Oh, what a lazy marsupial sloth!

While reading about koalas for Monday’s post, I came across this poem in Anette Benesch’s PhD thesis. I thought it was too nice not to share:

Photo by Glennis Tracey

In the heat of the day in an average zoo,

What’s a koala most likely to do?

In the fork of a tree, in a featureless heap.

It closes its eyes and endevours to sleep.

And the visiting children and parents, they both

Say, “Oh, what a lazy marsupial sloth!

It’s far to lethargic to romp or to toil,

Perhaps it is drunk on the eukalypt oil.”

But that isn’t fair; it’s improper to mock

A creature that works on a different clock.

The schedule koalas are destined to keep

Is to deal with their business while we are asleep.

And, up in the treetops, throughout the long night,

They climb and they feed and make love, or they fight.

And if they took notice of me or of you,

They’d probably think we’re a sluggardly crew

Ronald Strahan - The Incomplete Book of Australian Mammals”

Those sleepy koalas

A while ago, I wrote about sloths not being the sleepiest animal in the world after all. Another animal that is known for its sleepiness is the koala. I even heard an expert on a German zoo-television shows claim that koalas spend at least 24 hours of each day asleep. That is surely an impressive amount. Could the koala take the sloth’s place as master-sleeper?

Researchers from Germany and Australia recorded the behavior of koalas in 2 zoos (Vienna, Austria and Sidney, Australia) to see exactly what they were doing all day. The results are in the actigrams below.

Captive koalas sleep about 12-14 hours per day, which is less than a captive sloth and a little more than a lab rat. Then why are koalas still mostly known as the adorable, but slow, sleepy and lazy menthol-scented cousins of the dropbear? The answer lies in our different daily rhythms: koalas are nocturnal animals.

Graph 1: Activity of four koalas. Actigrams (double plot = 48 hours) for (a) Ken, (b) Yindi (male and female from Sydney), (c) Bilyarra and (d) Mirra Li (male and female from Vienna). Grey blocks indicate feeding, black blocks represent locomotion. Sydney: white arrow under x-axis of plot = time of morning cleaning; black arrow = feeding. Vienna: white arrow = morning cleaning and feeding; black arrow = weighing and main feeding. D = dark (night); L = light (day) (reference 3)

Each line shows when the koala was moving (black) or eating (grey). All* of them do most of their eating and moving in the dark. From these graphs you can also see that koalas do not move much; there are only a few black marks each day. This is not just the case in captive koalas, who don’t have many reasons to move around in the first place, but free living koalas are not very active either. This inactivity is an excellent behavioral adaptation for conserving energy, because their food source, eucalyptus leaves, is poor in nutrients.

The nocturnal activity pattern of the koala is more obvious in a graph that shows when he sleeps:

Graph 2: Chrono-ethogram of Bilyarra’s sleeping behavior. Double plot, 24 hr twice in one line; black, sleeping activity (reference 2)

In graph 2, the black marks show when the koala (Bilyarra, the Viennese male) was asleep (that is, motionless with closed eyes). Pretty much all the sleeping happens when it is light, which is also when the zoo-visitor, a mostly diurnal species, comes to take a look.

In the Vienna zoo, the koalas are weighed and given fresh food in the morning. This is the beginning of the koala’s resting phase, which is also the time when animals sleep deepest. Interruptions of their rest at these times can be particularly stressful. It would be better to feed and weigh in the afternoon, after the koalas have had the chance to recover from the night’s activities. It may even be a good idea to limit access of visitors until then. After all, koalas do not nap their lives away; they just don’t do much at the times when we happen to be awake.

*Mirra Li (from the Vienna Zoo) has a much less obvious rhythm than the others. This is probably because she was much more responsive to the caretakers, who came in twice a day to weigh and feed the animals (see arrows in graph 1). She was also the only animal who came and interacted with them.

References:

(1)  Chronoethological assessment of well-being and husbandry in captive koalas Phascolarctos cinereus, Goldfuss 1817; A Benesch;  Frankfurt 2007 (PhD Thesis)

(2) Chrono-Ethologic Investigations on the Queensland Koala (Phascolarctos cinereus adustus) in Captivity; C Richter; Zoo Biology 25:357–368 (2006) (link)

(3)  Chronoethology of captive koalas Phascolarctos cinereus; AR Benesch, U Munro, D Schratter and G Fleissner; Australian Mammalogy 29: 237-240 (2008) (link)

(4) The photos in this post were taken by Glennis Tracey (glenasena on flickr).

New year

Happy new year, internet!

I feel that the first week of this fresh new year must include a resolutions post, even though telling others about your resolutions is probably the first step towards failure. So, I will only share my goals for this year by the numbers: 2, 25, 13, 50, 15, 1, 2 and 4.Other than that, I just hope to continue to become a better scientist, teacher and writer. Nothing major…

I’ll let you know what numbers I ended up with by the end of the year.   Do you have any plans for this year?

I hope 2013 will be pleasantly eventful, happy and healthy for all of you. And for my fellow science people: may your differences be significant, your controls well behaved, your reviewers be helpful, your weekends off be guilt-free, and your references properly formatted.

