Presenting words as spoken narration rather than on-screen text alongside graphics improves learning¶

Claim · All claims
Evidence · 4 studies · 2 quant-synthesis r3, 1 causal r?, 1 review r2 · q2–q4 · i1–i2
Currency · newest study 2022 · 3 of 4 entries checked against their sources (automated) · not reviewed by a person

Presenting words as spoken narration rather than on-screen text — alongside graphics or animation — improves learning, because distributing information across the auditory and visual channels reduces overload in the visual channel.

Subclaims¶

q4 i2 A meta-analysis of 43 independent effects found that presenting words as audio narration rather than on-screen text alongside graphics improved learning, with the benefit moderated by element interactivity, pacing, and field of study. → Ginns 2005

q3 i? College students who heard words narrated aurally alongside computer-presented animation integrated words and pictures more easily than students who read the same words as on-screen text. → Mayer and Moreno 1998

q3 i1 The larger modality review found moderate learning benefits (g = 0.38), reduced to g = 0.20 after accounting for publication bias. → Noetel 2022

q4 i2 Across 39 between-subjects experiments, presenting textual information in auditory mode alongside visual graphics produced a moderate-to-large learning advantage (d = 0.72) over fully visual presentation. → Ginns 2005

q2 i? Combining a diagram with spoken rather than written text was more effective for learning, and the effect has been replicated in many experiments. → Sweller 2019

Evidence¶

Ginns 2005¶

Ginns, P. (2005). Meta-analysis of the modality effect. Learning and Instruction, 15(4), 313–331. doi:10.1016/j.learninstruc.2005.07.001

q4 · meta-analysis (43 independent effects: 39 between-subjects, 4 within-subjects) · i2 · moderate-to-large effect, d = 0.72 (95% CI 0.52–0.92) across 39 between-subjects experiments · n=43 independent effects · quant-synthesis · r?

A meta-analysis of the modality effect literature, pooling 43 independent effects (39 between-subjects designs, 4 within-subjects designs) comparing visual-plus-narration presentations against visual-plus-on-screen-text presentations. The major hypotheses about the instructional benefit of splitting text and graphics across the auditory and visual channels were supported, and the size of the benefit was moderated by level of element interactivity (material complexity) and by whether presentation was system-paced or learner-paced, and varied between fields of study. The strongest effect appeared under system-paced conditions, where visual-channel overload is most acute — consistent with Cognitive Load Theory's account of the effect.

Mayer and Moreno 1998¶

Mayer, R. E., & Moreno, R. (1998). A split-attention effect in multimedia learning: Evidence for dual processing systems in working memory. Journal of Educational Psychology, 90(2), 312–320. doi:10.1037/0022-0663.90.2.312

q3 · peer-reviewed experiment · i? · no effect size available in what was read · n=146 · causal · r?

An experiment with 146 college students learning from a computer-based animation of lightning formation. Learners who heard the explanatory words as spoken narration alongside the animation were able to integrate the verbal and pictorial material more easily than learners who read the identical words as on-screen text alongside the same animation — the split-attention/modality effect the meta-analysis above later synthesized across studies. The authors interpret the result as evidence for a dual-channel (auditory/visual) working-memory architecture rather than a single visual-processing bottleneck.

Noetel 2022¶

Noetel, M., Griffith, S., Delaney, O., Harris, N. R., Sanders, T., Parker, P., del Pozo Cruz, B., & Lonsdale, C. (2022). Multimedia design for learning: An overview of reviews with meta-meta-analysis. Review of Educational Research. https://doi.org/10.3102/00346543211052329

q3 · i1 · quant-synthesis · r3

Synthesis of two reviews of the modality effect (Ginns, 2005; Reinwein, 2012); the larger review (Reinwein, 2012, k = 86) found moderate benefits (g = 0.38), which remained significant but smaller (g = 0.20) after publication-bias correction. Eye-tracking studies showed omitting text for spoken words increased attention to images and learning.

