Representing Guided Interaction in a Sociocultural Framework

This paper considers some methodological problems in representing interactions between young children, technology and practitioners. It is based on work for an ESRC TLRP-funded project Interplay: Play, Learning and ICT in Pre-School Settings, the purpose of which is to identify ways of enhancing young children’s learning with a range of technologies. Sociocultural theory underpins the research design because children’s competences are not simply operational but encompass an understanding of the roles of technologies in work and play and are developed in a social context. This theoretical perspective needs to translate to empirical work but representations of interaction in the literature are not always consistent with a sociocultural account of the data. Guided interaction is central to this study as a way for us to conceptualise ways of supporting technology-mediated learning and as a heuristic for practitioners and its representation is an important part of both analysing and defining guided interaction. The video-recorded data serves as a stimulus for assisted reflection and is central to identifying what constitutes interaction (eg gesture, gaze, dialogue) with the artefact, the adult, the cultural setting and, in turn, the patterns and behaviours that constitute guided interaction. The problem is how to develop representations that i) are quick and easy to produce, ii) provide information which is quick and easy to understand, iii) provide information which would not normally be available from simple observation, iv) emphasise the context of use by providing information on the interrelationship of task, interface and social dimensions, v) are applicable for multiple users, and vi) are adaptable for other systems and contexts of use. The discussion leads to some suggested principles for representing guided interaction. Introduction In its analysis of situated social practices, the theoretical assumptions of a sociocultural perspective lead to a focus on activity rather than the focus on individuals’ internalised learning associated with a more cognitive or behaviourist approach. This focus on activity leads to interaction as a unit of analysis and the individual’s cognitive development is seen as the product of the activities and cultural practices that they engage in with others. Wertsch (1991, p.86) comments that the way in which this relationship between psychological processes and social context must ultimately be addressed is ‘in connection with the basic theoretical constructs and units of analysis’. However, this paper is concerned not so much with the unit of analysis as with its representation and how this relates to sociocultural theory, an aspect that is not generally attended to with the same level of detail. Researchers with a sociocultural orientation are, perhaps, more likely to be drawn to the use of video as a form of data capture because it appears to lend itself to capturing the Draft paper: not for citation Representing guided interaction 2 ‘big picture’ and is suited to representing some of the complexities of social life and activities that are mediated by artefacts. As part of Interplay: Play, Learning and ICT in Pre-School Settings, we wanted to identify ways in which practitioners could support children’s developing competences with ICT within the social context of the playroom. To this end, we needed to find a way to represent their interactions so that we could identify guided interaction and provide illustrations of it within the framework of playroom practice. This paper discusses some of the methodological issues we encountered in representing interactions between young children, technology and practitioners with the aim of elaborating our understanding of the concept of guided interaction. Following reflection on the provenance of the term and a description of the context of Interplay we discuss here i) the sociocultural framework of this study, ii) some of the issues that have emerged in using video as the source for representing multimodal interactions and iii) some principles for representing guided interaction. At the time of writing, we conceive of this paper as a work in progress, to be refined as analysis continues. Background to guided interaction Our contribution to the ESRC’s Teaching and Learning Research Programme goal of enhancing outcomes for learners is to identify ways of enhancing young children’s experiences with information and communication technologies (ICT) in pre-school settings. The concept of guided interaction was identified in earlier research studies, where it had been used to refer specifically to computers and the use of interactive media. It included the ways in which the interface can offer guidance (Plowman, 1992, 1996a) and the ways in which attention to narrative in the design of interactive media can guide navigation and comprehension (Plowman, 1996b). Guided interaction in this context was conceptualised primarily as a design issue relating to communication between the computer and the student. Observations in schools showed that children were reluctant to seek human or machine help, used random clicking, could not interpret dialogue boxes and lacked strategies for dealing with problems encountered at the computer. The agent guiding interaction was the computer responding to user input and, at this time, the concept of guided interaction did not explicitly encompass a pedagogical role as teachers were usually absent from the locations in which computers were used. Nevertheless, this scenario had parallels with our more recent studies in pre-school settings where adults are always present and in Interplay