Real scientists don’t have a life

I saw this advertisement from Swiss Cancerresearch (Krebsforschung Schweiz) in the tram this morning, and it makes me sad.

Missed 89 home matches.
Blew off friends 76 times.
Found 1 new treatment for a rare form of cancer.

They ran a similar one some time ago:

Cancelled 18 theater nights.
Did not celebrate 2 wedding anniversaries.
Developed 1 new method to diagnose cancer.

These ads are there to call our attention to the fact that this foundation uses your donation to finance dedicated researchers, so that more people can be cured of cancer. What makes me sad is not that a foundation wants to collect money to cure cancer, but their portrayal of what a dedicated researcher should be like. There are by now many initiatives to show the public (that is, all the non-scientist people) that scientists come in many different shapes and like many different things, including science. Krebsforschung Schweiz seems to have understood that, but also tells us that dedicated scientists don’t actually do any of those things.  Proper researchers become science monks and nuns, who sacrifice everything to find a cure. They don’t need lives, because the thanks of the patients is all they need. That, and funding from this foundation, probably.

I know ideas like these are not new, but they make me feel I don’t ever want to be a proper scientist. I love my work, but I also love my life. Please stop telling me I am not allowed to have one.

EDIT:

The pics don’t look very readable on my home monitor. In case you find them hard to read too, this is what they say:

Top ad:
“89 Heimspiele verpasst, 76 Mal die Freunde vertröstet, 1neue Behandlung einer seltenen Krebsart erforscht”

Bottom ad:
“18 Theaterabende abgesagt, 2 Hochzeitstage nicht gefeiert, 1 neues Diagnoseverfahren bei Krebs entwickelt”

Both:
“Mit Ihrer Spende fördern wir engagierte Forscherinnen und Forscher. Damit immer mehr menschen von Krebs geheilt werden”

The consciousness of the octopus and the ignorance of the field

A bunch of researchers have declared the octopus conscious! Huzzah for the octopus!

There’s a reason they don’t call it the Order of the Brass Fruitfly, you know.

They declare the following:
“The absence of a neocortex does not appear to preclude an organism from experiencing affective states. Convergent evidence indicates that non-human animals have the neuroanatomical, neurochemical, and neurophysiological substrates of conscious states along with the capacity to exhibit intentional behaviors. Consequently, the weight of evidence indicates that humans are not unique in possessing the neurological substrates that generate consciousness. Nonhuman animals, including all mammals and birds, and many other creatures, including octopuses, also possess these neurological substrates.”

Well, duh.

So all mammals and birds are conscious, because they are similar to humans. And some molluscs (of which the octopus is the coolest). And some insects too, perhaps (although they are not cool enough to get a specific mention). This is because they display planning behavior, or have different neurophysiological states with their behavior. And because they have affective states too, in their own specific ways. And all that even though not all these animals are human, or even have a brain with a proper cortex.

I might be too cynical, but if discovering that humans are not the only ones to have some complex behaviors and inner states is a major step for your field, your field has been a pretty ignorant one.

But I am sure the dinner and drinks at the signing ceremony were nice.

Read the Cambridge Declaration of Consciousness here:
http://fcmconference.org/img/CambridgeDeclarationOnConsciousness.pdf

FENS day 4: gamers pwn it all

I would have liked to file the following talk under ‘research I like because it reinforces my existing patterns of behavior’, but it turns out that I actually play the wrong kind of games. Nevertheless, Daphne Bavelier’s lecture about the awesomeness of gamers and how becoming a gamer can make you more awesome was a very fun one.

Her research shows that playing video games is good for you. Gamers who play action video games (like Medal of Honor, Call of Duty, Doom etc.) are better at a variety of learning tasks than non-gamers, even when the skills that those learning tasks test have nothing to do with skills that could have been learned in the game.

Does this mean that playing video games makes you a better learner, or simply that better learners tend to play video games? To figure this out, subjects were tested on a learning task before and after playing 50 hours of video games. The people that played the action game ended up learning faster than those that played another kind of game. So, it looks like video gaming can actually make you a better at learning tasks.

A problem with training for a specific task, is that the training normally improves performance only on that task. For example, training on a specific type of visual learning task improves your performance only on that task, not on other visual learning tasks. In the same way, expert tetris players are not automatically experts in playing Super Mario or chess. Nor are expert violinists always good piano players. Dr. Bavelier’s gamers, however, are better at a wide range of learning tasks. What’s up with that?

But I need it for my brain! Science says so!

The gamers are not only better, but also faster at learning tasks than other subjects, because they have better attention. They are better at filtering out irrelevant noise from the environment and focus on the things that really matter to learn something. fMRI studies also show that gamers can do the same task with less brain activation. In this sense, they are similar to expert meditators.

Gaming teaches you how to be a better learner. How could we use this for good things besides fun?

Apart from being better learners, which is good for anything that requires learning, gamers also have better contrast sensitivity when their vision is tested. Contrast sensitivity is a trainable skill, which means that in theory, gaming could be used as a training to improve vision. Where usual strategies for improving vision are generally focused on the eyes (think glasses, contact lenses etc.), games could change vision at the level of the brain. This would be useful for adults with amblyopia (lazy eye), for example, for whom the putting a patch on the eye has not helped to improve vision in one of their eyes. Bavelier hopes that games could be used to help reopen a critical period for acquiring stereovision for these people.

