Wednesday, May 4, 2011

Energetic theories of sleep function

A three part series on the function of sleep, considering three different types of models. I'm currently working on a thesis, and this is a snippet from that work.

One of the oldest proposals regarding the functional purpose of sleep is the role sleep plays in organizing behavior in a way to optimize energy expenditure; even a minor saving in energy expenditure could be preserved evolutionarily. At times optimal for action as determined by circadian factors and resource availability, waking function peaks (Mignot, 2008). This preserves energy while food would not be available, or the organism is not best suited for going about waking affairs, such as restricting hunting behavior to a time of day most optimal for vision. This would suggest, as observed, that sleep would have the strongest restrictive effect on animals most dependent upon these processes, while having a less restrictive effect on grazers and browsers (Hobson, 2005). With the brain as an incredibly energy-intensive organ, it would be a primary target of this process.

One point of benefit that these theories have is the link between circadian disruption and metabolic malfunction (Kohsaka & Bass, 2007). Further, mechanisms underlying feeding behavior have strong associations with the orexin system underlying the flip-flop switch (Saper, 2006). The primary proposed biochemical correlate in this model is adenosine, primarily released as a metabolite from glial cells (Benington & Heller, 1995). Adenosine has been found to increase after sleep deprivation, and serves as an inhibitory signal for cholinergic basal forebrain nuclei (McCarley & Massaquoi, 1992).

Unfortunately, this model leaves far too much unexplained. Although adenosine clearly serves a function in regulating sleep homeostasis in mammals, it’s far from clear how sleep energetically benefits the organism. The observations that local cell populations can enter synchronous firing indicative of sleep (Vyazovkiy et al., 2011) in addition to instances of unihemispheric sleep (Mukhametov et al., 1977) call into question any model that focuses too heavily on the functionality of sleep as being at the scale of the whole brain or whole organism. Also, regarding the diminished sleep of browsers and grazers, it still would seem unusual that sleep has not been totally eliminated in animals which would seemingly be suited for consuming throughout the 24-hour cycle.

Perhaps even more pressing are issues related to the energetics of sleep itself. Sleep is a very active process; although some energy is conserved during NREM sleep, some mammals have a much more efficient way of energy conservation: hibernation. Marked by a lowering of the core body temperature to a few degrees above ambient temperature, as well as incredibly slowed metabolic rate and dampened neural activity (Heller & Ruby, 2004), hibernation is an excellent tool for energy conservation. Were the primary function of sleep energetic in nature, it would much more closely resemble hibernation than it currently does (Mignot, 2008). Animals coming out of hibernation also experience sleep rebound, as if deprived of sleep during the time spent in hibernation (Heller & Ruby, 2004), which would not be expected if the two processes had overlapping functionality.

The model is also incredibly specific to endothermic organisms. While the model may serve as an explanation of the conservation of sleep among high-metabolism small mammals (Mignot, 2008), it does little to explain sleep in reptiles, and even less to explain observed sleep-like states observed in flies and flatworms (Hendricks et al., 2000; Shaw et al., 2000; Raizen et al., 2008).

Lastly, mammalian REM sleep again poses a particular problem for this model. REM sleep is marked by increased energetic demands, with wake-like neural activity and increased breathing and heart rate (Parmeggiani, 2003). Whatever function served by REM sleep, by this approach, would seem to be disconnected entirely from non-REM sleep.


Benington, J.H, & Heller, H.C. (1995) Restoration of brain energy metabolism as the function of sleep. Progress in Neurobiology 45: 347-360.

Heller, H.C. & Ruby, N.F. (2004) Sleep and circadian rhythms in mammalian torpor. Annual Reviews in Physiology 66: 275-289.

Hendricks, J.C., Finn, S.M., Panckeri, K.A., Chavkin, J., Williams, J.A., Sehgal, A., & Pack, A.I. (2000) Rest in Drosophila is a sleep-like state. Neuron 25:129-138.

Hobson, J.A. (2005) Sleep is of the brain, by the brain and for the brain. Nature 437:1254-1256.

Kohsaka, A. & Bass, J. (2007) A sense of time: How molecular clocks organize metabolism. Trends in Endocrinology and Metabolism 18: 4-11.

McCarley, R.W., & Massaquoi, S.G. (1992) Neurobiological structure of the revised limit cycle reciprocal interaction model of REM cycle control. Journal of Sleep Research 1: 132-137

Mignot, E. (2008) Why we sleep: The temporal organization of recovery. PLoS Biology 6(4): e106.

Mukhametov, L.M., Supin, A.Y., & Polyakova, I.G. (1977) Interhemispheric asymmetry of the electroencephalographic sleep patterns in dolphins. Brain Research 134:581-584.

Parmeggiani, P.L. (2003) Thermoregulation and sleep. Frontiers in Bioscience 8: s557-s567.

Raizen, D.M., Zimmerman, J.E., Maycock, M.H., Ta, U.D., You, Y.J., Sandaram, M.V., & Pack, A.I. (2008) Lethargus is a Caenorhabditis elegans sleep-like state. Nature 451:569-572.

Saper, C.B. (2006) Staying awake for dinner: Hypothalamic integration of sleep, feeding, and circadian rhythms. Progress in Brain Research 153: 243-252.

