary” link chain, which turns out to behave just the way you would expect.  In this case, the last link actually disconnects from the rest of the falling chain and doesn’t pass along any force, meaning the assumption used to solve the textbook problem is valid in this case.

Impact with a table doesn’t speed up an ordinary link chain, since the last link decouples from the rest of the chain (click picture for video) [credit: Ruina lab]

Given that we can measure and understand fluctuations in the universe’s infancy, it’s pretty striking that we’re still learning new things about something as familiar as a falling chain.  I often end posts by marvelling at the vast richness of the world, but this time its our ability to persistently question and precisely conceive the world that amazes me.  It’s wonderful that science — the same kind of science that enables us to build durable bridges and lightweight airplanes, to loft satellites to the edge of the solar system and to explore the deepest ocean valleys — enabled these researchers to predict and confirm such an odd, counter-intuitive result in something so seemingly mundane and straightforward.  Tiny mysteries and everyday wonders abound, just waiting to be discovered and described by an inquiring mind — and that’s part of what makes science so great!

Ref
Grewal, A., Johnson, P., & Ruina, A. (2011). A chain that speeds up, rather than slows, due to collisions: How compression can cause tension American Journal of Physics, 79 (7) DOI: 10.1119/1.3583481
(There’s also a preprint version of the article freely available on the Ruina lab webpage.)

Despite its title, this blog doesn’t answer a question that recently brought someone here: “Who is an inspirational scientist?”. I think many of the scientists I’ve met (or read) are inspirational, but I’m guessing “inspirational” is being used in a more public sense here.  Off the top of my head, Carl Sagan & Stephen Jay Gould certainly fit the bill. If you’ve got a suggestion, please share it in the comments! (Note: David Attenborough, though incredibly inspirational, isn’t a scientist.)

I was amused to find out that the question “Why do women in different areas like different guys?” led someone to this blog.  I guess I mention sexism & gender issues pretty frequently, and I did write a post entitled Why do men and women want different things?.  Unfortunately, I don’t have an answer for the original question.  Don’t different men/women generally like different people?  I’ve never thought of it as correlating with areas; I thought that’s just how people are.

Someone searching for “Did our ancestors have group sex?” found this blog — probably thanks to my old post Debating our ancestors’ sex lives.  I honestly don’t know the answer to that question (I’m not sure anyone really does!), but I can’t imagine why they wouldn’t have.  Modern humans have group sex, as do some of our closest living relatives, so I’m willing to go out on a limb and guess that our forebears did, too.

Whoever asked “Why is it always the most incompetent people promoted to leadership positions?” (or, similarly, “Why do incompetent people get promoted?”) may have been left unsatisfied by my post about the Peter principle.  If you were looking for emotional release and didn’t get it, I’m sorry.  I can only hope that learning about the inherent problems in how people are promoted made you feel better.

This one isn’t a question, but it was so odd that I decided to include it anyway: “peacock tail breast milk”.  I have no idea what that query is supposed to be about, but I hope my post about sexual selection and the brain satisfied whoever found it.  I’m also really glad I was able to tell someone the name for “the earthly smelling substance after first raining”: petrichor.

I’m guessing it was the discussion about parasites and identity that drew in the person who asked, “What is the natural habitat of an emerald cockroach?”.  I’ve never heard of an Emerald Cockroach (is there such a thing?), but if you were asking about the Emerald Cockroach Wasp, you’re in luck — I’m planning to write a whole post about it!

So, how about you?

• Have you done any unusual searches lately?  What did you find?
• What strange searches have led people to your blog (or website)?
• How did you find Inspiring Science?  What brought you here?

That’s all from me!

