Wandering Minds: The Default Network and Stimulus-Independent

Thought
Malia F. Mason et al.
Science 315, 393 (2007);
DOI: 10.1126/science.1131295

This copy is for your personal, non-commercial use only.

If you wish to distribute this article to others, you can order high-quality copies for your
colleagues, clients, or customers by clicking here.

Downloaded from [Link] on March 9, 2013

Permission to republish or repurpose articles or portions of articles can be obtained by
following the guidelines here.

The following resources related to this article are available online at

[Link] (this information is current as of March 9, 2013 ):

Updated information and services, including high-resolution figures, can be found in the online
version of this article at:
[Link]
Supporting Online Material can be found at:
[Link]
A list of selected additional articles on the Science Web sites related to this article can be
found at:
[Link]
This article cites 21 articles, 6 of which can be accessed free:
[Link]
This article has been cited by 189 article(s) on the ISI Web of Science
This article has been cited by 100 articles hosted by HighWire Press; see:
[Link]
This article appears in the following subject collections:
Neuroscience
[Link]

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
2007 by the American Association for the Advancement of Science; all rights reserved. The title Science is a
registered trademark of AAAS.
REPORTS
3. S. Grillner, in Handbook of Physiology, Sect. 1: The 15. J. S. Anderson, M. Carandini, D. Ferster, J. Neurophysiol. 27. K. Diba, H. A. Lester, C. Koch, J. Neurosci. 24, 9723
Nervous System, vol. 2, Motor Control, V. B. Brooks, Ed. 84, 909 (2000). (2004).
(American Physiological Society, Bethesda, MD, 1981), 16. Y. Shu, A. Hasenstaub, D. A. McCormick, Nature 423, 28. C. van Vreeswijk, H. Sompolinsky, Neur. Comput. 10,
pp. 1179–1236. 288 (2003). 1321 (1998).
4. E. Jankowska, M. G. Jukes, S. Lund, A. Lundberg, Nature 17. E. A. Stern, A. E. Kincaid, C. J. Wilson, J. Neurophysiol. 29. C. van Vreeswijk, H. Sompolinsky, Science 274, 1724
206, 198 (1965). 77, 1697 (1997). (1996).
5. P. S. Stein, J. Comp. Physiol. A 191, 213 (2005). 18. W. R. Softky, C. Koch, J. Neurosci. 13, 334 (1993). 30. J. Marino et al., Nat. Neurosci. 8, 194 (2005).
6. E. Marder, R. L. Calabrese, Physiol. Rev. 76, 687 (1996). 19. M. N. Shadlen, W. T. Newsome, J. Neurosci. 18, 3870 31. This study was supported by grants from the Danish
7. S. Grillner, Nat. Rev. Neurosci. 4, 573 (2003). (1998). Research Council and the Lundbeck Foundation (J.H.) and
8. O. Kiehn, O. Kjaerulff, M. C. Tresch, R. M. Harris-Warrick, 20. J. Hounsgaard, O. Kiehn, I. Mintz, J. Physiol. 398, 575 a NATO Reintegration Grant (A.A.). We thank J. F. Perrier
Brain Res. Bull. 53, 649 (2000). (1988). for critically reading the manuscript. Thanks to
9. R. Delgado-Lezama, J. Hounsgaard, Prog. Brain Res. 123, 21. G. L. Gerstein, B. Mandelbrot, Biophys. J. 4, 41 D. Kleinfeld and lab members for helpful discussions (R.B.).
57 (1999). (1964).
10. J. Keifer, P. S. Stein, Brain Res. 266, 148 (1983). 22. W. H. Calvin, C. F. Stevens, J. Neurophysiol. 31, 574 Supporting Online Material
11. A. Alaburda, J. Hounsgaard, J. Neurosci. 23, 8625 (1968). [Link]/cgi/content/full/315/5810/390/DC1
(2003). 23. A. Destexhe, M. Rudolph, D. Pare, Nat. Rev. Neurosci. 4, Materials and Methods
12. A. Alaburda, R. Russo, N. MacAulay, J. Hounsgaard, 739 (2003). Figs. S1 to S5
J. Neurosci. 25, 6316 (2005). 24. F. S. Chance, L. F. Abbott, A. D. Reyes, Neuron 35, 773 References
13. Materials and methods are available as supporting (2002).
material on Science Online. 25. E. Salinas, T. J. Sejnowski, J. Neurosci. 20, 6193 (2000).
14. L. J. Borg-Graham, C. Monier, Y. Fregnac, Nature 393, 26. B. Haider, A. Duque, A. R. Hasenstaub, D. A. McCormick, 11 September 2006; accepted 23 November 2006
369 (1998). J. Neurosci. 26, 4535 (2006). 10.1126/science.1134960