"The larger of these reviews found moderate benefits on learning (g = 0.38, 95% CI [0.33, 0.43], k = 86; Reinwein, 2012). Effects were smaller, but still significant, when statistically accounting for publication bias (g = 0.20, 95% CI [0.15, 0.25], k = 86; Reinwein, 2012)"

Sweller 2019¶

Sweller, J., van Merriënboer, J. J. G., & Paas, F. (2019). Cognitive Architecture and Instructional Design: 20 Years Later. Educational Psychology Review, 31, 261–292. https://doi.org/10.1007/s10648-019-09465-5

q2 · i? · review · r2

Mousavi et al.'s (1995) geometry-learning experiment, first test of the modality effect per this review, compared a diagram with integrated written text versus auditory spoken text; the spoken-text combination was most effective, and the review cites Ginns (2005a) meta-analysis.

"They indeed found the modality effect that has since been replicated in many other experiments."

Studies bearing on this claim¶

Should the words be spoken or printed? This table gathers the studies behind this claim and the 9 narrower claims that bear on it: 7 distinct studies (3 causal, 2 quant-synthesis, 2 review), q2–q4, 4 of 7 with an effect size coded. One row per study, codes as each entry records them; no effect is averaged and no verdict is drawn. Built by scripts/synthesize_claim.py from eval/synthesis/; a row that looks wrong is fixed on the claim it comes from.

Discussion¶

The modality effect is a core prediction of Cognitive Load Theory and its dual-channel account of working memory: when graphics and written words both compete for visual processing, narration offloads verbal material to the auditory channel, freeing visual resources for the diagram or animation. This mechanism is closely related to the dual coding tradition, though the two literatures make somewhat different predictions and should not be conflated. It is one of several working-memory interventions alongside chunking and broader cognitive load management.

Several boundary conditions are well established in the literature and should guide design. The effect is strongest for fast-paced, system-paced, graphics-heavy material (e.g., animations) where the visual channel is genuinely overloaded; with learner-paced or static graphics, learners can self-manage the split attention and the benefit shrinks or disappears. The effect also shows an expertise reversal pattern: for advanced learners, narration can become redundant with the graphic and depress performance — see Expertise Reversal Effect. Finally, the effect applies to explaining words accompanying graphics, not to all text: narrating everything on screen violates the redundancy principle when text duplicates narration verbatim.

Open questions include how the effect interacts with learner control over pacing, whether it holds for learners with auditory processing difficulties or non-native language proficiency, and how it manifests in mobile or noisy listening environments where narration quality degrades.

Merged from “Modality Principle Narration Beats On Screen Text” (modality-principle-narration-beats-on-screen-text): Mechanism. The modality principle is grounded in cognitive load theory and the dual-channel assumption of multimedia learning: the visual channel processes both graphics and text, so replacing on-screen text with narration offloads verbal material to the auditory channel and reduces the risk of visual overload. This is closely related to the broader account in Cognitive Load Theory and to Coherence Principle: irrelevant material hurts learning.

Boundary conditions. The principle applies when words accompany graphics that the learner must inspect simultaneously. It does not apply to text-only material, to content the learner must re-read at their own pace (e.g., complex definitions or technical terms), or to learners with limited listening proficiency or hearing impairments. Narration that disappears can impose memory demands that static text does not, so designers should keep narrated segments short and avoid redundancy with identical on-screen text.

Constraints on application. The effect is strongest for system-paced lessons where learners cannot control the flow of information; in learner-paced environments, learners can compensate by switching attention between text and graphics, which weakens or eliminates the narration advantage. Presenting identical words as both narration and on-screen text (redundancy) can actively depress performance relative to narration alone, particularly for novices. For learners who are non-native listeners, hearing-impaired, or working with dense technical terminology that must be consulted repeatedly, on-screen text is the better choice and narration can impose extraneous transience costs.

Open questions. Effect sizes vary with pacing (learner-controlled vs. system-paced lessons), learner expertise, and language proficiency. The entries on this page name these moderators without quantifying them; the narrower claims on pacing and field of study carry those figures.

