Frontiers

Frontiers

University of California, Los Angeles, United States

Anglia Ruskin University, United Kingdom

The editor and reviewers affiliations are the latest provided on their Loop research profiles and may not reflect their situation at the time of review.

The Role of Book Features in Young Childrens Transfer of Information from Picture Books to Real-World Contexts

Counselling and Psychology in Education, University of South Dakota, Vermillion, SD, United States

Department of Applied Psychology and Human Development, University of Toronto, Toronto, ON, Canada

Picture books are an important source of new language, concepts, and lessons for young children. A large body of research has documented the nature of parent-child interactions during shared book reading. A new body of research has begun to investigate the features of picture books that support childrens learning and transfer of that information to the real world. In this paper, we discuss how childrens symbolic development, analogical reasoning, and reasoning about fantasy may constrain their ability to take away content information from picture books. We then review the nascent body of findings that has focused on the impact of picture book features on childrens learning and transfer of words and letters, science concepts, problem solutions, and morals from picture books. In each domain of learning we discuss how childrens development may interact with book features to impact their learning. We conclude that childrens ability to learn and transfer content from picture books can be disrupted by some book features and research should directly examine the interaction between childrens developing abilities and book characteristics on childrens learning.

On the bookshelf of a pre-reader, one may find storybooks that take children to magical worlds with fantastical characters, to faraway lands with unique animals and customs, or keep them close to home with tales about backyard bullies or trips to the dentist. Alongside these, one may also find factual books about outer space, underwater creatures, or pre-historic dinosaurs. These books may differ from one another in a number of their features, including their genre, presence of fantastical elements, pictorial realism, and use of factual language. Children are expected to learn facts, concepts, or values and apply them to real life. The current body of evidence on whether children can learn and transfer new content from picture books suggests that it is important to consider both the dimensions on which the books vary and childrens developing abilities. In this review we summarize the existing evidence on the effect of book features on young childrens learning and transfer and outline three developmental abilities that may interact with whether childrens learning will be impacted by the presence or absence of those book features.

The majority of past research on picture books has focused on the nature of the book sharing interaction between adults and children (e.g.,Fletcher and Reese, 2005). This large body of research demonstrates that different picture book features shape the interactions that take place between dyads; for example expository texts lead to more maternal teaching during reading than narrative texts (Pellegrini et al., 1990), less specific language (Nyhout and ONeill, 2014), and more maternal feedback (Moschovaki and Meadows, 2005), whereas high quality illustrations lead to more child labeling of pictures (Potter and Haynes, 2000). Mothers are more likely to point and label letters for their young children when interacting with a plain book than a book with manipulative features and children also vocalize most often about the letters and pictures in the plain book (Chiong and DeLoache, 2012). Thus, aspects of the book can alter what both parents and children focus on. Recently the impact of book features directly on childrens learning from print picture books has also received increasing attention in developmental research. Two recent reviews have provided targeted overviews of features that support vocabulary learning (Wasik et al., 2016) and learning from fictional media more broadly (Hopkins and Weisberg, 2017). These reviews indicate that children are selective in their learning and that properties of media can affect childrens learning. In the current review, we focus specifically on learning from picture books, with the goal of outlining how three key developmental factors (symbolic development, analogical reasoning, and reasoning about fantasy) may influence young childrens learning and transfer from books that vary across various dimensions. We will focus on domains of learning where most of the research on picture book features so far has been conducted with pre-readers: learning of words and letters, science concepts, problem solutions, and morals.

One goal of educational book-sharing interactions is for children to build generalizable knowledge they can learn and transfer outside of storybooks to everyday situations. Bylearning, we refer to the childs ability to recognize or recite information presented in a book. Bytransfer, we refer to an ability that goes beyond such learning: the ability to apply newly-acquired information to new exemplars or contexts. Bypicture books, we refer to books designed for pre-readers that contain pictures and may also contain text. We first present three developmental factors that may constrain learning and transfer from picture books. They have been selected because of their importance in supporting transfer of information across contexts, which is the focus of the studies we review here. We then provide a summary of studies investigating how features of picture books influence childrens learning and transfer across a variety of educational domains by either reinforcing or working against the developmental processes presented. We conclude with ideas for new research and ways in which parents and educators can scaffold childrens learning and transfer from picture books.

