1. INTRODUCTION
2. PASSIVE TOUCH, STUDIES OF SEPARATE TOUCH SENSATIONS
3. ORGANISATION OF THE HAPTIC SYSTEM
The Skeletel System
The Neural System
4. DIFFERENCES BETWEEN PASSIVE AND ACTIVE TOUCH
5. THE THEORIES OF J.J. GIBSON
Perceptual Meaning
Information Pickup
Verbal Meaning
Evaluation of J.J. Gibson's Contribution
6. SOME EARLY INVESTIGATIONS INTO BRAILLE READING BEHAVIOUR
7. PERIPHERAL MECHANISMS
Structure of the Glabrous Skin of the Human Hand
Receptive Fields
8. PSYCHOLOGICAL STUDIES OF PERIPHERAL MECHANISMS
Roughness Discrimination
Pressure, Vibration, and Shear
9. CENTRAL MECHANISMS
Parts of the Brain Involved in Touch Perception
10. PSYCHOPHYSICAL STUDIES OF CENTRAL MECHANISMS
Hemisphere Asymmetry
Convergence
Memory
11. THE INTERLOCKING STRUCTURE OF VARIABLES
This short introduction to the psychophysiology of touch perception as it relates to using the braille code is included because an understanding of the functions involved in braille reading is incomplete without it. For many years educational psychologists and teachers investigated reading habits of blind children, and the studies of Ashcroft (1960) and Nolan and Kederis (1969), for example, showed the effects of different reading situations such as dot numerosity and the use of contractions in different parts of the words. Investigations, since approximately 1970, have highlighted a greater understanding of the peripheral and central mechanisms of neural input, which can be applied to the complex procedures involved in braille reading. However, as so often happens, this new knowledge leads to the realisation that so much more is still waiting to be discovered about tactual reading. Many investigations are being made, but direct application to braille reading still needs considerable expansion.
To many the code has proved difficult to learn and to use for several reasons. Amongst these are problems of perception, and the many rules made necessary because only 63 configurations are possible using Braille's 3x2 matrix. The same sign may have to represent up to 8 different meanings according to position within the matrix, within the word and when used for punctuation. As a result of such difficulties it is a very slow medium to use, for the average number of words read per minute is approximately only a third of those read in a visual medium. The past 50 years has seen a succession of attempts to alleviate this situation by means of observation, experiment, and training in endeavours to improve the accuracy, comprehension and rate of braille reading performance. The more important of these efforts will be discussed.
To be of most help these studies need to be read in the light of an understanding of touch perception as it applies to the reading of braille. Much has been written about the psychophysiology of visual and auditory systems, occasionally using the touch modality for contrast, but little has been understood about the latter until the writings of Katz in 1925. His work was not translated into English in its entirety so was comparatively little known. It was therefore all the more impressive when the works of J.J. Gibson (1962; 1966, published in Britain, 1968) became known in America and Britain. In a personal comment on Katz' monograph "Der Aufbau der Tastwelt", Gibson said "I owe more to it than I have recognised recently" (Krueger, 1982), but such a comment does not detract from J.J. Gibson's work for all research should be carried out with due recognition of what has already been discovered. Since then there have been several studies on different, specialised aspects of touch perception. Much work still needs to be carried out in these areas, but meanwhile the present knowledge needs to be linked up with what is known concerning the braille code and reading in braille.
"For a long time, two assumptions have been made about the senses, first that they are the only sources of knowledge about the world, and second that they are the channels for special qualities of experience" (Gibson, 1968, p.47), yet as late as 1973 Taylor, Lederman, and Gibson noted that "It is remarkable how little is known about perception by touch after more than a century of experimental psychology" (Krueger, 1982, p.4). Reasons that have contributed to this lack of knowledge include:
For a century from approximately 1830, tactual perception was studied as cutaneous sensitivity. Parts of the skin were probed in efforts to determine reactions such as awareness of pain or cold, and it therefore seemed reasonable to suggest that each sensation corresponded to a nerve ending which, when excited, would convey the information to the brain. An attempt to list the mosaic of sensations proved impossible because parameters of the sensations were difficult to define. For example, "It was argued that temperature was a different quality from touch, and that pain also was different" (Gibson, 1968, p.98), that warmth must be separated from cold and that pressure differed from prickly pain. Early in this century it was thought that touch could be divided into 5 senses of pressure, warmth, cold, pain, and kinesthesis, which is an awareness of movement (ibid., p.98). All these enquiries were confined to the study of passive touch rather than active touch.
