Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 64 of the 65 references it cites were found in a registry or an index; 13 of the 52 sources its claims cite were checked for support.

Rethinking pre-training: cognitive load implications for learners with varying prior knowledge

Open access
Gorbunova, Anna, Kapuza, Anastasiia, Chen, Ouhao, Costley, Jamie · 2025 · Frontiers in Psychology

Abstract

This study examines how prior knowledge and pre-training relate to cognitive load during problem-solving. Grounded in cognitive load theory, it investigates whether pre-training facilitates learning by reducing cognitive load or imposes redundant information for learners with higher prior knowledge. In an experiment with 136 university students, pre-training was implemented through concept maps and a glossary introducing essential terms and procedures before problem-solving. Results revealed that learners with higher prior knowledge experienced lower intrinsic and extraneous load and higher germane load during problem-solving compared to learners with lower prior knowledge, suggesting enhanced schema refinement rather than redundancy. Pre-training consistently reduced extraneous load across all learners, including those with higher prior knowledge, challenging the expected expertise reversal effect. While learners with lower prior knowledge did not show significant reductions in intrinsic load, they benefited from decreased extraneous load during problem-solving. These findings highlight the value of pre-training as an instructional strategy and underscore the importance of aligning instructional design with learners’ existing knowledge.

Overview

CoChat Summary

TL;DR

This experiment with 136 university students examined whether pre-training via concept maps and a glossary reduces cognitive load differently for learners with lower versus higher prior knowledge. It found higher prior knowledge was linked to lower intrinsic and extraneous load and higher germane load, while pre-training reduced extraneous load for all learners, contrary to expertise reversal/redundancy expectations.

Key findings

  • Learners with higher prior knowledge reported lower intrinsic load, F(1, 134) = 5.51, p = 0.02, η² = 0.03; lower extraneous load, F(1, 134) = 5.41, p = 0.02, η² = 0.03; and higher germane load, F(1, 134) = 8.07, p < 0.01, η² = 0.04, compared with lower prior knowledge learners.
  • Pre-training reduced extraneous load from pre-training to problem-solving for both lower prior knowledge learners (contrast = −1.21, p < 0.01) and higher prior knowledge learners (contrast = −1.13, p < 0.01); no prior knowledge × phase interactions were significant for ICL, ECL, or GCL (all p > 0.19).
  • For higher prior knowledge learners, germane load significantly increased during problem-solving (contrast = 0.63, p < 0.01), contradicting the expected redundancy effect; intrinsic load was lower during pre-training (p = 0.05). For lower prior knowledge learners, ICL and GCL changes were non-significant (p = 0.15 and p = 0.10).
  • Mean problem-solving score was 7.51 (SD = 2.43); lower prior knowledge group mean = 5.97 (SD = 2.79), higher prior knowledge group mean = 8.03 (SD = 2.08).

Methods

  • Participants were 136 second-year graduate students (73.6% women) from over 15 academic programs at a selective Russian university, recruited during a synchronous online lecture. Prior knowledge was measured with 8 multiple-choice questions (mean = 6.31, SD = 1.40); a median split of 6/8 classified 26.5% as lower prior knowledge and 73.5% as higher prior knowledge.
  • Pre-training consisted of a pre-constructed concept map and glossary on “Cognitive Load,” followed by problem-solving tasks. Cognitive load was measured with Krieglstein et al. (2023) scales for ICL, ECL, and GCL (5 items each, 0–9) after pre-training and after problem-solving.
  • Analyses used repeated-measures ANOVAs separately for ICL, ECL, and GCL, with phase (pre-training vs. problem-solving) as a within-subjects factor and prior knowledge group as a between-subjects factor; afex Type III sums of squares, emmeans with Bonferroni adjustment, and generalized eta squared effect sizes.

Limitations

  • The paper states the prior knowledge test was relatively easy, producing uneven lower/higher prior knowledge groups; the specific pre-training format (concept maps plus glossary) may not generalize, and the two components were not isolated. It also notes tasks were well-structured/rule-using rather than ill-structured, GCL measurement remains debated, and only immediate outcomes were assessed without delayed retention/transfer tests.
  • The paper states future work should compare pre-training against a control condition; the absence of such a control condition is evident from the design and limits causal isolation of pre-training effects from other instructional exposure.
  • The prior knowledge test had low internal consistency (Cronbach’s α = 0.44), which the authors describe as expected for short dichotomous tests but which is an evident measurement limitation; unequal group sizes also reduce power for the smaller lower-prior-knowledge group, as the paper acknowledges.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.3389/fpsyg.2025.1628047

Cites

64 of 65 references verified0 citation errors · 1 not checkable

Statistics

Figures

Figure 1 An overview of the procedure.
Figure 1 An overview of the procedure.CC BY · PubMed Central

Verification

Claims the paper makes

11 Valid · 2 Not valid · 28 Not checked · 53 Own results and interpretations

30 more claims

Create a free account to see all 94 claims and to verify the 2 references still open.

See all claims — free account
  • It may state two separate claims as one. 37
  • Probably not a substantial claim. 31
  • The quote only partly states this claim. 4

Create a free account to see these claims.

Statistics recomputed

Every reported test result, recomputed from its test statistic and degrees of freedom. A result shows as not adding up only once an independent check has confirmed it.

  • ConsistentF(1, 134) = 5.51reported p = .02 · recomputed p = .020

    “F(1, 134) = 5.51, p = .02”

  • ConsistentF(1, 134) = 5.41reported p = .02 · recomputed p = .022

    “F(1, 134) = 5.41, p = .02”

  • ConsistentF(1, 134) = 8.07reported p < .01 · recomputed p = .005

    “F(1, 134) = 8.07, p < .01”

3 of 3

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

1 of 1

Citations

Cites

Show all 63 references

Research works printed but not linked: 1

  1. Singmann, 2021, Afex: Analysis of factorial experiments (R package version 1. 0–1)

Cited by

See all 14

Metadata

Authors
Gorbunova, A., Kapuza, A., Chen, O., & Costley, J.
Affiliations
  • National Research University Higher School of Economics (RU)
  • University of Leeds (GB)
  • United Arab Emirates University (AE)
Venue
Frontiers in Psychology
Published
2025-08-07
Volume
16
Pages
1628047-1628047
ISSN
1664-1078 (ISSN-L)
OpenAlex
W4413034996
Cited by
14
References
63
Citations by year
2026: 13 · 2025: 1
FWCI
14.66
Type
journal article
Language
en
Open access
Locations
  • Frontiers in Psychology (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
National Research University Higher School of Economics
Topics
  • Social Sciences › Psychology › Experimental and Cognitive Psychology › Visual and Cognitive Learning Processes
  • Social Sciences › Psychology › Developmental and Educational Psychology › Innovative Teaching and Learning Methods
  • Physical Sciences › Computer Science › Artificial Intelligence › Intelligent Tutoring Systems and Adaptive Learning
Keywords
pre-training, cognitive load, prior knowledge, expertise reversal effect, concept maps, glossary, problem solving, germane load, extraneous load, intrinsic cognitive load, instructional design

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Rethinking pre-training: cognitive load implications for learners with varying prior knowledge. https://app.cochat.ai/papers/10.3389/fpsyg.2025.1628047
The paper
Gorbunova, A., Kapuza, A., Chen, O., & Costley, J. (2025). Rethinking pre-training: cognitive load implications for learners with varying prior knowledge. Frontiers in Psychology. https://doi.org/10.3389/fpsyg.2025.1628047

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.