With a little help from my friends: children’s interactions with interactive toy technology

Digital toys offer the opportunity to explore software scaffolding through tangible interfaces that are not bound to the desktop computer. In this paper we discuss the empirical work completed by the CACHET (Computers And Children’s Electronic Toys) project team investigating young children’s use of interactive toy technology. The interactive toys in question are plush and cuddly cartoon characters with embedded sensors that can be squeezed to evoke spoken feedback from the toy. In addition to playing with the toy as it stands the toy can be linked to a desktop PC with compatible software using a wireless radio connection. Once this connection is made the toy offers hints and tips to the children as they play with the accompanying software games. If the toy is absent, the same hints and tips are available through an on-screen animated icon of the toy’s cartoon character. The toys as they stand are not impressive as collaborative learning partners, as their help repertoire is inadequate and even inappropriate. However, the technology has potential: children can master the multiple interfaces of toy and screen and, when the task requires it and the help provided is appropriate, they will both seek and use it. In particular, the cuddly interface experience can offer an advantage and the potential for fun interfaces that might address both the affective and the effective dimensions of learners’ interactions. Introduction and Theoretical Background Information and Communication Technologies (ICT), and in particular the desktop computer,are now a part of classroom culture; the expectation of their use is crosscurricular and exists from an early age (see Plowman, L. & C. Stephen in this issue). Within the infant classroom and beyond, there is an increasing pressure to integrate ICT through both wired and wireless technologies. But how can this integration be pedagogically grounded, whilst at the same time innovative and engaging? In this paper we explore the use of digital toys and in particular their potential for offering collaborative support and engendering collaboration between peers. We conduct this exploration within the context of an educational theory that emphasises the importance of collaborative support and which acknowledges the current role of the computer as an alternative tool for communication and interaction (Tikhomirov, 1979). The image of the computer as a partner providing feedback and support has been presented by others, including Papert (1980) and Chan (1990). This collaborative partnership role is central to this paper, which considers how and why digital technology might provide support to young learners. Scaffolding is a term coined by Wood, Bruner and Ross (1976) from the ideas of Vygotsky (1978; 1986) to account for how a more knowledgeable partner can assist the cognitive development of a less able one, and gradually foster the development of successful independent task performance. The work of Vygotsky places emphasis upon interaction between a learner and her environment. The development of the individual is the result of her internalisation of this interaction: the relationship between development and learning was the object of Vygotsky’s attention when he proposed the Zone of Proximal Development (ZPD) as the essential 'ingredient' in effective instruction (Vygotsky, 1986). A fundamentally important feature of the ZPD is the necessity for collaboration or assistance from another more able partner. The need for this more able learning partner arises from the belief that the activities that form a part of the child’s education must be beyond the range of her independent ability. Teachers are able to fulfil the sort of collaborative partnership role envisaged within this theory (Plowman, Luckin, Laurillard, Stratfold, & Taylor, 1999). This paper explores whether digital toy technology provides collaborative support to young learners. If we start with the desktop metaphor and the design of Interactive Learning Environments (ILEs) we find that the scaffolding techniques proposed by Wood et al for face to face interactions have been used to implement software scaffolding and have offered designers one way of implementing flexible assistance for learners of different ages. Examples of software scaffolding can be found in the adaptation of Wood et al's original notion of scaffolding into the contingent teaching approach implemented in the QUADRATIC tutor (Wood, Shadbolt, Reichgelt, Wood, & Paskiewitz, 1992; Wood & Wood, 1996) This provides a series of graded help interventions that support the learner. Peer discussion is also one of the most powerful ways of implementing scaffolding approaches. Guzdial et al., (1996) and Luckin, Plowman, Laurillard, Stratfold and Taylor (1998), for example, describe an approach to scaffolding learners quite different to that of Wood. Assistance is tackled through support for peer collaboration rather than graded interventions by the system. There is a large literature on the benefits of peer collaboration in general (e.g Dillenbourg, Baker, Blaye, & O'Malley, 1995), in paired reading (Topping, 1988) and in learning through interactive multimedia (Jackson, Stratford, Krajcik, & Soloway, 1996). Of course, the question of effective collaborative assistance is not just about the content of the help provided by a collaborator, human or digital, it is also about how that help is made available to learners. There is much emphasis within education upon learners' metacognitive skill development that brings with it a need for system designers to explore how learners seek and use the help provided. Various recent studies have shown that learners do not always make effective use of the available help (Aleven & Koedinger, 2000; Luckin & du Boulay, 1999; Luckin & Hammerton, 2002; Wood & Wood, 1999 for example). However whether concerned with designing help, promoting peer collaboration or exploring how learners ask for help, the emphasis of the work on software scaffolding has been entirely directed at the desktop computer metaphor. So, what happens when you take the helper out of the box? In this paper we describe our empirical studies and discuss the ways in which children requested and used assistance from the digital toy, the accompanying software, their peers, parent or the adult researcher. The toys and software used in this work are not particularly sophisticated in terms of the range of support that they can offer. They do however offer a means of investigating how children conceive of and use these toys as potential helpmates. CACHET is a research project that aims to construct an explanatory framework for the interaction and mediation engendered by digital toys. The electronic toys used in this project are freestanding soft toys that can move, speak and respond to a child’s touch. They can also be ‘linked’ to a PC with a special wireless unit that transmits information between the toy and the computer. In freestanding mode (they are about 30 cm tall) these toys superficially appear like traditional soft toys but they have motors to provide movement and a ROM chip so they respond to inputs. The toys can gesture, using programmed motion; and speak, using a digitised vocabulary of more than 4,000 words, so they can play simple games. Interaction operates through sensors located in parts of the toy’s body, each of which controls a different function. When combined with compatible software, and operating via wireless connection with the PC, further interactioncan take place through educational software games. The software encourages basic language and number skills and the toy can comment on the child's interaction, provide feedback and give support. The child is therefore no longer interacting solely with the computer or solely with the toy, but is also interacting with a toy that, in turn, interacts directly with the computer and mediates the child’s actions. Figures 1 and 2 illustrate the DW(Arthur’s sister character) toy without the software and the Arthur toy being used in conjunction with the software. Figure 1 DW Toy Figure 2 Arthur Toy and software We are exploring how children interact with the toys and the associated software in the informal and formal learning contexts of children’s homes, out of school clubs and a primary school. Within these different contexts we are exploring and mapping interface and interactivity in order to describe and analyse what motivates emotional and cognitive engagement. This will enable us to address questions such as: Are the patterns of interaction goal-directed? To what extent do individual differences account for different patterns of interaction? In particular, we are interested in the nature of the assistance that the toy and/or software may afford the children as they complete the activities provided. The findings we present in this paper specifically address the following questions about how children ask for and use help as they interact with this digital toy technology: From where do children seek assistance, the toy, software, peer or researcher? Do children use any assistance offered without their specific request? If so is there any difference between their reactions to the different sources of assistance? Even if they take notice of the help, do children interpret it correctly? Have children sufficient mastery of the computer interface to implement help when given? If the toy is absent, the same hints and tips are available through an on-screen animated icon of the toy’s cartoon character, do children react in the same way to the same content delivered through different interfaces? How help is offered by Arthur and DW The software consists of a number of discrete games. Whilst engaged in the software activities, children are able to elicit help and useful information from the toy by squeezing its ear. If children are having difficulty progressing through a game, or persist in making the same mistake, the toy may