Revesz (1950) wrote "the sighted organize space mainly in terms of external spatial co-ordinates. The congenitally totally blind rely instead on haptic (touch and movement) space" (Millar, 1994, p.19). This seems highly probable and will be discussed later in the chapter.
Katz suggested that the hand should be regarded as the organ of touch (Krueger, 1982, p.17) because of its versatility. Its movements make possible the perception of tactual qualities, the manipulation of objects, and it is constantly in use in a variety of ways. More specifically, Gibson wrote (1962, p.479) "When the hand is feeling an object the movement or angle of each joint from the first phalanx of each finger up to the shoulder and backbone makes its contribution". That is, the skeleton is an organized system simultaneously and successively linked via joints and tendons to the central nervous system. It is "not a collection of sensations, but structured perception" (Gibson, 1968, p.118). In 1968 (p.110) Gibson included an illustration showing the innervation of the muscle of the upper arm resulting in the movement of the arm at the elbow. It would seem reasonable to infer that a similar mechanism must be responsible for the flex and stretch of the muscles in the fingers when reading braille.
The neural system is also arranged hierarchically comprising the peripheral nerves connected with the spinal column and thence to the brain. Cranial nerves are involved with vision, hearing, tasting, and smelling, all within the region of the head, but the touch mechanism is more complex and wide-spread. The receptive units, being affected by mechanical energy, are termed mechanoreceptors. They occur all over the body (ibid., p.108), in and below the skin, in joints and connecting ligaments between bones, in the muscles and tendons and also wrapped around blood vessels. The afferent (ingoing) nerves lead to nerve centres and some continue as far as the brain, and efferent (outgoing) nerves from the brain or nerve centres connect with muscles and joints responsible for movement (ibid., p.5). All of these parts are mobile. For example, the skin is deformed as it passes over a surface, the joints rotate in their sockets and muscles are contractile (ibid., p.118). Gibson (1988, p.5) regarded the system as a series of active neural loops which are "ways of seeking and extracting information about the environment from the flowing array of ambient energy" that seem to function at different levels. From Gibson's list (ibid., p.37) the following are extracted as being most involved in touch perception:
(a) lower proprioceptive systems responsible for posture and equilibrium linked with gravity;
(b) higher proprioceptive systems which are muscular and in which the receptors are probably excited by tension and probably register effort but not movement;
(c) articular, in which the receptors are in the joints and possibly the tendons;
(d) cutaneous in which the receptors are in the skin and perhaps also in most body tissue.
It is remarkable how the articular and nervous systems co-ordinate so perfectly that we are generally unaware of their activities.
Katz, a German psychologist, wrote "Der Aufbau der Tastwelt" (The World of Touch) in 1925, but only parts of it were translated into English. Meanwhile, J.J. Gibson, who agreed with many of Katz' theories, published his own radical ideas about perception in 1962 and 1966 respectively. (The latter was published in Britain 2 year later and references here are to this edition.) Both psychologists emphasised movement being necessary for active touch to take place. In touching, concentration can be of two kinds:
Gibson believed that the information that is detected via neural loops has intrinsic qualities which never vary; for example, a hard surface may always be recognised as such and a ball is always a spherical shape. He realised that there was insufficient evidence to be certain about how these invariances from the external world could get into the nervous system and that much more research was needed. He stated that the sources of stimulation are in the environment, but the actual stimuli are "patterns and transformations of energy at the receptors" (1968, p.28). He compared active awareness to tentacles or feelers and that "the function of the brain when looped into its perceptual organs is not to decode signals, nor to interpret messages, nor to accept images. Instead, the entire neural loop, including the brain, is involved in seeking and extracting information" (ibid., p.5). He went even further from orthodox thinking by suggesting that "perception requires neither memory of past events nor inference from sense impressions". We are left wondering how learning takes place, how it is remembered and how the information is retrieved.