Tetrahymena thermophila is a ciliate, a unicellular creature covered with short, hair-like cilia.  Although ciliates are single-celled, they’re more closely related to us than to bacteria.  Unlike bacteria, ciliates (and other single-celled eukaryotes) house their DNA in a nucleus the same way that plants, animals, and fungi do.  In fact, ciliates are remarkable for having two nuclei, a larger one and a smaller one.  While multicellular animals have special cells (sperm and eggs) for reproduction, ciliates, being unicellular, settled for having multiple nuclei instead.  The larger nucleus (macronucleus) is used for the day-to-day running of the cell, while the smaller micronucleus is only used for sexual reproduction.  When two ciliates reproduce, the macronucleus disappears from both partners, while the micronuclei undergo meiosis and then come together to form a new macronucleus.  It’s during this process that the sex of the new T. thermophila gets determined.

The researchers starved Tetrahymena cells to make them mate and then used modern sequencing techniques to track down the sex-determination genes.  In humans, the sex-determination genes are mainly active in controlling how the embryo develops — whether it becomes male or female.  Tetrahymena consist of only a single cell, so sex isn’t a major developmental choice; instead, the sex-determination genes are active during mating, where they control whether two Tetrahymena cells can get together.  Strictly speaking, ciliates, fungi, and organisms like them have “mating types” instead of sexes.  Unlike sexes, mating types look the same — they’re just incompatible when it comes to mating. Cells which express different sex-determination genes can mate, while those with the same gene can’t. While the seven different mating types of T. thermophila might sound impressive, some kinds of fungi have thousands of different mating types — what a bewilderingly different world!

The team identified a pair of genes, dubbed MTA and MTB, that are responsible for the sex of T. thermophila.  The proteins made by these genes are probably expressed on the surface of the cell, where they can interact with the MTA and MTB proteins of other T. thermophila cells to find out if they’re different.  By studying the macronucleus and micronucleus, the researchers discovered that MTA and MTB, which are always located side-by-side, come in different versions corresponding to the different sexes.  While the macronucleus only has one copy of MTA and MTB, corresponding to the sex of the cell, the micronucleus has copies of all the different versions.  These different versions are incomplete, lacking a fragment of the two genes which is common to all the sexes.  They’re strung together in a row which is flanked on either end by the remaining fragment.  When two micronuclei come together during mating, all but one of the copies of MTA and MTB are cut out and discarded; the remaining pair are stitched together with the flanking fragments to form complete copies of MTA and MTB in the new macronucleus.  The sex of the new T. thermophila cell depends on which copy of MTA and MTB it gets, which seems to be randomly decided.

Each coloured block represents an MTA/MTB pair of a specific type (II, V, VI, etc). The germline micronucleus at the top contains all the different versions. During formation of the macronucleus, all but one pair are cut out; the remaining MTA/MTB pair determine the sex (in this case, VI). The grey coloured regions are common to all the sexes, but the only complete copies are the larger versions (at the extreme left and right of the row in the germline nucleus); these are stitched to MTA/MTB during sex determination. (Reproduced from the article in PLoS Biology)

This is an amazing and intricate way to pick a sex!  It’s much more similar to aspects of our immune system than to our relatively simple XY sex-determination system.  That’s not entirely surprising, though — matching between mating types is more like self/non-self recognition than what we think of as sex.  Nevertheless, it’s a stunning piece of molecular biology.  The DNA rearrangements and editing involved are extremely precise, and we don’t understand how they’re carried out.  Fortunately, T. thermophila is commonly used in experimental biology; since it’s so easy to work with in the lab, this is a great opportunity for scientists to learn more about DNA editing and repair processes.  I have to admit, though, that the real reason I decided to write about these critters is just how amazingly strange the story is: two nuclei, seven sexes, and a weird, complicated, beautiful way of managing it all!

Ref
Cervantes, M., Hamilton, E., Xiong, J., Lawson, M., Yuan, D., Hadjithomas, M., Miao, W., & Orias, E. (2013). Selecting One of Several Mating Types through Gene Segment Joining and Deletion in Tetrahymena thermophila PLoS Biology, 11 (3) DOI: 10.1371/journal.pbio.1001518
(The PLoS journals are open access, which means the paper is freely available. Open access, which I support whole-heartedly, is the subject of an important and ongoing debate about science and science communication.)

