The maturing architecture of the brain's default network

Fair et al. 10.1073/pnas.0800376105.

Supporting Information

Files in this Data Supplement:

SI Text
SI Figure 4
SI Movie 1




SI Figure 4

Fig. 4. Full complement of LOWESS curves representing connections with significant differences between children and adults. Regions that increased correlation strength over age are on top. The one region that decreased correlation strength over age is on the bottom (of note: this connection between region homologues, despite decreasing, remained high in adults).





SI Movie 1

Movie 1. Dynamic representation of default network functional connectivity across age. At the beginning of the movie, regions are oriented in the same pseudoanatomical representation as in the main manuscript (for the youngest age group). A spring-embedding algorithm is applied to this initial orientation, such that regions can relax to their lowest energy state. The applied algorithm is known as the Kamada-Kawai method (1), one of the most commonly used strategies for displaying graph network data. This algorithm considers connected nodes as being attracted towards each other, whereas nodes that do not share a connection are pushed apart. The underlying model is that all nodes are connected by springs with resting lengths proportional to the shortest-path distance between them. The algorithm then iteratively adjusts the positions of each node to reduce the total energy of the system to a minimum. The movie includes data from all 210 subjects. Multiple graphs across age are created by using a sliding box car across the subjects in age-order (i.e., graph1: subjects 1-50, graph2: subjects 2-51, graph3: subjects 3-52, etc.). Correlation coefficients (r) across subjects are combined by using the Schmidt-Hunter method for meta analyses of r values (see Materials and Methods in SI Text), after which spring embedding is applied to graphs representing a temporal 6-month sliding box car to present the movie in regular time. The interpolations, algorithm application, and movie production were performed by using SoNIA (Social Network Image Animator) (2). Also see http://sonia.stanford.edu/.

1. Kamada K, Kawai S (1989) An algorithm for drawing general undirected graphs. Inf Processing Lett 31:7-15.

2. Bender-deMoll S, McFarland DA (2006) The art and science of dynamic network visualization. J Soc Struct 7.





SI Text

rs-fcMRI : The Method and Its Utility. Resting-state functional connectivity MRI (rs-fcMRI), the method used in this study, measures correlated spontaneous low frequency (usually <0.1 Hz) BOLD (blood oxygenation level dependent) signal fluctuations between brain regions occurring at rest (1-4). These low-frequency BOLD fluctuations are thought to be related to "spontaneous" neural activity (1, 4, 5). By cross-correlating the BOLD signal time series between different regions or voxels, one can determine which regions are "functionally connected."

Studies of spontaneous neural activity have been around for >100 years (6-8); however, recent theoretical and experimental findings using modern imaging techniques have fueled the expansion of resting-state functional connectivity studies (1-3, 8-17). Spawned by the initial observation by Biswal et al. (1), many scientists are now using rs-fcMRI to identify neural networks whose nodes spontaneously oscillate in synchrony with each other while subjects are at rest.

As highlighted in the main manuscript, one reasonable hypothesis regarding the nature of correlated spontaneous BOLD fluctuations is that they may, at least in part, reflect a longstanding history of coactivation (10, 11, 18, 19). In this sense, coactivation of brain regions across many tasks can lead to the Hebbian strengthening of the functional connections between them (20).

In addition, some evidence suggests that, along with experience dependent evoked activity, spontaneous waves of synchronized activity during prenatal and postnatal development support the integration (i.e., strengthening of interregional relationships) and segregation (i.e., weakening of interregional relationships) of neural networks (10, 18, 21-25). Such mechanisms may be important for gating information flow (21, 22), building internal representations (18, 21, 22), and developing and maintaining mature network relationships (10, 21).

Unburdened by experimental design, subject compliance, and training demands, rs-fcMRI has become particularly important for studies of development and clinical groups. Due to the ease of data acquisition and the ability to use these techniques during sleep (4) as well as anesthesia (4, 8, 9), some investigators have highlighted the potential use of rs-fcMRI as a biomarker in diagnostics and prognostics, as well as an efficacy gage for behavioral and medical interventions (4, 8).