The input from the external world is limitless so there must be some mechanism of selection. Gibson suggested that instead the perceptive system uses attention to explore and select, seeking for clarity, and ignores unwanted or unprocessed information. Perception also involves learning by association (Gibson, p.273). This ability to select and refine improves with use and with growth. Information that is adjacent can be detected within the span of attention, but a problem arises when information is successive beyond this span and therefore involves time. It would seem that the one involves only perceiving and the other both perceiving and remembering. The latter is often referred to as "short-term memory", but according to Gibson "learning does not depend on memory at all, at least not on the re-arousal of traces or remembering of the past". Instead, he thought (ibid., p.262) that "the development of this attunement, or education of attention, depends on past experience but not on the storage of past experiences". In other words, instead of dependence on memory, recurring attention will result in recognition, not only in the detection of finer details, and the span of attention will also be increased (ibid., p.270).
So far the references above have been to perceptual meaning where a stimulus extracts invariant information from the environment leading via resonance in neural loops to a percept of the environment. For learning to be possible there must also be indirect responses to sources produced by thought, and by responses to other people, either in speech or in the written word. This is second hand information about the world. Man has invented symbolic speech which has to be learned before communication can take place. Two stages are necessary, a knowledge of the coded signal and also what it represents. For verbal meaning the item referred to, by social convention, is given a symbol or word and by association this leads to thought.
Gibson's theories seem to suggest that experiences are built on as they occur, so that by repetition they become clearer. He believed that exterioception and proprioception work in conjunction in the same neural loops and referred (ibid., p.284) to "calibration" of the ranges of inputs from different perceptual organs, and that this might be learning of a higher order. The higher order invariants go direct to the brain.
Together with Katz, the main contribution made by Gibson to an understanding of touch perception was his belief that touch can be active as well as passive. For approximately a century before his work became known touch perception was thought of in terms of sensations produced by cutaneous stimulation, and this led to the supposition that there were separate nerve endings in the skin running direct to the brain. His work opened the way for psychologists to progress to the discovery of new facts about touch perception. Some investigations were concerned with the peripheral mechanisms of touch perception such as pressure, vibration and shear force leading to further information on discrimination of roughness (Lederman, 1982), while others specialized in aspects of the central mechanism of the nervous system, such as asymmetry (Hermelin and O'Connor, 1971), memory, convergence, and cross modal function (Millar, 1994). Both parts of the neural system are interdependent, and a greater knowledge of their functions should lead to a better understanding of the processes that underlie behaviour in braille reading. Where appropriate, some of the major braille studies will be included. Many of the latter have been carried out during the past 3 decades and give little or no indication of the underlying psychology of touch perception.
It is an interesting facet of the use of braille that there is a large body of experimental work, known to educational psychologists, some of which was carried out before the work of Katz and Gibson became more generally known (Burklen, 1922, translated into English in 1932; Holland and Eatman, 1933; Holland, 1934; Fertsch, 1946, 1947; Kusajima, 1961). These investigations mainly involved various aspects of braille reading behaviour, basically with the hope that such knowledge would help reading achievement. There was a substantial increase in the number of investigations during the 60's and 70's in America when grants were more readily available, and because the reading of braille is a multimodal activity, this continues to be an area where there is scope for further investigation. There must be some reason or reasons why the braille code has needed several reconsiderations whereas such is not the case for alphabet forms used in visual reading. In general, it is true to say that most of the investigations connected with braille reading until the 70's are geared towards the improvement of reading rate, which is measured in terms of words read per minute (w.p.m.). Part 1 has shown the constant need for revisions of the code to be made in endeavours to make the code easier to read and to use. Certain changes have been necessary from time to time for several reasons, viz. attempts to choose or alter the contracted versions of certain groups of letters in endeavours to help recognition, simplification of the many necessary rules, and an overview from time to time to keep braille usage compatible between countries using the same language.
