Home > Current, Discourse, Information, Science > Why Are We Conducting Obviously Flawed Science?

Why Are We Conducting Obviously Flawed Science?

October 16, 2013 Leave a comment Go to comments

A study making headlines today purports to conclude that Oreo cookies are “just as addictive as cocaine.” If a scientific study showed that a popular snack food had the addictive properties of a narcotic substance, popular press headlines would be appropriate. The study in question plainly does not support that conclusion, however.

The researchers conducted the study as follows:

On one side of a maze, they would give hungry rats Oreos and on the other, they would give them a control – in this case, rice cakes. . . . Then, they would give the rats the option of spending time on either side of the maze and measure how long they would spend on the side where they were typically fed Oreos.
. . .

They compared the results of the Oreo and rice cake test with results from rats that were given an injection of cocaine or morphine, known addictive substances, on one side of the maze and a shot of saline on the other. Professor Schroeder is licensed by the U.S. Drug Enforcement Administration to purchase and use controlled substances for research.

The research showed the rats conditioned with Oreos spent as much time on the “drug” side of the maze as the rats conditioned with cocaine or morphine.

From these two independent tests, it only seems possible to draw two independent conclusions: 1) rats like Oreos more than rice cakes, and 2) rats like cocaine or morphine more than saline. Plainly, because the testing did not directly compare Oreos and cocaine, it would be inappropriate to draw a conclusion that directly compares Oreos and cocaine.

From these two, independent tests, we do not know whether rats prefer Oreos in equal measure, for example, to cocaine. One seemingly easy way to find out would have been to ask them directly to choose between Oreos and cocaine, and it is strange that the researchers did not conduct such a test.

The testing conducted also appears to conflate preferentiality with addictiveness. Establishing that “hungry rats” consistently prefer one type of food over another does not necessarily mean that they are addicted to the preferred food option. The addictive force in a person would seem to be stronger than and fundamentally different from a mere preferential force; indeed, the power of addiction is that it can compel a being to act against its preferences in order to serve the addiction.

All we know from this research is that hungry rats would rather eat Oreos than rice cakes, not that the Oreos were “addicting” in a non-colloquial sense. A behavioral test for Oreos’ addictive properties might be whether rats choose Oreos over other, equally or more desirable food, or whether they eat Oreos even when they are not hungry, or otherwise consume Oreos to their detriment.

Addiction surely has a neurological component as well, but again, the difference between preference (or pleasure) and addiction (or need) would seem to be important. In follow-up research, one of the student-researchers conducted some neurological testing:

They used immunohistochemistry to measure the expression of a protein called c-Fos, a marker of neuronal activation, in the nucleus accumbens, or the brain’s “pleasure center.”

“It basically tells us how many cells were turned on in a specific region of the brain in response to the drugs or Oreos,” said Schroeder.

They found that the Oreos activated significantly more neurons than cocaine or morphine.

“This correlated well with our behavioral results and lends support to the hypothesis that high-fat/ high-sugar foods are addictive,” said Schroeder.

That we derive more pleasure from consuming Oreos than from consuming cocaine or morphine is interesting, but it does not necessarily mean that consuming Oreos creates the pervasive neurological shift that constitutes addiction. (This is probably why the researchers only describe a “correlat[ion]” on this point.)

As someone without formal neuroscience training, my assessment of this study and the conclusions drawn from it certainly may be incorrect, but my criticism seems obvious, appropriate, and easily addressed (and remedied, if necessary). I do not mean to suggest that this Connecticut College group is the only scientific research team susceptible to this critique, as the popular science news contains plenty of examples. Maybe something that seems obvious– Why not compare Oreos and cocaine directly?– to a lay reader like me would never occur to the trained researchers because it is not a scientifically relevant inquiry. If the scientific community wants to present its work to a popular audience, however, it should shed the thin veneer of social justice concerns, which the Connecticut College group attempted to apply, and focus on addressing that audience’s natural curiosities, which are particularly likely to arise in response to sensational headlines like “Oreos as ‘addictive as cocaine.’

  1. October 16, 2013 at 7:00 pm

    Agree of course. I should also point out that reading this made me want to acquire oreos for my own consumption.

    My understanding was that sugar and/or fat has already been shown to be addictive (resulting in quantifiable and permanent changes in the brain). Therefore all one should have to do with oreos is read the back, realize that they are mostly fat and sugar, and conclude that they are addictive.

    As for the greater point (the title) I think this is a disappointing experiment. Was their intention to probe every snack food? Every food? And if so they will need something more quantitative than “prefer to rice cakes” because I suspect that mice will end up preferring almost every sugary fatty snack food to rice cakes.

    My second point on the title has to do with what their goal was with this research. If it was to confirm that fatty, sugary foods are addictive then, well, they sort of kind of did that for one case (although you could imagine that oreos have some super special combination of regular stuff that is particularly addictive. I doubt it unless it has to do with crunchiness and squishiness, but I suspect that the mice are fed some kind of ground up stuff anyways). If it was to test oreos in particular then a selection rationale should be listed as to why the researchers believe that they are more problematic than other things. My problem is that it looks like they began this experiment without any thought as to what might happen with the outcome. It seems clear to a layperson that rats are going to target the sugary, fatty food. This is what they did. This is an effect that has already been (somewhat) quantified in actual humans. I don’t see this experiment as being “flawed” as you say, rather as irrelevant.

