NEUROSCIENCE: The Brain's Dark Energy Marcus E. Raichle, et al. Science 314, 1249 (2006); DOI: 10.1126/science.1134405 The following resources related to this article are available online at www.sciencemag.org (this information is current as of November 24, 2006 ): Updated information and services, including high-resolution figures, can be found in the online version of this article at: http://www.sciencemag.org/cgi/content/full/314/5803/1249 This article appears in the following subject collections: Neuroscience http://www.sciencemag.org/cgi/collection/neuroscience Information about obtaining reprints of this article or about obtaining permission to reproduce this article in whole or in part can be found at: http://www.sciencemag.org/help/about/permissions.dtl Downloaded from www.sciencemag.org on November 24, 2006 Science (print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by the American Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005. Copyright c 2005 by the American Association for the Advancement of Science; all rights reserved. The title SCIENCE is a registered trademark of AAAS. responding to. –10. Less than 10% of all synapses carry incoming information from the external world (3)—a surprisingly small number. the former has motivated most neuroscience research.edu At rest.5 to 1. 1267 (2006). While neither view is dominant. 558 (1998). Y. these studies have revealed that the additional energy required for such brain responses is extremely small compared to the ongoing amount of energy that the brain normally and continuously expends (2). L. particularly in neuroimaging.PERSPECTIVES The study of domains and domain walls has a long tradition in magnetism. Phys. Y. Maeno et al. D. Science 306.0 10. E. fMRI images of a normal human brain at rest. spontaneous cognition—our daydreams or. Strand. M. researchers have sought to explain the relative disproportion of connections (i. J. 10 times that predicted by its weight alone. Rev. M. P.1134921 2. yet accounts for about 20% of the body’s total energy consumption. St. E-mail: [email protected]% of the total energy budget (2). D. published online 26 October 2006 (10. the cortex may simply be more involved in intrinsic activities. (right) lateral and medial views of the right hemisphere. J. L643 (1995). 2006 1249 NEUROSCIENCE . 843 (1985). The observation of such a vortex with a fractional magnetic flux is a challenge to this emerging field. Condens. Kidwingira. MO 63110. 6. Maeno. MacKenzie. 63. involving the maintenance of information for interpreting. but active. From a brain energy perspective. Y.1126/science. Gor’kov. 97. whose primary function is to respond to external input to the retina. The Brain’s Dark Energy Marcus E. USA. first with positron emission tomography (PET) and now largely with functional magnetic resonance imaging (fMRI). and even predicting environmental demands. more technically. A..0 The author is in the Washington University School of Medicine. What fraction of this energy is directly related to brain function? Depending on the approach S used. M. Z. but it will not be easy in view of the short characteristic length scales that Kidwingira et al. driven by the momentary demands of the environment.. Phys. Q.sciencemag. however. Sov. Nelson.. But it is highly unlikely to account for more than that elicited by responding to controlled stimuli. Matter 7. and how do we study it? ince the 19th century. Maeno. What is this energy used for. (9): A novel magnetic vortex should accompany a singularity on a domain wall analogous to a Bloch line in magnetism. But technological advances. Volovik. The additional energy burden associated with momentary demands of the environment may be as little as 0. 167002 (2006). synapses) among neurons that appear to perform func- tions intrinsically within the cerebral cortex. Louis. Surprisingly. it is estimated that 60 to 80% of the energy budget of the brain supports communication among neurons and their supporting cells (2). D. have provoked a reassessment of these two perspectives. (center) dorsal view. J. 1727 (1989). Phys. F. K. Y. What is this intrinsic activity? One possibility is that it simply represents unconstrained. The brain apparently uses most of its energy for functions unaccounted for—dark energy. J. Mod. Human functional neuroimaging.1133239). They found that the domains should contain counterflowing supercurrents along the wall that generate a perpendicular magnetic dipole (see the figure). Sigrist. Rice. (Left) Lateral and medial views of the left hemisphere. Sigrist et al. Consideration of brain energy may thus provide new insights into questions that have long puzzled neuroscientists. 