Downloaded from [Link] on March 9, 2013

derpin the mind's wandering, then the magnitude
Wandering Minds: of neural activity in these regions should track
with people’s proclivity to generate SIT. Specif-

The Default Network and ically, individuals who report frequent mind-
wandering should exhibit greater recruitment of
the default network when performing tasks that
Stimulus-Independent Thought are associated with a high incidence of SIT.

Malia F. Mason,1*§ Michael I. Norton,2 John D. Van Horn,1† Daniel M. Wegner,3

Scott T. Grafton,1‡ C. Neil Macrae4

Despite evidence pointing to a ubiquitous tendency of human minds to wander, little is known
about the neural operations that support this core component of human cognition. Using both
thought sampling and brain imaging, the current investigation demonstrated that mind-wandering
is associated with activity in a default network of cortical regions that are active when the brain is
“at rest.” In addition, individuals’ reports of the tendency of their minds to wander were correlated
with activity in this network.

hat does the mind do in the absence conscious supervision (4, 5). But how does the

W of external demands for thought? Is it

essentially blank, springing into ac-
tion only when some task requires attention?
brain spontaneously produce the images, voices,
thoughts, and feelings that constitute stimulus-
independent thought (SIT)?
Everyday experience challenges this account We investigated whether the default
of mental life. In the absence of a task that re- network—brain regions that remain active
quires deliberative processing, the mind gener- during rest periods in functional imaging
ally tends to wander, flitting from one thought to experiments (6)—is implicated in mind-
the next with fluidity and ease (1, 2). Given the wandering (7). The default network is minimally
ubiquitous nature of this phenomenon (3), it has disrupted during passive sensory processing and
been suggested that mind-wandering constitutes attenuates when people engage in tasks with
a psychological baseline from which people de- high central executive demand (8, 9), which
part when attention is required elsewhere and to matches precisely the moments when the mind Fig. 1. Graphs depict regions of the default
which they return when tasks no longer require is most and least likely to wander (2, 4, 5). We network exhibiting significantly greater activity
thus trained individuals to become proficient on during practiced blocks (red) relative to novel
1 tasks (10) so that their minds could wander when blocks (blue) at a threshold of P < 0.001, number
Department of Psychological and Brain Sciences, Dartmouth of voxels (k) = 10. Mean activity was computed for
College, Hanover, NH 03755, USA. 2Harvard Business School, they performed practiced versus novel task se-
Harvard University, Boston, MA 02163, USA. 3Department of each participant by averaging the signal in
quences (11). Although previous research has
Psychology, Harvard University, Cambridge, MA 02138, USA. regions within 10 mm of the peak, across the
4 compared brain activity during rest to that during duration of the entire block. Graphs depict the
School of Psychology, University of Aberdeen, Aberdeen
AB24 2UB, Scotland. engagement in a task (12), the present investiga- mean signal change across all participants. (A)
*Present address: Martinos Center for Biomedical Imaging, tion assesses directly both the production of SIT Left (L.) mPFC (BA 9; –6, 54, 22); (B) Bilateral (B.)
MGH, Charlestown, MA 02129, USA. and activity in the default network during tasks cingulate (BA 24; 0, –7, 36); (C) Right (R.) insula
†Present address: Department of Neurology, University of that allow for varying degrees of mind-wandering. (45, –26, 4); and (D) L. posterior cingulate (BA
California, Los Angeles, CA 90095, USA.
‡Present address: Department of Psychology, University of
Despite its regular occurrence, not all minds 23/31; –9, –39, 27). Activity is plotted on the
California, Santa Barbara, CA 93106, USA.
wander to the same degree; individuals exhibit average high-resolution anatomical image and
§To whom correspondence should be addressed. E-mail: stable differences in their propensity to produce displayed in neurological convention (left hemi-
malia@[Link] SIT (1, 3). If regions of the default network un- sphere is depicted on the left).

