3.3.2    Children’s eye function, field of view and vision issues

This section begins with a reminder of the prior observations that began the research of the relevant literature.

  • During the period of contextual research children appeared to ignore icons in the corners of the screen. There appeared to be a preference for the middle-left and middle-right of the screen. They also appeared to press some buttons repeatedly, and ignore others completely. They clicked anywhere quickly and at random and frequently moved to the next screen using the hotspots without attempting to read the contents of the screen.
  • The interactive products in the contextual research had a cluttered interface. What was the significance of busy screens – a lot to look at, small pictures, and navigation buttons of all shapes and sizes?
  • In early drafts of the Research Tool children could not identify movement in areas far from the focus of activity on the screen.
  • The contextual research raised the concern that the computer was an adult product. The CD-ROM disc case, keyboard and the mouse were not designed for little hands. Even the best-designed trolleys were too small for the computer, keyboard and monitor. From these practical issues came serious concern that in the process of learning children had great difficulty holding the mouse, manipulating the cursor over hotspots, holding the mouse still whilst clicking on hotspots and carrying out click and drag movements.

The relevance of eye function in the literature review was central. First, the contextual research observations proposed a closer investigation into the factors influencing the way humans study visual material; second, the physiological aspects of eye search had not featured in educational research due to the experimental approach to IT and other limitations as indicated in the previous section of the literature review.

The section continues with a survey of writings on eye physiology, primarily the work of Bruner and Mackworth (1970) who used an eye-tracking camera. Later, Bruner and Mackworth’s findings are considered in the context of education by referring to the work of Piaget and Vygotsky.

In an educational context it was not common current practice to make connections between interactive design and children’s visual search abilities. In relation to still images only, not interactive images, Hubel (1988) offered a standard description of eyes fixation and saccades or short movements to a new position.  The eye was also alert to any object that asserts itself by moving slightly contrasting with a background, or presenting an interesting shape, either singularly or in combination. Bruner and Mackworth used an eye camera to record how the eye tends to oscillate around areas of visual interest of a photographic picture. However, they observed a difference in the eye movement between adults and children in the process of viewing still pictures. Pictures of a water hydrant (Figure 3.1) demonstrated the difference in terms of search patterns between children and adults.

Figure 3.1: Three individual eye-track records. Top: long eye track (66 inches from adult); Middle: extremely short eye-track (14 inches from a 6-year-old; Bottom: 5 to 6-year-old’s preoccupation with detail (27 inches). Large arrows added to illustrate the extent of eye-tracks for each age group. (Bruner and Mackworth, 1970, p.157)

The image shows how 6-year-old children’s eye movements oscillated around a small area of a picture. An adult tended to explore pictures systematically. The test was carried out using three levels of image sharpness. All results confirmed the same relationships. Yarbus (1967) attributed the differences to the apparent reluctance of children to make use of peripheral vision (p. 157) and lack of precision in the oculomotor system (p. 158). Mackworth (1965) also inferred that children fixated around a point when comprehending a difficult word and that the field of view was also reduced. He also reported (1970) that children make much larger leaps of eye tracking than adults do when the image was against a difficult background. The tendency to leap about extended to three inches or more as against one inch in favourable conditions. Other results of his research showed there was little consensus as to what was the agreed important feature of a scene. Also in terms of the direction of leaps of gaze, children failed to make vertical leaps, preferring horizontal search tendencies. Children also ended their gaze by making long leaps to large patches of colour, especially where these had a high brightness contrast with the background (p. 163).

Bruner and Mackworth showed eye search tracks in Figure 3.1 looking for edges. Hall (1969, p. 82) considered the brain in natural conditions automatically precues i.e. sees more clearly in terms of edges ‘Edges create a cortical jolt and straight lines as edges are beyond those experienced in nature.’ Gregory (1974) reported experiments in which individuals express a preference for images that use straight lines. A recognition of cultural factors where ‘carpentered’ features are rare was given by Deregowski (1980). A solution to the confusion of straight or curved line preferences for visual search stimulus appeared to be resolved by Hubel (1988) who described how the orientation of the end stopper cells in the eye are sensitive to both corner curvatures and to breaks in lines (Figure 3.2). They were excited by borders ‘We can thus view end-stopper cells that were as sensitive to corners, to curvature, or to sudden breaks in lines’ (p. 85).

