Method
Eight questions for thinking in time.
Every advance in the Observatory can be analysed with the same frame. It makes analyses comparable, separates facts from interpretations and shows what can still be decided.
- 01
What is claimed?
Describe the result in the authors' words, not the headline's.
- 02
What is the level of evidence?
Peer-reviewed, preprint, technical report, company statement or press report.
- 03
Who did it, when, and where was it published?
Team, institution, declared funding and date of publication.
- 04
Which concepts are at stake?
Mental privacy, consent, personal identity, agency, moral status, responsibility, risk and benefit.
- 05
Who is affected?
Participants, patients, users, vulnerable groups and people who took no part but feel the effects.
- 06
What scale, and how reversible?
One lab or millions of people? Can it still be undone, or are the code and data already circulating?
- 07
What window for intervention remains?
Standards, ethics review, regulation, technical design or teaching: where, and for how long, influence is still possible.
- 08
What remains uncertain?
Replication, the limits of the method and the questions the evidence does not answer.
Levels of evidence
- Peer-reviewed — published in a journal or conference with independent review.
- Preprint — publicly available, not yet peer-reviewed.
- Technical report — research published by an organisation, without formal peer review.
- Company statement — a result reported by those who produced it, without external verification.
- Press report — news coverage without access to the primary source; used only while the source is unavailable.
Types of contribution
- Science — results and methods with verifiable sources.
- Commentary — a reasoned analysis of a result, with named authorship.
- Opinion — a signed personal position, presented as such.
Corrections and retractions
When a source is corrected or retracted, the entry is updated with the date and the reason, and the previous version stays identified. Anyone can propose a correction through info@neuronet.pt.
Example: reading what the brain replays during sleep
- What is claimed: from the EEG of sleeping people, a model identifies which of 15 categories of learned material is being reactivated.
- Evidence: peer-reviewed (Neuron, 17 April 2026).
- Who and when: academic research published on 17 April 2026.
- Concepts: mental privacy and consent.
- Who is affected: participants in sleep studies and, if the method generalises, anyone whose brain activity is recorded during sleep.
- Scale and reversibility: today, one lab and categories defined by the researchers; a published method cannot be withdrawn from circulation.
- Window for intervention: consent rules for brain data collected during sleep and for consumer devices.
- What remains uncertain: whether it works outside the lab and with content that was not defined in advance.

