Bridging Content & Child Characteristics:
The Role of Epistemic Curiosity in Children’s Learning
from Educational Television and Apps
Research Master Thesis in Communication Science
Presented to the Graduate School of Communication at the University of Amsterdam Author Tara Kaldenbach
Student number 5898153
Under supervision of Jessica Taylor Piotrowski, Ph. D.
Abstract
Today, touch screen devices such as iPads (and tablets) are immensely popular among very young children. Their intuitive design, alongside their interactive features, seemingly makes them an ideal tool to aid children in academic or social learning processes. Indeed, many apps targeting young children claim to support young children’s educational needs. Yet, while there is a long research history which demonstrates the educational opportunities of educational television for youngsters, there is nearly no research on whether touchscreens may also meet similar educational needs. As such, it is unclear whether touchscreens may be effective educational tools – nor is it clear how they might compare to more traditional media like television. To address this gap, an experimental study was conducted with 36 preschool-aged children (M age = 60 months). Based on the affordances of touchscreens, it was
hypothesized that children would learn greater content from touchscreens than from television content. Moreover, it was hypothesized that the benefits of touchscreens would be particularly pronounced for children with a greater need for information (i.e., epistemic curiosity). In all, results indicate that young children can indeed acquire educational information from
touchscreen media – with knowledge acquisition almost twice as large as that from televised content. While epistemic curiosity did not seem to impact learning, patterns suggest that interactive platforms may be particularly beneficial for particular subsets of children. These findings encourage researchers to ask more questions about which type of children (i.e., which individual differences) can profit most from interactional educational (app) content. Keywords: app, educational media, epistemic curiosity, interactivity, iPad, learning, literacy.
Dedication
I dedicate this thesis to my dearest late father, Jan. Making you proud has given me the strength to pursue finishing my Research Master’s degree.
Acknowledgements
I want to express my gratitude to all of the participants in the study: teachers, parents and children. Without their participation, this research would not have been possible. I especially want to thank my supportive supervisor, dr. Jessica Piotrowski, who has inspired me,
provided valuable insights into my thesis writing and supported me throughout the process and completion of my Research Master thesis project.
Introduction
Since their first release in spring 2010, touch screen devices such as iPads and tablets and their applications (apps) have become universally widespread with a far-reaching range of users, including young kindergarten children (Costello, 2012). For example, in the
Netherlands, data suggest that by the age of three-and-a-half nearly half of all Dutch children are consuming touchscreen technology on a daily basis (Pijpers, 2011). Similar patterns are found throughout developed countries worldwide. Among other characteristics, the popularity of this technology for very young children is often attributed to the fact that touch screen devices are eye-catching, portable and intuitive to use (Neumann, 2014).
Presently, there are countless available apps for young children (both free and paid) which claim to offer educational benefits for their users. For instance, the App Store (by Apple) has more than 800,000 apps which claim to support the development of young children’s academic (e.g., literacy, mathematics and creativity) or prosocial (e.g., sharing, emotional understanding) skills (Vaala, 2014; Goodwin, 2012; Hatherly & Chapman, 2013; Henderson & Yeow, 2012). While this vast number of apps may seem surprising to some, in truth, it is actually quite reasonable that the children’s app landscape is becoming increasingly filled with apps which market themselves as educational for two main reasons. First, research consistently shows that parents often look for media content which can support their child’s educational needs – thus creating such content is likely to be profitable (Shuler, Levine & Ree, 2012; Chiong & Shuler, 2010). Second, there is a long history of successful media content when it comes to television – both in terms of beneficial effects on children as well as financial success (Revelle, Strommen & Medoff, 2001; Linebarger, Kosanic, Greenwood & Doku, 2004; Hisrich & Blanchard, 2009). This is especially true for television designed to support native and foreign language development (Rice & Woodsmall, 1988; Cummins, 1979; Lesaux & Siegel, 2003) where it seems that the visual-verbal redundancy of television is
particularly powerful (Collins, 1983; Strommen & Revelle, 1990; Fisch, 2004; Fisch, 2014). Considering that there is a (1) a long history of parents seeking educational television content for their youngsters, (2) a multitude of studies demonstrating the ability of educational television to beneficially support children’s development, and (3) documented financial success of such content, it is reasonable that app developers are anxious to recreate this “magic” of educational media via the creation of apps. Moreover, considering the strong body of work that argues that the visual-verbal redundancy of television makes it particularly well-suited for language support, it is not surprising that we are seeing a fast rise in the number of apps which claim to support language development. Similar to television’s visual-verbal redundancy attributes, apps have distinctive features which may be uniquely attuned to support language skills. For instance, they allow children to interact with oral, visual, and written communication (either unconnectedly or in combination), provide feedback or rewards during the learning process to encourage motivation and sustained engagement, and progressively increase in difficulty to maintain a child’s interest (Cohen, Hadley & Frank, 2011; Hutchison, Beschorner & Schmidt-Crawford, 2012; Hatherly & Chapman, 2013). That said, given the novelty of touchscreen media and the nearly non-existent empirical research on the topic, it remains less clear whether this medium is as effective as television at
supporting educational outcomes – particularly language skills.
On the one hand, it is possible that television and touchscreen media are equally powerful in supporting children’s learning. Both offer a unique set of attributes which may work to support children’s language learning. On the other hand, it may be that these attributes work differently for different children. Indeed, it is possible that certain children learn more by interacting with the content (i.e., via touch screen media) while other children may benefit more from a more observational way of learning (i.e., via television). For example, research indicates that children have different informational needs (i.e., epistemic
curiosity) with some children having a much greater need for information than their peers. One can imagine that such differences in informational needs may influence the extent to which children learn content from television versus touchscreen media.
