Posts Tagged ‘Playing Ben Goldacre’

Shedding light on astrological nonsense

Dec 17th, 2010

Last night I was directed towards an article claiming that scientists had inadvertently found evidence to support astrology:

Principle of astrology proven to be scientific: planetary position imprints biological clocks of mammals

(NaturalNews) Mention the word “astrology” and skeptics go into an epileptic fit. The idea that someone’s personality could be imprinted at birth according to the position of the sun, moon and planets has long been derided as “quackery” by the so-called “scientific” community which resists any notion based on holistic connections between individuals and the cosmos.

Skeptics must be further bewildered by the new research published in Nature Neuroscience and conducted at Vanderbilt University which unintentionally provides scientific support for the fundamental principle of astrology — namely, that the position of the planets at your time of birth influences your personality.

As a member of the “so-called “scientific” community” I admit that I immediately suspected quackery, or woeful misinterpretation on the part of the astrologers writing the article. However, were I to write off their claims entirely based on this suspicion, then I would be guilty of the close mindedness that they accuse me of. So, in order to dissect their claims, I had to go back to the primary source, the paper they were referencing1 .

What do the astrologers claim?

Before we take a look at the original paper, lets have a look at what the astrologers claim. I’ll only concern myself with the claims directly relevant to the paper they are discussing; their claims regarding scientific arrogance, their straw-man scientists, and cold fusion can be left for someone else to tackle.

The first claim is that “according to the conventional view, your genes and your parenting determine your personality, and the position of planet Earth at the time of your birth has nothing to do with it.” This attempt to set out the conventional scientific viewpoint ignores the large number of other environmental effects that scientists accept can influence development. These environmental effects can be influenced indirectly by the position of the earth, as it is the movement of the earth that is responsible for influencing the seasons.

The astrologers then make the following claims regarding the findings of Ciarleglio et al.:

  1. That the position of the planets at the time of your birth influences your personality.
  2. That mice born in the winter show a “consistent slowing of their daytime activity”
  3. That mice born in the winter are more susceptible to Seasonal Affective Disorder (SAD).

They then summarise by claiming that “one of the core principles of astrology [is] that the position of the planets at the time of your birth (which might be called the “season” of your birth) can actually result in changes in your brain physiology which impact lifelong behavior.”

So, does the paper actually support the claims they make of it?

The Paper

Unfortunately the writers of the article don’t actually provide a link to the original paper, something which is sadly all too common . However, they do provide enough information to let us find the paper:

Perinatal photoperiod imprints the circadian clock

Christopher M Ciarleglio, John C Axley, Benjamin R Strauss, Karen L Gamble & Douglas G McMahon
Nature Neuroscience (2010) doi:10.1038/nn.2699
Received 24 August 2010 Accepted 21 October 2010 Published online 05 December 2010

Using real-time gene expression imaging and behavioral analysis, we found that the perinatal photoperiod has lasting effects on the circadian rhythms expressed by clock neurons as well as on mouse behavior, and sets the responsiveness of the biological clock to subsequent changes in photoperiod. These developmental gene × environment interactions tune circadian clock responses to subsequent seasonal photoperiods and may contribute to the influence of season on neurobehavioral disorders in humans.

As far as I can tell, the paper is open access, which means that anyone should be able to read it. I’m afraid the same can’t be said for all the papers I’ve referenced here.

What do they do?

The Circadian Rhythm

That nature responds to the cycle of day and night is easily observable, flowers track the sun, people and other animals sleep and awake, and other behaviour seems to correlate with the time of day. What is perhaps more surprising, is many of these behaviours continue to show a daily cycle, even if the sun is removed from the equation. For example, flowers will continue to track the position of the sun for several days after being placed in a dark room. Typically the cycles shown during these ‘free-running periods’ (either of constant light, or constant dark) are either slightly longer or slightly shorter than twenty-four hours, and the regular light-dark cycle is necessary to maintain calibration.

The impact of circadian rhythms on human behaviours should be apparent to anyone who has ever experienced jet lag, and the depression that can follow the shorter winter days (SAD), has been linked to the circadian cycles. In humans and other mammals, this internal biological clock has been tracked down to a region of the brain known as the suprachiasmatic nuclei (SCN), in which the expression of a number of genes, or associated hormone production, vary on an approximately twenty-four hour cycle. These cycles continue, even when exposed to constant light, or constant dark, thus providing an internal ‘body clock.’