4. in astronomical terms. Xia et al. Nature 372. Science 314. 7. T. 1151 (2004). 3. 9.. 75.org on November 24.sciencemag. Volovik and Gor’kov (8) were the first to study the theory of these kinds of superconducting domain walls. What do we know about this dark energy? The adult human brain represents about 2% of the body weight. For example. Van Harlingen. All this remains to be verified experimentally. Confirmation of these exotic predictions will be a clear test of our understanding of these intriguing superconductors. that the brain’s operations are mainly intrinsic. the other. Rev. as does another effect predicted by Sigrist et al. Correlations are depicted by an arbitrary color scale. stimulus-independent thoughts. 10.org SCIENCE VOL 314 24 NOVEMBER 2006 Published by AAAS Downloaded from www. M. JETP 61. The images reveal the highly organized nature of intrinsic brain activity. Liu. and these experiments open this field to the study of chiral superconductors. allows the brain’s responses to controlled stimuli to be studied by measuring changes in brain circulation and metabolism (energy consumption). G. References 1. Mao. Phys.1126/science. Raichle Much of the brain’s enormous energy consumption is unaccounted for by its responses to external stimuli. 657 (2005). Rev. P. represented by correlated spontaneous fluctuations in the fMRI signal. which accounts for a very small fraction of total brain activity. Y. Phys. Take the visual cortex. Positive correlations reside in areas known to increase activity during responses to controlled stimuli. One view posits that the brain is primarily reflexive. Lett. Maeno. 8. deduce from their experiments. G. Luke et al. 5. [Reprinted from (12)] www. two perspectives on brain functions have existed (1). negative correlations reside in areas that decrease activity under the same conditions. This cost-based analysis implies that intrinsic activity may be far more significant than evoked activity in terms of overall brain function. 532 (1994). Nature 394.e.wustl. and possibly longer. Lett.0 0. M.org Published by AAAS Downloaded from www. Murayama. Natl. Petersen. Proc. 100.org on November 24. Future research should address the cellular events underlying spontaneous fMRI BOLD signal fluctuations. that whilst part of what we perceive comes through our senses from the object before us. Thus. 4535 (2006). Hahn. 13638 (2003). References 1. For example. 26. A prominent feature of fMRI is that the unaveraged signal is quite noisy. 422 (2003).sciencemag. Cambridge. James. Acad. 29. Neurosci. Acad. 2004). C. Hasenstaub. William James presciently suggested in 1890 (11) that “Enough has now been said to prove the general law of perception. However. and ice sheets (6).” It may be that in the normal brain. Peters. 9673 (2005). J. U.8 mm/year on average over the past 50 years (2. 10). A. Stumbling on Happiness (Knopf. I of the Vortex: From Neurons to Self (MIT Press. L. The “balance” of these stimuli determines the responsiveness (or gain) of neurons to correlated inputs and. 3. to make predictions about the future (6). S. B. A more expanded view is that intrinsic activity instantiates the maintenance of information for interpreting. tide gauges have detected global sea-level rises [~1. These figures constrain the contribution from ice sheets to less than 1 mm/year in the past decade. But even a small loss of ice mass from the ice sheets would have a great impact on sea level. Cereb. vol. global sea level has remained stable. A. E. as well as the space-borne Synthetic Aperture Radar Interferometry (InSAR) technique. A. dynamical instabilities in response to climate warming may cause faster ice-mass loss. 2006 . Satellite altimetry data document a rate of ~3 mm/year since 1993 (4).1 Hz) in neuronal membrane polarization—so-called up and down states—are intriguing (4. Llinas. J. During the past decade.A. it turns out that a considerable fraction of the variance in the blood oxygen level–dependent (BOLD) signal of fMRI in the frequency range below 0. A. ocean warming has contributed roughly half of the observed rate of sea-level rise (5). Cereb. H. D. Natl. D. which is this. Y. New York. Gilbert. R. includEnhanced online at ing those reported www. Present-day sea-level rise has several causes. D. Budd. New York. 1603–1615. element of brain function. responding to. Mehta. Sci. glaciers.A. McCormick. J. A. have provided the first observations of ice sheet mass balance (8–13). designed to generate predictions about the future (5). T. 12. M. S. 2. M. On Intelligence (Holt. is lost.cazenave@cnes. Blakeslee. prompting researchers to average their data to reduce this “noise” and increase the signals they seek. Hawkins. remote-sensing data based on airborne laser and satellite radar altimetry. a balance of opposing forces enhances the precision of a wide range of processes. it remains unclear whether the recent rate increase reflects an accelera- I The author is at the Observatoire Midi-Pyrenees. particularly on low-lying islands and coastal regions. 