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

The DOI matches its Crossref record; 83 of the 83 references it cites were found in a registry or an index; none of the 104 sources its claims cite is checked for support yet.

Learning new word meanings from story reading: the benefit of immediate testing

Open access

Published version of a preprint

Abstract

This study investigated how word meanings can be learned from natural story reading. Three experiments with adult participants compared naturalistic incidental learning with intentional learning of new meanings for familiar words, and examined the role of immediate tests in maintaining memory of new word meanings. In Experiment 1, participants learned new meanings for familiar words through incidental (story reading) and intentional (definition training task) conditions. Memory was tested with cued recall of meanings and multiple-choice meaning-to-word matching immediately and 24 h later. Results for both measures showed higher accuracy for intentional learning, which was also more time efficient than incidental learning. However, there was reasonably good learning from both methods, and items learned incidentally through stories appeared less susceptible to forgetting over 24 h. It was possible that retrieval practice at the immediate test may have aided learning and improved memory of new word meanings 24 h later, especially for the incidental story reading condition. Two preregistered experiments then examined the role of immediate testing in long-term retention of new meanings for familiar words. There was a strong testing effect for word meanings learned through intentional and incidental conditions (Experiment 2), which was non-significantly larger for items learned incidentally through stories. Both cued recall and multiple-choice tests were each individually sufficient to enhance retention compared to having no immediate test (Experiment 3), with a larger learning boost from multiple-choice. This research emphasises (i) the resilience of word meanings learned incidentally through stories and (ii) the key role that testing can play in boosting vocabulary learning from story reading.

Overview

CoChat Summary

TL;DR

Three experiments with adult native British English readers found that new meanings for familiar words can be learned incidentally from stories, though intentional definition training produced higher immediate accuracy and was more time-efficient. Immediate testing after learning produced a robust testing effect, improving 24-h retention for both incidentally and intentionally learned meanings, with either cued recall or multiple-choice sufficient.

Key findings

  • In Experiment 1, immediate cued recall was higher for intentional learning (85.2%) than incidental story learning (62.1%; χ²(1) = 14.32, p < .001); multiple-choice accuracy was 96.0% vs. 83.9%, a non-significant difference (p = .056). Intentional training was also faster (M = 5 min 28 s) than story reading (M = 12 min 30 s), t(39) = 11.43, p < .001.
  • After 24 h, intentional multiple-choice accuracy dropped significantly (96.0% to 87.9%; p = .014), while incidental items showed no significant forgetting (83.9% to 83.1%; p = .750); incidental cued recall improved slightly from 62.1% to 70.1%, though the interaction was non-significant.
  • Experiment 2 found a testing effect: immediate testing improved 24-h cued recall, boosting incidental learning by 28.8% and intentional learning by 10.6%; the main effect of test type was significant (χ²(2) = 25.78, p < .001), as was the delayed second vs. delayed first test comparison (χ²(1) = 18.83, p < .001).
  • Experiment 3 showed either immediate cued recall or multiple-choice alone was sufficient to produce a significant testing effect, with non-significantly larger benefits from multiple-choice. For delayed cued recall of previously tested items, immediate multiple-choice gave 58.3% vs. 49.5% for immediate cued recall (interaction p = .074); immediate multiple-choice significantly boosted delayed multiple-choice (p = .001), while immediate cued recall did not (p = .069), indicating cross-task transfer.

Methods

  • Three online experiments (Qualtrics) with monolingual native British English adults recruited via Prolific: Experiment 1 N = 40 (31 completed the 24-h follow-up), Experiment 2 N = 99, Experiment 3 N = 98. Participants learned artificial new meanings for 16 familiar English nouns, with eight exposures per item in incidental story-reading and intentional definition plus two-alternative multiple-choice training conditions.
  • Memory was assessed with cued recall of meanings and multiple-choice meaning-to-word matching immediately and 24 h later. Experiment 2 manipulated whether items were immediately tested vs. first tested at 24 h; Experiment 3 manipulated immediate test method (cued recall vs. multiple-choice) between participants.
  • Data were analysed with logistic mixed-effects models in R using lme4; Experiments 2 and 3 were preregistered.