we have expanded the concept from its earlier emphasis on design to include the role of situated human help in guiding children’s interactions, the focus of the discussion here. In this context, the agent guiding interaction is an adult. The context of Interplay Scottish pre-school provision is distinct from formal school education whether it is offered in nursery classes in primary schools or standalone nurseries. Practitioners do not view themselves as teachers but see their role as monitoring and responding to children’s progress and facilitating development through both the activities they provide and the conversations that prompt and extend children’s thinking. Sessions are characterised by extended periods of ‘free play’ with some brief adult-led group activities. The culture of pre-school settings values learning through play and child-initiated activities but this approach can lead to unproductive interactions when children choose Draft paper: not for citation Representing guided interaction 3 the computer as a free play activity. Although the use of the term ‘unproductive’ seems inappropriate in a free play context we found that children were not finding their computer play rewarding either in terms of pleasure or in terms of achievement. For example, children aged three and four are usually unable to follow text instructions and so can encounter basic operational difficulties which preclude even getting started if they do not have adult support. Earlier research (Plowman & Stephen, 2003, 2005; Stephen & Plowman, 2002, 2003) showed that practitioners focussed on computers, not recognising diverse technologies in the playroom as ICT and there was a dominance of reactive rather than proactive supervision. We found that in all settings practitioners had responsibility for a number of activities or a particular area of the room. This meant that supervision of children playing at the computer was limited, opportunistic and competed with other supervisory duties. Children rarely asked for help so adults became involved with computer play only if they noticed a need for intervention. Computers were often located in corners away from sand and water or high levels of traffic so were not within peripheral vision. Where they were visible, a casual glance was not adequate for assessing the state of play at a computer in the same way as it would be for other playroom activities. Within this complex environment, practitioners are familiar with providing highly contingent responses to children but this did not extend to children’s play with ICTs: most supervision consisted of overseeing turn-taking and ensuring that children did not cause damage rather than providing individualised assistance. Lack of intervention by practitioners was due to one, or a combination of, the following factors: • resistance to approaches that could be perceived as too instructional • a desire for children to find out for themselves • not being aware that children need help • limited confidence with ICT • priority being given to interventions required elsewhere in the playroom. Although we recognise the dominant role of play in young children’s learning, our focus on guided interaction challenges the widespread belief that free play is a sufficient condition for learning in the context of ICT. Play is young children’s characteristic mode of activity but it also offers opportunities for support for learning (Bennett, Wood & Rogers, 1997). Our aim, therefore, was to identify ways in which pedagogical actions by adults can support children’s encounters with ICT in the playroom and to conduct the study with the active participation of pre-school practitioners. This paper does not detail these processes, some of which are described in the companion paper, but focuses on the issues involved in finding ways to represent guided interaction. Children’s competences with computers and other technologies are not simply operational but encompass an understanding of the roles of technologies in work and play and are developed in the social context of the home and elsewhere. In a study related to Interplay, we found (McPake et al, 2005) that children at home were developing competences in three areas: technical, cultural and learning. Most children of three or four had more exposure to a broader range of technologies, used for both leisure and work-related purposes, at home than they did at preschool. Guided interaction here was typically provided by siblings, parents, other family members, peers and neighbours. Draft paper: not for citation Representing guided interaction 4 .Design of the study and sources of data Eight pre-school settings representing a range of types of provision and serving a broad socioeconomic range of children and families were formed into two clusters based on proximity. The research was undertaken in collaboration with two practitioners from each setting, at least one of whom had little or no previous experience with technologies in the playroom. Researchers visited each individual setting on seven occasions and produced baseline information, a technology audit, field notes, focused observations and video recordings. Over the course of the 2003-2004 school year, each cluster group additionally met with researchers four times to share observations based on video recordings and to identify ways in which practitioners could provide guided interaction and support to children using ICT within the playroom setting. During this period, each site identified two interventions for implementation and evaluation that would either address recognised problems with technology-based activities or allow them to explore new activities or pedagogical actions. One