7. 4. New York. H. pp. Payne. 103. f the ice sheets covering Greenland and Antarctica were to melt completely. Sci. Haider. a useful conceptual framework from theoretical neuroscience posits that the brain operates as a Bayesian inference engine. p.sciencemag. N. In doing this. In this regard. leaving the other half for ocean-mass increase caused by water exchange with continents. 1134405 ATMOSPHERE How Fast Are the Ice Sheets Melting? Anny Cazenave Remote-sensing data suggest that ice sheets currently contribute little to sea-level rise. 10. 2. potentially sculpts communication pathways in the brain (4). C. 11. in The Visual Neurosciences. R. K. Chalupa. 31400 Toulouse. “balance” might be viewed as a necessary enabling. Cortex 13.PERSPECTIVES Another possibility is that the brain’s enormous intrinsic functional activity facilitates responses to stimuli. 5. 127 (1985). 2001). the brain is then sculpted by worldly experience to represent intrinsically a “best guess” (“priors” in Bayesian parlance) about the environment and. Neurobiol. Werner. 8.fr tion in sea-level rise or a natural fluctuation on a decadal time scale. However. B. Annu.sciencemag. New satellite observations. they would raise sea level by about 65 m. Balance also manifests at a large systems level. In this regard. E-mail: anny. MA. which reflects fluctuating neural activity. Grinvald. another part (and it may be the larger part) always comes (in Lazarus’s phrase) out of our own head. It has long been thought that the ability to reflect on the past or contemplate the future has facilitated the development of unique human attributes such as imagination and creativity (7. descriptions of slow fluctuations (nominally <0. Eds. U. Logothetis. The contribution of mountain glaciers and small ice caps to sea-level rise in the past decade is estimated to be ~0.S. 4. (MIT Press.. Since the early 1990s. Sakmann. A. but since the end of the 19th century. 10). D. For the past 3000 years.1 Hz. Hum. 215 (1994). G. Raichle. Cortex 4. B. J.1126/science. Proc. 449 (2006). neurologists know that strokes that damage cortical centers that control eye movements lead to deviation of the eyes toward the side of the lesion. Duque. 1890). These observations indicate accelerated ice-mass loss in recent years in the coastal regions of southern Greenland. MA. 3)]. Neurosci.S. The challenge of neuroscience is to understand the functions associated with this energy consumption. B. Principles of Psychology (Henry Holt & Company. 8). 10. 6. In contrast. Rev. W. 102. but their functional consequences may be relevant to an understanding of the variability in task-evoked brain activity as well as behavioral variability in human performance. Beginning with a set of “advance” predictions at birth (genes).org/cgi/ by Luthcke et al. Mintun. Ingvar. Over Antarctica. Not only does their temporal frequency correspond to that of the spontaneous fluctuations in the fMRI BOLD signal. T. implying the preexisting presence of “balance. fMRI provides one important experimental approach to understanding the nature of the brain’s intrinsic functional activity without direct recourse to controlled stimuli and observable behaviors. remote sensing indicates 1250 24 NOVEMBER 2006 VOL 314 SCIENCE www. M.8 mm/year (7). content/full/314/5803/1250 on page 1286 of this issue (1). R. 2006). Leopold. now allow estimates of the mass balances of the ice sheets and their evolution through time. Studies likely will cover a broad range of approaches to the study of spontaneous activity of neurons (9. This activity exhibits striking patterns of coherence within known networks of specific neurons in the human brain in the absence of observable behaviors (see the figure). Olshausen. M. 9.” The brain’s energy consumption tells us that the brain is never at rest. D. but costly. 2003). and even predicting environmental demands. in so doing. Cambridge. in the case of humans at least. A. Fox et al. slight mass gain is reported in central high-elevation regions. France. Neurons continuously receive both excitatory and inhibitory inputs.