Thus, two key questions remain: does learning from a touchscreen device (such as an iPad or tablet) differ from learning from televised content? Or is this efficacy in part
dependent on user characteristics? This study is designed to address these questions by evaluating the extent to which preschool children (between three and five years old) can learn (foreign) literacy skills from (interactional) play-based app content and (observational) televised content on an iPad – with a particular focus on the extent to which learning is dependent upon relevant user characteristics (namely, epistemic curiosity). The results of this study provide information on the educational competence of different types of (interactional or observational) media with very young children.
Theoretical Insights on Children’s Learning from Media
Although research on children’s learning from new digital media remains rather limited, work with educational television highlights the complexity of the learning process and the key factors that influence whether children extract and recall information from the media they use. In particular, theories on children’s learning from media highlight the
criticality of content format and user background (both individually and conjunction) when it comes to children’s learning from education media.
In terms of content, one of the key theories to explain when and why children may learn from educational media is Fisch’s capacity model (Fisch, 2000). The model is based on the proposition that children’s working memory is limited, and for maximum learning to occur, the processing that the media content requires should not exceed the child’s available cognitive resources. Furthermore, the capacity model argues that all media content consists of two forms of content: narrative content and educational content and, importantly, the degree
to which educational content (e.g., identifying letters) is intertwined with the narrative (the storyline) will play a direct role in whether children are able to extract and retain the educational information (Fisch, 2000). Specifically, when both forms of information are integrated, comprehension of content is expected to improve as is overall attitude towards the content (Buijzen, Van Reijmersdal & Owen, 2010; Fisch, 2000). This is because, rather than overtaxing a child’s working memory, integrated content is sensitive to children’s limited working memory. The capacity model further argues that content features which are designed to improve processing of either the narrative or educational content (e.g., participatory cues in content, Piotrowski, 2014a) can aid processing and subsequent learning. Similarly, the model also argues that individual differences which may improve content processing (e.g., a child’s narrative understanding or size of working memory, Piotrowski, 2014b) can also aid
processing and subsequent learning.
While Fisch acknowledges that individual differences do influence learning from educational media, his model places a strong focus on content structure. More recent
conceptualizations of media effects, however, argue that individual audience differences are just as important as content structure. Indeed, the differential susceptibility to media effects model (DSMM, Valkenburg & Peter, 2014) persuasively argues that variances in personality (i.e., dispositional factors) and family and peer influence (i.e., social factors) may influence the degree to which one is influenced by media content. These factors may not only influence the type of media that one consumes but also how one is affected by media content. As such, just as the structure of educational media content can influence whether learning effects occur, so too can individual differences amongst users. Following this line of thought, any research which investigates children’s learning from media should pay careful attention not only to the media itself but also to relevant individual differences.
Bridging Media and Individual Differences
Even though young children increasingly use touch screen media for playing apps and viewing video content, in general, they still prefer televised content (Gutnick, Robb, Takeuchi & Kotler, 2011). With both medium types, there are distinct challenges and opportunities which have implications for the effectiveness of educational media (Fisch, 2004). Most notably, processing television content is significantly different than processing interactive content (Fisch, 2004). With television, the cognitive demands of processing a program are affected by the nature of television itself. Not only must the user integrate the visual and auditory information of the content, but the user must do this within the confines of the content pacing (Eckhardt, Wood, & Jacobvitz, 1991). In other words, the rapidness of the incoming information cannot be controlled and, as such, can result in difficulties in
understanding the content (Pace, 1980). Moreover, the content is static – it does not adjust in difficulty to meet the development needs of its user. Contrast this with interactive media which is both self-paced and typically more closely aligned with the user’s cognitive needs via scaffolded content – in other words, features which are associated with improved learning (Revelle, Strommen & Medoff, 2001; Schauble, 1990).
Although no existing studies have compared the processing of television with the processing of mediated touchscreen content, it seems likely that the structure of touchscreen content may lend itself to superior learning. This argument is bolstered by the fact that researchers have indeed shown that, when compared to simply observing content, being actively involved with mediated content (as is commonplace with touchscreen technology) can lead to improved understanding for young children (Calvert, Strong, Jacobs & Conger, 2007; Calvert, Strong & Gallagher, 2005; Lauricella, Pempek, Barr & Calvert, 2010; Feldman, 2004). As such, it seems reasonable to hypothesize that young children will
demonstrate greater learning gains by playing (touchscreen) app content than from viewing similar televised content (Hypothesis 1).
Of course, while the existing literature on young children’s learning from media would suggest that the active content engagement promoted by touchscreen apps will lead to
superior learning, it is important to remember that media effects theories (e.g., the capacity model and the DSMM) postulate that individual differences can augment these effects in unique ways (Fisch, 2000; Valkenburg & Peter, 2013). Although there are many relevant individual difference variables that likely influence the degree to which children learn from television and apps, one particularly relevant construct is epistemic curiosity (Piotrowski, Litman & Valkenburg, 2014).
Epistemic curiosity is defined as “the desire to obtain new knowledge” either “by producing positive experiences of intellectual interest” (I-type) or “by reducing undesirable conditions of informational deprivation” (D-type) (Piotrowski, Litman & Valkenburg, 2014, p.542). Children with heightened epistemic curiosity are said to have an increased “drive to know” new information, and as such, are expected to work harder and be more motivated to seek out information. Importantly, epistemic curiosity is associated with children’s preferred learning style – namely, whether they prefer a mastery-oriented approach to learn (e.g., goals that encompass expending effort in order to learn and appreciate new intrinsic interests) or a performance-oriented approach (where accuracy and a strong desire to solve information gaps to attain knowledge is seen).
