Every tester knows the situation: the test passes, and nobody can say what state the system was in when it started. The precondition goes into the header — „database empty“, „user logged in“ — and is never mentioned again. Until someone tries to verify it.
One of the most cited results of imaging neuroscience in the last twenty years has exactly such a header. It is called the resting state: a person lies in the scanner doing nothing, and from slow fluctuations of a signal that tracks blood oxygenation, maps of the brain’s „networks“ are computed. The claims about those networks rest on the assumption that the measured state is one and the same state — „the brain at rest“. We wanted to know how that precondition is actually specified.
What we did
We took the fifteen papers that carry the most citation weight for the concept (Raichle 2001 through Allen 2012; full texts openly available in PubMed Central) and looked for the five things a tester would want in a test header: the instruction given to the participant, the eyes (closed / open / fixation), the duration, verification of wakefulness, and other conditions (time of day, caffeine, prior task, medication). First a keyword screen over the whole text, then every returned sentence read. „Not found“ below means neither the screen nor the reading found it — not that the authors could not have stated it elsewhere.
What we found
Specified: instruction, eyes, duration. The instruction is stated in all fifteen where found, and it is similar throughout: „lie quietly with your eyes closed and do not fall asleep“, „think of nothing in particular“, „fixate on the crosshair“. Eyes are stated in fourteen: nine closed, five fixation; one paper (Fox 2005) compared all three variants. Duration is stated in ten — from four minutes to 37 minutes 44 seconds.
Not specified: wakefulness. Six papers instruct wakefulness („not fall asleep“, „stay awake“, „remain alert“). None of the fifteen verifies it — no eye tracking, no sleep scoring, no post-scan debriefing that we could find. The one paper that had wakefulness technically under observation (Mantini 2007, simultaneous EEG) does not state that the record was used to exclude anyone drifting off. The field itself knows this is an edge case: Seeley 2007 cites work finding the networks detectable in subjects „presumed to be in the early stages of sleep“; Biswal 2010 lists „arousal, and sleep“ among the known modulators of connectivity.
The condition varies and is pooled. Eyes closed and fixation alternate 9 : 5 across the set and are merged in the large pooled studies (Biswal 2010, 35 sites) — although Allen 2012, in the same literature, writes that eyes open versus closed changes the spectra and the connectivity patterns, and that participants engage in several kinds of mental activity during „rest“ that cannot be specified in advance.
The hypothesis actually under test. Rewrite „the brain at rest“ as it is operationalised and you get: a person lying still in a scanner for 4 to 37 minutes, eyes closed or on a cross, told not to sleep, with wakefulness unverified. Every claim about resting-state networks in these fifteen papers is a claim about that state.
What follows — and what does not
It does not follow that the networks are an artefact of the condition. On the contrary: two papers in the set treat the condition as a variable and test it — Fox 2005 finds the same networks under all three eye conditions, Buckner 2009 compares fixation with a task — and the patterns survive. That is a healthy habit and is credited as such.
What follows is more modest, and familiar to any tester: the result is reproducible across conditions that nobody specified exactly. That is a statement about the stability of a pattern, not about a „state of the brain“. If one goes on to infer what the brain does when it „does nothing“, one is inferring from a header that nobody checked.
In testing we would call this an unverified precondition shared by a whole suite. The field knows the remedy and sometimes applies it (eye tracking, EEG-based sleep scoring in the newer large datasets). The fifteen load-bearing papers do not — and that is exactly the kind of thing we want to be able to say aloud, precisely, and without exaggeration.
