Skip to main content
Radlor

The research

Why visual learning works.

People learn better from words and pictures together than from words alone. Not from every picture, and not for every task. Here is the evidence, with its limits and sources.

Try it: how do you clean a French press?

Pull the plunger all the way out. Tip the used grounds into the bin. Unscrew the filter. It comes apart in 3 parts. Wash every part in warm, soapy water. Let it dry, then put it back together. Make sure the mesh sits between the top plate and the bottom plate when you screw the filter back on, and that the plunger goes in through the lid.

Same five steps. Which one could you follow faster? The research below explains why.

Cognitive psychology

Words and pictures together beat words alone.

  • Series of experiments

    Words with pictures beat words alone, by a large margin.[1]

    d = 1.35

    Median effect size in 13 comparisons. Mayer counts above 0.4 as educationally important.
    • Words with graphics vs words aloned = 1.35
    • Educationally important0.4
    Details and limits

    The multimedia principle

    In 13 experimental comparisons, students who learned from words and graphics did much better on a test that asked them to use the idea in a new problem than students who learned from words alone. The median effect size was d = 1.35, a large effect.

    The limit. Most of these were short lessons for college students in one research group, tested soon after the lesson. Mayer also says that not all graphics are equally useful.

  • Theory

    Words and pictures go in through two channels, and each channel is small.[2][3][4]

    Details and limits

    Two channels: one for words, one for pictures

    Dual coding theory says that the mind keeps verbal and image-based representations, and that these two kinds work together. Multimedia learning research assumes that each channel can only process a small amount at a time.

    The limit. Dual coding is a theory, and it has critics: some researchers argue that all knowledge is stored in one abstract form.

  • Lab studies

    After 2,500 pictures, people still picked the right one most of the time.[5][6]

    Correct picks out of two (chance is 50%)
    • 92%
      Different object
    • 88%
      Same kind, other example
    • 87%
      Same object, other state

    Dashed tick: chance, 50%

    Details and limits

    Picture memory is very large

    People who saw 10,000 pictures once, a few seconds each, still recognized most of them about two days later. Memory for pictures was better than memory for words. In a later study, adults who saw 2,500 object pictures picked the right one from a pair 92%, 88% and 87% of the time in three tests, even when the other picture was very similar.

    The limit. These tests asked “have you seen this before?”, which is easier than understanding or using an idea. The groups were small (5 to 14 people).

  • Review

    We hold only about four new things in mind at once.[7]

    About 4 chunks at once (3 to 5), fewer than the older figure of 7
    Details and limits

    Working memory is small

    Adults hold only about three to five chunks of new information in mind at one time, about four on average.

    The limit. The number depends on the task, and other researchers debate it. It is the reason we teach one idea or one step at a time.

  • Series of experiments

    Clean, in-step, well-placed pictures win almost every test.[1]

    Tests where the design rule helped. One square is one test.
    • Remove extra material18 of 19
    • Say words with the picture8 of 8
    • Put words next to the part9 of 9
    • Split into learner-paced parts7 of 7
    Details and limits

    Design matters as much as the picture

    The same research found that people learn better when extra material is removed (18 of 19 tests), when spoken words come at the same time as the picture (8 of 8), when words sit next to the part they describe (9 of 9) and when a lesson is split into learner-paced parts (7 of 7).

    The limit. A picture that is decorative, crowded or out of step with the words can make learning worse, not better. The effects are medians from one research group.

Video, animation and how-to tasks

Moving pictures help most when you must do something.

  • Meta-analysis

    Animation helps a little in general, and a lot for hands-on skills.[8][9]

    Effect size of animation over still pictures (d)
    • 0.37All topics
    • 0.76Realistic, video-like
    • 1.06Hands-on skills

    ComparisonWith pictures or video

    Details and limits

    Animation beats still pictures, most for how-to tasks

    A meta-analysis of 26 studies found a medium-sized advantage of instructional animations over static pictures (d = 0.37). The advantage was much larger when people learned a procedural-motor skill, something they must do with their hands (d = 1.06), and when the animation was realistic, for example video (d = 0.76).

    The limit. A later and larger meta-analysis (61 studies, 7,036 learners) found a smaller overall advantage (g = 0.226).