Away from the orange glare of our cities, the night sky is a breath-taking sight, a velvet curtain richly adorned with countless specks of glimmering light and the great, starry dynamo of the Milky Way. A sensitive enough telescope would fill that blackness with countless stars and galaxies invisible to us, a night sky dense with celestial wonders. An even more sensitive telescope could probe the void between those stars and galaxies and find a faint background glow coming from every direction. This glow is the cosmic microwave background radiation, the oldest light in the universe.

The early universe was hot. It was so hot that even atoms couldn’t form and matter existed only as plasma, a soup of protons and electrons. Although plasma might sound exotic, it’s actually quite common. Lightning is made of plasma, and you can also find it inside fluorescent bulbs and plasma globes. Because the early universe was a hot, dense plasma, light couldn’t travel very far through it; the whole universe was effectively opaque. Light goes through things that are transparent (like glass or air), but the charged particles in a plasma scatter light so effectively that it doesn’t get very far. Only after other types of matter had formed could light travel through the universe. Around 380,000 years after the Big Bang, the universe had cooled enough for the protons and electrons in the plasma to come together and make hydrogen. Shortly afterwards, light could finally travel freely; that light is what we see as the cosmic microwave background radiation.

During the billions of years that it traveled towards us, the light cooled due to the expansion of the universe. From an initial temperature of around 3000K, it cooled over a thousand-fold to become the 2.73K (-270°C) glow we now detect. The cosmic microwave background (CMB) was first detected by Arno Penzias and Robert Wilson in 1964, a feat which earned them a Nobel prize in 1978. That wasn’t the last Nobel to come out of CMB research, though. In 1989, NASA launched the COBE satellite to investigate the early universe by measuring the CMB. COBE detected differences in the temperature of the CMB throughout the sky. Although the microwave background comes from every direction, the temperature of the radiation differs by fractions of a degree throughout the sky. These differences are because of quantum fluctuations in the earliest structure of in the universe, when it was expanding faster than light during the first fraction of a second of existence. These fluctuations are recorded by the last scattering of the light that makes up the CMB. The tiny variations in the temperature of the CMB enable scientists to peer back at the structure of the early universe — the seeds from which everything else came. This discovery earned a Nobel prize for lead investigators George Smoot and John Mather in 2006.

The famous map of the CMB variations from data produced by the COBE spacecraft. (Photo credit: Wikipedia)

COBE was followed by another satellite, WMAP, which launched in 2001 and measured the variation in the CMB even more precisely. The data from WMAP fit the Standard Model of cosmology quite well; based on these data, scientists calculated the age of the universe (13.772 billion years) and its composition: 4.5% ordinary matter, 22.7% dark matter (which doesn’t interact with light) and 72.3% dark energy (the mysterious explanation of the fact that the expansion of the universe is accelerating). The high-resolution data set also included a few surprising anomalies, such as an unusually cold spot in the CMB, unexpected correlations in the large-scale structure of the universe and the ISW mystery, an unexpected strong distortions of the CMB by matter in the late universe.

Image of the (extremely tiny) variations in the cosmic background radiation detected by WMAP (Photo credit: Wikipedia)

The Planck satellite was launched by the ESA in 2009 to measure the CMB at an even higher resolution and with more precision. The detectors on Planck are cooled to an incredible 0.1K, making it sensitive to variations as small as a millionth of a degree. This morning, the Planck team released their first set of results about the CMB, including a 50 million pixel map of the temperature variations.