References (the fifteen audited papers + cited limits)
- Raichle, M. E. et al. (2001). A default mode of brain function. PNAS, 98(2), 676–682. https://doi.org/10.1073/pnas.98.2.676
- Gusnard, D. A. & Raichle, M. E. (2001). Searching for a baseline. Nature Reviews Neuroscience, 2, 685–694. https://doi.org/10.1038/35094500
- Greicius, M. D. et al. (2004). Default-mode network activity distinguishes Alzheimer’s disease from healthy aging. PNAS, 101(13), 4637–4642. https://doi.org/10.1073/pnas.0308627101
- Beckmann, C. F. et al. (2005). Investigations into resting-state connectivity using independent component analysis. Phil. Trans. R. Soc. B, 360, 1001–1013. https://doi.org/10.1098/rstb.2005.1634
- Fox, M. D. et al. (2005). The human brain is intrinsically organized into dynamic, anticorrelated functional networks. PNAS, 102(27), 9673–9678. https://doi.org/10.1073/pnas.0504136102
- Achard, S. et al. (2006). A resilient, low-frequency, small-world human brain functional network. J. Neurosci., 26(1), 63–72. https://doi.org/10.1523/JNEUROSCI.3874-05.2006
- Damoiseaux, J. S. et al. (2006). Consistent resting-state networks across healthy subjects. PNAS, 103(37), 13848–13853. https://doi.org/10.1073/pnas.0601417103 · PMC1564249
- Seeley, W. W. et al. (2007). Dissociable intrinsic connectivity networks for salience processing and executive control. J. Neurosci., 27(9), 2349–2356. https://doi.org/10.1523/JNEUROSCI.5587-06.2007
- Mantini, D. et al. (2007). Electrophysiological signatures of resting state networks in the human brain. PNAS, 104(32), 13170–13175. https://doi.org/10.1073/pnas.0700668104
- Sridharan, D., Levitin, D. J., Menon, V. (2008). A critical role for the right fronto-insular cortex in switching between central-executive and default-mode networks. PNAS, 105(34), 12569–12574. https://doi.org/10.1073/pnas.0800005105
- Buckner, R. L. et al. (2009). Cortical hubs revealed by intrinsic functional connectivity. J. Neurosci., 29(6), 1860–1873. https://doi.org/10.1523/JNEUROSCI.5062-08.2009
- Honey, C. J. et al. (2009). Predicting human resting-state functional connectivity from structural connectivity. PNAS, 106(6), 2035–2040. https://doi.org/10.1073/pnas.0811168106
- Murphy, K. et al. (2009). The impact of global signal regression on resting state correlations. NeuroImage, 44(3), 893–905. https://doi.org/10.1016/j.neuroimage.2008.09.036
- Biswal, B. B. et al. (2010). Toward discovery science of human brain function. PNAS, 107(10), 4734–4739. https://doi.org/10.1073/pnas.0911855107
- Brookes, M. J. et al. (2011). Investigating the electrophysiological basis of resting state networks using magnetoencephalography. PNAS, 108(40), 16783–16788. https://doi.org/10.1073/pnas.1112685108
- Allen, E. A. et al. (2014; online 2012). Tracking whole-brain connectivity dynamics in the resting state. Cerebral Cortex, 24(3), 663–676. https://doi.org/10.1093/cercor/bhs352
Our reference documentation (working .md version; TeX typesetting in the manner of a scientific article in preparation): NA-D-01 neuro-assessment/layerD/CONDITION-AUDIT-rest-v0.1.md (per-paper table, versions, check state); charter NA-00 neuro-assessment/CHARTER-v0.2-EN.md (§3.7). The documentation currently lives in a private working repository.
Forthcoming. This text is a short output of a larger assessment of the basis of imaging neuroscience that we conduct under a public charter (NA-00): corpus mapping, physics of measurement, the statistics actually used, and a final synthesis with an independent second check. The full reference documentation — per-paper tables, data with checksums, inference reconstructions, and a typeset version with the apparatus of a scientific article — will be released in a public repository and announced here once it has passed the second check. Until then: every number in this text has its row in the documentation, and every row has its source. All DOIs verified via OpenAlex 2026-10-04.
Method and limits. One reader; keyword screen plus reading of returned sentences; PMC full texts; no paper after 2012 (later practice not represented). Per-paper table and document version: neuro-assessment/layerD/CONDITION-AUDIT-rest-v0.1.md (NA-D-01). Part of a wider assessment of the basis of imaging neuroscience whose charter and rules are public (NA-00).