The ‘clock genes’ are those whose expression (the level at which the cell makes protein from a given gene) varies over the cycle. For example, the gene per1 is most active during the day. By genetic manipulation, is is possible to ensure that per1 produces a product with a fluorescent tag, allowing its activity to be measured by monitoring cell florescence.

Mice were exposed to either short-day or long-day light cycles (8 or 16 hours of ‘daylight’ respectively) from before birth until weaning (The perinatal phase). Following weaning, mice were either maintained under the same light cycle, or were stitched to the opposite cycle for a further 28 days (The continuation phase, Figure 1). Experiments were repeated in both winter and summer, to take into account variations that may be induced by the actual season.

Figure 1: Mice were expose to either long or short day light cycles prior to weaning. Subsequently mice were either maintained under the same conditions, or were switched to the opposite cycle.

Following this treatment, the circadian rhythms of the rodents were studied, either by tracking the expression of specific ‘clock genes,’ or by monitoring rodent behavioural activity2.

What do they find?

The authors found that the ‘day length’ at birth continues to have an impact on the circadian rhythms of the mice, even after they have been shifted to a different cycle. In particular, long-day born mice, showed a narrower peak of activity in the SCN, and in addition showed a shorter internal clock cycle, this resulted in more consistency between subsequent exposure to either long or short day cycles. In contrast, short-day born mice generally showed wider peaks of activity, and longer internal clock cycles. They were also subject to more variation in circadian rhythm behaviour between subsequent ‘summer’ or ‘winter’ cycles.

What does this mean?

This indicates that the day length at birth may have long term effect on subsequent circadian rhythms, even when the day length changes. In particular, mice born during a ‘winter’ cycle seem to show more variation in subsequent seasonal changes. The authors indicate that this may have implications with respect to previous findings, which show that mice born during winter-like cycles show increased levels of depression like symptoms, and that winter born humans show increased levels of SAD, bipolar disorder and schizophrenia. However at this stage it is not possible to draw a firm link between the findings.

What are the limitations of the study?

One of the primary limitations of the study is that it only addresses a short time span, and the authors acknowledge that they have no indication as to how long the effects may last. This may be especially relevant when attempting to translate the results to humans: not only a different species, but one with a significantly longer lifespan.

Additionally, I am also concerned that the study only considered two phases, the perinatal phase, and then either a continuation of the initial levels, or a switch to the opposite seasonal cycle. The possibility remains that exposure to a long-day cycle outside the perinatal period is sufficient to, at least partially, imprint the same pattern as seen in the long-day born mice. In support of this possibility is the considerably lower variation between short-day and long-day born mice under long-day conditions. However it should be pointed out that there are still significant differences between shot-day and long-day born mice.

Does the paper support the claims of the original article?

In short no.

Firstly, the only planet whose position has effect on day length is planet earth, where the orientation of its axis relative to the sun determines the season, and hence the day length. The other planets have different orbital periods, which means that Mars won’t be at the same point in the sky on December 17th 2010, as it was on December 17th 2000. Furthermore, while the northern hemisphere is in the grip of winter, the southern hemisphere is in the middle of summer, meaning that any seasonal effects are going to depend on the hemisphere, as much as the location of the earth.

But importantly, the researchers have been careful to isolate day length as the variable, separating other seasonal effects, such as temperature, or indeed the earth’s location. Furthermore, by conducting the experiment in summer AND winter, the effectively prevent any actual seasonal variations from having an effect.

In short, the findings have nothing to do with astrology. Ironically, the ‘tabloid horoscopes’ derided in the article have more to gain; your star sign is correlated with day length when you were born, at least assuming we confine ourselves to a narrow latitude.

Oddly the article also ascribes the findings regarding depression like symptoms in winter-born mice to this study, when in fact they were conducted separately. But again, these findings were purely associated with day length, not planetary position.

So in summary we have either a gross misrepresentation of scientific findings, or a major misunderstanding. We also have a paper which is potentially interesting, although I’d like to see how stable the effects are through multiple changes in day-length. If the effect is isolated to a short period of possible imprinting, then it will be interesting to narrow this period down in different organisms.

  1. Ciarleglio, C. M., Axley, J. C., Strauss, B. R., Gamble, K. L., & McMahon, D. G. (2010). Perinatal photoperiod imprints the circadian clock. Nature neuroscience. doi:10.1038/nn.2699. []
  2. Wheel usage []

Bean had?