  • Lab studies

    People learned knots and puzzles better from animation.[10][11]

    Details and limits

    Hand tasks: knots and puzzle rings

    In two experiments, students who learned to tie knots or solve puzzle rings from an animation learned more than students who saw the same steps as a series of still diagrams.

    The limit. Small laboratory experiments with students. In an assembly study, animated instructions made the first build faster, but the advantage was gone by the third build.

  • Single study

    One clear action per step: faster assembly, fewer errors.[12]

    Average cut compared with factory and hand-designed instructions
    • Assembly time−35%
    • Errors−50%

    With clear stepsThe part that was cut

    Details and limits

    Clear step-by-step drawings cut assembly time

    In a furniture-assembly study, step-by-step instructions designed with cognitive principles (one clear action per step) cut assembly time by 35% on average and errors by 50%, compared with factory instructions.

    The limit. These were still diagrams, not video. The study shows that clear steps matter; it does not test video.

  • Systematic review

    Video helps students most when it is added to teaching.[13][14]

    Effect size on learning (g), 105 trials
    • Video replaces teaching0.28
    • Video added to teaching0.80
    Details and limits

    Video helps students, most when it adds to teaching

    A review of 105 randomized trials with 7,776 university students found small gains when video replaced other teaching (g = 0.28) and strong gains when video was added to it (g = 0.80). In a study with 111 pupils in grades 5 and 6, a video tutorial beat a paper tutorial for a software task.

    The limit. The review is about university students, and adding video also adds study time. The pupils’ study is one task, and all groups improved, the paper group too.

  • Meta-analysis

    Short parts beat one long video.[15][16]

    ≤ 6 min

    Median watch time, whatever the video length (6.9 million sessions)
    Details and limits

    Short, segmented videos

    A meta-analysis of 56 investigations found that people learn better when a lesson comes in segments than as one continuous unit. In a study of 6.9 million video sessions, the median watch time was at most 6 minutes, whatever the length of the video.

    The limit. Segments also make learning take longer. The watch-time study measured attention, not learning, in adult online courses.

  • Review

    Motion alone does not teach. Clear content and steps do.[17][18]

    Details and limits

    The sceptical view, and the best answer to it

    A well-known review found that many wins for animation came from unequal comparisons: the animation showed more information or let the learner interact. Video works best together with a few still key pictures, according to another study.

    The limit. This is why we do not claim that motion itself teaches. Content, clear steps and control over the pace do much of the work.

Education

US education research recommends visual math.

  • Strong evidence (IES)

    US education research rates visual math as strong evidence.[19]

    Details and limits

    Drawings, diagrams and number lines

    For elementary students who struggle with math, the guide recommends a well-chosen set of concrete and semi-concrete representations, and the number line. It rates the evidence for both as strong.

    The limit. The guide is about small-group teaching led by a teacher, not about apps or video.

  • Strong evidence (IES)

    Students should learn to draw the problem, not only watch it.[20]

    Details and limits

    Teach students to use visual representations

    For problem solving in grades 4 to 8, the guide recommends that teachers show students how to use visual representations such as tables and diagrams. It rates the evidence as strong.

    The limit. The recommendation is that students learn to draw and use the visuals themselves, not only watch them.

Medicine and health

Patients follow instructions better with pictures.

  • Review

    Pictures next to the words: more attention, better recall.[21]

    Details and limits

    Pictures linked to the text raise attention and recall

    A review of health-communication research found that pictures closely linked to written or spoken text can markedly increase attention to and recall of health information, compared with text alone. People with low literacy are especially likely to benefit.

    The limit. A narrative review with no pooled numbers. Pictures can also change adherence in either direction, because the emotional response to a picture matters.

  • Randomized trial

    Cartoon instructions: more patients understood and followed them.[22][23]

    Patients, text only vs text with cartoons. One square is 1%.
    Show the measure

    Read the sheet:

    • 79%Text only
    • 98%With cartoons

    All questions right:

    • 6%Text only
    • 46%With cartoons

    Daily wound care:

    • 54%Text only
    • 77%With cartoons

    ComparisonWith pictures or video

    Details and limits

    Cartoons on wound-care sheets

    In a randomized study of 234 emergency patients, those who got wound-care instructions with cartoons more often read them (98% vs 79%), answered all wound-care questions correctly (46% vs 6%) and did their daily wound care (77% vs 54%).