Image of the variations in the CMB detected by Planck (Photo credit: ESA)

Comparison of the variations detected by Planck and WMAP (Photo credit: ESA)

Planck confirmed the findings of WMAP, though some of the figures have changed. The age of the universe is now thought to be 13.81 billion years, so everyone is tens of millions of years older. There’s also more ordinary matter (4.8%) and dark matter (26.8%), meaning there’s less of the stuff we really don’t understand — dark energy (68.3%)! There’s also no evidence that any of a variety of extensions to the Standard Model — from new families of neutrinos to cosmic strings — is needed. Overall, the data from the Planck mission matches the predictions of the Standard Model of cosmology remarkably well, affirming our understanding of the cosmos — which is really amazing!

The Planck mission also confirmed the anomalies in the WMAP data. The researchers at the press conference highlighted the large-scale correlations in particular. Cosmologists can use the Standard Model to predict how similar the temperature of the CMB will be at two points based on the angle between them in the sky. Their predictions match the Planck data amazingly well, but only for small angles. For points farther apart in the sky (ie, separated by more than about 6°), the temperature of the CMB seems to be more similar than theory would predict. The same anomaly had been found by WMAP, but it might have been an artefact of the instruments or the analysis. The fact that Planck also found it makes it more likely that this is a real feature of the CMB.

Correlations of the temperature variation of the CMB at different distances in the sky. The predictions (green line) match the observations (red dots) extremely well for small angles, but not as well for large ones. (Photo credit: ESA)

Planck also confirmed other anomalies, such as the cold spot and the ISW mystery, and produced enough data to keep cosmologists and astronomers busy for years to come. These unexplained phenomena are a tantalizing reminder of how much we still don’t understand, despite the incredible accomplishments of cosmologists so far. “The universe is not really so simple,” said Paolo Natoli, one of the lead scientists of the Planck team. “Theories need to be improved.”

The solar wind raced towards the Earth, crossing the cold gulf of space at an incredible 700km/s — around 2.5 million kph!  Two and a half days later, these highly energetic protons and electrons smashed into the Earth’s magnetic field.  Charged particles can change direction when they fly through a magnetic field — that’s how the old CRT televisions worked — so the high-speed solar wind was deflected around the Earth by its magnetic field.  Some of the charged particles, instead of being deflected around the Earth and out into space, followed the magnetic field to the poles and down into the atmosphere.  Racing into the upper atmosphere, they crashed into oxygen atoms.  The impact made these atoms much more energetic; as they lost their extra energy, they fluoresced (just like a fluorescent bulb), producing a green glow that I was lucky enough to see: the aurora borealis.

The solar wind probably also energized some nitrogen particles to make red aurora, but I didn’t get to see any of those over Helsinki.  The solar wind has to penetrate deeper into the atmosphere to excite nitrogen and make a red glow that we can see, so red northern lights are more common further to the north.  The aurora don’t just happen over the North Pole; excitation of the atmosphere over the Southern Pole — called aurora australis or southern lights — happen together with those over the North Pole.  Here in Finland, the aurora are called revontulet, which means “fox fires”, because of old legends that they were made by an arctic fox.  One version claimed that the fox started fires by rubbing its bushy tail against mountains, sending up sparks that we see as the northern lights.

In addition to being beautiful, the aurora are immensely powerful, producing tens (or even hundreds!) of billions of watts (i.e., gigawatts).  A single show can produce as much energy as the North American power grid!  Unfortunately, that energy is spread over millions of square kilometers and released high up in the atmosphere, so we don’t really have a way of harnessing it.  Very powerful aurora can sometimes generate current in our electrical system; in 1859, an auroral storm produced enough power in some telegraph lines for operators to continue using them with the power switched off.

Earth isn’t the only planet in our solar system that gets auroral displays.  Jupiter, Saturn, Uranus and Neptune all have magnetic fields strong enough to produce aurora; here’s an amazing Hubble photo showing the aurora on Saturn:

Unfortunately the conditions tonight weren’t ideal for photography and we weren’t really prepared, so our pictures aren’t as impressive as I’d hoped.  What about you?  Have you seen the aurora (bonus points if you’ve seen the Southern Lights)?  Did you get any photos you’d like to share?  What are they called in your language and do you know any interesting or amusing stories about these mystical displays that light up our night sky?

