Do children learn from cartoons as well as from real people?

Background: why the question matters

Many resources for children's hard moments are animated: cartoon series about feelings, animated social stories, apps with drawn characters. Classic video modelling, by contrast, usually shows a real person on film doing the target behaviour (see our review of video modelling). When the model is drawn or animated rather than filmed, does a child learn as much, and does what they learn carry over to real people and places?

There are reasons to expect animation to help, and reasons to expect it to hurt. Animation lets designers strip away distracting detail, exaggerate the parts of a face or situation that matter, and make characters appealing. Some autistic children appear to process cartoon faces in a more typical way than real ones (Rosset et al., 2008). On the other hand, a cartoon is one step further from reality than a film, and many cartoons add talking animals, magic and impossible events. Young children have to work out which parts of a story apply to real life, and they do not always get it right.

This review brings together applied studies, mostly with autistic children, and developmental studies of how typically developing children learn from screens and fiction. The question matters to us directly: Cala Luna's cartoons use lynx characters.

How the research has been done

Single-case comparisons. Studies that pit an animated model against a human model usually involve three to eight children. Each child is measured repeatedly, and the two kinds of video are either alternated (an alternating treatments design) or introduced at staggered times (a multiple-baseline design). These designs can show which format works better for a particular child, but not how common that pattern is. Our methods glossary explains the terms.

Group trials of animated series compare children who watch the series with children who watch a control programme or nothing, sometimes with random allocation (a randomised controlled trial, or RCT). Outcomes are usually emotion-recognition tasks at increasing distance from what was taught (levels of generalisation).

Laboratory experiments on transfer read or show preschoolers a story in which a character solves a problem, then give the child a similar real problem. Others compare imitation after a video or live demonstration. Prosocial lessons are often measured by how many stickers a child gives away.

Face-processing studies compare how accurately, and with what strategy, children read emotions from photographs, cartoon human faces and cartoon animal faces.

What the studies found

Animated versus human video models

The direct comparisons are recent and small. Bloh and colleagues taught eight autistic children conversational responses, together with matching facial expressions and body language, using both human and animated videos. Seven children learned the target behaviours with one or both formats. Two learned more from the human video, three more from the animated video and three showed little difference; one child learned only from the human video and another only from the animation. The authors concluded that both could work and neither was clearly superior (Bloh et al., 2025). A study of three autistic children aged 4 to 6 found the same pattern for verbal expressions, facial expressions and gestures: both formats were effective, with no consistent difference, although which one was more efficient varied from child to child (Gokce & Ulke-Kurkcuoglu, 2026). An earlier set of studies with five autistic boys also found no clear difference between equivalent animated and human videos; children with stronger imitation skills responded better to video in general (Rayner, 2015).

Animated models also work without a comparison. In a multiple-baseline study of six autistic children aged 7 to 11, animated video modelling was followed by rapid gains in joint attention and social engagement for most children, which generalised across settings and people, and caregivers rated it acceptable (Ho et al., 2019). In a study of avatars animated live by a human operator, autistic children aged 8 to 12 were engaged on average 43% of the time with the avatar, against 24% with a human interventionist, and children aged 8 to 10 then learned to start conversations; the authors caution that novelty may explain some of the extra engagement (Kellems et al., 2020).

An animated series built to teach: The Transporters

The Transporters is an animated series designed to teach emotion recognition, and it has been tested in several trials. Its characters are vehicles with filmed human faces grafted on, so children watch animation but read real expressions. In the first evaluation, 20 autistic children aged 4 to 7 watched daily for four weeks and improved in emotion recognition more than a matched autistic group who did not watch, reaching the level of typically developing children (Golan et al., 2010).