Complementary Interpretations. As outlined in the main article, some have argued that the improvement of episodic memory (thought to depend, in part, on regions of the default network) over age is not solely due to improved abilities of encoding and retrieval per se, but is rather related to the incorporation of increasingly more complex strategies to encode and retrieve stored information (26, 27). Thus, the progressive integration of the default network may support the ability to incorporate a growing number of alternative strategies, or higher-order strategic organization, to improve memory processes over age (27). Although highly speculative, a complementary interpretation begins with the work of Piaget, who posited that cognitive development involves the acquisition and organization of increasingly more complex and interrelated schemata. Schemata (plural for schema) are defined as well established routines or mental representations of affiliated collections of images, memories, perceptions, ideas, and/or actions (for an example, see below) (28). The schema concept is now deeply rooted in cognitive psychology (26-29), but until recently (see refs. 30 and 31), it has received less attention in cognitive neuroscience. According to Piaget, deductive reasoning, systematic planning, and an overall behavioral maturation across the later stages of development involve the acquisition, assimilation, accommodation, and manipulation of existing schemata (28). This view is very similar to other perspectives in cognitive neuroscience (14, 32, 33) that suggest regions within the default network are important for the "retrieval of past events, both personal and general, in an effort to solve problems and develop future plans" (14).

The linking of memory processes and the adaptation of schemata to the regions of the default network has been implied in recent publications (30, 31, 33, 34). For example, recent work in rodents by Tse and colleagues (30) has shown that memory consolidation is not only greatly enhanced by a preexisting schema, but also that updating and adapting a schema is a process dependent on the hippocampus (30, 31), a brain component closely related to the default network (2, 14, 32, 35) (see also Fig. 2). In contrast, retrieval of past events or "autobiographical" information (32, 36, 37) (i.e., episodic memory) becomes less dependent on the hippocampus over time (30, 31, 38, 39) but continues to be closely related to other portions of the default network, including the mPFC, PCC, and lateral parietal regions (30, 31, 37, 38, 40-42). Hence, one might speculate that the functions of the individual default regions that support various aspects of mentalizing, self-projection, SITs, and/or introspectively oriented behavior, are intact and functioning at relatively young ages; however, the later-developing integrated network structure of the default system assists in the assimilation, accommodation, consolidation, and manipulation of experience-dependent information, recollections, or routines (i.e., schemata). Such internal, "offline" organization may be important for adult-level problem solving (34), remembering past events (27), thinking about the future (32), and the integration of these processes for memory-based decision making.

Schema Example. One particular useful example, illustrated in a seminal study by Bransford and Johnson (43), helps elucidate schema theory (below). Read the following passage, then re-read it after the "schema" is provided below.

"The procedure is actually quite simple. First, you arrange things into different groups. Of course, one pile may be sufficient depending on how much there is to do. If you go somewhere else due to lack of facilities, that is the next step, otherwise you are pretty well set. It is important not to overdo things. That is, it is better to do too few things at once than too many. In the short run this may not seem as important but complications can easily arise. A mistake can be expensive as well. At first, the whole procedure may seem complicated. Soon, however, it will become just another facet of life. It is difficult to foresee any end to the necessity for this task in the immediate future, but then one can never tell. After the procedure is completed, one arranges the materials into different groups again. Then they can be put in their appropriate places. Eventually they will be used once more and the whole cycle will then have to be repeated. However, that is part of life."

For those who have never read about or seen this passage, it may seem very confusing without the proper context or "schema." Rereading the passage with the schema, "washing clothes," in mind not only makes the passage more readable, but also aids in the recall of its details as illustrated by Bransford and Johnson (43).

Further explanations of the schema theory and more examples can be found elsewhere (26-28, 30, 31, 44).