Merged from “Modality Principle Spoken Narration Beats On Screen Text” (modality-principle-spoken-narration-beats-on-screen-text): Mechanism. The modality principle follows from the dual-channel assumption of Cognitive Load Theory: visual working memory must handle both pictures and written words, whereas spoken narration offloads verbal processing to the auditory channel. This reduces the risk of cognitive overload degrading learning, especially when graphics and words are presented simultaneously. It is closely related to cognitive load management as a design goal: the modality choice is one of the most direct levers for redistributing load across channels.

Boundary conditions. The benefit is strongest for novices and for fast-paced, system-paced multimedia where learners cannot control the pace. With learner-paced environments, long or complex text, or learners who need to re-read (e.g., second-language learners or those with hearing impairments), on-screen text can be equal or superior — an instance of the expertise reversal effect and a reminder that modality choices interact with learner characteristics. Narration should also respect the coherence principle: adding audio does not help if it introduces irrelevant material.

Design implications. Prefer narration over on-screen text when words accompany animated or system-paced graphics, and keep narration conversational rather than formal. Do not duplicate the same words in both narration and on-screen text — redundancy reintroduces the visual-channel competition the modality principle is meant to avoid. Where learners must consult reference text (definitions, code, formulas), retain it as text rather than reading it aloud. Where pacing control is available, chunking the material into segments gives learners the pause-and-replay capacity that narration otherwise removes.

Open questions. Most supporting evidence comes from short, lab-style lessons in well-structured domains; generalization to lengthy, complex, or self-paced online courses remains an active research question. The Ginns (2005) meta-analysis recorded above reports a pooled between-subjects effect of d = 0.72 (95% CI 0.52–0.92).