Childrens ability to transfer knowledge from picture books to the real world may be constrained by developments in their symbolic understanding, analogical reasoning, and their understanding of fantasy and reality. Although we discuss them separately, these areas of development are interwoven. As we will see, these developmental factors can be used to explain experimental findings on childrens learning and transfer from picture books, as well as identify areas for future research.

One particular challenge that children may face when learning and applying real-world information from picture books is that ofsymbolic insight(DeLoache, 1991). That is, children need to be able to think flexibly about books as entities in themselves as well as symbolic sources of information about the world. For example, when reading an informational book about new animals such as South American cavies, children need to realize that they are reading a book with pages that can be flipped and pictures that tell a story about 2-dimensional cavies. They also need to recognize that the cavies on the page are intended to be representative of animals in the real world that have the same name (cavies) and features. Understanding that a picture in a book is an object that represents another entity is a symbolic task. This may not be a straightforward task for children especially since pictures in childrens book can vary on the nature of their relation to the referent, that is, whether the picture represents real concrete (e.g., a cat) and abstract (e.g., letters and numbers) entities or imaginary entities (e.g., talking cats, talking pots, unicorns;Ganea and Canfield, 2015). Beyond the basic understanding that pictures are symbolic and stand for their referents, children will have to figure out what the nature of the referent is.

Young children often struggle with tasks that require symbolic reasoning. For example, 2-year-olds struggle to use information from videos and pictures of a room to help them find an object hidden in the real version of the room (Troseth and DeLoache, 1998). Despite the fact that these toddlers can easily point out and label the corresponding objects in the pictures and in the room, they do not transfer information from one to the other. Presumably this is because they think of the picture and the room each as a separate entity, and do not make the connection that the hidden object in the picture also represents a life-sized object hiding behind a pillow in the life-sized room. In addition, pictures in books are impoverished compared to information presented in real life because they provide only one visual perspective, lack depth cues like motion parallax and changing shadows, and may be low resolution.Simcock and DeLoache (2006)assert that perceptual differences between images in picture books and objects in the real world present a barrier to childrens ability to use picture books symbolically, as a source of information about the world. This problem is not specific to young childrens use of information from picture books, but from other symbolic media as well, such as videos (Anderson and Pempek, 2005Barr, 2013). There is some evidence that transfer difficulties are similar across different media (books vs. videos;Brito et al., 2012), although there is also evidence of medium-specific differences in transfer (books vs. touchscreens;Strouse and Ganea, 2017). For the remainder of this review we will focus specifically on factors influencing young childrens transfer from picture books.

Various features of picture books may differentially affect childrens ability to treat the information symbolically. For example, pictures that more clearly represent the objects they depict may support children in recognizing the link between book depictions and the real world (Ganea et al., 2008Ganea and Canfield, 2015). As such, unrealistic portrayals such as cartoonish images, fantastical settings, and depictions of animals with human characteristics may present particular challenges for children and will be reviewed below. Tactile features may pose a similar challenge, as they may highlight the book as an object, rather than as a symbol with information to be conveyed about the real world. These interactions between symbolic understanding and book features will be reviewed across various domains of learning below.

For successful transfer of complex information and concepts, children may need more than symbolic insight. To transfer basic information like the name of a novel animal from a picture book, children need to activate a representation of the animal in the book and remember details about its appearance to correctly apply the label to the real-world animal. To transfer more complex concepts, such as the ability for animals (in general) to use color camouflage to hide from predators, children must also recognize the abstract features of the depicted example and apply these to novel instances. Transferring conceptual information from one domain to anotherin this case, from the picture book to the real worldrequires children to recognize the abstract relational structure between the two domains (Gentner, 1989).

Childrens ability to reason analogically depends somewhat on the difficulty of the task and their existing knowledge of the relations used in the analogy (Goswami, 1991). When they have experience in a domain, children as young as 1 or 2 years can use deep rather than surface features to solve analogical problems (e.g.,Brown, 1990Chen et al., 1997). However, when domain knowledge is limited, children without prior conceptual knowledge may be reliant on surface-level features to help them look for commonalities across analogical cases (Brown, 1989). One benefit of picture books as an educational resource is that they can provide children access to content that they would not experience in their day-to-day lives. However, this very feature of picture books may make analogical transfer especially difficult. For example, if childrens understanding of color camouflage is tied to specific picture book illustrations (e.g., a frog) and surface features of that example (e.g., greenishness), they will likely fail to transfer the concept to other animals or contexts.