Later trials were more mixed. In a randomised study, children who watched The Transporters at home for three weeks got better at identifying facial expressions than children who watched Thomas the Tank Engine, but social behaviour improved in both groups (Young & Posselt, 2012). A trial with 55 autistic children aged 4 to 7, with a wider range of intellectual ability, found improvement only in recognising anger, with few gains kept at three months and no spread to theory of mind or social skills (Williams et al., 2012). A larger RCT in Israel with 77 autistic children without intellectual disability found gains at every level of generalisation, maintained at three months. Children whose parents gave more support generalised more, and children with more severe autism improved less (Gev et al., 2017). A Cantonese app version improved emotion recognition in autistic children aged 4 to 6 with and without ADHD, with gains extending to new situations and characters (Chan et al., 2024).

Cartoon faces and real faces

Several studies suggest that some autistic children and adolescents find cartoon and animal faces easier to read than human ones. Twenty autistic children read emotions in real and cartoon faces as accurately as controls, but used the typical whole-face strategy only for cartoons, switching to a feature-by-feature strategy for real faces (Rosset et al., 2008). Among 37 autistic adolescents aged 11 to 16, controls outperformed the autistic group on human faces but not on animated ones, and the autistic group did better than controls on still cartoon images (Brosnan et al., 2015). Fifteen adolescents with autism and intellectual disability recognised emotions better in animal faces than in human faces (Cross et al., 2019). The pattern is not limited to autism: among 107 typically developing children aged 5 to 10, the youngest group understood social blunders (faux pas) better when the characters were animals, while older children did better with human characters (Atherton et al., 2025).

Easier to read is not the same as better transfer: none of these studies tested whether practice with cartoon faces improves reading of real faces in daily life.

The video deficit

Very young children learn less from a screen than from a person in the room. A review of how infants and toddlers transfer learning between pictures or screens and real objects concluded that children learn less from television than from a live demonstration until at least age 3 (Barr, 2010). A meta-analysis of 122 effect sizes from 59 reports, covering children aged 0 to 6, found an average deficit of about half a standard deviation (Hedges' g = −0.53). An effect size expresses the gap between conditions in standard-deviation units, so this is a moderate gap. It shrank with age: g = −0.63 for children under 3 and −0.25 for children aged 3 and over, which was still statistically significant. Live and prerecorded video did not differ. The authors flagged small, low-powered studies and possible publication bias; among the most precise half of studies the deficit was only −0.16 (Strouse & Samson, 2021). Almost all the videos were made by researchers and none of the analyses compared cartoons with filmed people. A recent study of 80 toddlers found no video deficit at all (Sommer et al., 2023). For Cala Luna's age range of 5 to 9 there are few data.

Fantasy, realism and transfer to real life

Preschoolers are cautious about applying lessons from fantasy. In three experiments with children aged 3½ to 6, solutions were transferred to real problems more often when the story characters were real people than when they were fantasy characters (Richert et al., 2009), and a follow-up with children aged 3½ to 5½ found the same (Richert & Smith, 2011). That caution makes sense. When 108 preschoolers learned a new cause-and-effect relation from a storybook, they generalised it more when the story world resembled reality, and they got better at telling realistic from fantastical worlds between ages 3 and 5 (Walker et al., 2015). Similarly, better judgements of what was real on educational television partly explained why older preschoolers learned more from it (Mares & Sivakumar, 2014), and 4-year-olds tended to underestimate the reality of ordinary events shown in a cartoon (Li et al., 2015).

Fantasy is not always harmful. Making a story's setting more fantastical, such as a planet with orange grass and a green sky, did not change transfer, and adding a few impossible events before the lesson increased it (Hopkins & Lillard, 2021). Characters with impossible abilities who solved problems in realistic ways prompted more transfer than ordinary characters (Richert & Schlesinger, 2022). In two preregistered studies with 214 children, 5-year-olds learned and generalised a physical cause and effect from both realistic and fantastical books (Brenner et al., 2025). The common thread seems to be whether the lesson itself works by real-world rules.