    • AD
      October 17, 2013 at 8:55 am

      Thanks for reading and commenting.

      My understanding was that sugar and/or fat has already been shown to be addictive (resulting in quantifiable and permanent changes in the brain).

      That definitely sounds possible. It doesn’t help this study at all, but it’s believable.

      My second point on the title has to do with what their goal was with this research. If it was to confirm that fatty, sugary foods are addictive then, well, they sort of kind of did that for one case . . . .

      If we already know that the ingredients are addictive, what would this study add? As you mentioned in your comment, a review of the ingredients should be all that is necessary. Furthermore, as I wrote above, I think the most this study demonstrates is preference. I’m not convinced it can be used to prove addiction.

      I don’t see this experiment as being “flawed” as you say, rather as irrelevant.

      I think irrelevance counts as a flaw. Semantics aside, though, I think the main flaw is the first point I raised: the researchers’ failure to compare Oreos and cocaine. Failure to address an obvious element or test an obviously relevant relationship seems like a flaw to me, and it also seems to me like this sort of lapse happens pretty often, at least so far as popularly reported science goes.

  2. Chaz Michaleson
    October 16, 2013 at 8:36 pm

    A few hair-splitting comments about the larger issue here. First, science is inherently flawed – that is, no scientific experiment is perfect and there can be a good deal to be gained from findings from “flawed” science. There is a saying that no data is still data (cue uproarious laughter the from researchers in the room).

    This appears to be a worthy experiment, even if the methods and reporting of the findings could obviously be improved. Conn College is not a research university so anything coming out of there, save for perhaps an op-ed, should be taken with a grain of salt. Frankly, this is the type of research that comes from an undergrad at a school not really focused on research. That said, the professor should have had a bit more control over himself or the Coms office. And, while I share your view on the social commentary within this study, action research has its merits, and should not be overlooked.

    The release notes that the findings will be presented at The Society for Neuroscience conference, I imagine that the results will be reported much differently for that presentation – researchers usually don’t put up with this sort of thing.

    What I find troubling, and perpetuated by stories like this from Conn College, is the general distrust/lack of understanding among the general public about scientific research. Is there any other field where individuals who spend their entire life dedicated to education, study, and research are as frequently challenged by lay-people who lack those same years of education/experience? Maybe sports announcers of meteorologists? While there is plenty of biased research, it never ceases to amaze me how little trust there is in science and researchers. Perhaps its deserved, perhaps not.

    Thought provoking post.

    • AD
      October 17, 2013 at 9:27 am

      Thanks for stopping by to offer some thoughtful comments, Chaz. (Seriously, blink twice if you’re Don Rumsfeld.)

      I agree that there are some inherent limitations to science from practical/methodological and theoretical/conceptual standpoints. I’m hoping to use this post as a bit of a setup for a future one along those deeper lines.

      Conn Col is no main-campus UCONN, that much is clear, but I think that if you’re going to do science, you ought to do it rigorously or not at all, and if you want to be written up in major online and offline spaces (or present at major conferences hosted by groups like The Society for Neuroscience), the burden is on you to cover your bases. I don’t mean to diminish the social justice aspects of scientific research at all. I think areas like environmental justice and consumer economics are very important. Here, though, I thought it was raised in a trite manner, and without any apparent input from or claim to authoritative economic scholarship. At best, a missed opportunity.

      I’ll push back on your last main paragraph a little bit to suggest that what you see as “general distrust” is, across the population, more of a genuine curiosity. When science meets the political (i.e., broadly, when it calls into question someone’s financial interests), affected stakeholders are more likely to react irrationally negatively, but I don’t think that’s the general stance toward science, and I think part of it comes from a readily confessed lack of understanding. As to whether “there any other field where individuals who spend their entire life dedicated to education, study, and research are as frequently challenged by lay-people who lack those same years of education/experience,” here comes my own bias, but I don’t know of any licensed profession besides lawyers that have their own section in the joke-book library.

      I do think that the scientific community has suffered damage as a result of its entrance into the politics of climate science. Maybe that’s a special case, whether due to the political contentiousness of the subject or the existence of a vocal minority opposition, but I do think that the science community places little value on making efforts to communicate about their work and findings in a way the public can understand, and I think they pay a price for that from time to time, both in terms of lay perception and funding.

      Thanks again for reading and commenting.

      • Chaz Michaleson
        October 18, 2013 at 10:31 am

        Very valid points and follow up!

  3. AD
    January 25, 2014 at 1:54 pm

    Note also that this criticism extends to the social sciences, as illustrated by this response to a recent “study” of employment litigation outcomes. There, the answer to the above post’s question, Why?, is more readily apparent. The person conducting the study is a lawyer who represents employees in employment litigation. As such, she has a financial interest in presenting her (flawed) critique of the judicial landscape.

  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: