Science

Current Picks

A Spectrograph’s 2013 Calibration File Reshaped One Survey’s Galaxy Mass Estimates

By Alice Chen / Aug 10, 2026

A 2013 spectrograph calibration file introduced a subtle wavelength offset that silently shifted galaxy mass estimates in a major survey. New work traces the glitch and urges better calibration hygiene.

The Replication Rift Over Priming’s Invisible Threshold Split One Psych Lab’s 2012 Data

By Renu Shah / Aug 10, 2026

A 2012 priming dataset yielded opposite conclusions in one lab, exposing how procedural choices like stimulus timing and sample size shape results. A look at the replication rift and what practitioners now do differently.

Intervention Fatigue: How One Lab’s 2009 Framing Reversed Its Own Nudge Findings

By Karim Osman / Aug 10, 2026

A lab's 2009 nudge study showed a 4% uptake lift. Replications found near-zero effects. New research reveals intervention fatigue may explain the reversal.

A Drosophila Rearing Room’s 22°C Setpoint, Not Its Genotype, Drove One Lab’s Circadian Output

By Karim Osman / Aug 10, 2026

A circadian mutant phenotype in fruit flies vanished when another lab replicated the study, traced to a 22°C vs 25°C rearing room difference. A methodology breakdown.

A 1993 Fortran Routine Resurfaced to Recompile a Turbulence Simulator’s Core Loop

By Jonas Eriksen / Aug 10, 2026

A 1993 Fortran routine resurfaced to recompile a turbulence simulator’s core loop, revealing the challenges of preserving scientific software across decades of compiler and hardware changes.

The Isotope Ratios in 40 Penguin Feathers Flipped a Foraging Model After One Lab Rechecked Museum Dates

By Karim Osman / Aug 10, 2026

A lab recheck of museum specimen dates on 40 penguin feathers shifted a foraging model, showing how isotope ecology depends on provenance.

A Mouse's Two-Second Lick Delay Forced One Lab to Rewrite Its Reward Circuit Model

By Alice Chen / Aug 10, 2026

A two-second delay in a mouse's licking behavior upended a lab's dopamine reward model, prompting a methodological overhaul and a revised understanding of neural timing.

A 1912 Dust-Collection Log From Krakatoa’s Ash Reshaped One Volcanologist’s Eruption Timeline

By Jonas Eriksen / Aug 10, 2026

A forgotten 1912 dust-collection log from Krakatoa's ash led volcanologist Helen Whitfield-Brown to question the 1883 eruption timeline, suggesting earlier activity.

A Two-Decade Ice Core Dispute Settled When One Lab Reanalyzed Its Own Boron Isotope Data

By Karim Osman / Aug 10, 2026

A boron isotope claim from Antarctic ice cores survived 20 years of dispute until the original lab reanalyzed its own samples, revealing a calibration error that shrank the signal.

A Platform’s 1999 Archive Format Still Dictates How Two CERN Labs Share Code

By Karim Osman / Aug 10, 2026

Two CERN labs still rely on tar archives for code sharing. The hidden costs, why it persists, and what a modern replacement would need.

A Kitchen Timer’s Ring Tone, Not Its Duration, Drove One Lab’s Procrastination Puzzle

By Alice Chen / Aug 10, 2026

A lab found that changing a kitchen timer's alarm sound, while keeping duration constant, shifted procrastination rates. The finding highlights how sensory details often ignored in behavioral experiments can be critical.

The Shapley Supercluster’s 2017 Velocity Map Fractured When One Analyst Replotted Its Redshifts

By Renu Shah / Aug 10, 2026

A 2017 velocity map of the Shapley Supercluster claimed a 2,000 km/s flow. A reanalysis of the redshifts cut that by 15-20%. Here's why astronomers disagree.

Millikan’s Oil-Drop Apparatus Costs Now Outweigh Its Classroom Replication Value

By Alice Chen / Aug 10, 2026

Physics departments weigh the high cost of Millikan oil-drop apparatus against its pedagogical value, as replication failures and cheaper alternatives prompt a rethink.

A Standardized Fear-Potentiated Startle Protocol Flipped One Lab’s Threat Generalization Verdict

By Renu Shah / Aug 10, 2026

A lab's threat generalization finding reversed after adopting a standardized fear-potentiated startle protocol, highlighting how procedural choices shape neuroscience conclusions.

A Lithium Anode’s 2019 Promise Fractured When Four Groups Replotted Impedance Spectra

By Alice Chen / Aug 10, 2026

A 2019 preprint claimed 99.9% Coulombic efficiency for a lithium anode, but four groups re-plotting the impedance spectra found inconsistencies that fractured the promise, revealing deep pitfalls in battery characterization.

A Fortran 77 Routine From 1987 Still Gates Which Climate Models Get Archived

By Jonas Eriksen / Aug 10, 2026

A 1987 Fortran 77 routine still gates climate model archives, enforcing rigid metadata rules. Scientists debate whether it preserves trust or stifles innovation.

Behavioral Decisions Shifted When One Lab Swapped Its Survey From Likert Scales to Sliders

By Karim Osman / Aug 10, 2026

How a simple switch from Likert scales to sliders changed survey answers, shifted effect sizes, and rippled across psychology, economics, and political science.

The Physics Vault That Charged Labs for Peer Review Instead of Paying Them

By Alice Chen / Aug 10, 2026

How preprint servers and article processing charges inverted funding flows in physics, shifting costs to researchers and distorting incentives.

A Grant Reviewer’s Scoring Rubric Replaced One Lab’s Behavioral Dataset

By Renu Shah / Aug 10, 2026

Grant review rubrics favor feasibility, nudging labs to shrink behavioral datasets. Explore how funding incentives reshape scientific depth.

A Deep-Water Mooring’s Battery Budget Decided Which Paleoclimate Proxy Got Published

By Alice Chen / Aug 10, 2026

How a deep-water mooring's battery budget determined which paleoclimate proxy got published, shaping scientific consensus through infrastructure economics.