(But surely that shouldn’t be a struggle for such hyper-intelligent pan-dimensional beings…)
The first law of thermodynamics holds that
energy cannot be created or destroyed, merely changed from one form to another.
This is one of the most basic principles (so far as we know) of the physical universe. It’s why we can’t have a perpetual motion machine. It’s why a coal-fired power station may make electricity, but it doesn’t make energy – instead it changes the potential energy stored in coal into thermal energy (heat), thermal energy into kinetic energy (movement), and finally converts kinetic energy into electrical energy.
Looked at from a thermodynamic perspective, animals are machines for converting energy from one form to another, and so are not all that different to power stations. We consume energy in the form of food, then change it into other types of energy. Some animals, like humans, are also very good at storing energy, so if we consume more than we expend we store it within our bodies, mainly in the form of fatty tissue. However, if we expend more energy than we consume, that extra energy has to come from somewhere – because, as the first law of thermodynamics tells us, energy cannot be created. The excess energy comes, usually, from metabolising the fatty tissues we have stored up previously.
This is what gives rise to the diet-and-exercise approach to weight loss for lardy-arsed people like me. Reducing energy intake while simultaneously increasing energy expenditure means – thanks to the first law – that my body is forced to metabolise the wobbly protuberance I have in place of rock hard abs, and this means I lose weight. This in turn means that I become (by the current standards of our society) massively more attractive, acquire a radiantly beautiful and disgustingly rich boyfriend, and sail off into a blissful future. Or at least, that’s the theory. To be honest, I tend to live my life according to an alternative hypothesis – a chocolate bar in the hand is worth almost any number of hypothetical blowjobs – so I can’t offer personal testimony on whether it actually works or not.
This – eat less, do more – is a very simple idea to grasp, but it’s also hard, and slow, and boring to carry out in real life, which is why people like me tend to lose interest in diet and exercise long before we get to the stage where blowjob frequency starts to increase, let alone the bit where beautiful, rich guys start pledging their undying love. It’s also why all sorts of faddish diets come and go. Most of them, when it comes right down to it, involve drastically reducing food intake, sometimes by recommending that you cut out entire food groups, and are so radical they’re impossible to stick to long term.
Because of this, everyone is looking for the special, magic thing that will make weight loss easy. For a good while now, one of the special magic things that has been mentioned fairly often is the idea that weight-gain is less to do with what you eat than it is when you eat. The theory is that people who eat later in the day tend to put on more weight. This is, of course, a deeply seductive idea – it suggests you can have all the chocolate you can eat of a morning and the radiantly beautiful blowjob guy – so I’ve always been inclined to treat it with some caution. As the old cliché has it: if something seems to be too good to be true, then it probably is.
Specifically, I’ve always approached this idea by thinking about it in terms of the first law of thermodynamics. Assuming the same level of activity (i.e. identical energy expenditure) and the same level of eating (i.e. identical energy input), the net gain of body fat must be identical whenever you eat. Otherwise, where is the extra energy to make the extra body fat coming from? (If you’re thinking ‘efficiency’, please bear with me – I’ll be getting to that later.)
I’ve always assumed that the anecdotal accounts from people convinced that they put on more weight when they eat later in the day must actually be explainable by some other factor. Perhaps people who eat late in the evening get hungrier through the day, snack more, and as a result actually consume more calories as a whole. Perhaps, because their meal has been delayed, people who eat late in the day have emptier stomachs and lower blood-sugar levels, which means they feel hungrier, and so eat more. (This is the theory that lies behind the diet suggestion to eat little, often – your stomach is never completely empty, so you’re less likely to eat too much at any one sitting.) Perhaps, because it’s late in the day, people are more tired when they’re preparing food, and so are more likely to opt for low-effort, high-calorie ‘convenience’ foods. Perhaps being hungrier through the bulk of the day makes people feel less energetic, which makes them less active, and so they expend less energy.
Inasmuch as I thought about this at all – and I’m not going to claim it was ever a major preoccupation – I assumed that one of these factors, or a combination of them, was the cause of the reported difference in weight gain. But now it seems, according to the BBC, that my scepticism was misplaced, and that in fact ‘Eating late at night adds weight’.
The gist of the BBC article is that some researchers from Northwestern University, Illinois were interested in why shift workers tended to be overweight. They speculated that the time of day at which the shift workers ate their meals might be partially responsible. In order to test this hypothesis, the researchers designed an experiment involving mice. (The abstract of the research paper is available here – I have based some details in this summary on the abstract rather than the BBC story, which is a little vague in places.)
