A friend of mine at hopkins (several, actually) do lab experiments with mice. Recently one experiment a co-worker was doing was continually failing even though it had been proven to work in other cases. Several of the mice tested would behave fine, but one or two were in terrible health and became highly antagonistic. They couldn't figure out why the experiment was failing.
My friend took a look at their setup. During the day there was a machine that was turned on next to the unhealthy mice. It vibrated and emanated a faint noise. He told them to turn off the machine. In a week, the mice were back to normal.
Nobody had thought about it, but mice are much more sensitive to sound and vibration than humans and it was stressing them out so much they were going crazy. There's probably lots of other things that can be overlooked (like temperature), but it's hard for us to notice because we aren't mice.
Right. The point being, if we're talking scientifically here, you can't just say that the machine was the issue and be done with it.
What kind of machine was it? Was the machine emitting radiation? Was it emitting light? What time of day was is turned off? What color was it? Could the mice see if from their cage, or was it out of view?
The science i've seen done in university labs does not follow the kind of rigor you're talking about. Much of the time, you simply need to get "pretty close" to best practices.
The biggest indicator that the machine was the likely culprit (without considering any of the factors you mention) was that the experiment had been proven in the past to work, with a control group for comparison. Add to that the other mice that were doing fine that were away from the machine, and that the mice close to the machine got better when it was turned off. It's not "scientific" per se, but it makes more sense than any other reason that anyone could come up with for this scenario.
To deal with this in future projects, a researcher could take an approach that treats everything as a set of systems, and examining every part of each system for potential problems. But I doubt anyone's getting paid well enough to do all that for every experiment. But it is reasonable that if you're doing an experiment that depends on a bunch of other experimental results, you do your due diligence and vet all the information and results.
Actually none of the sub-questions you asked matter in this case.
The researchers already functionally conducted an AB study - http://en.wikipedia.org/wiki/Single-subject_research#A-B . In this case, they saw behavior with the machine turned on that went away when they turned the machine off. If they want to confirm that some aspect of the machine (and it doesn't matter if it's noise, smell, light, or some mousey-sense that we are completely unaware of) they could extend it to an ABAB study, by turning the machine on, noting the incidence of 'crazy' behaviors over a period of time, then turning it off again and noting the incidence of 'crazy' behaviors.
There's probably lots of other things that can be overlooked
rather than about machines and mice. It's a hard problem in science: how do you build a truly reproducible setup when you don't have a full causal understanding of the system?
I'm reminded of the study that found rats will choose heroin over food, over and over again, until they die; and the follow-up study that compared rats kept in isolation (matching the original study) with rats allowed to live more naturally in communities and found that rats living in communities try the heroin but then go back to eating food.
The hazard here is not inside a fully parallel blinded trial with controls in the same environment, but in comparison and replication of studies, as well as sequential studies where the lab temperature changed. If quite minor, to human senses, changes in temperature significantly affect tumor formation, then up until now that's been a huge noise source in cross-study comparison, and temperature should be at least recorded precisely over time for all future tox studies.
not necessarily so - the article mentions that the cold inhibits two specific immune cell types; discovered drugs may only be effective in combating tumors arising from these specific deficits, which aren't relevant in real scenarios, i.e. a systematic error from cold in all studies.
When I was in middle school my science fair project involved the temperature controls for experiments on electric fish. I ended up getting a byline in science mag for the project, as it had thrown into question a lot of existing studies that hadn't controlled for temperature.
Richard Feynman wrote about a very similar problem, that rat-maze experiments rarely controlled for the rats' ability to hear differences in the floor. He used it as an example of "Cargo Cult Science" (http://neurotheory.columbia.edu/~ken/cargo_cult.html) - if you aren't rigorous with your assumptions, you are not doing science, and the truth is not going to come.
Did anyone else assume that temperature would be something they'd realize they need to control for? I would even suggest they should have a control group with a fluctuating temperature, as humans go through temperature fluctuations - and perhaps even seasonally alter temperatures.
They did control for temperature. They mandated mice be kept between 20C and 26C, which is between 68F and 78F. Mice live in houses, that's the temperature of houses.
31C is 88F. It's not absurdly hot, but it's hot enough that it's surprisingly hot to me.
Control for temperature doesn't mean "control the temperature". It means to set up the experiments such that you can detect the effect of temperature. Basically, how did they decided that 20-26C was the right range? If it wasn't by testing differences between mice at colder and warmer temperatures, then they weren't controlling for temperature.