One of the things that makes this study different from earlier ones is where they collected their samples from.  In most of the work I’ve written about previously, the researchers took samples from air over land, usually just a few kilometers up.  In this case, though, Natasha DeLeon-Rodriguez and her colleagues took advantage of a NASA project researching tropical hurricanes over the Caribbean.  By hitching a ride, the team was able to sample a habitat that hasn’t been sampled before — cloud water 10km above the open ocean — and to collect samples before, after and during two hurricanes in 2010.  They also collected cloud water off the coast of California and on a flight from California to Florida, as well as from several flights over the Gulf of Mexico, the Caribbean Sea and the midwestern Atlantic.

To identify the kinds of bacteria and fungi in their samples, the team extracted DNA, sequenced a fragment of a gene, and compared the sequences against known sequences.  They also estimated the number of cells in their samples by dividing the number of copies of the gene they found by the average number in a bacterial or fungal cell.  This gave them an estimate of about 5000 bacteria per cubic meter and 10-100 times fewer fungal cells — probably because fungal cells are larger and so can’t stay in the air for as long — which were divided into around 300 species.  That’s not much when compared with communities on the surface, where a single gram of soil can have thousands of species of bacteria and millions (or even billions) of individual cells.

Even though the cloud-borne communities might not be as rich or complex as those down here, the researchers noticed a few interesting things about their make-up.  Samples from higher altitudes had a different composition from those at lower altitudes; the researchers suggest that this might be because species that are better at nucleating ice get washed out and don’t make it to higher altitudes.  The hurricanes also had a big impact on the community composition — different kinds of bacteria were found in samples before, after and during the hurricanes, which seemed to bring up many new microbial cells into the atmosphere.  By tracing back the path of the clouds and comparing their genetic data against the greengenes database, the researchers hoped to figure out where the bacteria in their samples came from.  They found bacteria originating from nearly all the habitats on Earth, but most of the bacteria came from aquatic habitats.  This highlighted another unique feature of the samples from hurricanes, which were the only ones with significant quantities of bacteria associated with human feces — probably because they passed over populated areas.

Perhaps the most striking finding was that, despite all of the variability in community structure, 17 species were present in all of the samples.  The researchers think that these might be the core members of the microbiome of the clouds, capable of surviving and persisting for long periods at the high altitudes of the clouds, where harsh conditions like UV light, dessication, and oxidation present daunting challenges to most life.  Some of these organisms can break down oxalic acid, a common chemical in clouds, and use it as an energy source.  “For these organisms, perhaps, the conditions may not be that harsh,” said Konstantinos Konstantinidis, an assistant professor at the Georgia Institute of Technology who led this study. “I wouldn’t be surprised if there is active life and growth in clouds, but this is something we cannot say for sure now.”

Part of my reason for writing about bacteria in the clouds again is to show how science proceeds.  Despite the popular image of dramatic breakthroughs, most science doesn’t happen that way.  Instead, we accumulate knowledge bit by bit and try to piece it together, asking questions and gleaning nuggets of insight as we go.  It’s only afterwards, looking back, that we can see how far we’ve come.  Every study of these microbes uncovers a bit more about them, slowly forming a coherent picture.  This study showed that bacteria aren’t just floating around in the clouds or using them to get from one place to another — some bacteria actually make their living up there.  That’s the other reason I wrote about this — I just love that image!  It’s such a wonderful example of how strange this rich and complex world can be.  The surface of the planet is teeming with life and we know that microbes play an extremely important role in every habitat down here.  With growing evidence that they also persist in aerial habitats and make cloud-borne ecosystems, it’s important to consider what impact they might be having.