While increased I-type epistemic curiosity is generally associated with both learning styles, D-type epistemic curiosity tends to distinguish itself more. Specifically, children with increased D-type epistemic curiosity tend to prefer a more performance-oriented approach to learning (Piotrowski, Litman & Valkenburg, 2014; Litman, 2008; Richards, Litman, & Roberts, 2013). This is a particularly relevant distinction when one considers the fact that the
majority of educational apps favor a performance-oriented approach (Markopoulos, Reed, MacFarlane & Hoysniemi, 2008). In apps, children have a greater chance to face
(challenging) information, and must try to make sense of this information and then devote considerable effort to solving the challenge presented. Apps typically require a high-level of action (i.e., interacting with game elements successfully to reach next level) and in doing so, rely heavily on repetitive input-based action from their users. These apps offer multiple new learning practices by encouraging interaction with educational content, while at the same time appealing to the natural epistemic curiosity of young children – particularly children who favor performance-based learning (i.e., children with heighted D-type epistemic curiosity, Cohen, Hadley & Frank, 2011). The same cannot be said of educational television content which, in general, requires a more passive learning experience and does not – by definition (or to a much lesser extent) – trigger the same performance-oriented goals.
While overall it seems fair to hypothesize that apps will lead to superior learning over television content, the literature on epistemic curiosity would suggest that these benefits may be particularly pronounced for children with heightened D-type curiosity. On the other hand, for television, it is likely that such differences would be minimal to non-existent. As such, it is hypothesized that children with a higher level of (D-type) epistemic curiosity (who desire to attain new knowledge by determined problem solving) will demonstrate greater learning from touchscreen content when compared to children with a lower level of (D-type) epistemic curiosity, whereas no such distinction is expected for television content (Hypothesis 2).
Method
To address study hypotheses, a within-subjects experimental design was conducted whereby all participants viewed an educational literacy-based television clip and played an interactive educational literacy-based app on an (second-generation) iPad. Counterbalancing was not implemented in this study (see stimuli section for detailed reasoning).
Sample
Children were selected from two elementary schools in a small town in The
Netherlands. A total of 36 children between 49 and 71 months (M = 60.41, SD = 7.48; 58% female) participated and completed the experiment. Detailed information on demographics, media use, and predominant language was gathered through parent surveys (n=19, 53% return rate). From all mothers who fulfilled the questionnaire, the minority finished secondary education (10,5%), followed by vocational education (e.g., MBO, 26.3%), University education (e.g., WO, 26.3%) and higher education (e.g., HBO, 36.8%). In 95% of the
households, Dutch was the predominantly language spoken (5% spoke Portuguese). For these children, almost half of them have one younger sibling (47.4%) or one older sibling (36.8%). On an average day, the majority of children (57.9%) reportedly spent 60 minutes per day on watching television programs, either on an iPad or another device. Nearly three-quarters of the participating children spent upwards of thirty minutes per day playing digital games. None of the participating children were familiar with the stimuli used in this study.
Stimuli
The experimental stimuli was an episode of a popular (English) educational children’s television program named Super Why! and the (English) Super Why! app with interactive literacy games. Super Why! is a program developed for 3- to 6-year-olds. Its first television season was appraised for the program’s impact on early literacy skills. Indeed, findings revealed that viewing Super Why! content effectively supports various English literacy skills including phonemic awareness, letter recognition, and speech to print matching (Linebarger, McMenamin, & Wainwright, 2009; Linebarger, 2015). This specific stimuli was chosen because similar content existed in both a television format and app format. With similar content, the likelihood that features such as characters or content design would differentially influence learning is minimized (since they are the same in each) and, as such, there is greater
confidence that it is the structure (television versus app) that is influencing content
comprehension. At the time of this study, no Dutch educational media offered comparable app and television content. Since research suggests that children may acquire literacy skills in English in a similar way as they do for their native language (Jongejan, Verhoven & Siegel, 2007; Chiappe, Siegel & Wade-Woolley, 2002), Super Why! was considered a reasonable selection in the context of learning secondary language literacy skills.
Although within-subjects designs generally recommend a counterbalancing approach to avoid any potential ordering effects, in this study, counterbalancing was not used because there was some concern that viewing the television program first would affect the experience with the app. In particular, it was assumed that all participants would be unfamiliar with the selected content. The television content introduces its viewers to the main characters in the show whereas the app content does not make such a formal introduction (but does rely on the characters). If the television content was featured after the app content, participants would likely have a qualitatively difference experience as a result of their unfamiliarity with the characters. To ensure a similar experience across participants, all children viewed the television content first and then continued with game play.
Television stimuli. In the (originally 24-minute) Super Why! Episode (episode 1 of
season 1), one of the characters comes across a problem. The characters then transform into literacy-powered super heroes (e.g., Alpha Pig with “Alphabet Power”). While playing reading games and activities, they practice crucial literacy skills in order to solve the story problem. In this particular episode, the activities are identifying letters (by both Alpha Pig and Princess Presto) and identifying words and picking words that fit into a sentence (by Super Why). To avoid participant fatigue, the episode was edited into a 15-minute version while still maintaining the narrative structure and participatory cues. This way the episode keeps the
premise of educational learning as the educational content (literacy message) and the entertaining content (storyline and characters) remained intertwined (Fisch, 2000).