[Link] SCIENCE VOL 315 19 JANUARY 2007 393

REPORTS
To investigate the relation between default the posterior lateral cortices (BAs 40 and 39), chological process), changes in default network
network activity and mind-wandering, we first the insular cortices, the cingulate (BA 24), and BOLD activity during practiced relative to novel
established high-incidence mind-wandering pe- aspects of both ventral and dorsal medial pre- blocks should be related to individuals’ minds
riods by training participants on blocks of verbal frontal cortex (mPFC) [BAs 6, premotor and propensity to wander. Voxel-wise correlations
and visuospatial working-memory tasks (days supplementary motor cortex; 8, including frontal were conducted on participants’ standardized
1 to 4), then verified that these frequent mind- eye field; 9, dorsolateral prefrontal cortex; and score on the daydream frequency scale of the
wandering periods were associated with in- 10, frontopolar area (most rostral part of superior Imaginal Processes Inventory (IPI) (14) and
creased default network recruitment as seen with and middle frontal gyri)] (8, 9) [table S1 (13)]. their practiced relative to novel contrast images
functional magnetic resonance imaging (fMRI) To determine whether a relation exists be- [threshold at r(14) > 0.50, P < 0.05] (table S3).
on day 5. Finally, we related participants’ pat- tween the default network and mind-wandering, Results revealed a significant positive relation
terning of default network activity to their self- we investigated how BOLD activity within this between the frequency of mind-wandering and
reported propensity to generate SITs (13). functionally defined network changed as a the change in BOLD signal observed when
On day 4, the proportion of sampled function of block type, by comparing activity participants performed WpracticedW relative to
thoughts participants classified as SIT varied when participants performed practiced (i.e., high- WnovelW blocks in several regions, including the
by block type (baseline, practiced, or novel), incidence SIT periods) blocks to activity during right SFG (BA 8; 12, 48, 36), the mPFC,
F(2, 34) = 81.49, P < 0.01. Participants reported novel (i.e., low-incidence SIT periods) blocks bilaterally (BA 10; –6, 51, –9), bilateral aspects
a greater proportion of SIT during the baseline (13). Default network recruitment was greater of the cingulate (BA 31; 7, –21, 51) and neigh-
blocks (mean = 0.93; SD = 0.16) than during during high-incidence SIT periods. Regions of boring precuneus (BA 31/7; 3, –45, 37), and the
both practiced blocks (mean = 0.32, SD = 0.20), the default network that exhibited greater activity left (BA 13; –36, –16, 17) and right insula (BA

Downloaded from [Link] on March 9, 2013

t(17) = 9.22, P < 0.01, and novel blocks (mean = during these periods included bilateral aspects of 13; 47, 0, 4) (Fig. 2). No region of the default
0.22, SD = 0.18), t(17) = 10.96, P < 0.01. Par- the mPFC (BAs 6, 8, 9, and 10); bilateral su- network exhibited a significant negative correla-
ticipants reported a significantly greater propor- perior frontal gyri (SFG; BAs 8 and 9); the tion with daydream frequency scores at this
tion of SIT during the practiced blocks than anterior cingulate (BA 10); bilateral aspects of threshold.
during the novel blocks, t(17) = 2.11, P < 0.05, the posterior cingulate (BAs 29 and 30) and We proposed that mind-wandering consti-
despite the fact that the tasks were identical. precuneus (BAs 7 and 31); the left angular gyrus tutes a psychological baseline that emerges
Thus, periods of reduced central executive (BA 39); bilateral aspects of the insula (BA 13); when the brain is otherwise unoccupied, sup-
demand were associated with a greater incidence the left superior temporal (BA 22), the right ported by activity in a default network of cortical
of mind-wandering. superior temporal (BA 41) and the left middle regions. Results demonstrated that reductions in
On day 5, we performed functional imaging. temporal gyri (BA 19) (Fig. 1 and table S2) (13). processing demands, that is, performing prac-
We first functionally defined the default network No single default network region exhibited ticed versus novel sequences of otherwise iden-
by comparing the BOLD response associated greater activity during low-incidence SIT pe- tical tasks, were accompanied by increases in
with baseline (i.e., fixation) to the response riods. These findings are consistent with the both the generation of SIT and activity in the
associated with task periods (i.e., novel and hypothesis that the tonic activity observed in the default network. Furthermore, the magnitude of
practiced working-memory tasks). This compar- default network during conscious resting states is BOLD increases that participants exhibited as
ison revealed significantly greater recruitment at associated with mind-wandering. they were able to generate increasing levels of
rest in a distributed network of regions that in- If recruitment of the default network dur- SIT was positively correlated with their self-
cluded aspects of the posterior cingulate and the ing tasks with low processing demands reflects reported daydreaming propensities. Other re-
precuneus [Brodmann areas (BAs) 23 and 31], mind-wandering (rather than some other psy- search provides further evidence for default