Ref:
Deleon-Rodriguez N, Lathem TL, Rodriguez-R LM, Barazesh JM, Anderson BE, Beyersdorf AJ, Ziemba LD, Bergin M, Nenes A, & Konstantinidis KT (2013). Microbiome of the upper troposphere: Species composition and prevalence, effects of tropical storms, and atmospheric implications. Proceedings of the National Academy of Sciences of the United States of America, 110 (7), 2575-80 PMID: 23359712

Nicolelis and his team implanted up to 30 tiny electrodes in rats’ brains to record their neural activity.  The rats were trained to press one of two levers whenever an indicator light in their cage went on. Once they had learned to pick the correct lever, the rats were separated into two groups, “encoders” and “decoders”. An encoder rat then repeated the lever-pressing test ten times during which the team then recorded its neural activity.  They used a computer to analyze patterns in the readings and build an average profile from the ten sessions.  Meanwhile, the decoder rats received further training.  They were taught to respond to electric pulses from the implants in their brain; many pulses meant to press the “correct” lever (the same one as in the light tests) while just a single pulse meant to press the other lever — the “wrong” one.

During the experiment itself, the encoder and decoder rats were in separate cages.  When the light went on in the encoder rat’s cage, the decoder rat couldn’t see it.  The well-trained encoder rat would press the correct lever and, as it did, its brain activity was fed into a computer which was connected to the implants in the decoder rat.  The computer calculated how similar the encoder’s brain pattern was to the average profile from earlier and used this to decide how many pulses the decoder rat received.  The more similar the profile was, the more pulses the decoder rat received.  If everything went smoothly, the encoder rat’s profile brain activity would match up with the average from the correct tests, so the decoder rat would get many pulses into its brain and would also make the right choice.  Things went that way in seven out of every ten trials — even though the decoder rat couldn’t see the light, it picked the correct lever because it was getting information from the encoder rat’s brain!

Information was transmitted from the brain of the encoder rat to the brain of the decoder, allowing them to influence one another’s behaviour. (Figure 1 from the paper)

To make things more interesting, the researchers also included a connection back to the encoder rat: if the decoder rat pressed the right lever, the encoder got an extra reward.  The effect was amazing.  If the decoder made a mistake, the encoder not only made a decision more quickly the next time, but also increased the clarity of the signal from its brain so it would be easier to detect!  “We saw that when the decoder rat committed an error, the encoder basically changed both its brain function and behavior to make it easier for its partner to get it right,” said Nicolelis. “Invariably, when the encoder made those adaptations, the decoder got the right decision more often, so they both got a better reward.”  Thanks to the feedback, the encoder rat’s brain was learning to improve its new-found ability to communicate!

The researchers also succeeded in making rats communicate about what they were sensing instead of what they were doing.  To do this, they recorded the activity from an encoder rat’s brain as it went through a gap rather than when it pressed a lever.  The rat’s whiskers would bend as it passed through the gap and, in a similar experiment, a decoder rat could receive information about how wide the gap was from the encoder’s brain.  In one experiment, encoder rats in Natal, Brazil communicated over the internet with decoders at Duke University in North Carolina.

So is this telepathy?  I don’t think so.  At the moment, the rats aren’t actually communicating directly, brain-to-brain.  There’s a computer in between them, interpreting the signal from the encoder’s brain and passing information to the decoder.  The computer judges whether the encoder’s brain activity looks right and informs the decoder, so the rats can only communicate about one thing.  The communication is also limited to a single binary signal for now — a left/right choice — but it’s not hard to imagine improving the translation program to include more information.  Semantic quibbles aside, though, this is exciting and amazing research!  Even if it’s just one bit of information about one fact, it’s communication from one brain to another.  This may not be a mind-meld yet, but it’s a first step…and I can’t wait to see where it leads!

Ref
Pais-Vieira, M., Lebedev, M., Kunicki, C., Wang, J., & Nicolelis, M. (2013). A Brain-to-Brain Interface for Real-Time Sharing of Sensorimotor Information Scientific Reports, 3 DOI: 10.1038/srep01319
(Scientific Reports is an open-access journal, so the original article is freely available to everyone!)