Figure 3.2: The best shape for stimulating activity in rods and cones. (Hubel, 1988, p.85)

An alternative reason for selection of areas of focussed attention was the spotlight model of visual attention, Eriksen (1988, p. 10), and since confirmed by others, which demonstrated that pre-cueing by 100-200m the location in a visual field, where a target subsequently appeared facilitated target identification.

In estimating the value of vision literature to the argument, Bruner and Mackworth’s observations were only made with still images and before computers with graphical user interfaces. Whether modern complex interactive screens reveal similar patterns was speculative. The contextual research observations of extended leaps in unfavourable conditions, the tendency to horizontal search, the focussing around small areas of interest appeared to be informed by these studies.  There were no references to Bruner and Mackworth’s physiological approach to the complexity of screen design issues in guidelines or current writings. It is conceivable that multimedia screen ‘pages’ may be much more difficult for children to search for these physiological reasons. Further research will be required to clarify the supposition. Unfortunately it was beyond the scope of this study to employ the use of an eye camera to demonstrate Bruner and Mackworth’s findings on current multimedia display conclusively. This was because of the time, cost of using an eye camera and the technical problems associated with the equipment.

Next, the role of the foveal oval was researched. The reason for the investigation was that the operation of the foveal oval may inform interface design because of the fovea’s role in directing user’s gaze at elements on a complex images on computer screens. Of particular interest was the narrow area of clear focus and further subconscious narrowing of vision as a reaction to movement of the mouse pointer arrow or moving events on an interactive screen.

The fovea is a small depression in the central region of the retina containing only cones, which are colour sensitive and capable of finer resolution of images. Gregory (1974) described the foveal area as 1 degree of visual field and being of greater acuity consisting of 2,000 cones, with the edge of the retina sensitive to movement, noting that when movement stops the object becomes invisible. The foveal oval was also termed the foveal lobe. Hall (1969) described the two fields of view – the foveal oval and the wide-angle view – as ‘together creating an illusion of broad band clear vision.’ The foveal oval represented an area in the scene which was: ‘quarter of an inch wide at a distance of 12 inches from the eye. It extends 12??q-15??q in the horizontal and 3??q in the vertical within the larger area of the ocular field of vision covering 90??q.’ Figure 3.3 simulates the area of sharp focus indicated by the centre circle and unconscious effect to the blurring of the edges. The average person, and especially a child is unaware of the phenomenon. The illustration does not fully accurately visualise the true oval shape of the fovea’s area of acute vision as described by Hall or the extent of the field of view.

Figure 3.3: Double Portrait Tate Exhibition Wellcome Trust Sci-Art Project: The Painter’s Eye Movements, 1991. (Ocean and Tchalenko, 1998)

The effect of a concentrated field of clarity that the foveal oval provides was further enhanced by the shape of the eyeball and a vertical plane such as a computer screen. First, the spherical plane of field of view in focus created by the shape of the eyeball, ensures that, as demonstrated in Figure 3.4, the edge of computer screen surface was naturally out of focus when a user was focussing on a narrow central area of the screen owing to the curvature of the eye lens. (See alsoFigure 3.7)

Figure 3.4: Diagram showing the limit of the area of focus at the computer screen.(Adler, 1997)

Another aspect – the ‘outer’ limitation of visual field – may account for children tending to ignore corners of the computer screen; the human field of view was not rectangular. As shown in Figure 3.5, the top left and right of the stereoscopic wide-angle view of the visual field were not scanned in single fixed gaze.

Figure 3.5: The ocular field of view as defined in standard eye physiology references. (Kalawsky, 1993, p. 50)

Children staring at the screen may have ‘blind’ spots in these areas. Interface designers appeared to assume that the whole screen fills the field of view.