App stimuli. The Super Why! App called “SUPER WHY!” (by PBS KIDS) features
four interactive literacy games. In this study, three games were selected that mimicked the same skills taught in the television episode. The first game is “Alpha Pig’s Lickety Letter Hunt” where a user needs to identify letters. The second is “Princess Presto’s Wands-Up Writing” where a user needs to help identify letters, letter sounds, and write words. And lastly, “Super Why’s Story Saver” requires the user to identify and select words that fit into sentences in order to complete story sentences (http://pbskids.org/apps/super-why-app.html). The mini-games contain a very simple, short narrative.
Measures
A selection of measures was used to evaluate study hypotheses. In particular, measures were used to ascertain learning from the television and app content as well as evaluate the child’s epistemic curiosity. Moreover, several variables were evaluated as potential covariates.
Television-specific content comprehension. To measure the dependent variable,
recall after televised content (i.e., content comprehension), 10 questions were developed based on the key objectives of the content. The objectives in the episode were to learn how to identify letters and words, and to select words that fit into sentences in order to complete the story sentence. These goals were operationalized (example: “Alpha pig needed to find all of the letters in the word ‘wolf’, which one is the letter ‘w’?”) with large plasticized printed screen images to support questions. The scores across the 10 items (0= wrong answer, 1= correct answer) were summed to create a percentage score with higher scores representing greater comprehension of the educational content (see Appendix A for all of the administered items).
App-specific content comprehension. As with the television-specific content
comprehension, each participant answered researcher-designed questions on recall (e.g., content comprehension) after playing the literacy game stimuli. The objectives in the mini-games were to learn how to identifying letters, words and sounds, write words, and select words that fit into sentences in order to complete story sentences. These goals were similar to those in the television content – thus helping to ensure that the challenge of the questions for both content types was similar. In all, the learning goals for the app content were assessed via 9 items, with sub-items, (example question: “So you just had to find the word ‘BOX’. We need to find all the letters in the word ‘BOX’. Do you see the letter ‘B’?”) with large plasticized printed screen images to support questions. The scores (0= wrong answer, 1= correct answer) were summed to create a percentage score with higher scores representing greater
comprehension of the educational content (see Appendix A for all of the administered items).
Epistemic curiosity. Children’s level of epistemic curiosity (D-type) was evaluated
using 5 items (4 point scale) from the epistemic curiosity measurement tool by Piotrowski, Litman & Valkenburg (2014). These items were administered as part of the teacher survey. The scale was internally consistent, with a higher score representing a higher level of (D-type) epistemic curiosity (α = .92, M = 2.29, SD = .87).
Working memory. Children’s working memory was evaluated using the 10-item
scale (response ranges from 1 through 3) from Smidts & Huizinga (2009). These items were administered as part of the teacher survey. The scale was internally consistent, with a higher score representing greater working memory challenges (α = .95, M = 1.56, SD = .57). This variable was included as a potential covariate.
Native (Dutch) literacy skills at pretest. To identify a child’s level of emergent
(native) literacy skills, a one-minute standardized task was assessed - the Picture Naming Test (PNT)(Missall & McConnell, 2004). This task is designed to indicate signs of expressive
language development (i.e., by showing multiple random pictures –binocular, horse, apple etc.- and ask “What is this?”) with a higher score representing a greater knowledge of expressive language development (M = 21.18, SD = 5.88). This variable was included as a potential covariate.
Foreign (English) literacy skills at pretest. To identify a child’s level of emergent
(foreign) literacy skills, a one-minute standardized task was performed - the DIBELS Initial Sound Fluency (ISF) (Good & Kaminski, 2002). The task requires the child to produce initial sounds in an orally presented word. The questions were posed in Dutch, with English words being pronounced in English, and English phonemes to be answered. (e.g., “This is the word ‘skate’... Which picture begins with the /sk/?”) A higher score represented a higher level of foreign literacy skills (M = 3.47, SD = 1.76). This variable was included as a potential covariate.
Procedure
Prior to data collection, ethical approval from the Institutional Review Board at the University of Amsterdam was received. Following this, two schools were recruited for participation. Parents were informed about the research by mail (from the elementary school management) with an information sheet and passive informed consent form. Parents also received a questionnaire with a series of questions regarding household demographics,
children’s home media use, and language. In addition to parent report, all elementary teachers completed a brief survey regarding each participating child whereby information on the child’s epistemic curiosity and working memory could be ascertained.
The experiment was conducted by one researcher within an available space at the school (i.e., an empty classroom). The researcher was introduced in the classroom and the teacher told the children that they would “accompany the researcher to help out while watching television clips and playing games”.
At the start of each session, the researcher broke the ice by providing a coloring page, crayons, and a small talk. Then, two literacy tasks were completed to assess the level of native and foreign literacy skills. Following this, the television stimuli was shown. Viewing was completed individually with only the researcher present. Children were then asked to answer questions about the content of the clip in order to assess recall in terms of educational content comprehension. Subsequently, children were asked to play three rounds of each of three literacy games of an app. After completing three rounds of each mini-game, children completed questions on educational content-specific comprehension. Crayons and color sheets were available during the television viewing as well as the game playing. In total, the child visit lasted approximately 45 minutes per child.
Participating schools were compensated with age-appropriate educational books and children received a small gift to thank them for their time.
Analytic Approach
Descriptive statistics were calculated for all variables to ascertain normality and identify potential outliers. Relevant statistics for all model variables can be found in Table 1. Following this, a bivariate correlation matrix was used to examine relationships between planned model variables and model covariates (see Table 2). To evaluate hypothesis 1, a Repeated Measures Analysis of Covariance (ANCOVA) was implemented in which televised content and app content comprehension were treated as within-subjects variables. To evaluate hypothesis 2, a Repeated Measures ANCOVA was again implemented while including (D-type) epistemic curiosity as a between-subjects factor. Both models included four control variables: age, native and foreign literacy skills, and level of working memory. Gender was evaluated as a potential covariate but did not correlate with either dependent variable and thus dropped for model parsimony. Models met the required statistical assumption of parametric testing. All analyses were computed in SPSS (version 23.0).