intervention was designed to include computers and one involved an alternative form of ICT. Practitioners reflected on these interventions in the cluster meetings and the discussions were recorded. Practitioners were also interviewed individually about how they conceptualised their practice in the playroom both before and after data collection and a questionnaire on competence and attitudes was distributed to all forty practitioners in these settings. This data has been supplemented by a survey of the views of over 200 parents of pre-school children and case studies of sixteen children which explored their exposure to, and developing competences with, technologies in the home as well as in the nursery. This paper does not refer to the survey or case study data. Sixteen hours of video recordings made at the eight nurseries were coded to capture episodes in which children interacted with technologies and to provide examples of guided interaction. The video data is central to identifying what constitutes interaction (eg gesture, gaze, dialogue) with the artefact, the adult, the cultural setting and, in turn, the patterns and behaviours that constitute guided interaction. The study was thus designed to i) give us multiple indices of dispositions and behaviours, ii) take account of the interplay between various factors, including the different roles of peers and practitioners in mediating content, and iii) use authentic settings. The sociocultural framework Guided interaction is located within a Vygotskian tradition of understanding supported learning and relates to other concepts such as scaffolding (Wood, Bruner & Ross, 1976), sustained shared thinking (Siraj-Blatchford & Sylva, 2004) and guided participation (Rogoff et al, 1993). The study also draws on Tharp and Gallimore (1998) for their analysis of the interactions between adults and children that facilitate learning and Rogoff (2003) for her focus on the interaction as the appropriate level of analysis and its role in supporting children’s development. Although these sources do not address the role of digital technologies they all focus in one way or another on some of the issues at the heart of sociocultural thinking about technologies and the adult’s role in mediating learning, particularly: • the verbal and nonverbal means by which adults orientate children • the verbal and nonverbal ways in which children seek guidance Draft paper: not for citation Representing guided interaction 5 • the emphasis on language as a tool for learning • the role of artefacts in the mediation of learning. Interaction, then, whether child to child, child to adult or between people and artefacts (in this case, ICTs) is central to understanding activities from a sociocultural point of view. Our interest is in looking in more detail at one realm of interaction and thinking about how it can be better understood by finding ways to translate observations or records of interaction into representations that provide useful information. Scaffolding and guided interaction Pea (2004, p. 423) wonders whether ‘scaffolding has become a proxy for any cultural practices associated with advancing performance, knowledge, and skills’ and clearly there are overlaps between our concept of guided interaction and the concept of scaffolding. He (op. cit., p. 439) goes on to comment that ‘scaffolding is not at all a theoretically neutral term’ as it is rooted in the zone of proximal development and the relations among people, tools and environment. It is activity and performance centred and implies a developmental trajectory for task performance. A similar theoretical perspective informs Interplay’s research framework, given our goal of enhancing young children’s encounters with ICT, although the language of task performance is inappropriate in a free play context. Wood, Bruner and Ross (1976) first used the term ‘scaffolding’ to describe how a tutor and child jointly constructed a wooden puzzle. It was described as a process that enables a child or novice to solve a problem, carry out a task or achieve a goal which would be beyond their unassisted efforts. There has been considerable research exploring this concept in relation to technology-supported learning (e.g. Wood & Wood, 1996) but, whether concerned with designing help, promoting peer collaboration or exploring how learners ask for help, the emphasis of this work has been directed at the personal computer. The main mode of scaffolding usually referred to is language, although Klerfelt (in press) analyses the role of gesture in the support of pre-school children’s storymaking at the computer. Guided interaction is distinctive in uniting the social and technological dimensions of learning, illustrated by the two stages in the evolution of its meaning: its original use being to describe the ways in which design of the interface could support learning and its later development focusing on the role of situated human help. Our recent work suggests that planning, the provision of resources and the role of the environment constitute critical elements of guided interaction and can also be encompassed by the concept. This attention to the broader context of learning differentiates guided interaction from scaffolding, which tends to have a close focus on the activity itself, but they are similar inasmuch as they both describe a dynamic, reciprocal process which is multicausal and highly contingent. A description of some of its characteristics is provided in Appendix A. Representing guided interaction Although it is now commonplace to take account of social, cultural and institutional settings when theorising interaction this does not always translate to empirical work. This is a