    The limit. One hospital, cuts only, and wound care was self-reported by telephone three days later. An earlier study of 101 patients found a smaller gain: 65% vs 43% gave five or more correct answers out of 10.

  • Randomized trial

    Picture instructions cut medicine dosing errors.[24]

    Dosing errors for a daily medicine, more than 20% off the dose (lower is better)
    • 47.8%
      Usual instructions
    • 5.4%
      With pictograms

    ComparisonWith pictures or video

    Details and limits

    Picture-based medicine instructions cut dosing errors

    In a randomized trial with 245 caregivers of young children, plain-language instructions with pictograms cut dosing errors for a daily liquid medicine from 47.8% to 5.4%.

    The limit. The pictograms came together with plain language and counselling, so the effect of the pictures alone is not known. One hospital emergency department.

  • Meta-analysis

    Across many trials: a moderate gain, and results vary.[25][26]

    Details and limits

    The overall picture: a moderate gain, and it varies

    A meta-analysis of randomized trials found that pictures in health information moderately improved knowledge and recall, with a larger gain for people with lower health literacy. A separate review found that pictures help most when they are used together with written or spoken instructions.

    The limit. The results differ a lot from study to study, and the authors ask readers to interpret them with caution. Some icons, for example clock faces, were too complex to help.

  • Meta-analysis

    Video did best on recall, but the gap was not significant.[27][28]

    Average correct recall, from 0 to 100% (differences not significant)
    • Spoken47%
    • Written58%
    • Video67%
    Details and limits

    Video instructions: promising, not proven

    Across emergency-department studies, patients correctly recalled on average 47% of spoken discharge instructions, 58% of written ones and 67% of video ones.

    The limit. These differences were not statistically significant, and only 7 of 51 studies used video. A review of video health education found that it changed some health behaviors but not others.

  • Guidance

    US health agencies tell writers to add a clear visual.[29][30]

    Details and limits

    US health agencies tell writers to use visuals

    The US Agency for Healthcare Research and Quality says videos are useful to demonstrate self-care, such as how to use an inhaler. The CDC scores health materials on whether they have at least one visual that supports the main message.

    The limit. This is guidance, not a study. Both agencies also warn that a visual that is unclear or unrelated to the words can confuse people, and AHRQ says not to drop paper materials.

Neuroscience

The brain stores a picture differently from a word.

  • Brain-imaging study

    Pictures use more of the brain's visual memory areas than words.[31]

    Details and limits

    Pictures use the visual memory areas more

    In a brain-imaging study, memorizing pictures used visual and memory areas of the brain more than memorizing words did. The authors suggest that this may explain why pictures are remembered better than words.

    The limit. A small study of young adults with single objects and single words, not lessons or videos. “May explain” is the authors’ wording.

Honest limits

What we do not claim.

  • We do not say there are “visual learners”

    The idea that each person has a learning style, and learns best when teaching matches it, is not supported by evidence. A major review found no adequate evidence base for using learning-style tests in education. Visuals help most learners; they are not for one type of person.[32]

  • We do not use the “10% of what we read, 80% of what we see” numbers

    These percentages are not research findings. A study of where they came from found that they do not make sense as data and were added later to Edgar Dale’s “cone of experience”, which had no numbers.[33]

  • We do not say the brain reads pictures “60,000 times faster”

    We could not find any study behind this number, so we do not use it. One classic study even found that people retrieve words faster than pictures.[5]

  • We do not say pictures always help

    Decorative or confusing pictures can distract. Health agencies give the same warning. The research supports pictures that show the idea, close to the words that explain it.[1][30]

What we do with it

How our products use this.

This research is about learning in general. It does not test our products, and we do not say that it does.

Sources

Every source on this page.

We opened each source and compared it with our sentence. Last check: 22 November 2026. Many journal articles need a subscription for the full text; the abstract is free to read. If you find a mistake, write to us on the contact page.