Fig. 2. Graphs depict regions that exhibited a

significant positive relation, r(14) > 0.50, P <
0.05, between the frequency of mind-
wandering and the change in BOLD signal
observed when people performed practiced
relative to novel blocks. Participants’ BOLD
difference scores (practiced – novel) are
plotted against their standardized IPI day-
dreaming score. BOLD signal values for the
two blocks were computed for each participant
by averaging the signal in regions within 10
mm of the peak, from 4 TRs (10 s) until 10 TRs
(22.5 s) after the block onset. (A) B. mPFC (BA
10; –6, 51, –9; k = 25). (B) B. precuneus and
p. cingulate (BA 31, 7; –3, –45, 37; k = 72).
(C) R. cingulate (BA 31; 7, –21, 51; k = 73).
(D) L. insula (BA 13; –36, –16, 17; k = 10). (E)
R. insula (BA 13; 47, 0, 4; k = 13). Activity is
plotted on the average high-resolution ana-
tomical image and displayed in neurological
convention (left hemisphere is depicted on the
left).

394 19 JANUARY 2007 VOL 315 SCIENCE [Link]

REPORTS
network involvement in the production of SIT. of arousal, thereby facilitating performance on 14. J. L. Singer, J. S. Antrobus, in The Function and Nature of
First, damage to parts of the network (e.g., mundane tasks (4). A second possibility is that Imagery, P. Sheehan, Ed. (Academic Press, New York,
1972), pp. 175–202.
mPFC) is associated with Wmental emptinessW SIT—as a kind of spontaneous mental time 15. A. Damasio, G. Van Hoesen in Neuropsychology of
and an absence of spontaneous speech and travel—lends a sense of coherence to one’s past, Human Emotion, K. Heilman, P. Satz, Eds. (Guilford, New
thought (15). Second, aging is associated with present, and future experiences (26–29). Finally, York, 1983), pp. 85–110.
the development of plaques in default network the mind may generate SIT not to attain some 16. L. M. Giambra, Psychol. Aging 4, 136 (1989).
17. R. L. Buckner et al., J. Neurosci. 25, 7709 (2005).
regions and a corresponding reduction in SIT extrinsic goal (e.g., staying alert) but simply 18. M. E. Raichle et al., Proc. Natl. Acad. Sci. U.S.A. 98, 676
(16, 17). Taken together, these findings provide because it evolved a general ability to divide (2001).
converging evidence for the role of the default attention and to manage concurrent mental tasks. 19. J. W. Schooler, Trends Cognit. Sci. 6, 339 (2002).
network in mind-wandering. Although the thoughts the mind produces when 20. A. D. Craig, Curr. Opin. Neurobiol. 13, 500 (2003).
21. A. R. Damasio et al., Nat. Neurosci. 3, 1049
Of course, mind-wandering is not the only wandering are at times useful, such instances do (2000).
cognitive process that ensues when tasks cease not prove that the mind wanders because these 22. D. A. Gusnard et al., Proc. Natl. Acad. Sci. U.S.A. 98,
to require conscious supervision. Reductions in thoughts are adaptive; on the contrary the mind 4259 (2001).
task difficulty are also likely accompanied by may wander simply because it can. 23. W. M. Kelley et al., J. Cognit. Neurosci. 14, 785
(2002).
qualitative changes in attention and, perhaps, the
24. J. W. Schooler, E. D. Reichle, D. V. Halpern, in Thinking
implementation of general “housekeeping” References and Notes and Seeing: Visual Metacognition in Adults and Children,
functions (18). It is likely that activity in the 1. J. L. Singer, Daydreaming (Plenum Press, New York, 1966). D. T. Levin, Ed. (MIT Press, Cambridge, MA, 2004),
default network is associated with a range of 2. J. S. Antrobus, J. L. Singer, S. Greenberg, Percept. Mot. pp. 203–226.
Skills 23, 399 (1966). 25. D. M. Wegner, in Scientific Approaches to Consciousness,
cognitive functions. For example, although we