Animal characters or human characters

This is where the evidence is most relevant to Cala Luna and least settled. In one experiment, a sharing story with human characters increased preschoolers' giving, whereas the same story with anthropomorphised animals (animals that talk and act like people), or a control story, decreased it. The authors concluded that realistic stories were better for prosocial lessons (Larsen et al., 2018). A larger conceptual replication with 179 children aged 3 to 7 did not confirm this: children gave more after every story, and the human sharing story was no more effective than the animal one (Russell & Cain, 2022).

For factual learning about animals, anthropomorphic language made children more likely to give animals human traits, and anthropomorphic books reduced younger children's learning of new facts (Ganea et al., 2014). But in a registered study of children aged 5 to 8 learning about viruses at home, anthropomorphic and realistic books taught equally well, and children who read the anthropomorphic version reported being less afraid of viruses (Menendez et al., 2025).

Familiar characters, responsive characters and talking together

A child's relationship with a character matters. Toddlers learned a sequencing task better from a video when the demonstrator was a character meaningful to them (Lauricella et al., 2011), and toddlers who got to know a new character through videos and toys over three months later learned more from it (Howard Gola et al., 2013). Among 217 children with an average age of almost 5, the strength of their bond with an interactive on-screen character, and how much maths talk they had with it, predicted faster and more accurate answers in the game; transfer to real objects was better when the character was embodied or gave socially contingent replies (Calvert et al., 2020). Adults matter too. Three-year-olds whose parents paused story videos to ask questions understood the stories better and learned more of their vocabulary than children whose parents only commented or who simply watched (Strouse et al., 2013), and preschoolers showed benefits from an animated series about feelings when their viewing was usually accompanied by adult talk (Rasmussen et al., 2016).

For whom and under what conditions

• Age. The video deficit is largest before age 3 and much smaller afterwards (Strouse & Samson, 2021). Children's ability to tell realistic from fantastical story worlds grows between 3 and 5 (Walker et al., 2015), so school-age children should be better at drawing real-world lessons from cartoons, although direct studies with 5- to 9-year-olds are scarce.
• The individual child. In head-to-head comparisons some children learned from only one format (Bloh et al., 2025), and children with stronger imitation skills responded better to video in general (Rayner, 2015).
• Autism and ability. Gains from The Transporters were smaller for children with more severe autism (Gev et al., 2017) or lower cognitive ability (Williams et al., 2012).
• Adult support. The amount of parental support predicted how far emotion learning generalised (Gev et al., 2017), and adult questioning during viewing improved learning (Strouse et al., 2013).
• Familiarity. Characters that children know and care about seem to teach better (Howard Gola et al., 2013).

What this means in practice

• Do not rule out cartoons. In the small studies available, animated models taught as well as filmed ones for most children. Neither format is reliably better.
• Watch your child, not the format. If a child is not learning from one kind of video after a fair try, switching format is reasonable.
• Prefer realistic worlds and solutions. Choose stories where the setting, routine and solution work as they would in real life, even if the characters are drawn or are animals.
• Build the bridge out loud. Pause, ask what the character did, and connect it to your child's own situation: “That's like our school gate.”
• Practise for real. A video prepares; small real steps afterwards turn it into a skill.
• With under-3s, watch together. The youngest children learn least from screens on their own.

What this means for short personalised cartoons

Cala Luna's personalised cartoons show a cartoon lynx version of one real child getting through one hard moment. Nobody has trialled our cartoons, so what follows is reasoning from other studies, not evidence about ours.

What plausibly transfers. Animated models can teach specific behaviours, at least to autistic children learning discrete skills (Bloh et al., 2025). Showing the child's real situation, routine and setting fits the finding that children generalise more from story worlds that resemble reality (Walker et al., 2015), and solutions that follow real-world rules are the ones children carry over (Richert & Schlesinger, 2022). A parent who pauses, asks questions and links the cartoon to the real moment is doing what the co-viewing studies support (Strouse et al., 2013). Personalisation itself is reviewed in our article on personalised stories.