Results Descriptive Statistics & Variable Intercorrelations
The sample involved 20 girls (58%) and 16 boys (42%) with an average age of 60.4 months (SD =7.84). Descriptive statistics indicated that children score higher on app recall (e.g., content comprehension) (M = 62.56%, SD = 16.04) than television recall (M = 38.24%, SD = 19.92).
A Pearson correlation coefficient matrix was computed to assess possible relationships between all model variables (see Table 2). Unsurprisingly, television and app comprehension are moderately correlated with one another (r = .56, N = 36, p < .001). In addition, as is
reasonable to expect, age correlated positively with both television (r= .44, N = 36, p < .001) and app-specific content comprehension (r = .44, N = 36, p < .001). Concerning prior literacy skills, the level of native language skills has a positive correlation with both
television-specific comprehension (r = .44, N = 34, p < .001) and app-television-specific content comprehension (r = .47, N = 34, p < .001) while foreign language skills only correlated with app-specific content comprehension (r = .37, N = 36, p < .005). D-type epistemic curiosity was positively associated with television comprehension (r = .38, N = 36, p < .005) whereas it was,
somewhat surprisingly, unrelated with app comprehension. Lastly, increased problems with working memory was associated with decreased app comprehension (r = -.36, N = 36, p < .005) but unrelated to television comprehension.
Hypothesis 1: Television versus Touchscreen Performance
The first hypothesis predicted that children would demonstrate greater content comprehension when playing with a touchscreen app than from viewing similar televised content. Overall, results of the Repeated Measures ANCOVA model controlling for child age, pre-test literacy skills, and working memory indicate that the differences between television and app recall are not statistically significant, F(1,29) =1.25, p = .27 (multivariate main
effect). However, in line with hypothesis 1, a follow-up pairwise comparison between television recall (covariate adjusted M = 38.23, 95% CI [32.11, 44.36]) and app recall (covariate adjusted M = 62.56, 95% CI [57.61, 67.50]) does suggest a significant difference between means (p < .05). While caution should be used when interpreting mean differences within the context of a non-significant omnibus multivariate effect, the size of the difference between app and television recall alongside the small sample (and, thus, underpowered analysis) suggests findings in support of hypothesis 1.
Hypothesis 2: Epistemic Curiosity as Moderator
The second hypothesis anticipated that, while app recall would be greater than
television recall overall, children with heightened D-type curiosity would particularly benefit from app content. Results from the Repeated Measures ANCOVA with D-type curiosity as a between-subjects factor indicated that D-type curiosity did not statistically moderate app recall, F(1,28) = 1.15, p = .29. Interestingly, while not statistically significant, follow-up pairwise comparisons suggest a somewhat reverse pattern than hypothesized such that children with heightened D-type EC (covariate adjusted M = 58.13, 95% CI [49.59, 66.66]) seem to recall less content from apps than their lower EC peers (covariate adjusted M = 66.50, 95% CI [58.59, 74.41) (See Figure 1). Hypothesis 2 is rejected.
Discussion
The objective of this experimental study was to examine whether (literacy) learning from (interactional) play-based app content on a touch screen device (i.e., an iPad) would differ from learning from more traditional (observational) televised content. Moreover, guided by the propositions of the capacity model (Fisch, 2000) and differential susceptibility to media effects model (DSMM, Valkenburg & Peter, 2013), the study also asked whether the efficacy of learning from app content would be in part dependent on a young child’s D-type epistemic curiosity. It was hypothesized that youngsters would, in general, comprehend
greater content from the app content than from television content, and further, that these gains would be particularly apparent among children with heightened D-type epistemic curiosity.
In all, results offered both predicted and unforeseen findings. Most importantly, the study does lend support to the expectation that children would recall more content from apps than from television. Given limitations associated with statistical power, it is certainly advised to replicate this study with a larger sample to ensure that effects remain. That said, the fact that comprehension of content was nearly two times as large from app content as from television content suggests that touchscreen technology may be a worthy addition to the educational media landscape. Apps on a touch screen device offer children the ability to interact with and engage with educational content, providing them valuable opportunities to rehearse the content in visual, verbal, and even motoric ways. It is logical that this varied repetition, alongside a more individualized experience whereby content is adapted to user performance and where performance feedback is directly available, results in a powerful learning experience. Earlier research indeed concurs that active involvement with educational content can lead to superior learning (Calvert et al., 2007; Strong & Gallagher, 2005;
Lauricella et al., 2010; Feldman, 2004). Given the potential power of apps for education, it is certainly worthwhile to replicate this study to determine the stability of this effect as well as the potential boundaries of this effect.
While the main effect in terms of apps versus television recall did fall in line with expectations, findings for the moderating role of epistemic curiosity were non-significant – with means actually suggesting a reverse pattern than what was hypothesized. While a replication study with a larger sample is needed before too much is made of this finding, it is possible that this reverse pattern emerged for one of several reasons. For instance, D-type epistemic curiosity is associated with a strong desire to attain new knowledge by resolute problem solving (Piotrowski, Litman & Valkenburg, 2014). In general, play-based apps
require a high(er) amount of input-based action as users are more likely to face problems (e.g., in terms of game elements to overcome to succeed to the next level) and solve these. It is possible that, by virtue of the game mechanics, children with lower D-type epistemic curiosity are being “forced” to expand their effort to face challenges and obtain new
knowledge beyond what they would normally do. In doing so, they are being confronted with the requirement to exercise a skill they do not normally use and may actually experience a differential benefit from practicing this typically dormant skill. If such a finding is in fact replicated in a sufficiently powered follow-up study, this would certainly have important implications for both the consideration of D-type curiosity as a trait (given that this would suggest this construct is, at least temporarily, malleable) as well as the implication for ways to activate interest in knowledge seeking among young children.