recognised omission: Wertsch (1991, p.86) comments that social interaction is often investigated as if it occurs outside social structure and Cole (1996, p.220) refers to a ‘reliance on face-to-face, two-person interactions as a major shortcoming in the empirical foundation of a cultural-historical activity theory’. Heath et al (2002, p.11) consider that Draft paper: not for citation Representing guided interaction 6 the means by which objects and artefacts feature in the production, coordination and intelligibility of conduct are largely disregarded and comment that ‘studies of social interaction remain curiously dislocated from the material circumstances in which it is accomplished’. Our emphasis on the artefacts, the playroom situation and the broader context combined with the use of video go some way to remedy this whilst simultaneously introducing problems relating to how to represent these diverse elements. Using video Studies that are informed by a sociocultural framework frequently draw on transcripts of speech or ready-made scripts of computer-mediated communication such as the content of message boards. In an educational context, researchers usually aim to use naturally generated speech in the interests of authenticity; this could involve a group of students collaborating on a task (although the task itself is frequently a researcher intervention) or recording teacher-student classroom dialogue. There are many studies of collaboration at computers (eg Littleton & Light, 1999) and these have the benefit of naturally generated talk between the students: this was less often the case in Interplay because children either used the computers on their own or communicated with each other nonverbally (eg taking control of the mouse, moving in closer when they wanted to join in or upending the sand timer used to time sessions when they wanted to prolong their turn). Language is not as dominant in our analysis as it is in these examples because it is not necessarily the main mediator of learning for children of this age. Talk is a key medium for school classroom interaction but in pre-school settings responsiveness to children’s emotional states is central to the practitioners’ role, so touch, gesture and eye contact are also important forms of communication. This makes video the obvious medium for data collection but it is not well suited to capturing unobservable processes such as thoughts, attitudes, feelings and perceptions. In other studies this shortcoming can be addressed by using interviews and questionnaires to supplement the video data but, in this case, analysing observable behaviours is generally more appropriate for pre-school children as articulating responses for strangers can be difficult. Suchman (1987, p.109) claims that video offers a means of considering 'just those fleeting circumstances that our interpretations of action systematically rely upon, but which our accounts of action routinely ignore' and she points out the advantages of using video as a means of providing a record of the action which is independent of the analysis, thus avoiding reliance on field notes. Video is relatively straightforward to use when recording people using a photocopier, as in Suchman’s study, but playrooms are noisy, busy places where the children are highly mobile. It is difficult to capture the many events going on at the same time and the researcher needs to make constant judgements about the focus of attention and what should be recorded (Plowman, 1999). For instance, we have already remarked that staff are well practised at regulating children’s behaviour by means of glances or gestures from a distance. Focusing the camera on the child will capture their response to this intervention but not the intervention itself. The danger is that these decisions about the focus of attention are made instantaneously, judgements are implicit and moments cannot be recaptured if they are missed. In these respects, video is not independent of the analysis, as Suchman claims, because some degree of analysis is taking place even as the material is being recorded. The use of video does not offer an unmediated reality and the presence of a video camera, Draft paper: not for citation Representing guided interaction 7 inevitably introduces an element of artifice. It is difficult, therefore, to represent video data in a way that is meaningful for the purpose of interrogation whilst maintaining verisimilitude. Transcription translates the ‘raw’ data into an easily accessible, paperbased tool for analysis but produces an incomplete representation of the video recording. Analysis then entails extrapolating from the transcript to an even more selective version of what was happening, given that the transcript will be interrogated through a particular theoretical lens (Ochs, 1979). It was important in Interplay to identify naturally occurring examples of guided interaction because we did not want to impose pedagogy on practitioners. Rather, we wanted to be able to provide ‘real life’ illustrations drawn from their current practice to serve as a stimulus for reflection and to enable us jointly to identify guided interaction and how this might support children’s learning. This meant that our video data is not systematic, but was recorded on an opportunistic basis, and has led to a preponderance of computercentred activities rather than those using the broader range of ICTs that we had encouraged with limited success. The free play ethos also meant that activities at the computer were not structured (other than by use of the sand timer, a mechanism for controlling turn-taking which was often subverted by the children) and the boundaries of interaction episodes