  1. Mayer RE. Research-based principles for designing multimedia instruction. In: Overson CE, Hakala CM, Kordonowy LL, Benassi VA, eds. In Their Own Words: What Scholars and Teachers Want You to Know About Why and How to Apply the Science of Learning in Your Academic Setting. Society for the Teaching of Psychology; 2023:143–157. www.unh.edu/teaching-learning-resource-hub/sites/default/files/media/2023-06/itow-research-based-principles-for-designing-multimedia-instruction-mayer.pdf
  2. Clark JM, Paivio A. Dual coding theory and education. Educational Psychology Review. 1991;3(3):149–210. doi.org/10.1007/BF01320076
  3. Paivio A. Dual coding theory: retrospect and current status. Canadian Journal of Psychology. 1991;45(3):255–287. doi.org/10.1037/h0084295
  4. Mayer RE, Moreno R. Nine ways to reduce cognitive load in multimedia learning. Educational Psychologist. 2003;38(1):43–52. doi.org/10.1207/S15326985EP3801_6
  5. Standing L. Learning 10,000 pictures. Quarterly Journal of Experimental Psychology. 1973;25(2):207–222. doi.org/10.1080/14640747308400340
  6. Brady TF, Konkle T, Alvarez GA, Oliva A. Visual long-term memory has a massive storage capacity for object details. Proceedings of the National Academy of Sciences. 2008;105(38):14325–14329. pmc.ncbi.nlm.nih.gov/articles/PMC2533687/
  7. Cowan N. The magical number 4 in short-term memory: a reconsideration of mental storage capacity. Behavioral and Brain Sciences. 2001;24(1):87–114. pubmed.ncbi.nlm.nih.gov/11515286/
  8. Höffler TN, Leutner D. Instructional animation versus static pictures: a meta-analysis. Learning and Instruction. 2007;17(6):722–738. doi.org/10.1016/j.learninstruc.2007.09.013
  9. Berney S, Bétrancourt M. Does animation enhance learning? A meta-analysis. Computers & Education. 2016;101:150–167. archive-ouverte.unige.ch/unige:92234
  10. Ayres P, Marcus N, Chan C, Qian N. Learning hand manipulative tasks: when instructional animations are superior to equivalent static representations. Computers in Human Behavior. 2009;25(2):348–353. doi.org/10.1016/j.chb.2008.12.013
  11. Watson G, Butterfield J, Curran R, Craig C. Do dynamic work instructions provide an advantage over static instructions in a small scale assembly task? Learning and Instruction. 2010;20(1):84–93. eric.ed.gov/?id=EJ863665
  12. Heiser J, Phan D, Agrawala M, Tversky B, Hanrahan P. Identification and validation of cognitive design principles for automated generation of assembly instructions. Proceedings of the Working Conference on Advanced Visual Interfaces (AVI ’04). ACM; 2004:311–319. doi.org/10.1145/989863.989917
  13. Noetel M, Griffith S, Delaney O, et al. Video improves learning in higher education: a systematic review. Review of Educational Research. 2021;91(2):204–236. eric.ed.gov/?id=EJ1290063
  14. van der Meij H, van der Meij J. A comparison of paper-based and video tutorials for software learning. Computers & Education. 2014;78:150–159. doi.org/10.1016/j.compedu.2014.06.003
  15. Rey GD, Beege M, Nebel S, Wirzberger M, Schmitt TH, Schneider S. A meta-analysis of the segmenting effect. Educational Psychology Review. 2019;31(2):389–419. eric.ed.gov/?id=EJ1217373
  16. Guo PJ, Kim J, Rubin R. How video production affects student engagement: an empirical study of MOOC videos. Proceedings of the First ACM Conference on Learning @ Scale (L@S ’14). ACM; 2014:41–50. doi.org/10.1145/2556325.2566239
  17. Tversky B, Morrison JB, Bétrancourt M. Animation: can it facilitate? International Journal of Human-Computer Studies. 2002;57(4):247–262. doi.org/10.1006/ijhc.2002.1017
  18. Arguel A, Jamet É. Using video and static pictures to improve learning of procedural contents. Computers in Human Behavior. 2009;25(2):354–359. doi.org/10.1016/j.chb.2008.12.014