Burklen, a German psychologist, contemporary with Katz, published a study on touch reading by blind people (1922), but it was not translated into English until 1932. His observations comprised the first detailed research on braille reading behaviour since the less scientific, but nevertheless worthy attempts to study embossed reading, carried out under the auspices of the American Association of Workers for the Blind from 1907 through 1913. Burklen's studies covered such aspects as the characteristics of symbols, pressure, the use of left and right hands and speed of reading. His work was not replicated because the braille reading was tested under rather artificial conditions. For example, he used nail heads for braille dots, and later ones made of tin to save continual replacement of embossed paper because of the dots becoming pressed by constant use. In addition, each student wore a "tastschreiber" on the reading finger which may have caused some discomfiture. The device was bent round the reading finger and extended beyond the embossed material to smoked paper on which finger movements were recorded. Even so, Burklen's work provided the fillip which encouraged further experimental studies on braille usage.
Holland and Eatman (1933) compared the silent reading habits of good and poor readers in school grades 3 through 11. These basic general observations were needed before more detailed examinations of reading habits could be carried out, hopefully leading to some means of improving performance, particularly in the field of rate of reading. Moving pictures were obtained by mounting a camera above the subject's hands to photograph the fingers as they moved along the line of braille, and by means of a projecting device the records were later superimposed upon the material read. Some of the time-consuming complexities of braille reading are indicated by the following list of information obtained: the total number of exposures per line, the average number of braille cells read by the left and right hand independently, the time taken by the subjects at the beginning and end of each line, the number of regressive movements made when facing difficulties of interpretation, and the time taken in making "return sweeps" to find the beginning of the next line. More detailed reference will be made to hand use in the section on asymmetry later in this chapter.
Holland (1934) next investigated the relation of pressure to the rate of reading by good and poor readers using a device for measuring pressure, a timing mechanism and a kymograph which recorded results on smoked paper. The apparatus was not intrusive to the reading situation. The results showed (ibid., p.17) that fast readers tended to use less pressure than slow readers, the amount of pressure varies within a given line, and poor readers showed a tendency to increase the amount of pressure as they read from the beginning to the end of a given paragraph. The cause of these variations by slow readers was regarded as being largely due to difficulties in interpretation of meaning. Holland regarded the study as "an hypothesis rather than an absolute truth" (ibid., p.17) because of the small size of student participation.
Similar general observations of braille reading to those of Burklen were carried out by Kusajima in Japan in 1961. He recorded observations on the movements of the reading finger, and the function of the accompanying finger of the other hand. As in Burklen's experiments, the students wore a tactual recorder on the reading finger. In addition, he compared the differences between visual and touch reading. He replicated his experiments with further detail in 1974.
It is noticeable that because so little was understood about touch perception before 1960, all these early braille experiments observed reading behaviour and the knowledge was important, but there was little psychological explanation given on why such behaviour occurred. In many instances the link still has to be made.
During the 70's investigations were carried out to determine in detail what contribution is made by the neural units in human fingers. In this area the finger-pad is thicker than other cutaneous surfaces of the body (Quilliam, 1978, p.5). Quilliam also gave the following information (ibid., p.12). The skin consists of several cellular layers, and the surface has ridges, well-known because their impressions are the finger prints used for legal purposes. Sweat glands occur along the upper surfaces of the ridges in greater profusion than anywhere else on the body surface, and because the ridges are arranged parallel to each other, the "channelling effect" distributes the sweat evenly over the fingers. Elsewhere on the body shearing force when applied to the skin results in wrinkling, but in the fingers (and toes) the outer and inner layers of the skin are attached, and it is possible that the sweat glands between the layers also help to bind the surfaces together. Fat cells also contribute to make the firm, cushioned surface typical of the finger-pads. These factors combined with the presence of an accumulation of sensory units concerned with temperature, pain, and touch perception, demonstrate the high quality of fingers as sensing mechanisms. The implications for braille reading are obvious.