are not clearly defined. Multimodal interactions We have already made the point that the interactions associated with guided interaction are multimodal in nature, encompassing language, gesture, touch, gaze, and physical movement. These interactions are highly contingent on the activity, the site of learning and the learner’s level of competence and they have been shaped by cultural and social processes. A number of different approaches to representing multimodal interactions that have been recorded on video are presented here. The first three use relatively conventional transcripts. This is followed by a description of a Media Interaction Chart, a time-based graphical representation of interactions between users and computers, and, finally, by a discussion of pictorial representations. The first example (Cole, 1996, pp.244-245) is from a transcript derived from video of an argument over how many rooms the children have in their apartments. The focus is on speech, with occasional references to gesture or to whom an utterance is directed. 4 Dolores (to Jackie) Count them all (Dolores holds up pinky). 5 Jackie Three bathrooms (holds up three fingers). 6 Dolores That’s not a room. 7 Reggie They’re rooms. Bath rooms. 8 Jackie No, I’m counting bathrooms. For my (holds out arms) whole house we got thirteen rooms. There are two (inaudible). 9 Dolores (holds up 3 fingers) Three, two (adds 2 more fingers). Danby and Baker (2000, pp.115-116) also use conventions that place gesture and body deployment in brackets, integrated with the speech. The episode demonstrates the pragmatic and strategic moves that children make to build and co-construct their social positions. The following brief example, one of many provided, illustrates the method used for representing the interactions: Draft paper: not for citation Representing guided interaction 8 60 David I DON’T WANT HIM DOWN HERE ((to teacher)) >61 T ((leaning over block shelf)) Well can you find a space to build [away* ( ) 62 Andrew HE W-WANTS TO BUILD SOMETHING(.) A-A-AND HE (reaches towards teacher and brushes David’s arm, Alan is building with the blocks.)) DAVID WON’T LET HIM BUILD WITH US The transcript is very detailed, in line with the principles of conversation analysis, and emphasises information relating to intonation, gaps and overlaps because it is relevant to the strategies deployed by the boys to maintain or gain social presence. The representation therefore needs to highlight those aspects of interaction which are most relevant to the focus of study. Bourne and Jewitt (2003, p.67) use video data to analyse the ways in which the interpretation of a literary text is constructed through social interaction in a classroom and the teacher is talking in the following excerpt. They develop standard techniques by adopting a four-way division of information for each utterance. This has the benefit of assigning equal status to verbal and non-verbal interactions, but it also makes the speech track difficult to follow and it is difficult to form a reconstituted image of the interaction once it has been fragmented like this. (1) Speech: So you see how men then draw a – a kind of – you know Gaze: at Lizzy Gesture: right hand in air, pointing brings hands together, palms flat fingers touching [to form a line] Posture: sitting on edge of desk, one leg on floor, back upright, angled towards Peter (2) Speech: There’s a line between male and female Gaze: at Peter Gesture: moves right hand few inches in front of left, palms flat, moves hands away from one another and back again Posture: sitting on edge of desk, one leg on floor, back upright, angled towards Peter Each example is derived from video recordings and uses transcription conventions of varying degrees of detail to represent the range of interactions. It is noticeable that the transcriptions are not very different from those used for audio recordings, so even though gestures and gaze are more easily perceived visually, they are represented verbally in these examples. Video is presumably selected as a method of data collection for its audio-visual nature. The audio track can be represented in a way that is adequate for most analytical purposes by the verbal format of the transcripts but the visual nature of video is not adequately represented by the same verbal format. Reading these transcripts, which are further decontextualised here, it is difficult to visualise the context they are designed to represent because the information is verbal. This should not necessarily cause a problem, as the written word is used to generate images from novels to newspapers, but these scripts serve a different explanatory purpose from the transcript, for which interpretation is usually more controlled. Draft paper: not for citation Representing guided interaction 9 Plowman (1992) developed primitive time-based graphical representations of interactions between users and computers in Media Interaction Charts based on video recordings (see Appendix 2 for example). These charted guided interaction in its earlier form, showing the interrelationship of design features with group interactions and providing information which allows researchers to get a panoramic view of on-screen events and the situated action of learners both synchronically (at the same time) and diachronically (over a period of time). Analysis of the charts led to the identification of task structure, navigation, and forms and functions of machine interaction as being of central importance for supporting group interaction with the technology. These were later developed with Luckin et al (1998) for other cases of group use of interactive media in schools but, although they provided information on screen