Shaw, P.J., Cirelli, C., Greenspan, R.J., Tononi, G. (2000) Correlates of sleep and waking in Drosophila melanogaster. Science 287:1834-1837.

Vyazovkiy, V.V., Olcese, U., Hanlon, E.C., Nir, Y., Cirelli, C., & Tononi, G. (2011) Local sleep in awake rats. Nature 472: 443-447

Saturday, March 5, 2011

In which we can learn a few lessons from philosophy professors

Firstly, I apologize for the lack of posting, for anyone that may be reading. I would like to hear from anyone reading, if at all possible; it's hard to have the will to keep up a blog if the only comments I get are link-spamming bots.

In any case, I'm going to talk (vaguely) about politics here. Specifically, I'm going to note where political discourse could be informed by philosophical literature. So be warned, this is not my typical fare, and I'm not going to draw any strong conclusions either.

There's been quite a bit of noise about budget reductions at the federal level. Republicans in particular are talking of 2010 as a 'mandate' to reduce federal spending and to cut certain programs. There's a rather sharply critical piece by Paul Krugman that tackles the silliness of this mandate talk directly by contrasting it with polling evidence suggesting that voters are split over which programs are over-funded, with none (except foreign aid) showing any clear support:

"The moral is clear. Republicans don’t have a mandate to cut spending; they have a mandate to repeal the laws of arithmetic.

How can voters be so ill informed? In their defense, bear in mind that they have jobs, children to raise, parents to take care of. They don’t have the time or the incentive to study the federal budget, let alone state budgets (which are by and large incomprehensible). So they rely on what they hear from seemingly authoritative figures."

Bruce Bartlett pointed out this problem in November of last year. His suggestion:

"We get the Congressional Budget Office to set up an official deficit simulation model... They [House representatives] can do the simulation any way they want, but at the end of the day ANY PROVISION GETTING MORE THAN 50 PERCENT IS DEEMED TO HAVE PASSED THE HOUSE." [Emphasis Barlett's]

Here's the problem faced: individuals in aggregate are not rational. What we are seeing is a breakdown in group agency. The idea is, approximately, that we are considering 'voters' (in Krugman's case) or 'the House' (in Bartlett's case) as an agent, having beliefs and taking actions. This is clear, especially given the talk of mandates. There exists a fair bit of philosophical work detailing this problem, actually; some of the more prominent voices on the topic are Christian List and Philip Pettit.

Groups, in this tradition, are defined as a collection of agents with a constitution of sorts (either implicit or explicit), any sort of rule-set governing the transition from individual agents to the group's agency. For example, in a majoritarian democratic group, the rule-set could be as simple as 'the belief given by the majority of members is that which is taken by the group'. The majoritarian group is the most often considered in this tradition, because the rule is rather simple to work with and still clearly displays some of the properties of groups that are important for consideration.

From this tradition, there is one clear result: groups are not rational in the same way that individuals can be. This manifests in results like Arrow's impossibility theorem, among many other things. The clearest example in the literature is given in List and Pettit's paper above, citing Kornhauser and Sager in 1986:

Suppose there is a jury which is functioning on a majoritarian rule-set, and they are to determine guilt in a breach of contract case.

The case comes down to three questions/propositions:

P: The defendant did action X.
Q: The defendant had a contractual obligation not to do action X.
R: The defendant is liable for a breach of contract.

In the case of an individual agent, it's clear that (P&Q) -> R, or in plain-speak that if one assumes P and Q, then R is the necessary outcome. However, given we're dealing with a group, an interesting problem can arise.

Suppose that we have three members on this jury, Amy, Beth, and Carl.

Amy and Beth believe P, that the defendant did action X. Therefore, the group 'belief' by a strict majoritarian sense is that P is true. Beth and Carl believe Q, that the defendant had a contractual obligation not to do X. That leaves Beth as the only person that believes R, that the defendant is liable for a breach of contract. So, as a group, the position is P & Q... but not R. By the language of the literature, the result is then, that group rationality does not exhibit proposition-wise supervenience, which is effectively a way of saying that the rationality of individual agents concerning the relations between propositions doesn't follow when mapped on to groups. (I would actually recommend reading List and Pettit's paper; it's written in fairly plain language, even for the non-philosopher.)

The problem arises more often than one would think. Remember McDonald v. Chicago, the Supreme Court case dealing with Chicago's handgun prohibition? There's an interesting note to the ruling in the case. Four court members, lead by Justice Alito, decided that the 2nd Amendment was incorporated by the Due Process clause of the 14th Amendment. Justice Thomas concurred, but in a different opinion, decided its incorporation through the Privileges or Immunities clause of the 14th Amendment. In practical terms, this means that there was a majority court decision that the 2nd Amendment applies to state governments. However, there is no majority concurrence on a mechanism for incorporation.

So what does this mean for the budget debate? Well, it means that when considering the 'will' of the voters, there should always be the realization that groups are not always rational. I suspect this was Bartlett's underlying point, too, maybe slightly removed. Much like SCOTUS in McDonald v. Chicago, in spite of yells for cuts, there would be very little agreement in what specifically to cut. And, according to work on group agency, this shouldn't be surprising at all; it should be exactly what we expect.