Limitations

  • Stated: the incidental and intentional learning paradigms differed in several ways (attention to meanings, spacing, and possible internal testing/feedback in intentional training), so the cause of the intentional advantage cannot be firmly identified.
  • Stated: key interactions in Experiment 2 (less forgetting and a larger testing effect for incidental items) were non-significant; the authors call for replication and designs matching initial performance. Experiment 2’s immediate vs. delayed multiple-choice results were not comparable because the immediate test had four alternatives and the delayed test had eight, a deviation from preregistration.
  • Evident: participants were adult monolingual native British English speakers learning artificial new meanings in custom short stories, so findings may not generalise to children, second-language learners, or real-world vocabulary. Attrition and exclusions were also substantial (e.g., Experiment 2 had 36 non-returners; Experiment 3 had 18 non-returners plus 35 comprehension exclusions), and online testing may limit experimental control.

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

Integrity signals

Registry

Crossref · 10.7717/peerj.11693

Cites

83 of 83 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Experiment 1. Mean percentage of correct responses on the cued recall test for each learning condition, when tested on day one (immediately after learning) and 24 h later (N = 31). The LME analyses were carried out on the raw binary accuracy data, however mean percentage accuracy data are displayed in the graphs. Error bars show standard errors for the means, adjusted for the within-participants design (Cousineau, 2005).
Figure 1 Experiment 1. Mean percentage of correct responses on the cued recall test for each learning condition, when tested on day one (immediately after learning) and 24 h later (N = 31). The LME analyses were carried out on the raw binary accuracy data, however mean percentage accuracy data are displayed in the graphs. Error bars show standard errors for the means, adjusted for the within-participants design (Cousineau, 2005).CC BY · PubMed Central
Figure 2 Experiment 1. Mean percentage of correct responses on the multiple-choice test for each learning condition, when tested on day one (immediately after learning) and day two (24 h later; N = 31). Error bars show standard errors for the means, adjusted for the within-participants factor of learning condition (Cousineau, 2005).
Figure 2 Experiment 1. Mean percentage of correct responses on the multiple-choice test for each learning condition, when tested on day one (immediately after learning) and day two (24 h later; N = 31). Error bars show standard errors for the means, adjusted for the within-participants factor of learning condition (Cousineau, 2005).CC BY · PubMed Central
Figure 3 Experiment 2. Mean percentage of correct responses in the cued recall test for each learning condition and for the three different test types in the experiment. Error bars show standard error of the means adjusted for the within-participants design (Cousineau, 2005).
Figure 3 Experiment 2. Mean percentage of correct responses in the cued recall test for each learning condition and for the three different test types in the experiment. Error bars show standard error of the means adjusted for the within-participants design (Cousineau, 2005).CC BY · PubMed Central
Figure 4 Experiment 2. Mean percentage of correct responses in the multiple-choice test for each learning condition and for the three different test types in the experiment. Note that the results from the immediate test (with lighter shading) are not comparable to those from the two delayed test types due to an underlying difference in test difficulty. Error bars show standard error of the means adjusted for the within-participants design (Cousineau, 2005).
Figure 4 Experiment 2. Mean percentage of correct responses in the multiple-choice test for each learning condition and for the three different test types in the experiment. Note that the results from the immediate test (with lighter shading) are not comparable to those from the two delayed test types due to an underlying difference in test difficulty. Error bars show standard error of the means adjusted for the within-participants design (Cousineau, 2005).CC BY · PubMed Central
Figure 5 Experiment 3. Mean percentage of correct responses in the cued recall test measured at the delayed test for participants whose immediate test was cued recall, and for those whose immediate test was multiple-choice when items were or were not previously tested. Error bars show standard error of the means adjusted for the within-participants factor of whether items were or were not previously tested (Cousineau, 2005).
Figure 5 Experiment 3. Mean percentage of correct responses in the cued recall test measured at the delayed test for participants whose immediate test was cued recall, and for those whose immediate test was multiple-choice when items were or were not previously tested. Error bars show standard error of the means adjusted for the within-participants factor of whether items were or were not previously tested (Cousineau, 2005).CC BY · PubMed Central
Figure 6 Experiment 3. Mean percentage of correct responses in the multiple-choice test measured at the delayed test for participants whose immediate test was cued recall, and for those whose immediate test was multiple-choice, when items were or were not previously tested. Error bars show standard error of the means adjusted for the within-participants factor of whether items were or were not previously tested (Cousineau, 2005).
Figure 6 Experiment 3. Mean percentage of correct responses in the multiple-choice test measured at the delayed test for participants whose immediate test was cued recall, and for those whose immediate test was multiple-choice, when items were or were not previously tested. Error bars show standard error of the means adjusted for the within-participants factor of whether items were or were not previously tested (Cousineau, 2005).CC BY · PubMed Central

