REVIEW LANE — 214 PENDING, ONE AT A TIME

A stack, not a mailbox. One card is in hand, the rest are underneath, and the whole apparatus is built so a decision costs one key. Ranking sorts the stack and never hides anything — what is underneath is listed in order beneath the card, and the reserved outside-the-usual slot is placed into that order by the run rather than earned by it. Grouping picks the mental context: one Scout, one question, or one topic at a time.

consolidation-vault — scouts
⌘K
Scout Queue
{{ totalLabel }}
{{ l.label }}
GROUP BY
{{ g.label }}
{{ g.label }} {{ g.n }}
MUTED
3 rules · 47 items held
Muted items are kept in their own view, never dropped — a rule that is quietly wrong is findable.
{{ groupTitle }} {{ groupSub }} SORT
{{ s.label }}
{{ card.whyTag }} {{ card.why }} {{ card.pos }}
{{ card.title }}
{{ f.k }} {{ f.v }}
{{ k.label }}
{{ card.abstract }}
abstract as published · nothing on this card is generated
{{ a.key }} {{ a.label }}
{{ lastAction }}
UNDERNEATH {{ underneathLabel }} {{ b.label }}
{{ u.rank }} {{ u.title }} {{ u.venue }} {{ u.whyShort }} {{ u.age }}
{{ underneathNote }}
WHY THIS IS IN REVIEW
{{ r.label }} {{ r.val }}
rank {{ card.rank }} of {{ groupCount }} · sort only, nothing is filtered out
ORIGIN
{{ card.origin }}
{{ card.matched }}
OUTSIDE THE USUAL
One slot per run is reserved for well-corroborated work from authors and venues you have never accepted from. It is placed into the order, not appended to the end.
{{ oddOne.title }}
{{ oddOne.meta }}
MUTE FROM HERE
author venue keyword
Muting moves matching items to the muted view and shows how many it caught, so an over-broad rule announces itself.
THIS SESSION
{{ session }}
Review items skipped three times drop back to Skim on their own. The queue shortens whether or not you decide.
A acceptR rejectD deferJ nextM mute…⇧R reject group {{ footer }}

SKIM FEED — SCROLLING PAST IS THE INTERACTION

The same proposals, different physics. No cards, no borders, no stack, no accept or reject: a continuous column of single lines that thins out as it ages and disappears on its own after thirty days. The only affordance is promotion upward — anything you touch moves into Review. Nothing here is owed a decision, and the design says so by giving you nothing to press.

skim
Skim {{ skimLabel }} ↑ on any line moves it to Review · nothing else to do
{{ s.title }} {{ s.authors }} {{ s.venue }} {{ s.age }}
{{ s.dividerLabel }}
older items fade and are removed after 30 days · none of this is a backlog

ONE PROPOSAL, IN FULL — AND WHAT IS MISSING

Left: a journal article with every field populated. Right: a lab blog post with no venue, no keywords and no DOI. Both are valid cards. Missing fields are printed as missing in the same slot they would occupy, so the shape of the card tells you what kind of thing it is before you read a word of it — and nothing is inferred to fill a gap.

MATCHED 2 prior accepts from Schreiner · corroborated by 3 scouts · matches an open question
Cue-locked reinstatement during slow-wave sleep predicts overnight retention independently of encoding strength
AUTHORSSchreiner, Petzka, Staudigl, Rasch
DATE11 Sep 2026
VENUENature Communications 17
DOI10.1038/s41467-026-4471
targeted memory reactivation slow-wave sleep EEG declarative memory
Whether the overnight retention benefit reflects consolidation during sleep or differences in encoding at learning remains contested, in part because the two are confounded in correlational designs. Here we cue a subset of learned associations during slow-wave sleep and show that trial-level reinstatement strength predicts retention for cued items within participants, while encoding-phase measures predict retention equally across cued and uncued sets. The dissociation holds after matching sets on pre-sleep performance.
abstract as published·open in browser ↗·found by scout tmr-watch, 12 Sep
Aaccept → source stub Rreject Ddefer
WATCHED PAGE new post on a lab page you follow
What we got wrong about spindle density — a postmortem on our 2021 paper
AUTHORSD. Denis
DATE16 Sep 2026
VENUE— not stated
DOI— none
KEYWORDS— none supplied
We re-ran the 2021 analysis after a reader pointed out that our density measure was sensitive to the artefact rejection threshold. The direction of the effect survives; the magnitude does not.
first paragraph in place of an abstract — the page has none
source: sleeplab.example.org/blog · found by scout density-watch
Aaccept → source stub a stub with no DOI is still a stub

THE TRIAGE LOOP

Four keys carry almost all the traffic, and none of them opens anything. The card is replaced in place, the decision is reversible for as long as the session lasts, and accepting writes a Source stub — a bibliographic record and a link, no file.

KEYS
{{ k.key }}
{{ k.name }} {{ k.what }}
WHAT “ACCEPT” WRITES
sources/schreiner-2026-cue-locked.md
type: stub
title: Cue-locked reinstatement during…
authors: Schreiner, Petzka, Staudigl, Rasch
venue: Nature Communications 17
doi: 10.1038/s41467-026-4471
url: https://…
found_by: scout tmr-watch
matched: is awake replay measurable…
accepted: 2026-09-19
no file attached
The stub appears in the Vault immediately and in the matched question's supporting column as unsorted. Downloading the PDF is a separate decision you make later, if the stub survives contact with your reading list.
WHY THIS IS NOT A MAIL CLIENT
No read/unread, no threads, no sender column, no search-first layout. The unit is a stack you work down, the default posture is one card at hand, and the two lanes have different physics: Review is a stack that shortens, Skim is a river that flows past. Volume is handled by grouping and ranking before you arrive, not by you filing faster.