Materials and Methods

Subjects. Subjects were recruited from Washington University and the local community. Participants were screened with a questionnaire to ensure that they had no history of neurological/psychiatric diagnoses or drug abuse. Informed consent was obtained from all subjects in accordance with the guidelines and approval of the Washington University Human Studies Committee.

Data Acquisition and Processing Continued. fMRI data were acquired on a Siemens 1.5 Tesla MAGNETOM Vision system. Structural images were obtained using a sagittal magnetization-prepared rapid gradient echo (MP-RAGE) three-dimensional T1-weighted sequence. Functional images were obtained by using an asymmetric spin echo echo-planar sequence sensitive to blood oxygen level-dependent (BOLD) contrast. Whole brain coverage was obtained with 16 contiguous interleaved 8 mm axial slices acquired parallel to the plane transecting the anterior and posterior commissure (AC-PC plane). Steady state magnetization was assumed after 4 frames (~10 s) (see ref. 10 for further details).

Functional images were processed to reduce artifacts (3, 45). These steps included: (i) removal of a central spike caused by MR signal offset, (ii) correction of odd vs. even slice intensity differences attributable to interleaved acquisition without gaps, (iii) correction for head movement within and across runs, and (iv) within run intensity normalization to a whole brain mode value of 1,000. Atlas transformation of the functional data were computed for each individual via the MPRAGE. Each run then was resampled in atlas space (46) on an isotropic 3 mm3 grid combining movement correction and atlas transformation in one interpolation. All subsequent operations were performed on the atlas-transformed volumetric time series.

Functional Connectivity Preprocessing. For rs-fcMRI analyses as previously described (2, 3), several additional preprocessing steps were used to reduce spurious variance unlikely to reflect neuronal activity (e.g., heart rate and respiration). These steps included: (i) a temporal band-pass filter (0.009 Hz < f <0.08 Hz) and spatial smoothing (6 mm full width at half maximum), (ii) regression of six parameters obtained by rigid body head motion correction, (iii) regression of the whole brain signal averaged over the whole brain, (iv) regression of ventricular signal averaged from ventricular region of interest (ROI), and (v) regression of white matter signal averaged from white matter ROI. Regression of first order derivative terms for the whole brain, ventricular, and white matter signals were also included in the correlation preprocessing. These preprocessing steps likely removed any developmental changes in connectivity driven by changes in respiration and heart rate over age.

Computation of Mean Regionwise Correlation Matrix for Graph. The resting-state BOLD time series were correlated region by region for each subject across the full length of the resting time series, creating 210 square correlation matrices (13 ´ 13).

Because of the potential effects of head movement on rs-fcMRI data, even after movement correction (Cohen AL, Fair DA, Miezin FM, Dosenbach NUF, Wenger KK, Fox MD, Snyder AZ, Vincent JL, Raichle ME, Schlagar BL, and Pertersen SE, 36th Annual Meeting of the Society for Neuroscience, Oct. 14-18, 2006, Atlanta, GA, abstract), the child and adult groups were matched for movement to limit its effects. From a sample of 210 subjects, 96 movement-matched subjects (48 children aged 7-9 years; 48 adults aged 21-31 years) were used for the voxelwise mapping, graph visualization, and subsequent direct comparisons. Within-run subject motion was <1 mm rms for all matched groups and not significantly different between groups (rms values: children, 0.625 mm; adults, 0.622 mm).

For graph analyses the correlation coefficients (r) across matched subjects were combined by using the Schmidt-Hunter method for metaanalyses of r values (11, 17, 47). Graph visualization (Fig. 2) was performed by using NetDraw, a social networks visualization package found at http://www.analytictech.com/Netdraw/netdraw.htm.

1. Biswal B, Yetkin FZ, Haughton VM, Hyde JS (1995) Functional connectivity in the motor cortex of resting human brain using echo-planar MRI. Magn Reson Med 34:537-541.