Jun 9th, 2009

You may have noticed reports in several newspapers and tech sites today of the Heinz Beanzawave, a USB powered microwave. In total, according to Google News, the story appears to have been covered by over fifty different outlets including the UPI, CrunchGear, Cnet and the Daily Mail. Not to mention a number of popular blogs, such as BoingBoing. While some of these articles have users raising criticisms in the comments, none of the news sources I checked bothered to run a critical eye over the story. (If you find one, please post it in the comments so the good guys can get some acknowledgements.)

The news seemingly originated from a press-release form the ‘Microwave Association,’ working in association with Heinz. Unfortunately for fans of beans and fancy USB gadgets the story appears to be, well, a load of old beans1.

Even a quick critical eye will spot some key flaws. The average modern microwave has an output of 850W, with even weaker models outputting at least 650W. By comparison, the USB 2.0 standard provides a maximum current of 500mA, at 5V; which works out as a maximum power output of just 2.5W2. The former figures should be familiar to anyone who has cooked something in the microwave, as the output is printed on the front, and you are expected to adjust cooking times accordingly. The latter meanwhile should be obvious in part to anyone who has hooked up a printer. While simple devices like mice can obtain sufficient power from a USB socket, the same can’t be said for more power-hungry devices such as printers, which need their own power supplies. Indeed, the power supply units of most computers aren’t certified to deliver 850W of power, with possibly only high quality gaming rigs fitted with a hefty enough PSU; even then, powering a microwave would leave scant remainder for the processor and graphics card.

So we’ve already seen that such a device would have to operate at a significantly lower power output than most modern microwaves. Such a device would be unlikely to be marketed as a novelty, and instead would be positioned to replace most standard microwaves. However is it possible that such improvements have been made, and we are in fact looking at vastly superior technology. In short, not likely, unless they have also managed to break the laws of thermodynamics in creating this miracle microwave.

In demonstrating how truly impossible their claims are we need to consider some physics.

The specific heat capacity of a substance is used to describe how much energy is required to raise its temperature; for water, this is given as approximately 4.2 joules per gram per kelvin. This means that for every gram of water, you must supply 4.2 joules of energy, to raise the temperature by 1 kelvin, or 1 degree Celcius3. Therefore, for a 200g tub of beans (which we shall approximate by assuming it is all water.) it takes 4.2 x 200 = 840 joules to raise the temperature by a single degree.

So how does this translate to our microwave? Well fortunately, the measurement watts tells us how many joules of energy are transferred each second. In other words 1 watt = 1 joule per second. So if we are to assume that our microwave is 100% efficient, that all energy it uses goes directly into heating the beans, we can discover how long it will take to deliver the required 840((And here is a good illustration of the inefficiency of standard microwaves. As by this calculation the beans should be ready in just under a second.)) joules to raise the temperature of the beans by a single degree. Simply divide 840 by 2.5 and… oh dear… 336 seconds. Five minutes, 36 seconds to warm the beans by a single degree. If you want them boiling hot4, then you are going to be waiting over 7 hours.

What bothers me most about this story isn’t the dubious nature of the original press-release. It was clearly constructed by marketing bods in an attempt to gain free column inches. In that respect it worked, and I’m only adding to the effect by writing this. What bothers me is the way the press regurgitated it, unthinkingly, unquestioningly, delivering advertisements as news. Not only this, but it is clear that in most cases, they didn’t even stop to pass a critical eye over it. This isn’t just churnalism, this is factually incorrect churnalism. When the media sacrifices its credibility in terms of fact checking, and ends up falling slave to marketing, what does it have left? And when we lose one of the key methods of fact distribution, of investigation and exposure, what do we have left? Blogs and citizen journalism go so far, but an effective and trustworthy media is is important for everyone; this story is only one of many that makes me wonder how much of one we have left.

I have already contacted the Microwave Association in the E-mail provided in the press-release, and have invited them to respond. I’ll update this entry as soon if I hear anything from them, and leave the comments open if they wish to contribute there. (Although I encourage you do do so via my E-mail, as that way I can be sure the response is genuine. Also, please bear in mind that comments from new users will be held for moderation, and may not get published immediately.)

  1. See, you don’t have to be part of the tabloid press to make terrible puns []
  2. Power is the product of current and voltage []
  3. The scale of Celcius and kelvin is identical. Only the position of the zero position changes, with 0 kelvin being equal to -273.15 °C []
  4. Assuming room temperature 20 °C and a boiling temperature 100 °C []