Controlling a variable means setting up the experiments such that the variable doesn't influence the results. This is typically done by making sure it is the same in both the experimental and control group. Experiments typically control a large number variables, and only vary a small number (ideally, one).
Mice in a particular experiment tend to be housed in the same room or same rack, with enough circulation in the room to hopefully assure a homogeneous temperature in the room. Therefore, the control animals were likely right next to treatment animals. In this way, one could reasonably say that variations between the groups is not due to the way they were housed. Many other things that are controlled in housing include their feed, bedding, access to water, number of animals in a cage, etc.
Thank you both, I wasn't thinking clearly. I was thinking of the sort of "control for" that involves accounting for differences between groups, but totally spaced on the part where you can just "control for" something by ensuring it's the same.
Perhaps I just don't know the correct terminology, though you can have multiple control groups, no? That is what I was meaning, I realize I wasn't very clear.
Yeah, you can have multiple groups in the study, but you have to pick them somehow. Since most scientists aren't just sadistic mouse-torturers, you're picking groups that you think will tell you something. If you're testing a medicine, you might have one control (no-treatment) group, one group which receives whatever the current conventional/best drug is for the particular ailment, and one group which receives your new therapy. It's useful to see if your new therapy is any more effective than the old, in addition to seeing if it is effective at all.
On the other hand, prior to this experiment, there was no reason to suspect that mice at 31C would be significantly different than mice at 25C. So why waste time, money, and mice by re-doing everything at 31C? Remember, things don't end well for the control group just because they're the no treatment group: they've got to be killed and autopsied at the same rate as the experimental group. When the study says "chemical X resulted in a 35% increase in tumors", they know that because they cut open both the experimental and control group, counted their tumors, weighed them, compared them, and came up with that number of 35%.
Prior to this experiment, there was no reason to think that mice at 31C would be any different to mice at 25C, any more than you would expect mice wearing funny hats to be any different than mice without. Would you expect scientists to run a second control group of mice wearing funny hats, and slice them open and count their tumors, just in case it ended up being different?
All other things being equal, yeah. But while a six foot tall mouse would lose heat way slower than a normal sized mouse, people and mice have different amounts of fat and fur. It's not obvious to me a priori that mice would have a different optimum temperature range.
I mean, England barely ever has a high of 31C, and there are plenty of mice there. I wouldn't suspect that every mouse in England is immunocompromised.
Other points - mice rarely have bedding in their cages which they use to increase their body heat, and also to sleep in, which allows them to lower their temperature without ill affect (why many animals sleep through the night.
Nests in cages mean mice can't be observed, so are usually not provided.
Mice are kept lean, so have limited body fat for warmth.
Mice are kept in overly-well ventilated cages, which adds to the cold felt.
Many mice are bred for limited or low fur, which again means they need a higher mean temperature to stay warm.
Lastly the cages encourage non wild behaviour, and are rarely enriched with toys, or have long spaces in which to run, which the animals also use to mediate their body temperature.
Basically, doing "normal" biology/psychology experiments "lab" mice is sort of like doing those experiments on astronauts living on the moon. A rather horrid mis-approximation of the experience/behavior of normal population.
The moden scientific laboratory design puts so much emphasis on precision (cages, diets, etc, for consistent experimental outcomes), that it fails miserably at accuracy (similarity to what non-experimental individuals)
We've always known that lab animals are not perfect models of human biology. For instance, human livers -- for our body weight -- are several times larger than mouse livers. This has important consequences on how we deal with toxins.
Mice have been valuable tools not because of how good a model they are but because they're cheap and fast. Apes are a much better animal model for most purposes, but they're expensive and slow, and raise way more moral questions. And people as test subjects are all that and a bag of chips.
Instead, researchers using animals work to understand how an animal model compares to humans. They pick their animal test subjects based on what they're researching; for instance, rabbits might be better for researching a particular disease, and cats might be better for researching a particular cognitive disorder.
So -- assuming this new information holds up to replication -- what this means is that we potentially have a more accurate, cheaper animal model on hand than we previously thought. If mice at 31C behave significantly differently than mice at 25C, then researchers will be able to use mice in cases they previously weren't, eliminating the need for other animal models or even some human testing in some cases (eg, the experiments will fail at mice and never make it to human testing). This will save money, and speed up research.