events and user activities, these representations were labour intensive and suitable only as research tools. The charts had a number of shortcomings, including too much emphasis on screen events compared to the students’ experiences and a complex system of coding talk. Where events are broken down into too many component activities like this there is also a risk of losing the ‘big picture’. Although the chart is more visual than the excerpts from the transcripts above and the verbal mode does not dominate in the same way, it is diagrammatic rather than pictorial and does not solve the problem of how to facilitate visualisation of the context being represented. Rogoff (2003, p.52) avoids this problem by showing a series of images in which she offers ‘a sociocultural transformation of participation perspective in which personal, interpersonal, and cultural aspects of human activity are conceived as different analytic views of ongoing, mutually constituted processes’. The images (pp.53-61) are stills taken from a video which has been manipulated to emphasise different foci of attention. Thus, in figure 2.3A a solitary boy is shown with the background bleached out. In the next image, other people are included but artefacts are not shown and she demonstrates, through the sequence, how the interpersonal, personal and cultural-institutional aspects of the event constitute the activity. Nevertheless, she makes clear that ‘it is usually necessary to foreground some aspects of phenomena and background others simply because no one can study everything at once. However, the distinctions between what is in the foreground and what is in the background lie in our analysis and are not assumed to be separate entities in reality’ (op. cit. p.58). This approach seems closer to what we want to achieve and is consistent with a sociocultural account of the data. The pictorial illustrations that she provides are effective and more accessible than the standard transcriptions and charts exemplified above. By capturing still images from the video recording it is possible to analyse in more detail the non-verbal interactions which can be so fleeting but which are so important to an analysis of guided interaction. Principles for representing guided interaction This discussion leads to some suggested principles for representing guided interaction, some of which may have applicability for representing other forms of social interaction and some of which may be inappropriate. The overarching principles which are relevant regardless of context are i) having an explicit rationale for the format, ii) maintaining integrity between the theoretical and empirical approaches and iii) fitness for purpose. Making principled decisions about the chosen format for representations, whether transcript, chart or picture, strengthens the link between theory and its empirical Draft paper: not for citation Representing guided interaction 10 foundation, avoids using a particular format by default and reduces the potential for unknowingly valorising one mode of interaction over another. Representations may be very complex or quick and dirty but they need to serve the purpose for which they have been produced. For many researchers, the purpose is primarily analytical and representations are likely to be very detailed. For us, they need to be dual purpose: to assist in the process of identifying and analysing guided interaction but also to serve as a heuristic for practitioners. As such, they need to facilitate a process by which practitioners can identify what guided interaction is, recognise opportunities to enhance children’s encounters with ICT and apply this knowledge in ways that seem natural and do not disrupt playroom routines. At the time of writing we are uncertain whether we can find one representational format that serves both purposes or whether different formats are required. The overall aim is to represent the technological and interpersonal dimensions of using ICTs that illustrate the complex, dynamic and reciprocal relationship between practitioner and child/ren. To this end, representing guided interaction has been an important part of the process of defining guided interaction. For our purposes, the representations need to: • integrate social, cognitive, affective and sensori-motor responses with the ecology of the playroom • be applicable for multiple participants • demonstrate the role of artefacts in mediating interactions • acknowledge the multimodal nature of social interactions and give equal priority to non-verbal means of communication as to language • balance decomposition of activities with a holistic account of events • be pictorial rather than based on charts or transcripts • capture dynamic processes within still images. Additional challenges include how to produce such representations without being too labour intensive and how to provide information which would not normally be available from simple observation. In terms of their reception, the representations need to provide information which is easy to understand and to function both as research tools and as explanatory tools for practitioners. Ideally, they will also be adaptable for other systems and contexts of use. These principles for representing guided interaction provide the foundation for our current work of developing these representations. We hope to demonstrate the outcomes of this work in the next few months. 1 Some of this content overlaps with another contribution to this conference as both papers draw on the same research study. Using Sociocultural Theory to Develop Pedagogy in the Early Years (Stephen & Plowman) discusses guided interaction in the context of theorising support for learning and its role in developing pedagogy. 