The argument in the context of this research was confined only to drawing attention to an effect on visual focus caused by movement of the mouse or movement of objects on a busy computer screen. The process of a logical visual scanning process of a visual field in a natural or print environment has been the subject of many studies summarised by Rayner (1998) and there was still no consensus of how the human scanning process of static visual content works, due to the complexity of factors involved. Noton and Stark (1974) suggested a sequence of eye search in adults with a broad A-shaped sequence as shown in Figure 3.6. But Noton and Stark had no reference to children’s patterns of search.

Figure 3.6: Modified Feature Ring takes into account less regular eye movements that do not conform to a scan path. Several movements which appear in 35% of recognition viewings, are in the centre of this ring. Outside ring, consisting of sensory (colour) and motor memory traces (black), represent scan paths and remains the preferred order of processing. (Noton and Stark, 1974, p. 122)

However, in a paper soon to be published Reichle et al., (2002) attempt to draw together the complex information processes involved in reading text, including the recent advances in cognitive neuroscience to propose a time sequence for eye movements, cognitive processes, and brain areas involved during reading a word.

In this section, research into vision; visual search patterns, how the eye tends to focus on small areas due to the physical structure of the eye, preference for edges and response to objects that move informed the design of the Research Tool by indicating ways to improve the quality of interaction. These discoveries suggested a simple interface centred in the screen with few features was an advantage. Awareness of these physiological factors could be of use to graphical user interface designers of educational products.

The next section of the literature review investigated another conventional assumption of educationalists, that a book – as an object – could be easily transferred to a computer. The need to analyse this assumption was that many early educational CD-ROMs became multimedia because the original resources were books and were developed by book publishers and their designers. The challenge to this assumption in this review was not on academic or literary principles, but the visual relationship of the reader to the book in a physiological context. The origin of the enquiry was the researcher’s observations of children not reading the actual words on the screen when the book assets had been transferred to the computer screen, but to click on hotspots or buttons that had been added to the text by designers. The superficial cause was the propensity for children to click on highlighted, animated buttons. A deeper significance was proposed. The proposal was that a book held in the hand should be recognised as a three-dimensional object; the ‘on screen’ version was two-dimensional. The hypothesis was that a three-dimensional object transferred to a two-dimensional medium incurs physiological ‘penalties’. The penalties might be the cause of children’s observed problems. The purpose of this section was to illuminate the problem of the processes involved in loss of information between 3-D and 2-D elements on a computer screen.

Beck (1992, p. 4) in a study of visual processing in textural segregation demonstrated how diagonal slanting lines 18 degrees from the vertical embedded in a sequence of vertical lines were immediately easier to identify when the whole design was rotated 75 degrees backwards to the reader (to a perspective view). The easier recognition of diagonals suggested that textual images (letters) in a multimedia screen ‘book page’ display in a vertical screen may be easier to recognise on a screen sloping away from the reader with or without 3-D perspective visual clues. Enns and Rensink (1992) showed how improvement in visual search was also possible when shading was added to images. Enns (1988, p. 721) listed further advantages of a pseudo 3-D perspective field of view. The features that aid ‘pop-up’ resulting in faster identification times were not just line orientation, but also length, width, curvature, number, terminators, intersection, closure, colour (hue), brightness (greyness), flicker, direction of motion, binocular lustre and stereoscopic depth.

Two features aiding perception: orientation of the object – looking down, and light shading besides colour and motion stimulate pre-attentive vision or pre-cueing as identified by Gregory (1974). Smets et al., (1994) have recorded that to really have effect, shape and depth perception in 3-D displays should also have movement ‘The movement of the observer has to be linked to motion at the workstation.’ These features aiding perception, of looking down and pre-cueing were present when reading a real book but absent from the on-screen version; a real book held in the hand has print viewed at an angle, not at the vertical as on a computer screen. The current on-screen ‘book’ has none of these features of optimum viewing.

The potential significance of the child in relation to the computer and the orientation of the computer screen itself was considered next. At this stage in the argument, in an educational context, the concept of an optimum viewpoint was tentative.

Children reading text on a CD-ROM ‘book’ were in a quite different configuration to a normal book reading position as illustrated in Figure 3.7 by Mach .