What is uncertain. Animal characters are the honest weak point. One experiment found human characters promoted sharing better than animal characters (Larsen et al., 2018); a larger replication found no difference (Russell & Cain, 2022). No study has tested whether a child recognises an animal character as a version of themselves, or whether that changes how much they learn. On the evidence, fantasy shortcuts such as magic fixes would be a poor choice in a preparation cartoon.

What has not been studied. We found no trial of short, personalised, animated videos for a single hard moment in children aged 5 to 9, with or without animal characters. Most animated-versus-human comparisons involve autistic children learning discrete skills, not anxious moments such as separating at the classroom door or having a haircut.

Limits of the evidence

• The direct comparisons of animated and human models together include fewer than 20 children, all autistic, in single-case designs.
• Results for The Transporters vary between trials. Several of the positive trials share authors with the original evaluation, and independent trials gave mixed results.
• Most developmental studies used storybooks, not video, and measured transfer minutes later in the laboratory.
• Video-deficit research mostly involves children under 3, researcher-made videos and small samples.
• Samples often come from well-educated families in a few countries; in one large character study, 96% of families were college-educated (Calvert et al., 2020).

Open questions for research

1. In 5- to 9-year-olds facing a real hard moment, such as separating at the school gate, does an animated model change behaviour as much as a filmed model of a similar child? A randomised comparison with behaviour observed at the moment itself would answer this.
2. Does an anthropomorphic animal model teach a coping behaviour as well as a cartoon human model? The two sharing studies disagree, and nobody has tested coping behaviours on video.
3. Which matters more for transfer: a realistic setting or a realistic character? A factorial design could vary each separately.
4. Do explicit bridges, such as a parent prompt or a photo of the real place at the end of the video, increase transfer from cartoons?
5. Does the video deficit persist after age 5 for social and emotional learning, rather than object tasks?
6. For autistic children who read cartoon faces more easily, does that advantage predict better learning from animated models?
7. Do gains from animated models last beyond a few weeks, and do they generalise to new people and places?