Downloaded from [Link] on March 9, 2013

3. J. L. Singer, V. McRaven, Int. J. Soc. Psychiatry 8, 272 J. Cohen, J. Schooler, Eds. (Lawrence Erlbaum, Hillsdale,
interpret results from our correlational analyses (1962). NJ, 1997), pp. 295–316.
as evidence that cortical regions in the default 4. E. Klinger, Structure and Functions of Fantasy (Wiley, 26. E. Tulving, Can. Psychol. 26, 1 (1985).
network play a general role in the production of New York, 1971). 27. E. A. Maguire et al., Philos. Trans. R. Soc. Lond. B Biol.
5. J. Smallwood, J. W. Schooler, Psychol. Bull. 132, 946 (2006). Sci. 356, 1441 (2001).
SIT, it is possible that some of these regions 6. fMRI evidence suggests that, when deprived of variable 28. R. Cabeza et al., J. Cognit. Neurosci. 16, 1583
mediate the meta-awareness of SIT (19), such as sensory input, the mind recruits a discrete network of (2004).
the insular cortices, which subserve interocep- brain regions that includes the medial posterior cingulate 29. J. L. Vincent et al., J. Neurophysiol. 96, 3517
tion and self-awareness (20, 21), and regions of (BAs 30 and 31); the precuneus (BA 7); paracentral (2006).
lobule (BA 5); inferior parietal regions (BAs 40, 39, and 30. This research was based on M.F.M.’s doctoral dissertation,
the mPFC, which are involved in self-referential
7); the angular gyri (BAs 19 and 39); the inferior frontal which was completed at Dartmouth College. The authors
mental activity (22, 23). In light of behavioral cortices (BAs 10, 47); superior and middle frontal gyri thank H. Gordon, D. Turk, J. Smallwood, T. Heatherton,
evidence suggesting that people are frequently (BAs 8, 9, and 10); and a cluster spanning dorsal medial M. Bar, J. Fugelsang, and three anonymous reviewers for
unaware that their mind is wandering (19, 24), it frontal regions (BAs 8, 9, 10, and 32) (8). These their advice and assistance. C.N.M. was supported by a
may be the case that the daydream frequency distributed foci have temporal coherence and constitute a Royal Society–Wolfson Fellowship; M.F.M. by NIH grant
tightly coupled, organized neural network. R01 NS050614; and D.M.W. by NIH grant MH49127.
scale (14) used in the current investigation 7. P. K. McGuire, E. Paulesu, R. S. Frackowiak, C. D. Frith,
assesses people’s awareness of their mind's Neuroreport 7, 2095 (1996).
wandering rather than their propensity to engage 8. D. A. Gusnard, M. E. Raichle, Nat. Rev. Neurosci. 2, 685 Supporting Online Material
(2001). [Link]/cgi/content/full/315/5810/393/DC1
in SIT.
9. K. A. McKiernan, J. N. Kaufman, J. Kucera-Thompson, Materials and Methods
The purpose of the current inquiry was to ex- J. R. Binder, J. Cognit. Neurosci. 15, 394 (2003). SOM Text
plore how and when the mind generates SIT. A 10. J. D. Teasdale et al., Mem. Cognit. 23, 551 (1995). Figs. S1 to S7
more intractable question, however, is why these 11. R. M. Shiffrin, W. Schneider, Psychol. Rev. 84, 127 Tables S1 to S3
thoughts emerge at all. What is the functional (1977). References
12. K. Christoff, J. M. Ream, J. D. E. Gabrieli, Cortex 40, 623
significance of a system that wanders from its (2004).
current goals (25)? One possibility is that SIT 13. Materials and methods are available as supporting 14 June 2006; accepted 1 December 2006
enables individuals to maintain an optimal level material on Science Online. 10.1126/science.1131295

[Link] SCIENCE VOL 315 19 JANUARY 2007 395