Figure 3.7: Mach observing visible parts of his own body and the surroundings (Held and Durlach, 1991, p. 238)

The artist illustrated one of the natural positions for reading; looking down, comfortable, secure and with a field of view with indistinct periphery images. Mach illustrated a natural reading position which was similarly a desirable position of a computer user reading in a natural posture for reflective, engaged activity. The ergonomic aspects of the reclining figure in contrast to the British Standards regulations was considered further in the ergonomics section of this chapter. At this stage of the argument for a revision of the way children used computers in an education context, the scene had three vision issue elements of significance. First, the centre of the field of view was in greater acuity for the reasons already accounted for in Figure 3.4. Second, the natural field of view also takes in an awareness of one’s own body as part of the picture. Kalawsky (1993) considered the definition of virtual presence to be greatly enhanced by seeing parts of one’s own body:

A very strong feeling of presence can be achieved if the visual system allows the actual lower part of the operator to come (sic) into the field of view at the appropriate time. (p. 81)

Third, not clearly visualised in the drawing was that according to recent research by Intriligator and Cavanagh (2001) the lower visual field was physiologically more effective in terms of greater acuity and faster response times than the upper visual field. Therefore greater acuity in the lower visual field was reviewed for implications to this research.

In commercial marketing studies of supermarket purchases, it was reported that shoppers show a preference for items that were below the horizontal eye-level position indicate in the visible surface area in Figure 3.8. Stoper and Cohen described that in the normal terrestrial environment humans required three types of viewpoint: surface relative, gravitationally relative and head relative eye levels, corresponding to the angle at which the head was held.

Figure 3.8: Diagram: Three types of eye-level viewpoints in the normal terrestrial environment. (Stoper and Cohen, 1991, p. 391)

This phenomenon has been known for some time (Enns and Rensink, 1992) where ‘objects are more easily apprehended when they are below the line of sight and also when they are lit from the same direction.’ (p. 722). Maddess et al., (2000) also found, in a study of diagnosis of glaucoma, that there were finer responses to stimuli in the lower visual field.

In a computer user context, Cochrane (1996) proposed a general preference for head-down operations at the computer as being more efficient:

If I gave you a sheet of paper and a pen and asked you to write a note, I would be amazed if you held it at head ‘home page’ height and proceeded to write. I would expect you to sit down and crouch and write looking down. For various reasons locked into our distant past, we happen to be about 20 per cent more efficient when we read and write looking down on a sheet of paper than when we look straight ahead. (p. 10)

Intriligator and Cavanagh (2001) identified specific advantages of downward gaze in terms of finer resolution in the lower field of the visual field – by 50% in a radial dimension and by 17% in a tangential dimension (p. 200). They proposed their observations challenge the spotlight theory of attention by Eriksen and St James (1986) and others. Intriligator and Cavanagh drew attention to the over presentation of the lower visual field in the occipital-parietal region of the brain which Van Essen et al., (1984) argued ‘is required for the control of hand and arm movements’. The same part of the brain was the likely site for control and selection of visual attention.

In terms of not only visual acuity, but also speed of accessing information, the lower visual field has advantage. Sheedy (1990) showed an optimum response time (4% improvement) to photographs from a computer screen at 50 cm for 10 degrees of depression when a head-and-chin-rest fixed the head location with the frontal protrusion of the chin and a point just above the brow line aligned in the vertical plane. Mamassian and Landy (1998) reported humans preferred to interpret ambiguous line drawings as representing objects seen from above rather than seen from below; and Rubin et al., (1996) indicated the task of identifying the segmentation of an image into figures and background was shown to be performed much better in the lower visual field.

Another, additional factor that influenced the preference for viewing objects in a downward gaze – the vertical horopter – was proposed by Ankrum et al., (1995). The horopter is a vertical line starting from between the viewer’s waist and feet and projecting outwards. The stereoscopic mechanism operated in conjunction with eye cell sensitivity to colour in the red and green spectrum, consistently placed elements in a vertical plane below the line of gaze in front of the vertical plane, and above the line beyond the vertical plane as shown in experiments by Cogan (1979). A viewer identified the most efficient angle of downward gaze by fixating on the centre of a vertical wire. The vertical wire was seen as double until the wire was tilted backwards to a point where it appeared as a single line. At this point the optimum angle of view was defined for that user. A user may vary in their preferred angle of tilt. Researchers collated data from verbal responses to three screen angles – tilted towards the user (negative) 40??q, 15??q away and 40??q from the user (positive). A user was placed in high screen position – eye level horizontal to the top of the screen and low screen position – eye level 20??q above the top of the screen. Users reported significant preferences for the positive 40??q angle at both high and low position (p. 134). The authors considered a vertical screen orientation to be inconsistent with the human visual system and proposed ‘the vertical horopter may play a role in visual and postural discomfort at computer workstations’ (p. 135).