Might as well add magnetic field to the list. In the odd circumstance it's possible there could be a large amount of power / electricity in wiring nearby, creating a magnetic field that's different than other test groups, etc..
"Apparent gravity on the earth's surface varies by around 0.7%, from 9.7639 m/s2 on the Nevado Huascarán mountain in Peru to 9.8337 m/s2 at the surface of the Arctic Ocean"
I think you've been over-sensitized by spending too much time in certain regions of the Internet. When the only tool in your possession is a hammer, everything looks like a nail. When most of your informational input is of a certain nature, everything else tends to relate to it.
No. There is only one way to pronounce "PNAS" (other than spelling it), and it sounds exactly like "penis". Not "cock", "dong", "member", "wang", or any other arguably childish euphemism. It sounds exactly like the proper English noun for male genitalia.
And this has nothing to do with you me or the parent thinking it sounds like "penis". EVERYBODY WE TALK TO will think we're saying "penis". Believe me, I get this enough with Coq (the theorem prover) and "cock"'s merely slang.
I'd pronounce PNAS with a hard a, like in ass, not like the end of penis, which has the i in kiss, so personally I don't find it that much of a homophone. Though I hate to think what that sentence has just done to you, given your sensitivity in such matters.
Though I hate to think what that sentence has just done to you, given your sensitivity in such matters.
Beside being unnecessarily snide, you missed my point, like, entirely.
I don't give a rat's ass that it sounds like penis. But I couldn't say it in public without eliciting childish sniggers from whomever I'm trying to talk to.
Thermoneutral means zilch energy consumption at basal metabolic rate when motionless in a windless environment. Its a narrow temp range. For bare humans its about 80F which seems warm because we're usually not motionless and usually wear clothes.
You can make a joke out of the whole thing in that they're trying to turn the mice into couch potatoes who don't burn a single excess calorie to better match the human couch potatoes.
Its not totally ridiculous to think a drug could be developed that would have different effects on human couch potatoes vs the now somewhat unusual athletic humans. We already have identified compounds like that, take all the roids you want, no exercise means no gains. Off the top of my head there's some medications that screw up human thermoregulation, like prevent or reduce sweating, so exercise for someone on those meds equals heatstroke (well, not guaranteed, but close)
I'd be super uncomfortable with anybody asserting that conclusion.
What the quote in the original article says is that they have data that in a preliminary fashion supports the hypothesis that being cold (and trying to keep warm) suppresses a mouse's ability to fight off tumor growth.
That is very different from saying cold/stress causes tumors.
"causes" is the troubling word there. From what I've read, in most people and animals tumors are growing all of the time due to cell replication defects, but normally the immune system gets rid of them. Cancer, as a disease, isn't "your body is making tumors", it's "your immune system isn't keeping up with your tumors".
From that point of view, anything which makes your immune system less effective "causes" cancer by making it harder for your immune system to kill tumor cells faster than they occur. Being cold all of the time, or being subjected to any type of chronic stress, certainly seems like it could do that.
depends. a lot of these mice are specific breeds designed to make it at least feasible to perform experimentation. It's common to use mouse lines that have extremely deficient immune systems. Otherwise there would be almost zero chance of tumor cells transferred from a different species (i.e. human) being able to grow on the mouse for study.
This is a very good point. It is important for non-researchers to know that when you want to do a mouse (or rat, dog, or non-human primate) study, you have to decide what particular strain of mouse to buy. There are MANY options from Harlan, Charles River, etc, and each strain has its own genetic composition. Some of the strains are intentionally inbred, or intentionally outbred.
I think this variation might also be more important than some people are giving credit for. It is entirely possible to run a mouse study twice, confirming the results. Then, you can run the same protocol in a different strain and get no results or different results. Then, you have to try to understand which strain might be more applicable in a human population.
I'll have to double check, but I believe we keep our mouse room at ~25ºC (+/- 5%). 31C for housing seems very high and probably well outside mean temperatures for native habitat
Even putting aside the moral and ethic issues, how can animal testing be allowed to continue when the studies are rife with bias and error? Officials at America's National Institutes of Health (NIH) state over three quarters of all published biomedical findings are irreproducible due to fraud and incompetence.