2 In the original research design we had planned to investigate guided interaction initiated by both adults and peers. We found very little evidence of peer support for learning of this kind and so we have focussed on the role of adults. 3 We have defined ICT to include activity centres, musical keyboards, tape recorders and electronic toys as well as everyday items such as remote controls, telephones and toys that simulate appliances such as mobile phones, laptops, cash registers and barcode readers. This broad definition contrasts with a widely held view amongst pre-school practitioners and parents that ICT consists mainly of desktop computers and peripherals. Draft paper: not for citation Representing guided interaction 11 4 As the home is an important site of learning our fieldwork encompassed domestic settings but this dimension of our study is not discussed here. See Ways with New Technologies: pre-school children learning to become e-literate by Joanna McPake in these conference papers for a discussion of children’s competences with ICTs at home. Technical competence refers to the ability to switch items off and on, and conduct other necessary operations for the desired activity. Cultural competence refers to children’s understanding of the social roles which ICT plays, and to their ability to harness ICT for a range of social and cultural purposes, such as communication, work, self-expression or entertainment. Learning competence is a subset of cultural competence, but one of particular significance to young children as ICT was being used at home to support early literacy and numeracy, communication and musical skills, and also had a role to play in helping children learn how to learn. In each case, the degree of competence children had acquired was dependent on a number of factors, including access to equipment, support for learning to use it, and the particular interests and aptitudes of older family members. 5 This implies that we knew what we were looking for at this early stage. Our original concept of guided interaction was based on our earlier studies but it has since been developed iteratively. The first stage in Interplay was to present naturally generated examples of guided interaction to groups of practitioners as a means of stimulating reflection on practice. We found that, although it was new to them, this was a term with which they felt comfortable and which could be applied to their practice, so they were able to design interventions which would provide opportunities for them to explore what guided interaction might look like. During these interventions, practitioners produced their own records of observations (typically on sticky notes or written in progress files, but occasionally using photographs or video) and we collected more video data on an opportunistic basis. This new data was discussed in further group sessions and the definition of guided interaction developed as a result of our own analysis and practitioners’ experiences. (We refer to this process as guided enquiry. A fuller explanation is provided in the companion paper.) So, as well as providing a theoretical underpinning for the research, guided interaction was also heuristically useful for practitioners and they endorsed the concept as valuable. There is thus a circular relationship between our development and testing of theory and its application which takes account of the situated nature of practice. AcknowledgementsInterplay: Play, Learning and ICT in Pre-School Settings is funded by the ESRCTeaching and Learning Research programme (RES-139-25-0006). More informationabout the project is available from www.ioe.stir.ac.uk/Interplay/. ReferencesBennett, N., Wood, L. & Rogers, S. (1997). Teaching Through Play: Teachers’ thinkingand classroom practice. Buckingham, Open University Press. Bourne, J. & Jewitt, C. (2003). Orchestrating debate: a multimodal analysis of classroominteraction. Reading – Literacy and Language 37 (2) 64-72. Cole, M. (1996). Cultural Psychology: A once and future discipline. Harvard UniversityPress, Cambridge MA. Danby, S. & Baker, C. (2000). Unravelling the social order in block area. In Hester, S. &Francis, D. (eds.) Local Educational Order: Ethnomethodological studies of knowledge inaction 91-140. Amsterdam: John Benjamins. Heath, C., Luff, P. vom Lehn, D., Hindmarsh, J. & Cleverley, J. (2002). Craftingparticipation: designing ecologies, configuring experience. Visual Communication 1 (1)9-33. Klerfelt, A. (in press). Gestures in conversationthe significance of gestures andutterances when children and preschool teachers create stories using the computer.Computers & Education. Draft paper: not for citationRepresenting guided interaction 12Littleton, K. & Light, P. (eds.) (1999). Learning with Computers: Analysing productiveinteractions. London: Routledge. Luckin, R., Plowman, L., Gjedde, L., Laurillard, D., Stratfold, M., & Taylor, J. (1998). AnEvaluator's Toolkit for Tracking Interactivity and Learning. In M. Oliver (Ed.), Innovationin the Evaluation of Learning Technology (pp. 42 64). London: University of NorthLondon. McPake, J., Stephen, C., Plowman, L., Sime, D. Downey, S. (2005). Already at adisadvantage? ICT in the home and children’s preparation for primary school. BritishEducational Communications and Technology Agency, Coventry.www.becta.org.uk/research/research.cfm?section=1&id;=4821 Ochs, E. (1979). Transcription as theory. In E. Ochs and B. Schieffelin (eds.),Developmental Pragmatics, pp. 43-71. New York: Academic Press. Pea, R.D. (2004). The social and technological dimensions of scaffolding and relatedtheoretical concepts for learning, education, and human