2. Fox MD, et al. (2005) The human brain is intrinsically organized into dynamic, anticorrelated functional networks. Proc Natl Acad Sci USA 102:9673-9678.

3. Fair DA, et al. (2007) A method for using blocked and event-related fMRI data to study "resting state" functional connectivity. NeuroImage 35:396-405.

4. Fox MD, Raichle ME (2007) Spontaneous fluctuations in brain activity observed with functional magnetic resonance imaging. Nat Rev Neurosci 8:700-711.

5. Leopold DA, Murayama Y, Logothetis NK (2003) Very slow activity fluctuations in monkey visual cortex: Implications for functional brain imaging. Cereb Cortex 13:422-433.

6. Berger H (1969) On the electroencephalogram of man. Third report. Electroencephalogr Clin Neurophysiol Suppl 28:95-132.

7. Berger H (1969) On the electroencephalogram of man. Second report. Electroencephalogr Clin Neurophysiol Suppl 28:75-93.

8. Buckner RL, Vincent JL (2007) Unrest at rest: The importance of default activity and spontaneous network correlations. NeuroImage 37:1091-1096

9. Vincent JL, et al. (2007) Intrinsic functional architecture in the anesthetized monkey brain. Nature 447:46-47.

10. Fair DA, et al. (2007) Development of distinct control networks through segregation and integration. Proc Natl Acad Sci USA 104:13507-13512.

11. Dosenbach NU, et al. (2007) Distinct brain networks for adaptive and stable task control in humans. Proc Natl Acad Sci USA 104:11073-11078.

12. Hampson M, Driesen NR, Skudlarski P, Gore JC, Constable RT (2006) Brain connectivity related to working memory performance. J Neurosci 26:13338-13343.

13. Lowe MJ, Dzemidzic M, Lurito JT, Mathews VP, Phillips MD (2000) Correlations in low-frequency BOLD fluctuations reflect cortico-cortical connections. NeuroImage 12:582-587.

14. Greicius MD, Krasnow B, Reiss AL, Menon V (2003) Functional connectivity in the resting brain: A network analysis of the default mode hypothesis. Proc Natl Acad Sci USA 100:253-258.

15. Fransson P (2005) Spontaneous low-frequency BOLD signal fluctuations: An fMRI investigation of the resting-state default mode of brain function hypothesis. Hum Brain Mapp 26:15-29.

16. Kelly AMC, Uddin LQ, Biswal BB, Castellanos FX, Milham MP (in press) Competition between functional brain networks mediates behavioral variability. Neuroimage 39:527-537

17. Salvador R, et al. (2005) Neurophysiological architecture of functional magnetic resonance images of human brain. Cereb Cortex 15:1332-1342.

18. Miltner WH, Braun C, Arnold M, Witte H, Taub E (1999) Coherence of gamma-band EEG activity as a basis for associative learning. Nature 397:434-436.

19. Seeley WW, et al. (2007) Dissociable intrinsic connectivity networks for salience processing and executive control. J Neurosci 27:2349-2356.

20. Hebb DO (1949) The Organization of Behavior (Wiley, New York).

21. Raichle ME, Snyder AZ (2007) A default mode of brain function: a brief history of an evolving idea. NeuroImage 37:1083-1090.

22. Salinas E, Sejnowski TJ (2001) Correlated neuronal activity and the flow of neural information. Nat Rev Neurosci 2:539-550.

23. Varela F, Lachaux J-P, Rodriguez E, Martinerie J (2001) The brainweb: Phase synchronization and large-scale integration. Nat Rev Neurosci 2:229-239.

