← What we believe Evidence library

Home · Sources · MEDIA-009

Reading from paper, computers, and tablets in the first grade: The role of comprehension monitoring

Authors: Florit, E.; De Carli, P.; Rodà, A.; Cain, K.; Mason, L. · Year: 2025 · Tier: RCT · DOI/URL: https://doi.org/10.1016/j.caeo.2025.100243

Primary school (grades 1–6) Paper vs screen Reading and writing Analog vs digital

Links to claims

Limits (1)

ClaimBasis
C3Grade-1 within-subjects: no screen-inferiority main effect; tablet ≥ paper on some superficial descriptive measures. Narrows average paper advantage away from beginner readers.

Review state: first-pass draft, pending human review.

Source note (Swedish, verbatim from the Kunskapsbank)

Syntes: Florit et al. (2025) — first-grade paper vs computer vs tablet

Källa: Florit, E., De Carli, P., Rodà, A., Cain, K., & Mason, L. (2025). Reading from paper, computers, and tablets in the first grade: The role of comprehension monitoring. Computers and Education Open, 8, 100243. https://doi.org/10.1016/j.caeo.2025.100243

Metadata

Finding

  1. No screen-inferiority main effect of medium on main-point, literal, or inferential comprehension (ps > .15).
  2. For descriptive texts, tablet > paper on main-point questions (p = .01, d = −0.57 favouring tablet) and on literal questions (p = .01, d = −0.59); tablet also beat computer on main-point.
  3. For narrative texts, paper beat computer on literal questions (p = .02, d = 0.51); other narrative medium contrasts ns.
  4. Inferential comprehension: narrative > descriptive independent of medium; no medium × genre interaction at the inferential level.
  5. Comprehension monitoring related to comprehension across media; relation not moderated by medium.

Mechanism

Authors argue shallowing / screen-inferiority accounts do not hold for beginner readers with short linear static texts; familiarity with handhelds and lower accumulated leisure-screen habits may matter.

Limits

Small single-school sample; short grade-1 texts (~100 words); individual quiet sessions; linear static digital texts (not hypertext); calibration-as-overconfidence not the metacognitive measure (inconsistency detection instead).

This claim dies if …

(For C3) Meta/RCT in the relevant older age band where screen ≥ paper for deep comprehension with equal time and the calibration gap disappears — Florit alone does not kill the university/older average.

Swedish implication

Do not export the adult paper-advantage average unchanged to year-1 classrooms; for short linear texts with beginners, medium is not the lever Delgado/Clinton describe for undergraduates.

Links

Metadata as recorded

ID
MEDIA-009
Tier
RCT
Closed-book follow-up
n/a (comprehension assessed on same medium as reading; no GenAI)
Titel
Reading from paper, computers, and tablets in the first grade: The role of comprehension monitoring
Författare
Florit, E.; De Carli, P.; Rodà, A.; Cain, K.; Mason, L.
År
2025 (online 16 Jan 2025)
Källa/DOI
https://doi.org/10.1016/j.caeo.2025.100243
URL
https://www.sciencedirect.com/science/article/pii/S2666557325000023
Typ
within-subjects experiment (N=58 first graders; paper vs laptop vs tablet; narrative + descriptive)
Åldersgrupp
grundskola-lag (grade 1, Mage ≈ 6.8)
Teman
papper-vs-skarm; lasning-och-skrivande; analogt-vs-digitalt
Sparade fil
Florit-etal-2025-first-grade-paper-screen.pdf (OA CC BY-NC-ND)
Hämtad
2026-09-27
Paywall
OA (Computers and Education Open)

Original files held in the project (not published): Florit-etal-2025-first-grade-paper-screen.pdf