Implications
In all, the findings concur with the educational media effects literature – highlighting once again that children can and do learn from developmentally-appropriate educational media (Beilin & Fireman, 1999). From a practical perspective, the fact that app recall was dramatically greater than television recall should not be taken lightly. As such, it seems that efforts to merge best practices in educational media with best practices in interactivity are an important continued avenue for educational media developers. This means reflecting on not only the existing literature with traditional educational media (e.g., literature on educational television or educational radio such as Fisch, 2004) but also reflecting on what we know about interactive media affordances. For example, James Paul Gee (2007) has written
eloquently about what makes a game “good” and offers a set of guidelines on how to translate best practices in entertainment games to educational games. Alongside this, developers must keep in mind what we know to be true about media effects in general – namely, that
such, developers must carefully consider the relevant individual and contextual variables that their media product will be situated in from the outset. Of course, this sets the bar quite high for app developers –developing an entertaining app which successfully integrates educational content while meeting the individual needs of its target audience. This is most definitely no easy task, but if done right, it is certainly a worthwhile one.
Moreover, moving beyond the structure of the content for a moment, practically-speaking, it is also worth noting that this study adds to the literature which shows that young children can learn language skills (even second language skills) from media. Indeed, in this study, it is clearly evident that young Dutch speaking children were able to accurately learn a host of English language phonemes and English letters after a relatively short exposure. And importantly, thanks to the highly engaging nature of both the television and app content (Linebarger, McMenamin, & Wainwright, 2009), all participants in the study seemed to enjoy the content. While no formal appeal data was collected, anecdotally, children were visibly pleased with the television content (e.g., laughter, orientation of body towards television, disregard for available coloring task) and similarly were visibly pleased with the app content. Considering that there is much conversation currently ongoing in the Netherlands about the state of English education (Jansen, Bosman & Leseman, 2013; Persson, 2012) and the need to update this language education to meet the demands of today’s youth, it seems that reflecting on the opportunities of both traditional and digital media may be a worthwhile initiative.
Future Research
As this study was a first look at the differences in television and app comprehension, the most important next step is a replication study. In particular, a larger sample that has greater power to detect the small effects that are customary in communication science is crucial. Moreover, it would be quite reasonable to replicate this study with different content types as well as different outcomes to identify the boundaries of these findings. In doing so,
relying on either a between-subjects design using randomized assignment to condition or using a counterbalanced approach to a within-subjects design (assuming this would be possible with the selected stimuli) would be reasonable. In this study, it was not feasible to use a counterbalanced approach and, as such, it is possible that the findings for app
comprehension over television comprehension may reflect an ordering finding as opposed to true content differences. Even more, given that research shows that the process of media-based learning benefits from greater exposure and repetition (Lesaux & Siegel, 2003), extension of this work into a longitudinal framework would provide relevant information for those wishing to use media content as potential intervention materials.
Methodologically speaking, particularly in light of the confusing findings for hypothesis 2, it is important to keep in mind that epistemic curiosity was measured by teachers instead of parents (for which the measure was designed). Parents were also asked to report epistemic curiosity, but unfortunately, the response rate was too low for useable data. Though it is believed that teachers have an accurate interpretation of their students’ behavior, efforts to encourage parental report or perhaps even measurement of such behavior via observation could provide important clarity as to this construct and its relationship with content comprehension.
Beyond extension, replication, and measurement refinement, it will also be important for future work to consider other relevant individual difference variables. This study focused upon D-type epistemic curiosity, but as the DSMM acknowledges (Valkenburg & Peter, 2013), there are many individual differences which may influence how a child experiences media content. For instance, parents or peers can play a very important role in limiting or motivating exposure to educational content as well as the comprehension of this content (Nathanson, 2001). In particular, their norms and beliefs may motivate children to use and benefit from particular media or may, alternatively, lead a child to avoid or suspend such
media use. Thus, it may be the case that if a child’s social environment encourages the use of interactional play-based apps on a touchscreen device, their attitude towards the medium and its content can be positively related to (educational) learning by this medium (Valkenburg & Peter, 2013; Neumann & Neumann, 2013). Considering the context of use is certainly a relevant and important next step in this research line.
Conclusion
To date, touch screen devices as iPads and tablets are very popular among young children. Apps targeted at young children often claim to support young children’s educational needs, yet evidence to support this claims remains limited to non-existent. This experimental study provides a first look as to whether educational apps do indeed support learning, and more so, the extent to which such learning compares with more traditional media content and whether and how children’s innate need for information (epistemic curiosity) may moderate this process. In all, the results show that young children can indeed acquire educational information from educational apps – with knowledge acquisition nearly twice as large as that from television content. While epistemic curiosity did not seem to influence learning, in line with existing theory, patterns do suggest that interactive platforms may be particularly beneficial for particular subsets of children. As our children continue to be raised in an ever digitally-mediated world, it is crucial that we continue to ask questions not only about these effects but also reflect on how we can maximize these opportunities. This study would
suggest that educational apps can be one way to maximize such benefits, and pushes us to ask more questions about which children can benefit most from interactional educational content.
References
Beilin, H., & Fireman, G. (1999). The foundation of Piaget’s theories: Mental and physical action. Advances in Child Development and Behavior, 27, 221-246.