The mice were divided into two groups, both of which were fed the same high-fat diet, and both of which engaged in the same level of activity. One group was fed at times when the mice would normally be asleep, while the other was fed at times when they would normally be awake. It was found that the mice in the group fed while they should be asleep gained twice as much weight as those that were fed when they should be awake. From this the researchers have drawn the conclusion that a better understanding of the impact of the circadian system on weight gain might be of use in combating the obesity epidemic.
I guess the first thing I want to say here is that, according to the abstract, the researchers do not seem to make the conclusions that the BBC article suggests they do. The BBC article opens by saying
Late-night snackers are more likely to gain weight, research suggests.
A team from Northwestern University, Illinois, found that when you eat, not just how you eat, could make a big difference.
The story is illustrated by a photograph which shows a woman in a dressing gown standing in front of an open fridge in a darkened room, and stuffing food into her mouth. Taken together, the opening sentence of the article and the photograph create the distinct impression that the research looked at ‘late-night snackers’ – that is, people who eat their main meals at ordinary times, but then consume additional food late at night – and found that this activity is associated with an enhanced risk of gaining weight. What the research actually demonstrated is that mice who only eat when they should be asleep put on more weight than mice who eat when they should be awake. These mice weren’t the equivalent of ‘late-night snackers’, they were the equivalent of shift-workers, which shouldn’t come as a surprise, given that it was shift-workers the researchers were hoping to investigate. It’s hard to escape the thought that presenting the story as if it applied to late-night snackers (of whom there are many) rather than shift-workers (of whom there are rather less) may have had something to do with creating a more high-impact story.
I should say that I don’t hold the BBC journalist exclusively responsible for this. In fact, I am struck by the remarkable similarity between the BBC story and a press-release written by Megan Fellman, the Science and Engineering editor of Northwestern University’s impressively large and multidisciplinary Media Relations team. The press-release – which is written and formatted exactly like a news report, presumably to make churnalism just that little bit easier – appears under the headline ‘Late-Night Snacks: Worse Than You Think’, and begins like this:
Eat less, exercise more. Now there is new evidence to support adding another “must” to the weight-loss mantra: eat at the right time of day.
I should perhaps note that, in addition to the same apparently mistaken assertion regarding the applicability of the research to late-night snacking, Fellman’s piece also contains all of the quotations from the researchers used in the BBC article. Oh, dear, it looks like we’re in cut-and-paste territory again.
There are differences between the two articles, however. The BBC’s headline is less idiotic than the one produced by the university representative – Fellman doesn’t know how bad I think late-night snacks are (hey, maybe I think they cause 95% of all cancers),* so how can she tell me that they’re worse than I thought? The BBC have also sourced a quote from a representative of a UK-based special interest group, who helpfully provides the implied criticism of scientists that’s a standard feature of these types of BBC story:
Tam Fry, from the National Obesity Forum, […] said: “It is groundbreaking. It really gets you thinking why this has not been done before.”
Another difference is that the BBC add this disclaimer towards the end of their piece:
At this stage, the results could still be interpreted as controversial when applied to humans.
On one hand, the inclusion of this sentence is rather heart-warming – it suggests that the BBC journalist is at least aware that extrapolating from results in mice to possible results in humans is problematic. On the other hand, I don’t really know what the phrase ‘interpreted as controversial’ means. I know what it implies – that applying the results to humans isn’t really controversial, but that some people might interpret it that way. It also strikes me that the question of whether or not the findings from this research apply to humans isn’t so much controversial as it is unanswered. As things stand, no-one can say whether this research applies to humans or not, just that it may do. Mice are usually pretty good models for humans (hence why they’re used a lot in medical research), but until the results have been replicated in humans, it’s not definite.
Of course, as with any research, there are some residual question-marks hanging over it until it has been replicated by another group of researchers. This is always the case, but it seems to me that it applies here with particular force because of the extraordinarily dramatic nature of the findings. Mice that could only eat when they should have been sleeping didn’t just put on a bit more weight than their counterparts who ate at usual times, they put on more than twice as much (48% above baseline versus 20% above baseline).** That’s really a lot of extra weight. We also need to remember that the extra weight is really extra energy – it represents the difference between the energy the mice were expending and the energy they were consuming.
Now, we’re told that both groups of mice – the slimmer, normal-mealtimes ones and the fatter, sleepy-eating ones – had an intake and expenditure of energy that were similar enough for the differences not to be statistically significant. So here’s the thing – despite eating and exercising about as much as the mice who followed a normal pattern, the abnormal mice somehow found enough extra energy to put on twice as much weight. We know from the first law of thermodynamics that the mice can’t just have conjured this extra energy out of thin air, so the only possible explanation, at least so far as I can see, is greater efficiency. The mice who ate when they should have been sleeping must either have been more efficient at turning food into energy in the first place (so the same amount of food resulted in a greater excess of energy), or more efficient at turning the same amount of excess energy into fatty tissue (so the same amount of extra energy resulted in fatter mice), or both.