  19. Fuchs LS, Newman-Gonchar R, Schumacher R, et al. Assisting Students Struggling with Mathematics: Intervention in the Elementary Grades (WWC 2021006). Institute of Education Sciences, US Department of Education; 2021. ies.ed.gov/ncee/wwc/PracticeGuide/26
  20. Woodward J, Beckmann S, Driscoll M, et al. Improving Mathematical Problem Solving in Grades 4 Through 8 (NCEE 2012-4055). Institute of Education Sciences, US Department of Education; 2012. ies.ed.gov/ncee/wwc/PracticeGuide/16
  21. Houts PS, Doak CC, Doak LG, Loscalzo MJ. The role of pictures in improving health communication: a review of research on attention, comprehension, recall, and adherence. Patient Education and Counseling. 2006;61(2):173–190. doi.org/10.1016/j.pec.2005.05.004
  22. Delp C, Jones J. Communicating information to patients: the use of cartoon illustrations to improve comprehension of instructions. Academic Emergency Medicine. 1996;3(3):264–270. pubmed.ncbi.nlm.nih.gov/8673784/
  23. Austin PE, Matlack R, Dunn KA, Kesler C, Brown CK. Discharge instructions: do illustrations help our patients understand them? Annals of Emergency Medicine. 1995;25(3):317–320. pubmed.ncbi.nlm.nih.gov/7532382/
  24. Yin HS, Dreyer BP, van Schaick L, Foltin GL, Dinglas C, Mendelsohn AL. Randomized controlled trial of a pictogram-based intervention to reduce liquid medication dosing errors and improve adherence among caregivers of young children. Archives of Pediatrics & Adolescent Medicine. 2008;162(9):814–822. pubmed.ncbi.nlm.nih.gov/18762597/
  25. Schubbe D, Scalia P, Yen RW, et al. Using pictures to convey health information: a systematic review and meta-analysis of the effects on patient and consumer health behaviors and outcomes. Patient Education and Counseling. 2020;103(10):1935–1960. pubmed.ncbi.nlm.nih.gov/32466864/
  26. Katz MG, Kripalani S, Weiss BD. Use of pictorial aids in medication instructions: a review of the literature. American Journal of Health-System Pharmacy. 2006;63(23):2391–2397. pubmed.ncbi.nlm.nih.gov/17106013/
  27. Hoek AE, Anker SCP, van Beeck EF, Burdorf A, Rood PPM, Haagsma JA. Patient discharge instructions in the emergency department and their effects on comprehension and recall of discharge instructions: a systematic review and meta-analysis. Annals of Emergency Medicine. 2020;75(3):435–444. pubmed.ncbi.nlm.nih.gov/31439363/
  28. Tuong W, Larsen ER, Armstrong AW. Videos to influence: a systematic review of effectiveness of video-based education in modifying health behaviors. Journal of Behavioral Medicine. 2014;37(2):218–233. pubmed.ncbi.nlm.nih.gov/23188480/
  29. Agency for Healthcare Research and Quality (US). Assess, Select, and Create Easy-to-Understand Materials: Tool 11. Health Literacy Universal Precautions Toolkit, 3rd edition. www.ahrq.gov/health-literacy/improve/precautions/tool11.html
  30. Centers for Disease Control and Prevention (US). The CDC Clear Communication Index: item on visuals that support the main message. www.cdc.gov/ccindex/tool/page-4.html
  31. Grady CL, McIntosh AR, Rajah MN, Craik FIM. Neural correlates of the episodic encoding of pictures and words. Proceedings of the National Academy of Sciences. 1998;95(5):2703–2708. pmc.ncbi.nlm.nih.gov/articles/PMC19469/
  32. Pashler H, McDaniel M, Rohrer D, Bjork R. Learning styles: concepts and evidence. Psychological Science in the Public Interest. 2008;9(3):105–119. pubmed.ncbi.nlm.nih.gov/26162104/
  33. Subramony DP, Molenda M, Betrus AK, Thalheimer W. The mythical retention chart and the corruption of Dale’s Cone of Experience. Educational Technology. 2014;54(6):6–16. ERIC record EJ1057239. eric.ed.gov/?id=EJ1057239