In contrast to the evidence in this section for information to be presented in the lower visual field was the actual viewing position of children looking at a CD-ROM ‘book’ in an educational environment. The contrast was illustrated in Figure 3.9, photographed by the researcher at a school during the main study. Children were looking at the screen at varying vertical and horizontal angles. The classroom conditions illustrated in Figure 3.9 are still by no means uncommon in schools.

Figure 3.9: Children viewing the computer screen on a special commercial computer trolley (Howarth, 1995a, p.1).

The viewing configuration was less than visually efficient for the three reasons made in reference to Mach’s image illustrated in Figure 3.7 above. The cause was often due to the limited space on a computer trolley. A further visual problem was added by the common practice of more than one child looking at the screen at once. Sedgwick (1992) reported close viewing distorts geometrically specified virtual space. For example, a square grid was distorted into a diamond. A side viewpoint created a shearing of virtual space, so a square grid had its upper side shifted laterally to form a parallelogram. A similar effect occurred when the viewpoint was too high or too low.

The literature review in this section informed the Research Tool to the extent that eye search in children suggested simple, central screen images were preferable. Evidence about the preferences of downward gaze and sloping visual field suggested changes to existing height and orientation of computer screens in the classroom could also lead to greater efficiency in vision. However, the findings were not incorporated in the Research Tool because of limitations in computer screen configuration. Finally, the pattern of sharing three children to a computer was likely to remain in schools. In 1998 there were twelve computers per primary school (DfEE, 1998) and in 2000 seventeen computers per primary school (DfEE, 2000).

In the next section, the issues surrounding existing height and orientation of computer screens in the classroom – the field of ergonomics and the origin of the official standards for optimum computer viewing conditions – were studied in greater detail.


3.3.3    The role of ergonomic and human factors

In the contextual research the following aspects were observed:

  • Children appeared to have had a series of physical problems associated with managing the computer, the keyboard and the computer desk itself.
  • The classroom was a crowded place. The computer and keyboard were usually put on a narrow school desk, with the monitor on top of the box leaving just enough room to fit a keyboard on the space in front and no room to use the mouse.
  • The computer may be put in full sunlight, in a corridor, anywhere there was room. What was the significance of the poor ergonomic conditions of screen, desk, mouse design and general usability in a busy, crowded environment?

During the huge growth and interest in ergonomics over thirty years there appeared to be, as far as the research reveals, no specific references to computers in education of children or education generally, except by Pheasant (1986) who observed:

Once common, sloping desks have all but vanished from our schools and offices. Should the ergonomist mount a campaign for their re-introduction? (p. 194)

He reported Bendix and Hagburg (1984) who, in an adult context, evaluated horizontal desks with slopes of both 20??q and 45??q. They found that trunk posture improved with desk slope during a reading task. However, whilst subjects clearly preferred the steep slope for reading they preferred a horizontal plane for writing. Pheasant suggests the best solution was to allow for the option. He concludes:

Despite the extensive research a surprising number of controversies remain unresolved. Should the seat slope forward, should the writing surface slope, should the operator sit upright or recline? (p. 196)

These very early, unresolved controversies were not referred to in the set of guidelines for computer users, an example of official guidelines for one person and one computer by organisations such as the British Standards Institute (BSI) illustrated in Figure 3.10.

Figure 3.10: British Standards Instituteaverage man computer station measurements. Shown for general orientation only. (Karwowski and Marras, 1999, p. 25)

The BSI standards in Figure 3.10 indicated users should have a near straight back position in contrast to a natural reclining reading position of the back, and to be looking downwards at 20??q.