As with symbolic reasoning, various features of picture books may differentially affect childrens ability to analogically transfer conceptual information in books. For example, given that perceptual similarity between transfer contexts facilitates analogical reasoning (Crisafi and Brown, 1986Brown, 1989), childrens transfer of new content from books with fantastical contexts and characters should be more impacted than transfer from books with realistic contexts and characters (Richert et al., 2009). If we expect children to learn and transfer novel content from picture books to a real-world context, stories that are more similar in surface structure to the real world would be easier for children to use a source of information about the world. Interactions between book features and analogical reasoning will be reviewed below.

Children also have the challenge of determining which information in picture books should even be transferred. Anthropomorphism, or animals with attributes characteristic of humans, may be especially confusing when some information is meant to generalize and other information is meant to be true only in the story world. For example, if the cavies in a story talk and wear clothes, children must separate this anthropomorphization of cavies from factual information, inhibit transferring the unrealistic attributes, and selectively transfer only the factual information presented. Childrens learning from picture books must be selective in that they have to separate what information is fictional versus what could be true in reality, which is generally referred to as the readers dilemma (Potts et al., 1989Gerrig and Prentice, 1991).

The process of keeping real-world knowledge separate from fictional or false information encountered in a story context may be especially difficult in early childhood because children between the ages of 3 and 8 are just beginning to differentiate fantasy and reality (Woolley and Cox, 2007). According toWoolley and Ghossainy (2013)young children are nave skeptics when it comes to judging the reality status of fictional information. Instead of over-incorporating fantastical information into their real-world concepts, children err on the side of rejecting factual information presented. For example, 4- to 8-year-olds were more likely to state that an improbable event is impossible than to accept an impossible event as possible (Shtulman and Carey, 2007). A bias toward skepticism may impede transfer of educational information, as children may tend to not transfer details they are uncertain are real.

The ability to accurately distinguish reality and fantasy may also be related to childrens representational rriveau and Harris (2015)found that 3- to 4-year-olds accurately distinguished historical and fantastical characters in narratives at the same time that they started passing false belief and false signs tasks, suggesting that an understanding of representation (both mental and symbolic) may underlie the ability to distinguish fantasy from reality. Picture books, both in terms of their prose and illustrations, may be designed to represent reality or to represent make-believe.Corriveau and Harris (2015)argue that children may have difficulty deciding which of the two functions a particular story may fulfill. Thus, childrens ability to separate fantasy from reality may depend both on their recognition that a story stands forsomethingand their ability to judge what that something is (reality or pretend). In addition, childrens own experiences and background knowledge may influence the aspects of stories they view as realistic versus fantastical (Corriveau et al., 2015).

Books with unrealistic content, such as impossible events or anthropomorphic depictions of animals, may present a challenge to children in separating which aspects of the book apply to the real world and which belong only in the book. Therefore, we again expect books with realistic content to be more supportive of learning transfer, especially when learning conceptual information such as scientific facts and concepts. Although these book features interact with the two other developmental factors discussed abovesymbolic development and analogical reasoningwe also expect the developing ability to reason about what is real and what is fantastical to constrain or enable learning and transfer.

The following sections provide a review of how particular aspects of picture books (such as genre, pictorial realism, and the presence of manipulative features) interact with the three developmental factors we have proposed to influence childrens transfer from picture books. We chose not to present this review as systematic or definitive, as research in many areas is in its early stages (see Table1). Rather, we present information about how our identified developmental factors inform our understanding about childrens learning from various book features and areas for further consideration in picture book research. We focus predominately on pre-readers who are listening to an adult read while they view the books pictures.

Table 1. Summary of book features impact on learning and transfer in each learning domain.