Moreso, the number of retractions has grown over tenfold in the past decade - but it's only 0.2% of the 1.4 million papers published in academic journals. The studies that were retracted were only done so when fundamental flaws were exposed by others. Many more false studies go on unnoticed and continue to be referenced.
Worse of all, 90 percent of drugs that were "successfully" tested on animals then failed in human trials.
EDIT: It should be a criminal offence for researchers to subject animals to harm and death and produce fraudulent papers.
What makes you think that they aren't trying to get it right? The animal to human failure rate isn't really related to scientist's insight, skill, or knowledge, but to the fact that biology is really fucking complicated. A researcher can pour their life and soul into their work, watch it work its way through all the layers of testing, showing consistent and sencial results, and then watch it blow up when it reaches a human, cause surprise! Third and higher order effects can't be predicted!
Also, your line of logic seems like a non-sequitur. The poor overall quality of papers (irreproducability for example) has nothing to do with animal testing. If you put aside moral and ethical issues of animal testing, the question isn't 'how can animal testing continue given how crappy their results are' its 'how can testing continue given how crappy their results are'.
What makes you think that they aren't trying to get it right?
There's countless examples of fraud, complacency and bias in scientific research, but I'll just quote the wise (albeit fictional) Jedi Master Yoda: "Do, or do not. There is no try." Especially when the fate of sentient life hangs in the balance.
Still doesn't make sense. There are countless examples of shit gone wrong, but until you can prove that it is systemic and/or that the responsibility of the fraud, complacency, etc lie in the majority (or hell, say significant minority) of researchers and scientist, it's really not fair to paint them all with the same brush.
Certainly there are countless examples of fraud, complacency and bias in all human endeavors, but that hardly indites all of humanity to "not trying to get it right".
Furthermore, "do or do not" is childish when applied out of its proper context. The intention of "do or do not" is to remind you that in terms of your effect on the world, the effort you put into an unsuccessful act doesn't get you any closer to the actual act. What "do or do not" does not make -any- sense in is in situations with many factors out of your control (especially factors you don't even know about!), and especially when trying to judge someone or a group of people.
I'm not sure exactly what point you're trying to make, but I think its that "when the fate of sentient beings hang in the balance, how hard you tried to do it right doesn't matter in my judgement of your ability to conduct your work". Which is completely nonsense. Go tell that you a paramedic, or a doctor right after they had someone die on them.
Submitting and publishing dodgy, fraudulent or sloppy studies isn't trying, it undermines all of science. And when animals are subjected to pain and death for fame or career advancement, it's morally corrupt, even criminal.
Science is the art of improving successive approximations. If you think that a result is useless unless it is absolutely perfect, you are simply wrong.
Putting ethical and philosophical issues front-and-center, absolutely. Why is this a difficult argument to make? It's our disease. It affects our bodies. If we want a cure then we should put our own skin in the game.
Traditionally, we would be exclusively using "others" usually identified by race and religion, or perhaps socioeconomic class, so that's a pretty big wrench in the ethical / philosophical works. Not as simple ethically as rodents vs humans, more like rodents vs (most likely) non-white humans. Probably better off killing rodents than deciding which religion to genocide.
The Germans did some scientific research under this protocol in the early 40s, its still highly controversial WRT should the results be destroyed and ignored out of respect or kept around as a memorial or ...
Suppose it costs the animal kingdom 1,000,000,000,000 microbial animals of various sorts, 1,000,000,000 macroscopic arthropods, 1,000,000 vertebrates, and 10,000 mammals to live one life according to the lifestyle we expect in western society.
If that's so, is it your moral/ethical/philosophical duty to kill yourself, and to kill other humans, whose environmental footprint is so great? Why not?
You who would see species as peers, as independent entities worthy of a respect for their collective rights, I accuse you of being insufficiently imaginative, and lacking any sense of perspective. Your footprint on this earth is FAR costlier than the 1/1000th of a cow you didn't consume for dinner.
Human exceptionalism, as a formal ethical principle, is pragmatically required if you wish your system of philosophy to be persuasive to essentially anyone. If you can't see this, you are either entirely ignorant of the processes of industrialized civilization, or you're exhibiting magical thinking. The species boundary forms a convenient natural Schelling Point at which it is particularly useful to distinguish between things deserving of empathy on equal footing with us, and things which are subhuman, inferior, not capable and thus not useful to incorporate into our model. Yes, things beyond this boundary still think and learn, still feel pain, and still form social relationships... but they're not, they can't be, the same as us, because they're not capable of fulfilling the roles of, nor exerting the pressure on our social contracts necessary to change them.