Buijzen, M., Van Reijmersdal, E. A., & Owen, L. H. (2010). Introducing the PCMC model: An investigative framework for young people's processing of commercialized media content. Communication Theory, 20(4), 427-450.
Calvert, S. L., Strong, B., & Gallagher, L. (2005). Control as an engagement feature for young children's attention to, and learning of, computer content. American Behavioral
Scientist, 48, 578−589.
Calvert, S. L., Strong, B. L., Jacobs, E. L., & Conger, E. E. (2007). Interaction and
participation for young Hispanic and Caucasian children's learning of media content. Media Psychology, 9, 431−445.
Chiappe, P., Siegel, L. S., & Wade-Woolley, L. (2002). Linguistic diversity and the development of reading skills: A longitudinal study. Scientific Studies of
Reading, 6(4), 369-400.
Chiong, C., & Shuler, C. (2010). Learning: Is there an app for that. In Investigations of young
children’s usage and learning with mobile devices and apps. New York: The Joan
Ganz Cooney Center at Sesame Workshop.
Cohen, M., Hadley, M., & Frank, M. (2011). Young children, apps & iPad. Michael Cohen
Group LLC, New York, NY, USA (sd).
Collins, W.A. (1983). Interpretation and inference in children's television viewing. In
Bryant, J., & Anderson, D.R. (Eds.), Children's understanding of television: Research on attention and comprehension. New York: Academic Press.
Costello, S. (2012). How many apps are in the iPhone app store. About. com Guide.
Culén, A. L., & Gasparini, A. (2011). iPad: a new classroom technology? A report from two pilot studies. INFuture Proceedings, 199-208.
Cummins, J. (1979). Linguistic interdependence and the educational development of bilingual children. Review of Educational Research, 49, 222–251.
Eckhardt, B. B., Wood, M. R., & Jacobvitz, R. S. (1991). Verbal Ability and Prior Knowledge Contributions to Adults' Comprehension of Television. Communication
Research, 18(5), 636-649.
Feldman, R. S. (2004). Child development (3rd). Prentice Hall: Pearson Education, Inc. Fisch, S.M. (2000). A capacity model of children's comprehension of educational content on television. Media Psychology, 2, 63-91.
Fisch, S. M. (2014). Children's learning from educational television: Sesame Street and
beyond. Routledge.
Fisch, S. M. (2004, June). What's so new about new media?: comparing effective features of children's educational software, television, and magazines. In Proceedings of the 2004
conference on Interaction design and children: building a community (pp. 105-111).
ACM.
Gee, J. P. (2007). Good video games and good learning.
Good, R. H., & Kaminski, R. A. (Eds.). (2002). Dynamic Indicators of Basic Early Literacy Skills (6th ed.). Eugene, OR: Institute for the Development of Educational
Achievement.
Goodwin, K. (2012). Use of tablet technology in the classroom. NSW Department of
Education and Communities.
Gutnick, A.L., Robb, M., Takeuchi, L., & Kotler, J. (2011). Always connected: The new digital media habits of young children. Joan Ganz Cooney Center at Sesame
Workshop.
Hatherly, A., & Chapman, B. (2013). Fostering motivation for literacy in early childhood education using iPads. Computers in New Zealand Schools: Learning, teaching, technology, 25(1–3), 138-151.
Henderson, S., & Yeow, J. (2012, January). iPad in education: A case study of iPad adoption and use in a primary school. In System Science (HICSS), 2012 45th Hawaii
International Conference on (pp. 78-87). IEEE.
Hisrich, K., & Blanchard, J. (2009). Digital media and emergent literacy.Computers in the Schools, 26, 240–255.
Hutchison, A., Beschorner, B., & Schmidt‐Crawford, D. (2012). Exploring the use of the iPad for literacy learning. The Reading Teacher, 66(1), 15-23.
Janssen, M., Bosman, A. M., & Leseman, P. P. (2013). Phoneme awareness, vocabulary and word decoding in monolingual and bilingual Dutch children. Journal of Research in
Reading, 36(1), 1-13.
Jongejan, W., Verhoeven, L., & Siegel, L. S. (2007). Predictors of reading and spelling abilities in first-and second-language learners. Journal of Educational
Psychology, 99(4), 835.
Lauricella, A. R., Pempek, T. A., Barr, R., & Calvert, S. L. (2010). Contingent computer interactions for young children’s object retrieval success. Journal of Applied Developmental Psychology, 31, 362-369.
Lesaux, N. K., & Siegel, L. S. (2003). The development of reading in children who speak English as a second language. Developmental psychology, 39(6), 1005.
Linebarger, D. L. (2015). Super Why! to the Rescue: Can Preschoolers Learn Early Literacy Skills from Educational Television?.
Viewing the Television Program Between the Lions on the Emergent Literacy Skills of Young Children. Journal of Educational Psychology, 96(2), 297.
Linebarger, D. L., McMenamin, K., & Wainwright, D. K. (2009). SUMMATIVE
EVALUATION OF SUPER WHY!. Outcomes, dose and appeal. A final report prepared
for the Corporation for Public Broadcasting. Philadelphia, PA: Annenberg School for
Communication, University of Pennsylvania.
Linebarger, D. L., & Walker, D. (2005). Infants’ and toddlers’ television viewing and language outcomes. American Behavioral Scientist, 48(5), 624-645.
Lorch, E. P., & Castle, V. J. (1997). Preschool children's attention to television: Visual attention and probe response times. Journal of Experimental Child Psychology, 66(1),
11-127.