Related Claims¶

• Cognitive overload degrades learning. — the overload mechanism the modality effect is designed to prevent
• Coherence principle: irrelevant material hurts learning. — a companion multimedia principle governing what to exclude alongside how to present words
• Chunking reduces working memory load. — an alternative route to the same working-memory bottleneck
• Cognitive load reduction improves learning. — the broader family of load-reducing design interventions
• Cognitive Load Theory — the theoretical framework from which the modality effect is derived
• Dual Coding Theory — theoretical background on combining verbal and visual channels
• Expertise reversal effect. — why the narration advantage fades or reverses for advanced learners
• Adding relevant pictures or graphics to text improves learning from it, with a small pooled effect on reading comprehension, though whether dual coding explains the benefit is disputed — related
• Audio narration with finger-tracking animation directs bilingual preschoolers' attention to the target-language print in dual-language e-books, including the nondominant language — related
• Expertise Reversal Guidance Hurts Experts — related
• Redundant on-screen text impairs learning when it competes with a visualization, though written text duplicating narration alone can help — related
• On-screen text that duplicates the narration of a diagram can hurt learning, but spoken–written redundancy is not harmful on average and helps over spoken-only presentation in some conditions — related
• Screencast design findings: static vs. dynamic screen movement, explicit vs. implicit narration, and handwriting preferred though typefaces judged more legible — related
• Most surveyed Algerian university teachers of English (61.9%) report presenting words and corresponding graphics simultaneously — related
• In one middle-school quasi-experiment on electricity, spoken-text groups outperform written-text groups with both abstract and concrete animation (d = .35) — a narrower finding that bears on this claim
• In one middle-school quasi-experiment, adding written text to spoken narration makes no significant difference to achievement with either abstract or concrete animation — related
• Spoken text modality does not significantly differ from signaled written text modality, or from signaled written plus spoken text, for either abstract or concrete animation — a narrower finding that bears on this claim
• Avoiding double load on the visual channel improves visual knowledge (channel-overload effect) — related
• Students find reflective prompts accompanied by visual elements more actionable and easier to interpret than plain-text prompts — related
• Degree of anthropomorphism attributed to the agent was shaped by both interaction modality and, within audio conditions, voice accent — related
• Voice modality roughly doubled dialogue turns and question-asking on equal engaged time within the same students — related
• Modality principle: graphics with spoken text beat graphics with printed text for transfer — a narrower finding that bears on this claim
• Redundancy principle: animation plus narration beats animation, narration, and on-screen text for transfer — a narrower finding that bears on this claim
• Verbal redundancy does not facilitate learning when materials include images or animation — related
• The spoken–written advantage over spoken-only presentation appears for system-paced but not learner-paced materials — related
• Verbal redundancy (spoken–written presentation) yields a small overall learning benefit with large between-study heterogeneity — related
• Design principle effects are larger in system-paced and in complex media, and learner prior knowledge was not a significant moderator (p = 0.14) — a broader claim this one bears on
• Multimedia design interventions yield small average improvements in cognitive load management (g = 0.22), with personalisation and modality showing the largest effects — related
• Dual-mode (auditory plus visual) presentation of worked examples enhances subsequent problem-solving efficiency, except under high visual-search conditions without attention cues — related
• Presentation pacing moderates the modality effect: large under system-paced conditions and not significant under self-paced conditions, where it may disappear or reverse — a narrower finding that bears on this claim
• In one meta-analysis, modality-effect studies using virtual reality head-mounted displays show larger effects than studies using computer screens or audio-tapes — a narrower finding that bears on this claim
• Within-subjects designs also show a moderate-to-large modality effect (d = 0.73) — related
• Split Attention Effect Degrades Learning — related
• Multimedia Principle Improves Learning — related
• The animation advantage is larger when animation is coupled with auditory commentary — related
• The animation advantage is largest when instruction includes no accompanying text — related
• Animation produces a small overall learning advantage over static graphics in multimedia instruction — related
• Matching modality by conveying complementary information through auditory and visual channels increases retention, transfer, and engagement — related
• Working memory includes partially independent subcomponents for auditory/verbal and visual/2- or 3-dimensional information — related
• Animated personas produce better learning than static ones, and agent voice contributes significantly to learning — related
• Pedagogical agents enhance learning compared with learning environments that do not feature agents — related
• APA presentations tend to overload the visual channel, especially with high element interactivity content — a narrower finding that bears on this claim
• No difference in intrinsic cognitive load between visual-only and audio-visual text conditions in VR — related
• Oral non-interactive teaching can enhance learning more than written teaching for complex materials — a narrower finding that bears on this claim
• Reading text and presenting graphics on handheld screens are more difficult than on paper, and interactivity may be compromised — related
• In an immersive VR learning environment, visual-only text yields higher recall than audio-visual text (reverse modality effect) — reports the opposite
• Symbolic training transfers to the auditory modality, improving spoken number-word magnitude comparison — related
• Combining voiced auditory descriptions with tactile graphics increased braille readers' likelihood of answering assessment items correctly — a narrower finding that bears on this claim
• The talking head contributes little to AutoTutor's learning gains; the content of the conversation matters — related
• The review reports that the modality effect reverses with expertise: audio-visual presentations helped inexperienced learners, but after further training a diagram-only condition showed a substantial advantage — related
• A practice article reports that the studies it cites found better post-test outcomes from words with relevant pictures than from words alone, especially with spoken words — related
• Lecture modality (written, audio, video, live) shows no consistent effect on memory for lecture content or on the note-taking effect — related
• Transient information effect: lengthy spoken text can reverse the modality effect, while shorter spoken text restores the conventional modality advantage — related
• Synced auditory and visual information in narrated presentations produces better learning than single-channel or out-of-sync presentation — related
• Transient information in animations poses a serious challenge to learning, except for animations of human movement procedures — related
• Element interactivity moderates the effectiveness of cognitive offloading, with greater benefit for high element interactivity materials — related
• The modality effect may disappear or reverse for lengthy, complex spoken text — related
• Working-memory limitations can be circumvented by schema construction, automation, and multiple presentation modalities — a broader claim this one bears on