Verification

Claims the paper makes

0 Valid · 0 Not valid · 68 Not checked · 187 Own results and interpretations

107 more claims

Create a free account to see all 255 claims and to verify the 96 references still open.

See all claims — free account
  • It may state two separate claims as one. 73
  • Probably not a substantial claim. 101
  • 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.

  • Consistentt(39) = 11.43reported p < .001 · recomputed p < .001

    “t(39) = 11.43, p < .001”

  • Consistentχ²(1) = 14.32reported p < .001 · recomputed p < .001

    “χ2(1) = 14.32, p < .001”

  • Consistentχ²(1) = 1.23reported p = .268 · recomputed p = .267

    “χ2(1) = 1.23, p = .268”

  • Consistentχ²(1) = 0.83reported p = .362 · recomputed p = .362

    “χ2(1) = 0.83, p = .362”

  • Consistentχ²(1) = 1.57reported p = .21 · recomputed p = .210

    “χ2(1) = 1.57, p = .210”

5 of 63 · the rest aren’t loaded

58 more recomputed tests

Create a free account to see every test the Stats check recomputed.

See all tests — free account

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 83 references

Research works printed but not linked: 2

  1. Hulme, 2017, The impact of test method on the testing effect in long-term retention of new word meanings learned incidentally from stories, Preregistration
  2. Lehmann, 2007, Is intentional or incidental vocabulary learning more effective?, The International Journal of Foreign Language Teaching, 3, 23

Cited by

See all 23

Metadata

Authors
Hulme, R. C., & Rodd, J. M.
Affiliations
Aston University (GB)
Venue
PeerJ
Publisher
PeerJ
Published
2021-08-10
Volume
9
Pages
e11693
ISSN
2167-8359 (ISSN-L)
OpenAlex
W3163865001
Cited by
23
References
83
Citations by year
2026: 5 · 2025: 3 · 2024: 4 · 2023: 7 · 2022: 3 · 2021: 1
FWCI
1.69
Type
journal article
Language
en
Open access
Locations
  • PeerJ (publishedVersion, cc-by)
  • PsyArXiv (OSF Preprints) (submittedVersion, cc-by)
  • PubMed (publishedVersion)
  • PubMed Central (publishedVersion, cc-by)
  • Aston Publications Explorer (Aston University) (acceptedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • UCL Discovery (University College London) (submittedVersion)
  • PsyArXiv (OSF Preprints) (submittedVersion)
Funders
  • Doctoral Studentship (1473923)
  • Economic and Social Research Council (ES/K013351/1)
Topics
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Memory Processes and Influences
  • Social Sciences › Psychology › Developmental and Educational Psychology › Second Language Acquisition and Learning
  • Social Sciences › Psychology › Developmental and Educational Psychology › Reading and Literacy Development
Keywords
incidental learning, intentional learning, vocabulary acquisition, story reading, retrieval practice, testing effect, long-term retention, forgetting, cued recall, multiple-choice tests

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.

A language model drew the claims from the paper on 2026-11-18, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-11-18 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

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: Learning new word meanings from story reading: the benefit of immediate testing. https://app.cochat.ai/papers/10.7717/peerj.11693
The paper
Hulme, R. C., & Rodd, J. M. (2021). Learning new word meanings from story reading: the benefit of immediate testing. PeerJ. https://doi.org/10.7717/peerj.11693

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.