Frequently asked questions

References

1. Bloh, C., Bacon, L., Begel, B., Madara, K., & Koller, B. (2025). Comparing human video modeling to animated video modeling for learners with autism. The Analysis of Verbal Behavior, 41(2), 262–279. doi.org/10.1007/s40616-025-00224-y · PubMed
2. Gokce, S., & Ulke-Kurkcuoglu, B. (2026). A comparison of video modeling and video-animated modeling in teaching social-affective behaviors to children with ASD. Disability and Rehabilitation: Assistive Technology. Advance online publication. doi.org/10.1080/17483107.2026.2686778 · PubMed
3. Rayner, C. (2015). Video-based intervention for children with autism: Towards improved assessment of pre-requisite imitation skills. Developmental Neurorehabilitation, 18(2), 113–121. doi.org/10.3109/17518423.2014.890959 · PubMed
4. Ho, T. Q., Gadke, D. L., Henington, C., Evans-McCleon, T. N., & Justice, C. A. (2019). The effects of animated video modeling on joint attention and social engagement in children with autism spectrum disorder. Research in Autism Spectrum Disorders, 58, 83–95. doi.org/10.1016/j.rasd.2018.09.004
5. Kellems, R. O., Charlton, C., Kversøy, K. S., & Győri, M. (2020). Exploring the use of virtual characters (avatars), live animation, and augmented reality to teach social skills to individuals with autism. Multimodal Technologies and Interaction, 4(3), 48. doi.org/10.3390/mti4030048
6. Golan, O., Ashwin, E., Granader, Y., McClintock, S., Day, K., Leggett, V., & Baron-Cohen, S. (2010). Enhancing emotion recognition in children with autism spectrum conditions: An intervention using animated vehicles with real emotional faces. Journal of Autism and Developmental Disorders, 40(3), 269–279. doi.org/10.1007/s10803-009-0862-9 · PubMed
7. Young, R. L., & Posselt, M. (2012). Using The Transporters DVD as a learning tool for children with autism spectrum disorders (ASD). Journal of Autism and Developmental Disorders, 42(6), 984–991. doi.org/10.1007/s10803-011-1328-4 · PubMed
8. Williams, B. T., Gray, K. M., & Tonge, B. J. (2012). Teaching emotion recognition skills to young children with autism: A randomised controlled trial of an emotion training programme. Journal of Child Psychology and Psychiatry, 53(12), 1268–1276. doi.org/10.1111/j.1469-7610.2012.02593.x · PubMed
9. Gev, T., Rosenan, R., & Golan, O. (2017). Unique effects of The Transporters animated series and of parental support on emotion recognition skills of children with ASD: Results of a randomized controlled trial. Autism Research, 10(5), 993–1003. doi.org/10.1002/aur.1717 · PubMed
10. Chan, J. K.-Y., Cheung, T. C.-K., Chan, C.-W., et al. (2024). Enhancing emotion recognition in young autistic children with or without attention-deficit/hyperactivity disorder in Hong Kong using a Chinese App version of The Transporters. Autism, 28(4), 945–958. doi.org/10.1177/13623613231187176 · PubMed
11. Rosset, D. B., Rondan, C., Da Fonseca, D., Santos, A., Assouline, B., & Deruelle, C. (2008). Typical emotion processing for cartoon but not for real faces in children with autistic spectrum disorders. Journal of Autism and Developmental Disorders, 38(5), 919–925. doi.org/10.1007/s10803-007-0465-2 · PubMed
12. Brosnan, M., Johnson, H., Grawmeyer, B., Chapman, E., & Benton, L. (2015). Emotion recognition in animated compared to human stimuli in adolescents with autism spectrum disorder. Journal of Autism and Developmental Disorders, 45(6), 1785–1796. doi.org/10.1007/s10803-014-2338-9 · PubMed
13. Cross, L., Farha, M., & Atherton, G. (2019). The animal in me: Enhancing emotion recognition in adolescents with autism using animal filters. Journal of Autism and Developmental Disorders, 49(11), 4482–4487. doi.org/10.1007/s10803-019-04179-7 · PubMed
14. Atherton, G., Robinson, L., Smith, L. G., & Cross, L. (2025). The wind in the willows effect: Does age affect human versus animal faux pas recognition? Journal of Experimental Child Psychology, 249, 106116. doi.org/10.1016/j.jecp.2024.106116 · PubMed
15. Barr, R. (2010). Transfer of learning between 2D and 3D sources during infancy: Informing theory and practice. Developmental Review, 30(2), 128–154. doi.org/10.1016/j.dr.2010.03.001 · PubMed
16. Strouse, G. A., & Samson, J. E. (2021). Learning from video: A meta-analysis of the video deficit in children ages 0 to 6 years. Child Development, 92(1), e20–e38. doi.org/10.1111/cdev.13429 · PubMed
