We at bayblab have made the case for a distributed web of amateur scientist working together to do real science. Think of it as blogging is to the written press, podcasting to radio, or youtube to television: user-generated science. We also warned in our podcast (episode 15), that it comes with a great risk of animal cruelty, and limited access to proper equipment to get the job done safely. This couldn't have been more prescient when it comes to this story of Victoria Iness-Brown who decided to test aspartame on 108 rats for over 2 years. Any scientist looking at this experiment can clearly see the merit. The study was surprisingly well thought out. The experimenter used mixed strain and missed the point on why we use inbred for experiments such as this but otherwise the design was good. Yet a study like this could have never passed the ethics board in our university. The rats were given aspartame in their water throughout their lifetime, and allowed to carry out their lives. The problem is that there was precious little monitoring, no weighing of the rats, and more importantly, many rats developed tumours, which were left to ulcerate, with the obvious implications on the animal's well-being. The intentions were good, but the experimenter lacked the knowledge on how to reduce pain and discomfort on the animals. The other problem is that there are no results or statistical analysis. There are lots of pictures of rats with tumours, but no dissection, no measurement of tumour burden, no survival curves. How can we avoid rogue experiments like these but foster amateur science?
Friday, February 15, 2008
Amateur scientist: the aspartame experiment
We at bayblab have made the case for a distributed web of amateur scientist working together to do real science. Think of it as blogging is to the written press, podcasting to radio, or youtube to television: user-generated science. We also warned in our podcast (episode 15), that it comes with a great risk of animal cruelty, and limited access to proper equipment to get the job done safely. This couldn't have been more prescient when it comes to this story of Victoria Iness-Brown who decided to test aspartame on 108 rats for over 2 years. Any scientist looking at this experiment can clearly see the merit. The study was surprisingly well thought out. The experimenter used mixed strain and missed the point on why we use inbred for experiments such as this but otherwise the design was good. Yet a study like this could have never passed the ethics board in our university. The rats were given aspartame in their water throughout their lifetime, and allowed to carry out their lives. The problem is that there was precious little monitoring, no weighing of the rats, and more importantly, many rats developed tumours, which were left to ulcerate, with the obvious implications on the animal's well-being. The intentions were good, but the experimenter lacked the knowledge on how to reduce pain and discomfort on the animals. The other problem is that there are no results or statistical analysis. There are lots of pictures of rats with tumours, but no dissection, no measurement of tumour burden, no survival curves. How can we avoid rogue experiments like these but foster amateur science?
Subscribe to:
Post Comments (Atom)
Podcast
6 comments:
So, to summarize, she gave rats the human equivalent of 20 cans of diet soda a day (with nothing else to drink) and they didn't fare well? No shit.
And don't you think she would need yet another control? A sugar-control? Maybe the rats simply drank *more* aspartame laden water because it was sweeter. I think this is a major confound in addition to those brought up above. No analysis, no significance testing, and no IRB! I have issues with the strain of rats used, as well as lack of monitoring and lack of control for other factors. Rats consume more when they're stressed - what was the environment like? Were the rats handled daily, and equally? I don't think she teased out (or at least randomized) the confounds.
This is more an observational study of rats in an X-type of environment over anything remotely experimental, in my opinion.
I'm not backing aspartame (I do believe it causes tumors and such) but I am just being critical on her experimental design and procedure.
My 2-cents.
Actually a proper control is very difficult for this study. You need something which tastes sweet, but with no calories, and no side effects. Sugar will have broad consequences, think about caloric intake, diabetes etc... The best I can come up with is making it 3 groups: water / sugar / aspartame or changing the route of administration (iv/ip), which would be hard to do at home.
This experiment has made its way to both Reddit and Digg, and it's interested to read the comments there.
On the one hand it's great that people take an interest and talk about science, on the other hand it's a bit distressing to see what a lot of these people think passes as a well-executed and well-presented experiment.
That's not just an amateur experiment, it's a deliberate hack job. It's also a rather severe case of animal cruelty. I have done animal research, involving tumor development, and this would have never flown with the IRB. If those had been dogs or hamsters the woman would be in jail right now. I hope someone notifies the authorities so she doesn't try this again.
There's no way to stop people from doing stuff like this, we just have to try to educate people about what is proper science and what isn't.
I'm not surprised at all that the comments on digg are full of rampant stupidity. Isn't that where Stupid Filter got a lot if its seed corpus?
Post a Comment