The regulations are based on ergonomic considerations, which are related to back posture. The school IT bench design are also defined by: standard school door width, and traditional (woodworking) workbench dimensions. In practice, general rules are plainly ignored as in Figure 3.8 and in general by the variable situations of room and lighting availability in schools. Figure 3.8 is a current dedicated school computer trolley design, yet it only allows the monitor to be placed on top of the PC, forcing the child to look upwards.

Hill and Kroemer (1986) questioned early ergonomists’ suggestions for an optimum viewing angle and suggesting a variety of viewing angles from 10 to 38 degrees below the horizon, indicating the possible cause as problems associated with defining what was horizontal. They confirmed ‘a basic preferred declination of the line of sight’ (p. 129), the cause being the six muscles that hold the eyeball in tension and which are in minimum contraction when the eye was in a slight downward direction. They indicated that there was a range of angles preferred by subjects, especially as they lean forward when a steeper angle of view has been observed. They proposed ‘one should place the monitor closely behind the keyboard, not on top of the disk drives (CPU)’ (p. 134). Kroemer et al., (1994) reported not one but several optimum visual positions when operating different kinds of computer equipment. They again referred to ‘putting the monitor on a CPU box is rather uncomfortable for the viewer’ (p. 444).

Figure 3.11: Suggested angles of line of sight with current computer workplace technology(Kroemer et al., 1994, p. 144)

The variation of backrest position was rigorously investigated (Burgess-Limerick et al., 2000) the results indicating that a computer workstation should be 15 degrees below horizontal eye level. They confirmed the findings of Bergqvist et al., (1995) that there was an association between eye-level computer monitor heights and neck discomfort.

Two studies showed a different research approach into identifying the preferred angle for screen orientation. Instead of controlling the head position with rests and restraints and being concerned with the quantitative data of the body orientation, these reports focus on subjective responses to comfort and viewing angle preferences. Highest scores were achieved at a monitor angle of 35% with a research rig placing the TV style monitor next to the keyboard (in the similar position to the current laptop) (Quaranta Leoni et al., 1994). A reduction in fatigue was attributed to the fewer vertical saccades required between keyboard and screen and they suggest a ‘lowering of the VDU actually results in less visual fatigue’ (p. 196).

The exploration of the role of ergonomics in this section identified unresolved controversies about an agreed standard for writing and reading positions at the computer. The evidence was still not conclusive as to the preferred angle of downward gaze or for the slope of the screen in relation to the viewer. The research indicated that individuals vary. In this section of the literature review the research found no studies of visual or ergonomic efficiency of the same tasks on a head-down laptop and a head-up screen on CPU box in a real world environment – the conditions in many UK schools and colleges.

In addition, there appears to be no evidence discovered in the literature review that educationalists recognised the possible educational implication of: user’s head position in relation to the computer, the orientation of the screen, the distortion of images or less than optimum viewing positions of children. The current review of ergonomics descriptions of the relationships between a child and a computer appeared limited in scope. Changing computer screen orientations and ergonomic factors to achieve optimum viewing conditions in the main study for the Research Tool were outside the scope of the study. The next section examined other methods available of improving optimum conditions for computer interaction in the classroom.

3.3.4    Pleasure in learning

The potential for learning and pleasure at the computer was studied in this section. If the optimum position of a user and computer were difficult to obtain in terms of viewing angle of screen and its relationship to the keyboard and mouse and the images on the screen in the Research Tool what other features of interface design might provide optimum conditions in terms of a deeper sense of engagement? The emotional content of interaction was considered. The relationship to the previous section’s literature review research arose from the educationalists’ recognition of the pleasure children had when using computer games at home and in the classroom and from the contextual research.

In the contextual research children were observed:

  • Working away without any sense of time.
  • They chatted, concentrating intently.
  • They expressed their pleasure in the CD-ROM educational materials in relation to their quality of experience with computer game playing.

The role of pleasure and fun in learning focuses on the physical activity, the relationship of the mind to body and the sense of well-being that results from the whole body (holistic) involvement.