Particular features of picture books, such as the specific content they incorporate, or the way in which the content is presented, may influence childrens tendency to learn and transfer the educational content to real-world situations. Below we review studies that investigate some of these features, organized by the domain in which the educational content is presented. We have chosen this organization because particular features may be more influential in some learning domains than others. For example, visual features may be important when learning vocabulary, where children may be fairly successful at transfer on the basis of matching up perceptual features of objects. However, contextual information may be more important in science domains where transfer often takes place on a conceptual level. The domains we have chosen are primarily the domains in which the impact of picture book features on transfer of information presented in books have been studied. In each section, we address the book features that have been studied in that domain, interpreted with regard to our three developmental factors. Future work is needed to address how book features influence transfer in other domains such as math and the arts, as well as how additional book features impact transfer.

Picture books expose children to rich language. For example, picture books contain a richer diversity of words (Montag et al., 2015) and a greater incidence of rare grammatical constructions (Cameron-Faulkner and Noble, 2013) than child-directed speech. In addition, caregivers use a larger number and wider variety of words during reading than other activities (Hoff-Ginsberg, 1991). It is not surprising, then, that joint reading has been associated with a variety of later language outcomes, including vocabulary growth and early literacy skills like letter knowledge (e.g.,Bus et al., 1995). Here we are interested in particular features of books that may support the process of language learning from picture books on a less protracted scalewords and letters learned from individual reading sessions. We expect that symbolic understanding plays an especially important role in this domain, as transfer of a new word to a new context heavily depends on recognition of the labeled item in the book as representing objects in the real world to which the label also applies (Preissler and Carey, 2004Ganea et al., 20082009). Thus, features of books that make the link between depicted objects and real world referents clearer or easier to discern should support transfer, whereas features of books that make these links more difficult to recognize may make transfer more difficult. The book features that have been most studied in this domain include pictorial realism, manipulative features, and fantastical contexts.

Picture books vary in the degree to which their pictures represent reality, from photographs to illustrations to cartoonish line drawings. An image that is highly iconic, or visually very similar to its referent, may highlight the relation between the picture book image and real-world instances. As such, we might predict photographs to be the most supportive of childrens transfer of knowledge from books to reality.

Newborn infants perceive and distinguish the dimensional nature of pictures from real objects. If presented with a complex object and a photograph of it, they clearly prefer the real object (Slater et al., 1984). However, when presented with photographs alone, 9-month-olds interact with them in ways similar to how they would interact with the real object they representby hitting, rubbing, and grasping the photographs (Pierroutsakos and DeLoache, 2003). Their behavior suggests they have not yet grasped the symbolic function of pictures.

As infants reach the middle of their second year, they begin to treat pictures referentially, by pointing and labeling the depicted objects (DeLoache et al., 1998). Research also indicates that in their second year of life children understand the representational status of pictures (Preissler and Carey, 2004Ganea et al., 2009). Yet, childrens transfer of novel words from picture books to the real world referent can be impacted by pictorial realism at these ages.Ganea et al. (2008)showed 15- and 18-month-olds picture books presenting both familiar and novel objects in the form of photographs, realistic color drawings that closely resembled the photographs, or color cartoons which were less detailed and more distorted in appearance. After being read the book by a researcher (told the names for the pictured objects), children of both ages were able to recognize the labeled object they had seen in the book regardless of the type of image. However, children who were read the cartoon book did not generalize to a picture of a new exemplar different in color. Eighteen-month-olds transferred the label to its physical real-word referent across all three conditions, but 15-month-olds did so only in the photograph and drawing conditions. Taken together, these findings suggest that transfer from the photographs was easiest for children, and transfer from cartoons the most difficult. With age, children get better at transferring from perceptually dissimilar depictions to real objects, although there is evidence that the iconicity of pictures continues to play a role in some picture transfer tasks even at 3 years of age (Callaghan, 2000Mareovich and Peralta, 2015). The impact of iconicity on young childrens learning from picture books has also been found with other measures, such as imitation (Simcock and DeLoache, 2006). Thus, at young ages, when children are first beginning to think symbolically, their understanding that pictures stand for real objects interacts with the type of depictions in books.