Apart from that, gamers also have a better number sense than non-gamers. Kids with better number sense are better at learning more abstract math later on. If we could use games to train number sense, this could lead to improved math performance in school, which is good for a whole lot of other things that require some grasp on mathematics.

Other uses for video games could be training the workforce, improving education systems, help remedy some of the cognitive symptoms of aging, and perhaps even to aid rehabilitation. Although I have to say that improving general learning does in patients undergoing some sort of rehabilitation is only of limited use when they lack the means to do anything with those skills. A stroke patient may benefit from becoming a better learner, but if she has no working motor cortex, this will not be enable her to move her arm any sooner or better.

To be able to widely apply of games to help us all improve, it is important to find out which aspects of the games are relevant and to apply them in a variety of different game settings. Not everyone enjoys playing first-person shooters and I suspect that training will be a lot less effective if the persons that are to be trained resent playing.

Although all these benefits of playing action video games sound almost too good to be true, I think they give some reason to think about the way we study and train in a new way. If we can use the games to help ourselves learn to learn, without feeling like we are doing ‘the boring learning thing’, this is good. After all, who doesn’t like to play?

More from FENS: 
FENS day 1: All the cells!
FENS day 2: Dendritic computation

FENS day 2: dendritic computation

One of the things I really enjoyed was the plenary lecture given by Michael Häusser. What Dr. Häusser wants to figure out is basically how brain activity produces behavior. And because this is a pretty big question, he looks at some really small things.

The brain is really great at processing information: it receives all kinds of input from inside and outside of the body and then produces a specific and (generally) appropriate response.

Since the brain is made up out of individual cells, it stands to reason that these cells are the fundamental compartments where processing of signals, or computation, takes place in the brain. However, neurons are not just little blobs where every part does pretty much the same thing. Most neurons have one or more dendrites that receive incoming signals from other neurons, a cell body where integration of these signals and general cell housekeeping happen, and one axon that passes the output of this cell on to others.

A fluorescently labelled neuron. The thicker branch that exits the frame on the bottom is the axon, all the others are dendrites.

The dendrites are not just passive wires that conduct a signal to the cell body, though. They are active and can react to incoming signals in different ways. This means that single dendrite can already act as a computation unit. And because the dendrite is the first ‘processing stop’ that an incoming signal reaches, it can have a major effect on further transmission of signals.

Dr. Häusser’s lab combines single cell and single dendrite recordings, and even multiple recordings in different spots on the same cell with a really neat method that allows them to locally release glutamate  and activate very small parts of the dendrites.  By measuring what happens in the dendrites and in the cell body of the same cell, he and his colleagues are able to measure exactly what the responses are to different incoming signals.

So how are incoming patterns of activity interpreted by single neurons? The brain has two major mechanisms to encoding messages:  the firing rate, where stronger signals give more activity, and spike timing, where activity from two inputs at the same time leads to a response. The first is good for the integration of activity over time, while the latter is useful to detect coincidences, for example. Neurons generally use a mix of the both mechanisms, depending on the cell and circumstances.

Recording from the cell body (lower electrode, in blue) and dendrite (upper electrode, in green) of the same neuron.

Dr. Häusser’s research shows that the tips of the dendrites are more sensitive to rate coding, while the parts that are closer to the cell body are more sensitive to spike timing coding.  This means that single dendrites already process an incoming signal and carry out a part of the neuronal computation that eventually leads to a response (or not).

Generally, a neuron does not receive one, but many  incoming signals at the same time or shortly after one another. The pattern of these incoming signals is relevant to form an appropriate response.  Neurons are indeed able to distinguish different patterns of incoming signals on various parts of their dendrites, and generate different outputs based on these input patterns. Dr. Häusser and his colleagues found that dendrites are also able to respond in different ways depending on the order of activation of parts of that dendrite.

How do individual neurons interpret patterns of incoming signals?

The big question that remains is of course: does this kind of dendritic computation also happen in vivo? Most of the experiments I descibed above are carried out in cultured cells or in brain slices, and would extremely difficult to do in intact animals. The awesome people of Michael Häusser’s lab gave it a go anyway.  They recorded the dendritic activity from cells in the visual cortex. Cells in the visual cortex react specifically to different directions of stripes. Some cells respond to horizontal stripes, while others are activated by vertical stripes, for example. This is essentially what enables us to see different shapes. So we know that the cell bodies of these cells react in specific ways; what about the dendrites? It turns out that the dendritic activity of cells in the visual cortex is also tuned to the orientation of visual input, but their responses are very heterogenous. A dendrite will respond mostly to one vertical stripes for example, and not so much to horizontal stripes. However, activation patterns from dendrites of two different vertical neurons can look completely different. So, it looks like the dendrites do act as computational units in real(er) life too, but as with many things in real life, things are a lot more complicated than in vitro.

More from FENS:
FENS day 1: All the cells!
FENS day 4: Gamers pwn it all