Vygotsky (1962) stated ‘In play a child is always above his average age, above his daily behaviour, in play it is as though he were a head taller than himself’ in considering that conceptual abilities of children were enhanced through play. There were respectable educational precedents in the use of simulation in a historical IT context (see 3.2.1). The value of the emotional learning content of the Turtle in Logo was also described by Taylor (1980). Malone (1981) explored the reasons for the attraction of games and how features could be transferred to computer-aided learning. He defined three elements: challenge (goals and scoring), fantasy (features such as ‘explore within’ arrows and drama) and curiosity (music and forms of feedback) (p. 81). Malone referred to the physical and emotional processes involved in traditional learning by doing evidenced in pertinent titles such as Learning Through Activity, (Anzai and Simon, 1979), and The Use of Gaming for Motivation, (Atkinson and Birch, 1978). The roles of mental imagery and associative learning have been established from the psychologist’s perspective in the work of Bower (1972) and Buzan (1989). However, Malone considered that visual effects alone may entertain but be less educational but that environments varying in difficulty level ‘increase both challenge and potential for learning’ (p. 82). He also posited, that the real potential for computer learning was in games he calls isomorphic, where cognitive curiosity was fully engaged.

Malone’s observations have informed the Research Tool. The educational potential of gaming and particularly the ‘varying levels of difficulty’ concept were employed in the designing of a variety of tasks and the order of the activity pages. However, he was only able to predict the potential value for captivating sensory effects using computers because the level of interactivity at the time he was writing was so limited. The special value attached to isomorphism in the Research Tool was the connections made between the physical manipulation of the mouse as beater and the percussion instruments.

Csikszentmihalyi (1992) took a more comprehensive view than Malone and summarised, in his flow theory, as a result of a prolonged study of enjoyment and quality of life, eight components that provided the necessary phenomenology of enjoyment in the following list:

1.  The experience usually occurs when we confront tasks we have a chance of completing.

2.  We must be able to concentrate on what they are doing.

3.  Concentration is possible because the task has clear goals.

4.  Concentration is possible because the task has feedback.

5.  One acts with a deep but effortless involvement that removes from awareness the worries and frustrations of everyday life.

6.  Enjoyable experiences allow people to exercise a sense of control over their actions.

7.  Children lose concern for the self disappears yet paradoxically the sense of self emerges stronger after the flow experience is over.

8.  The sense of duration of time is altered; hours pass by in minutes.

The combination of all these factors caused a deep sense of engagement through enjoyment that is so rewarding that people feel that expending a great deal of energy is worthwhile just to be able to feel it. (p. 49)

Csikszentmihalyi identified these features of depth of engagement to be experienced by children and adults alike in any activity and walk of life. He drew on evidence for children’s pleasure in activity (p. 47) from Buhler (1982) using one’s body in such activities as running and swinging. Also from Piaget (1952), who viewed that one of the sensory motor stages of an infant’s development was characterised by the ‘pleasure of being the cause’. Berlyne (1960) sought neurological explanations in terms of balance between incoming stimulation and the nervous system’s ability to assimilate it. The approach of Deci and Ryan (1985) was to look at the pleasure aspect because it gave a person a feeling of competence, efficacy, or autonomy. In a description of enjoyment that suggests a resonance with Tuan (1977) about the loss of front/back body orientation being described as ‘lost in the forest’, Deci and Ryan described how a forward movement, when a person has achieved something but gone beyond what was expected, may not have been enjoyable at the time. But the novelty and the accomplishment override the moment and on reflection, become a memory of fun. Pleasure in itself is different to enjoyment through activity in that it requires no effort and concentration.

To summarise the relevance of this section to the thesis; the universal human qualities of Csikszentmihalyi’s phenomenology of enjoyment suggested no conflict in relating elements of concentration and pleasure to deeper engagement through physical activity at the computer using the mouse as a manipulation tool or in varying the level of challenges in the computer activities.  Csikszentmihalyi’s work contributed by informing how the coherence of these elements improved the quality of interaction and created activities which can be replicated using the interactive technology of the Research Tool. In the next section education theory concerning the physical activity during learning was studied in more detail to inform the advantages or otherwise of the replication of physical manipulation on the computer.