The term manipulative features has been used to refer to features that are designed to increase childrens physical interaction with [a] book, like lift-a-flap, scratch-and-sniff, and other three-dimensional add-ons (Tare et al., 2010, p. 396). These features may be entertaining for children, but research suggests they may not be optimal for learning. One reason they may not be optimal for learning is that they may draw attention away from links between the book and the real world. For young children who are still learning to use pictures in books as standing for real objects, this may distract from the insight necessary for transfer of learned information.

Using books designed to teach children animal names,Tare et al. (2010)tested the helping or hindering influence of manipulative features on 18- to 22-month-olds learning and transfer of the animal names. Children were read a book by a researcher featuring 9 animals either using a commercially presented manipulative book (with flaps and pull tabs) or a scanned copy of the book (without manipulative features). At test, children who had seen a copy of the book without manipulatives correctly generalized a new animal name to new pictures and a replica of the animal. Children who read the book with manipulative features did not perform above chance. In another study, researchers compared 30- to 36-month-olds learning of letters from a manipulative alphabet book with pulls, flaps and textures to a book without these features (Chiong and DeLoache, 2012). Children learned more letters from the simple alphabet book than the manipulative one. The authors argued that the salience of manipulative features may render them more like objects themselves and less like symbols that stand for other objects than their 2D counterparts. Childrens difficulty transferring labels from manipulative books may therefore stem from a difficulty in seeing past the fancy features to realize that the content is representational, meaningful, and applicable to other contexts.

Another possibility is that childrens mental effort is engaged with interaction with the features rather than attending to the content. For example, pulling a tab in an alphabet book to make a truck move does not help to emphasize the correspondence between the letter T and the first sound in the word truck. There is other evidence that features that require additional mental effort, like having multiple large pictures on each page, can result in cognitive overload, disrupting learning (Flack and Horst, 2017).Flack and Horst (2017)read 3- to 5-year-olds books with one or two regular-sized illustrations per page spread or one large image per spread. New objects in the pictures were labeled with new words during reading. At test, children were asked to identify the referent of the labels by pointing to the correct objects on a book page. Children were more successful when they had seen one illustration, regardless of size, indicating that two illustrations may have resulted in cognitive overload. The researchers did not assess transfer of learning. In a follow-up study, a hand gesture that directed children to the correct illustration supported learning from the book with two pictures per spread. In light of these effects of cognitive overload on childrens learning, more research is needed to determine whether manipulatives are particularly disruptive of symbolic insight, whether they result in cognitive overload, or both.

Research shows that not all manipulative features are detrimental to childrens learning. Multimedia researchers have argued that extra book features that engage children with the educational content of books (called considerate,Labbo and Kuhn, 2000) can support learning. A recent meta-analysis of studies involving electronic books with considerate enhancements like animated pictures, music, and sound effects were supportive of vocabulary learning for preschool and elementary children (Takacs et al., 2015). While we know of no similar results with manipulative features of print books, one study suggests that manipulatives designed to draw attention to the educational content, in this case the shape of letters, did not distract 3-year-olds from learning the letter names (Chiong and DeLoache, 2012).

For both word and letter learning, the manipulative features traditionally found in print books do not appear to facilitate learning and transfer, and in cases when the features are irrelevant to the books educational content, may even interfere with it. Content-central manipulatives that highlight educational content, such as highlighting the visual shape of a letterthe crucial component for transferring the letter name to new instances of the lettermay hold promise in facilitating symbolic insight, and thus transfer. Research in this area will become especially crucial as the features available in digital books continue to expand.

In picture books both fantastical and realistic, children may encounter new and unusual vocabulary. However, we might predict that realistic story contexts provide more cues to children that they can use to match story depictions and contexts with real-world situations. The similarity between the learning and transfer contexts can provide support for symbolic insightrecognizing the similarity between a symbol and its referentas well as for analogical transfer. A recent intervention with low-income preschoolers investigated the effect of fantastical or realistic content on childrens word learning (Weisberg et al., 2015). Children were presented with a set of realistic or fantastical commercial picture books and toys. The researchers measured childrens comprehension of the vocabulary presented in the books and toys receptively and asked them to tell everything they knew about the tested word (e.g., What are weeds?). Across both conditions, children showed similar gains in identifying the tested objects. However, children in the fantastical condition were able to provide more information about the objects when given open-ended prompts. This study suggests that children learned more about the target objects in the fantastical contexts. Importa