24. Honey C, Kotter R, Breakspear M, Sporns O (2007) Network structure of cerebral cortex shapes functional connectivity on multiple time scales. Proc Natl Acad Sci USA 104:10240-10245

25. Johnson MH, Munakata Y (2005) Processes of change in brain and cognitive development. Trends Cognit Sci 9:152-158.

26. Kail RV (1990) The Development of Memory in Children (Freeman, New York).

27. Schneider W, Pressley M (1997) Memory Development Between Two and Twenty (Lawrence Erlbaum Assoc, Mahwah, NJ).

28. Flanagan OJ (1984) The Science of the Mind (MIT Press, Cambridge, MA).

29. Bartlett FC (1932) Remembering: A Study in Experimental and Social Psychology (Cambridge Univ Press, Cambridge, UK).

30. Tse D, et al. (2007) Schemas and memory consolidation. Science 316:76-82.

31. Squire LR (2007) Neuroscience. Rapid consolidation. Science 316:57-58.

32. Buckner RL, Carroll DC (2007) Self-projection and the brain. Trends Cognit Sci 11:49-57.

33. Maguire EA, Frith CD, Morris RG (1999) The functional neuroanatomy of comprehension and memory: The importance of prior knowledge. Brain 122 (Pt 10):1839-1850.

34. Shallice T, Burgess P (1996) The domain of supervisory processes and temporal organization of behaviour. Philos Trans R Soc London Ser B Biol Sci 351:1405-1412.

35. Greicius MD, Srivastava G, Reiss AL, Menon V (2004) Default-mode network activity distinguishes Alzheimer's disease from healthy aging: Evidence from functional MRI. Proc Natl Acad Sci USA 101:4637-4642.

36. Tulving E (1983) Elements of Episodic Memory (Oxford, New York).

37. Frith U, Frith CD (2003) Development and neurophysiology of mentalizing. Philos Trans R Soc London Ser B Biol Sci 358:459-473.

38. Squire LR, Knowlton BJ (2000) in The New Cognitive Neurosciences, ed Gazzaniga MS (MIT Press, Cambridge, MA), pp 765-779.

39. Maguire EA, Frith CD, Rudge P, Cipolotti L (2005) The effect of adult-acquired hippocampal damage on memory retrieval: An fMRI study. NeuroImage 27:146-152.

40. Wagner AD, Shannon BJ, Kahn I, Buckner RL (2005) Parietal lobe contributions to episodic memory retrieval. Trends Cognit Sci 9:445-453.

41. Vincent JL, et al. (2006) Coherent spontaneous activity identifies a hippocampal-parietal memory network. J Neurophysiol 96:3517-3531.

42. Maguire EA, Mummery CJ, Buchel C (2000) Patterns of hippocampal-cortical interaction dissociate temporal lobe memory subsystems. Hippocampus 10:475-482.

43. Bransford JD, Johnson MK (1972) Contextual prerequisites for understanding: Some investigations of comprehension and recall. J Verbal Learning Verbal Behav 11:717-726.

44. Neath I, Surprenant AM (2003) Human Memory (Thomson Wadsworth, Toronto, Canada).

45. Miezin F, Maccotta L, Ollinger J, Petersen S, Buckner R (2000) Characterizing the hemodynamic response: Effects of presentation rate, sampling procedure, and the possibility of ordering brain activity based on relative timing. NeuroImage 11:735-759.

46. Talairach J, Tournoux P (1988) Co-Planar Stereotaxic Atlas of the Human Brain (Thieme, New York).

47. Field AP (2001) Meta-analysis of correlation coefficients: a Monte Carlo comparison of fixed- and random-effects methods. Psychol Methods 6:161-180.

This Article

  1. Published online before print March 5, 2008, doi: 10.1073/pnas.0800376105 PNAS March 11, 2008 vol. 105 no. 10 4028-4032
  1. OA Abstract
  2. Figures Only
  3. OA Full Text
  4. Full Text (PDF)
  5. » Supporting Information
About Our New Site Design >>
Link: Advanced Search

This Week's Issue

  1. July 1, 2008, 105 (26)
  1. Current Issue
  1. Alert me to new issues of PNAS

Most