Cross-Platform Learning and the Role of Transfer
Together, the above data point to two types of transfer of learning that can operate in cross-platform learning: (1) applying content acquired from educational media to yield greater performance in posttest assessment tasks and (2) applying content acquired from one form of educational media to produce richer engagement with a second educational media platform. The first type of transfer is not unique to cross-platform learning. It is often found in research on learning from a single media platform, where (for example) exposure to a television series results in significantly more sophisticated performance in novel posttest tasks (Fisch, 2004). However, the present data demonstrate that such transfer effects may be stronger when children are exposed to multiple treatments of educational content in related media platforms. In the present study, many pretest-posttest effects were stronger among the DVD + Web group than among either the DVD Only or the Web Only group.
Notably, contrary to our expectations, the same was not true of the All Materials group, which used all of the same materials as the DVD + Web group plus teacher-led hands-on materials. We cannot be certain why the All Materials group did not perform at the same level as the DVD + Web group, but we hypothesize that it may be because the All Materials group was the only one that used Cyberchase
materials every day; perhaps this schedule was excessive in light of the other constraints on teachers’ schedules, and was simply too much for participants to integrate effectively. Further research is necessary to determine whether an “optimal level” of media use exists. Nevertheless, data from the DVD + Web group suggest that cross-platform learning can hold added benefits for transfer of learning.
The second type of learning transfer—transfer from one educational medium to another—is unique to cross-platform learning. In the present study, online tracking data revealed that children who used multiple media employed more sophisticated strategies while playing three online games, and produced more correct responses while playing two of the three games. Just as in the posttest tasks, it appears that children took the educational content they encountered in one medium (television and/or hands-on activities) and applied it while engaging with mathematics content in another medium (online games). This transfer of learning supported their interaction with the second medium, allowing children to apply more sophisticated approaches and producing a richer, more successful engagement with the material.
How, then, did cross-platform learning contribute toward transfer of learning—and toward greater transfer to posttest tasks? One possible explanation is simply that children who used multiple Cyberchase media spent more time engaging with their embedded mathematics content. To some degree, this explanation is probably at least partially correct. Indeed, one of the chief purposes of informal education is precisely that—to encourage children to spend more time with educational content than they would otherwise. However, time clearly cannot explain the present findings by itself, because the benefits found for the DVD + Web group were not equaled by the All Materials group, which devoted even more time to Cyberchase activities. If time were the sole explanation, the gains shown by the All Materials group would have been at least as large as those of the DVD + Web group, if not larger.
A more promising explanation may lie in the concept of varied practice discussed in the educational research literature on transfer of learning (e.g., Gick & Holyoak, 1983; Salomon & Perkins, 1989; Singley & Anderson, 1989). As noted earlier, in varied practice, learners are provided with multiple examples of the same concept or repeated practice of a skill in multiple contexts, which increases the likelihood that the learner will apply the material in new tasks or situations. When children in the present study encountered mathematics and problem-solving content in multiple Cyberchase media, they were clearly engaged in varied practice, especially in those instances where there was close alignment among the content of a related television episode, hands-on activity, and online game. Effects within the online tracking data attest to children’s connecting the content of the different media, and even applying the content learned from one medium while they were learning from the other. Not only did children gain additional, varied practice by using multiple media, but their engagement with the latter medium was richer and more sophisticated. In this way, cross-platform learning has the potential to support learning by contributing to two types of transfer: transfer across educational media platforms (resulting in richer engagement and understanding), and transfer from educational media to new problems or situations encountered subsequently (such as our posttest assessments).
Moreover, it is quite possible that transfer may even be facilitated by the presence of the same characters and contexts across media. Past research on transfer of learning has shown that transfer is more likely to occur when two situations appear similar on their face (surface structure similarity) than when they are dissimilar on their surface but rest on similar underlying principles (deep structure similarity; e.g., Bassok & Holyoak, 1993; Gentner & Forbus, 1991). Thus, for example, encountering Cyberchase characters in an online game might lead children to think of other times when they saw the same characters (e.g., on television). This situation could facilitate the transfer of information and skills from one medium to another, in a way that seeing different characters on television and in a game might not. (For an extended discussion of these issues that draws on data from several contemporaneous studies as well, see Fisch, 2013).