So basically you want to treat humans like animals?
I assume you think you want to treat animals like humans, but that's not going to happen. If people listened to you it would end up going the other way.
So I'm glad your opinion is a marginal one and that virtually everyone on the planet disagrees with you.
A lot of animal testing is general research, not drug testing. For example, it's fairly impossible to do brain research on anything but a brain. Also, a lot of research is basically infinite trial-and-error until you find something that works; doing this on a scale large enough for humans would kill lots of people, versus lots of mice. It's also faster with mice because they breed quickly and prodigiously, so we don't have to wait 300 years for enough testing to find a result.
Your basic argument is "it's error prone so we shouldn't do it." But that's exactly why we do testing - to work out the errors! The fact that all of these failures are occurring is proof that doing human research would simply take more time, not less.
"Organ on a chip" is also only useful to test for specific response. It does not show how the experiment works when introduced to an entire system of organs.
yep, the same logic of vivisection, concentration camp hypothermia experiments, Tuskegee syphilis experiment, etc... - "it advances science"
>Also, a lot of research is basically infinite trial-and-error until you find something that works; doing this on a scale large enough for humans would kill lots of people, versus lots of mice.
Doing research on humans vs. mice would [these days, though not even 40-60 years ago] force to do the research differently, mindfully and efficiently, instead of meat chopping factory it is now.
Of course it would force them to do the research differently. Like, not at all, for example. Some research involves seeing processes which can only come after death. I can't even begin to explain all the other research that wouldn't happen, because I am not a bioresearcher.
It's really easy to judge an entire range of research with zero insight on what actually goes on or why. I would suggest you go ask a bioresearcher to explain what they do to you, and why it is necessary to do it that way on animals, before you call it "a meat chopping factory".
The sad thing about most of the experiments you list is that we gained a lot of useful knowledge from them. Sure it's easy to now look back and be horrified by them because we have the luxury of being able to do so, a luxury that is (in part) derived from the knowledge those experiments give us.
And yes, I think 100 years from now people WILL look back at us and be horrified about how do experiment on animals. But they'll be able to do so because they can afford that luxury. A luxury that's being built now, using those horrible methods.
I envision a future when we can grab a handful of organs on a chip, then decide on what kind of experiment we want to run. Want to see if something affects the cells in bone marrow? Bone chip. Want to understand if a drug is cleared exclusively through the kidney? Bone chip + kidney chip vs bone chip + kidney + liver chip. Want to test subcutaneous vs intravenous delivery kinetics? Put together a bunch of chips: bone, kidney, liver, gut, and skin.
Hell, if you're going that way, let's go the whole nine yards and grow mindless clone bodies for testing and transplants. It'd still probably be more expensive and time-consuming than animal experimentation, but it would allow for "human-like" testing faster than we can do today.
I'm pretty sure "the future" is going to be based on time, money and technology. The organ-on-a-chip thing is just one of many possibilities. But until it's cheaper and more efficient than the cost of feeding and keeping rats, keep on dreaming.
Science is a process and not a state. It doesn't need to be perfect, as long as you take slightly more steps forward than backwards the extent of knowledge progresses.
I think virtually no one likes experimenting on animals and scientists are acutely aware of all sorts of problems with these systems, but what else should humankind do? just twiddle our thumbs until better alternatives come around? The better alternatives also often come out of this type of research.
I generally agree with your criticisms, but what alternative do you propose? Of course we should have better vetting of studies before publication, more replication effort, and ideally a process that doesn't incentivize fraud/incompetence. But how to implement such a system? Also, I don't see how these issues are limited to animal studies. Your (entirely valid) criticisms of bias, uncaught error, and irreproducability apply much more broadly, to the vast majority of published modern experimental science, again raising the question of how we can address those issues so that the scientific literature is reliable.
My friend took a look at their setup. During the day there was a machine that was turned on next to the unhealthy mice. It vibrated and emanated a faint noise. He told them to turn off the machine. In a week, the mice were back to normal.
Nobody had thought about it, but mice are much more sensitive to sound and vibration than humans and it was stressing them out so much they were going crazy. There's probably lots of other things that can be overlooked (like temperature), but it's hard for us to notice because we aren't mice.