Does redundancy in instructional materials increase or decrease learning and cognitive load?

Why repeating the same words can make learning harder

The problem is working memory. When a learner reads on-screen text while hearing a narration of that same text, both streams of words must be processed through the same limited verbal channel. Kalyuga's review of the redundancy effect describes exactly this setup — narrated explanations shown simultaneously with written text — and notes that presenting material in two modalities this way may overload working memory and produce negative learning effects [3]. The redundancy is not helping the learner rehearse; it is splitting attention between two versions of one message.

This is why the effect is often framed as a design warning rather than a general rule about repetition. The concern is specifically about redundant information that adds no new meaning while consuming the same cognitive resources the learner needs for the actual lesson [3].

When the extra input carries real information

Redundancy behaves differently when the second stream is not simply a duplicate. A study of cross-language instructional design with foreign-language learners compared materials that followed the redundancy principle against materials that violated it, and found that redundant input may aid comprehension without increasing cognitive load [1]. For a learner still decoding a foreign language, the extra input appears to function as support rather than noise.

A study of segmentation and redundancy methods in multimedia English lessons points the same way: the group taught with redundancy methods reported significantly lower cognitive load and higher learning outcomes [5]. In both cases the redundancy is doing work — reinforcing meaning the learner is still building — rather than restating something already understood.

Not all redundancy is the same kind of redundancy

The type of duplication matters. An experimental study by Trypke and colleagues set out to separate modal redundancy from codal redundancy and to examine how the two interact in instructional design [2]. That distinction is a reminder that "redundancy" covers several different design choices, and lumping them together can hide why one version helps and another hurts.

Age and prior knowledge shift the balance too. Leslie's experiments with primary school students compared audio-only and audio-visual presentations and examined how redundant representations load working memory [4]. The results indicated that presenting students with a redundant version changed outcomes — evidence that the same design can land differently depending on who is learning from it [4]. A learner who already knows the material has less to gain from a second pass at it, while a younger or less experienced learner may still be building the representation the redundancy reinforces.

How to judge whether a design is helping or hurting

The practical question is whether the second stream adds meaning or merely repeats it. If the extra text, narration, or image tells the learner something new — a translation, a worked step, a visual anchor for an unfamiliar concept — it can support comprehension without taxing working memory, as the cross-language and English-lesson studies suggest [1][5]. If it restates the same words in the same verbal channel, it competes with the message it is meant to reinforce [3].

The evidence also cautions against treating any single finding as universal. The studies differ in learners, languages, and subject matter, and the experimental work on redundancy types is still working out how modal and codal redundancy interact [2]. What holds across them is the underlying logic: redundancy helps when it carries information the learner needs, and hurts when it spends working memory on information the learner already has.

Sources used in this answer

Related research questions