activity. Journal of the LearningSciences, 13 (3), 423-451. Plowman, L. (1999). Using Video for Observing Interaction in the Classroom. Spotlight72, Scottish Council for Research in Education: Edinburgh.http://www.scre.ac.uk/spotlight/spotlight72.html Plowman, L. (1996a). Designing interactive media for schools: a review based oncontextual observation. Information Design Journal, 8 (3) 258-266. Plowman, L. (1996b). Narrative, linearity and interactivity: making sense of interactivemultimedia. British Journal of Educational Technology 27 (2) 92-105. Plowman, L. (1992). An ethnographic approach to analysing navigation and taskstructure in interactive multimedia: some design issues for group use. In People andComputers VII, eds. A. Monk, D. Diaper, MD Harrison, 271-287. Cambridge UniversityPress, Cambridge. Plowman, L. & Stephen, C. (2005). Children, play and computers in pre-schooleducation. British Journal of Educational Technology 36 (2) 145-158. Plowman, L. & Stephen, C. (2003). A ‘benign addition’? Research on ICT and pre-schoolchildren. Journal of Computer-Assisted Learning. 19 (2) 149-164. (Special issue onChildren and New Technologies.) Rogoff, B. (2003). The Cultural Nature of Human Development. New York: OxfordUniversity Press. Rogoff, B., Mistry, J., Goncu, A. & Mosier, C. (1993) Guided Participation in CulturalActivity by Toddlers and Caregivers, Monographs of the Society for Research in ChildDevelopment, 58 (8) Serial No. 236. Draft paper: not for citationRepresenting guided interaction 13Siraj-Blatchford, I & Sylva, K. (2004). Researching pedagogy in English pre-schools.British Educational Research Journal 30 (5) 713-730. Stephen, C. & Plowman, L. (2003). ‘Come back in two years!’ A study of the use of ICTin pre-school settings, 42pp. Learning & Teaching Scotland, Dundee.http://www.ngflscotland.gov.uk/earlyyears/ComeBackinTwoYears.aspStephen, C. & Plowman, L. (2002). ICT in Pre-School: A ‘Benign Addition’? A review ofthe literature on ICT in pre-school settings (31pp). Dundee: Learning & TeachingScotland. http://www.ngflscotland.gov.uk/earlyyears/BenignAddition.asp Suchman, L. (1987). Plans and Situated Actions: The problem of human-machinecommunication. Cambridge: Cambridge University Press. Tharp, R., G. & Gallimore, R. (1998) Rousing minds to life: teaching, learning andschooling in social context. Cambridge: University of Cambridge. Wertsch, J. (1991). A sociocultural approach to socially shared cognition. InPerspectives on Socially Shared Cognition, eds. Resnick, L., Levine, J. & Teasley, S.,pp.85-100. American Psychological Association, Washington DC. Wood, D. J., Bruner, J. S., & Ross, G. (1976). The role of tutoring in problem solving.Journal of Child Psychology and Psychiatry, 17 (2), 89-100. Wood, D., & Wood, H. (1996). Vygotsky, tutoring and learning. Oxford Review ofEducation, 22 (1), 5-16. Draft paper: not for citationRepresenting guided interaction 14Appendix 1 Characteristics of guided interaction Through a practitioner-generated cycle of planning, action and review we have begun toidentify the characteristics of guided interaction. These have been categorised into thepurpose of the intervention (operational, cognitive, social, affective) and the mode(physical contact, oral, gestural, gaze), although these categories may be revised. Thefirst seven examples below are taken from children’s use of computers; the otherexamples are drawn from children’s use of other types of ICT. Interventions are rarelyexplicitly cognitive in orientation, ie with the aim of developing learning in terms of thesubject matter such as numeracy skills, and most interactions are operational. Aspractitioners move on from guidance to support operational interactivity to learningconversations to prompt thinking and shared exploration we would expect to seeevidence of more guided interaction that is cognitive in purpose. We did not analyse interactions in this level of detail in the home settings but they werenot dominated in the same way by operational interventions. This is one reason why wehave encouraged practitioners to explore the use of other types of ICT, but thisemphasis on operational skills is probably to be expected in the early stages – childrenneed to learn how things work before they can play in more sophisticated ways. Guidedinteraction may not be needed as much for other forms of ICT where the affordancesmake the modes of interaction more visible. Form of guided interaction Mode Purposeplacing a hand over child’s hand as they movethe cursor or click on an iconphysical contact operational demonstrating how to use a tool such as thepaintbrush or eraserphysical contact; oral operational moving child to appropriate level of difficulty gesture; oral operationalreading instructions oral operationalproviding positive feedback oral affectiveintervening in turn-taking oral; gaze socialsharing pleasure in features such as animation oral; laughter affectivehelping with typing in names (typically to start anew game)oral; typing operational;cognitiveshowing how to hold a digital camera physical contact; oral operationalshowing how to view pictures on a digitalcameraphysical contact; oral operational holding earphones so that child can hear tape atlistening centrephysical contact; oral operational showing children how to plug in electronickeyboardphysical contact; oral operational asking what the music on a Coomber CD playermakes them think oforal affective turning over pages of story as children listen onaudio tapegesture operational using a play phone to order a taxi oral social Draft paper: not for citationRepresenting guided interaction 15Appendix 2 Media Interaction Chart for part of lesson using North PolarExpedition from Plowman (1992)