Markopoulos, P., Reed., J.C., MacFarlane, S. & Hoysniemi, J. (2008). Evaluating children's interactive products: principles and practices for interaction designers. Morgan Kaufmann, Burlington.
Martin, E., Morgan, M. C., & Rukobo, E. (2014). Play-Based Technology Platforms and Early Learners. Retrieved on December 9, 2015, from
http://www.northeastcompcenter.org/wp-
content/uploads/2014/03/PlayBasedTechnology-FINAL.pdf
Missall, K. N., & McConnell, S. R. (2004). Psychometric characteristics of individual growth and development indicators: Picture naming, rhyming, and alliteration. Minneapolis,
MN: University of Minnesota Center for Early Education and Development.
Moss, G. (2007). Literacy and gender, researching texts, contexts and readers. Oxon: Routledge.
Nathanson, A. I. (2001). Parents versus peers: Exploring the significance of peer mediation of antisocial television. Communication Research, 28, 251–274. doi:
10.1177/009365001028003001. In: Valkenburg, P. M., & Peter, J. (2013). The differential susceptibility to media effects model. Journal of Communication, 63(2), 221-243.
Neumann, M. M. (2014). An examination of touch screen tablets and emergent literacy in Australian pre-school children. Australian Journal of Education, 0004944114523368.
Pace, A. J. (1980). The ability of young children to correct comprehension errors: An aspect of comprehension monitoring. Paper presented at the annual meeting of the American Educational Research Association, Boston, MA.
Penny, R. K., & McCann, B. (1964). The children’s reactivity curiosity scale. Psychological Reports, 15, 323–334. In: Piotrowski, J. T., Litman, J. A., & Valkenburg, P. (2014). Measuring epistemic curiosity in young children. Infant and Child
Development, 23(5), 542-553.
Persson, L. (2012). Engels voor kleuters. Een eerste verkenning naar de vormgeving van vroeg vreemdetalenonderwijs in Nederland. Levende Talen Tijdschrift, 13, 25-32. Pijpers, R. (2011). App noot muis: Peuters en kleuters op internet. Retrieved on December 1, 2015, from http://www.appnootmuis.nl
Piotrowski, J. T. (2014a). Participatory cues and program familiarity predict young children's learning from educational television. Media Psychology, 17(3), 311-331.
Piotrowski, J. T. (2014b). The Relationship between Narrative Processing Demands and Young American Children's Comprehension of Educational Television. Journal of
Children and Media, 8(3), 267-285.
Piotrowski, J. T., Litman, J. A., & Valkenburg, P. (2014). Measuring epistemic curiosity in young children. Infant and Child Development, 23(5), 542-553.
Revelle, G.L., Strommen, E.F., & Medoff, L. (2001). Interactive technologies research at the Children's Television Workshop. In Fisch, S.M., & Truglio, R.T. (Eds.), "G" is for
growing: Thirty years of research on Sesame Street . Mahwah, NJ: Lawrence Erlbaum Associates
Rice, M. L., & Woodsmall, L. (1988). Lessons from television: Children's word learning when viewing. Child development, 420-429.
Richards, J. B., Litman, J. A., & Roberts, D. H. (2013). Performance characteristics of measurement instruments of epistemic curiosity in third-year medical students. Medical Science Educator, 23, 355–363. In Piotrowski, J. T., Litman, J. A., & Valkenburg, P. (2014). Measuring epistemic curiosity in young children. Infant and
Child Development, 23(5), 542-553.
Shuler, C., Levine, Z., & Ree, J. (2012). iLearn II An analysis of the education category of Apple’s app store.
Smidts, D. & Huizinga, M. (2009). BRIEF executieve functies gedragsvragenlijst: Handleiding. Amsterdam: Hogrefe Uitgevers.
Strommen, E.F., & Revelle, G.L. (1990). Research in interactive technologies at the Children's Television Workshop. Educational Technology Research and
Development, 38, 4, 65-80.
Vaala, S. (2014). What’s in store today: A snapshot of kids’ language and literacy apps. Retrieved on November 30, 2015, from
http://www.joanganzcooneycenter.org/2014/12/02/whats-in-store-today-a-snapshot-of- kids-language-and-literacy-apps-part-3/
Valkenburg, P. M., & Peter, J. (2013). The differential susceptibility to media effects model. Journal of Communication, 63(2), 221-243.
Table 1
Relevant statistics for all model variables Overall range
Min. Max. M SD
TV condition 0 100 38.24 19.92
App condition 0 100 62.56 16.04
Age (in months) 49.90 71.60 60.41 7.48
Native language skills 21.18 5.88
Foreign language skills 3.47 1.76
Epistemic curiosity (D-type) 1 4 2.29 .87
Correlation matrix for all model variables
TV condition App condition Gender Age Native language skills Foreign language skills Epistemic Curiosity Working Memory
TV condition r - .56** -.14 .44** .44** .21 .37* -.36* App condition r .56 ** - .04 .44 ** .47** .37* .15 -.20 Gender r -.14 .04 - -.19 -.08 -.11 .15 -.25 Age r .44** .44** -.19 - .65** .07 .26 -.29
Native language skills r .44 **
.47** -.08 .65** - .00 .28 -.36*
Foreign language skills r .21 .37* -.11 .07 .00 - .15 -.29 Epistemic curiosity r .38* .15 .15 .26 .28 .15 - -.74** Working memory r .36* -.20 -.25 -.29 -.36* -.29 -.74** -**. Correlation is significant at the 0.01 level (2-tailed).
0 10 20 30 40 50 60 70
Television Recall (%) App Recall (%)
Low Epistemic Curiosity High Epistemic Curiosity