17. Sommer, K., Slaughter, V., Wiles, J., & Nielsen, M. (2023). Revisiting the video deficit in technology-saturated environments: Successful imitation from people, screens, and social robots. Journal of Experimental Child Psychology, 232, 105673. doi.org/10.1016/j.jecp.2023.105673 · PubMed
18. Richert, R. A., Shawber, A. B., Hoffman, R. E., & Taylor, M. (2009). Learning from fantasy and real characters in preschool and kindergarten. Journal of Cognition and Development, 10(1–2), 41–66. doi.org/10.1080/15248370902966594
19. Richert, R. A., & Smith, E. I. (2011). Preschoolers' quarantining of fantasy stories. Child Development, 82(4), 1106–1119. doi.org/10.1111/j.1467-8624.2011.01603.x · PubMed
20. Walker, C. M., Gopnik, A., & Ganea, P. A. (2015). Learning to learn from stories: Children's developing sensitivity to the causal structure of fictional worlds. Child Development, 86(1), 310–318. doi.org/10.1111/cdev.12287 · PubMed
21. Mares, M.-L., & Sivakumar, G. (2014). “Vámonos means go, but that's made up for the show”: Reality confusions and learning from educational TV. Developmental Psychology, 50(11), 2498–2511. doi.org/10.1037/a0038041 · PubMed
22. Li, H., Boguszewski, K., & Lillard, A. S. (2015). Can that really happen? Children's knowledge about the reality status of fantastical events in television. Journal of Experimental Child Psychology, 139, 99–114. doi.org/10.1016/j.jecp.2015.05.007 · PubMed
23. Hopkins, E. J., & Lillard, A. S. (2021). The Magic School Bus dilemma: How fantasy affects children's learning from stories. Journal of Experimental Child Psychology, 210, 105212. doi.org/10.1016/j.jecp.2021.105212 · PubMed
24. Richert, R. A., & Schlesinger, M. A. (2022). Relations between fantasy and transfer of learning from storybooks. Journal of Experimental Child Psychology, 222, 105474. doi.org/10.1016/j.jecp.2022.105474 · PubMed
25. Brenner, J., Steele, K., & Woolley, J. D. (2025). Children learn cause-and-effect relations from fantastical and realistic storybooks. Journal of Experimental Child Psychology, 253, 106198. doi.org/10.1016/j.jecp.2025.106198 · PubMed
26. Larsen, N. E., Lee, K., & Ganea, P. A. (2018). Do storybooks with anthropomorphized animal characters promote prosocial behaviors in young children? Developmental Science, 21(3), e12590. doi.org/10.1111/desc.12590 · PubMed
27. Russell, S. J., & Cain, K. (2022). The animals in moral tales: Does character realism influence children's prosocial response to stories? Journal of Experimental Child Psychology, 219, 105392. doi.org/10.1016/j.jecp.2022.105392 · PubMed
28. Ganea, P. A., Canfield, C. F., Simons-Ghafari, K., & Chou, T. (2014). Do cavies talk? The effect of anthropomorphic picture books on children's knowledge about animals. Frontiers in Psychology, 5, Article 283. doi.org/10.3389/fpsyg.2014.00283 · PubMed
29. Menendez, D., Richardson, E., McNeil, K. M., & Gelman, S. A. (2025). Discovering the world of viruses: Testing the influence of anthropomorphic representations on children's learning about COVID-19. Developmental Psychology, 61(3), 513–529. doi.org/10.1037/dev0001882 · PubMed
30. Lauricella, A. R., Gola, A. A. H., & Calvert, S. L. (2011). Toddlers' learning from socially meaningful video characters. Media Psychology, 14(2), 216–232. doi.org/10.1080/15213269.2011.573465
31. Howard Gola, A. A., Richards, M. N., Lauricella, A. R., & Calvert, S. L. (2013). Building meaningful parasocial relationships between toddlers and media characters to teach early mathematical skills. Media Psychology, 16(4), 390–411. doi.org/10.1080/15213269.2013.783774
32. Calvert, S. L., Putnam, M. M., Aguiar, N. R., Ryan, R. M., Wright, C. A., Liu, Y. H. A., & Barba, E. (2020). Young children's mathematical learning from intelligent characters. Child Development, 91(5), 1491–1508. doi.org/10.1111/cdev.13341 · PubMed
33. Strouse, G. A., O'Doherty, K., & Troseth, G. L. (2013). Effective coviewing: Preschoolers' learning from video after a dialogic questioning intervention. Developmental Psychology, 49(12), 2368–2382. doi.org/10.1037/a0032463 · PubMed
34. Rasmussen, E. E., Shafer, A., Colwell, M. J., et al. (2016). Relation between active mediation, exposure to Daniel Tiger's Neighborhood, and US preschoolers' social and emotional development. Journal of Children and Media, 10(4), 443–461. doi.org/10.1080/17482798.2016.1203806

Spotted an error, or know a study we should add? Email hello@calalunakids.com. Every study we draw on is in the research library; research terms are explained in our methods glossary.