The trouble with this is that I find it a very odd idea (particularly given that the mice were not just a bit more efficient at energy conversion, but substantially more efficient.) For a start it seems fairly counter-intuitive. I’ve watched science documentaries which have shown that all kinds of human functioning are less efficient when people are struggling to fight against their circadian rhythm – that is to be awake when they should be asleep, and vice versa. Yet this experiment showed that mice were more efficient at processing food when they should have been sleeping than they were at the times when they should have been awake. This is pretty much the opposite of what I would expect, which is why I say it’s counter-intuitive.
The other major problem I have with the idea is – why? Why would mice (and, potentially, humans) be better at processing food at a time when they wouldn’t normally have to process it? Presumably, in order for the trait to have survived and prospered, it must have offered some kind of evolutionary advantage (or, at the very least, no significant disadvantage), but I’ve struggled to think what that might be.
To start with I wondered if eating at a time when one should be sleeping might be something that, in nature, would be associated with an abundance of food. If so, then I could see there might be an advantage in storing the energy very efficiently in order to make sure of a good store of body-fat for the times when food was less abundant. The problem with this hypothesis is that there would also be a clear advantage in processing food very efficiently when it’s in short-supply. So then I wondered if the opposite was true – might eating at a time when one should be asleep have been associated with a scarcity of food? I could see that if an animal were on the verge of starvation it might search for food even when it should be asleep, and if this were the case then it would certainly make sense for any food consumed in this way to be very efficiently processed. The problem with this hypothesis is that, if an animal is on the verge of starvation, then there’s a clear advantage for it to process food at peak efficiency, even if it eats during normal waking hours.
In fact, this seems to be the major stumbling block with thinking about this trait in evolutionary terms. It seems fairly clear that humans (I don’t know about mice, but I would guess it’s much the same) evolved to deal better with conditions in which food was scarce rather than abundant. Humans will carry on converting excess energy to fat, even when they are so obese that adding extra fat impacts on their ability to perform physical activity, and substantially increases their risk of ill-health or death. This is clearly counterproductive in evolutionary terms – you would expect evolution to select for traits which diminish the chances of illness and premature death – which seems to suggest that humans (and our precursor species) did not often find themselves in circumstances in which there was an over-abundance of food. If humans evolved to deal with circumstances in which scarcity of food was a more significant problem than an over-abundance of food, then this in turn suggests that there would have been a distinct evolutionary advantage in processing food efficiently whenever it was consumed.
As a result of all this, I find myself wondering if, perhaps, there might be some other factor that would explain these results. I wonder about the statistical methods used to analyse the data (these aren’t reported in the abstract), particularly in light of the assertion that the differences in energy intake and activity between the two groups of mice were not statistically significant. I wonder about how the activity of the mice was assessed (this is also not reported in the abstract), and particularly whether a distinction was made between high- and low- intensity activity, since the former would be associated with higher energy expenditure. I wonder why both groups of mice were allowed to eat as much as they wanted, rather than being assigned a particular caloric intake – an experiment in which time of eating was the only variable would produce more robust results, after all. Above all, I find myself wondering if the results from this research will be replicated, or if they will turn out to have been some kind of anomaly or fluke
It is possible, of course, that they will be replicated. It’s possible that we stand on the threshold of a much deeper understanding of the causes of obesity, in which food consumption and physical activity are seen to be only part of the problem. But it doesn’t seem to me that we’re quite there yet. It seems instead like we’ve had some intriguing and thought-provoking results in mice presented as though they offer dramatic insights into human problems. It also seems like the results we do have suggest either that, for reasons it’s not immediately easy to understand, mice are substantially more efficient at processing food they eat when they should be sleeping, or, alternatively, that mice have found a way of overcoming the first law of thermodynamics.
Neither seems entirely plausible to me, so I think I’ll stick to the ‘eat less, do more’ approach to weight loss for the time being. Or at least I will, just as soon as I’ve started to care about the fact that I’m a little bit tubby…
* – I don’t.
** – I agree, percentage above baseline isn’t a very informative way of expressing the change. Unfortunately, this is the way it’s presented in the press release, and I don’t have access to the full paper, so I can’t get at the raw figures.
[Disclaimer: in this post I have commented on scientific matters, which I am far from expert on – my highest scientific qualification is a D at A-level biology. Just to be clear: I am not claiming to be an expert, just an interested and (I hope) reasonably well-informed layman. Following a previous little contretemps, I’ve decided to add this disclaimer to all the posts where I venture into the scientific domain.]