Practically speaking, sensory units concerned with temperature also play a part in this activity. From comments made by blind children and blind colleagues, it is known that braille dots cannot be successfully sensed when fingers are cold, and likewise, braille reading becomes difficult when fingers become hot and sweaty. A cold surface also impedes reading for it "feels smoother than neutral or warm ones (Lederman, 1982, p.141).
Knibestöl and Vallbo (1970) were the first to demonstrate that there are 4 main types of mechanoreceptors in the glabrous skin of the human hand (Vallbo and Johansson, 1978, p.32). So far the function of these four types of nerve endings is not fully understood, though there have been several tentative suggestions (Vallbo and Johansson, 1978, p.33, p.36; Lederman, 1982, 143-145).
Receptors respond to mechanical energy, on/off units firing bursts of impulses at the beginning and end of excitation. Two types of measurements can be made. Mechanoreceptors can be rapid or slow responding and also vary according to the size of their receptive field. Rapidly adapting units with small receptive fields are appropriate for braille reading, and indeed, the finger pads have been shown to be rich in these particular units (Vallbo and Johansson, 1978, p.44). The second type of measurement registers their sequence of responses resulting from sustained indentation. These investigators commented (p.48) that "particularly striking is the very high density of these two unit types at the finger tips, indicating that this is a skin area with outstanding qualities for tactile spatial analysis". It would seem that together with the versatile movements of the hand as a whole, movements in this part of the body are well adapted to tactile activity.
Because braille consists of raised patterns of dots it is possible to think of the reading finger moving along a line of characters in a continuum of changing textures. Lederman (1982, p.131) wrote, "the perception of texture may be thought of as a microcosm of the entire spectrum of perceptual activities", and (ibid., p.135) "texture perception by touch still remains relatively unexplored to-day".
Roughness is an aspect of texture discrimination and this fact was used by Nolan and Morris (1965) in the realm of braille reading. They published a test which was intended to show the development of the ability of young blind children "to utilize the tactual receptors and hands in a co-ordinated fashion", this being critical for the reading process. The test included comparisons between sandpaper of different grades of grit. No relation was found between ability to discriminate degrees of roughness and chronological age, but was positively associated with level of grade assignment. An important finding was that growth in this ability appeared to level off after Grade 3. The test was used in schools in America to predict likely ability in the use of braille for reading and writing, and was one of the tests used by Nolan and Kederis (1969, p.88) when selecting subjects for testing "the influence of number of dots and position of dots on recognition thresholds for braille words". Sandpaper shows an unregulated mass of texture compared with the more regular positions of dots and spaces which make up braille configurations.
It has already been shown that pressure applied to a surface results in a passive impression, but that movement in active touch is dynamic and includes vibration. Katz was more interested in surface structure rather than its shape (Krueger, 1982, p.41) and during investigations of movement considered that "the vibration sense represents temporal holism. The hand as a unitary organ ... represents spatial holism (ibid., 16-17). That is, both time and space are involved. Put another way, 3 aspects may be recognised, the 2 surfaces involved, finger force, and speed of movement.
Vibrations are set up in the skin when movement takes place. As the finger traverses a line of braille, the skin is squashed upwards towards the deep-seated part of the finger-pad and sideways as a result of lateral movement. Quilliám (1978, p.12) suggested that when this occurs a secondary type of vibration may also be present caused by movement across the ridges and spaces on the cutaneous surface, thus producing multiple exposure to the stimulus (ibid., p.12, 1978, p.12). However, the clarity of perception is blurred to a certain extent by the "oiling" by the sweat glands whose function affects the shearing action between the finger and surface being explored. Katz (Lederman, 1982, p.133) showed that by smearing the sensing finger(s) with collodion perception was improved, and Lederman (ibid., p.138) obtained a similar result when thin paper was placed between the finger and the rough surface. Both methods would neutralise the effect of the layer of sweat. The degree of callus on elderly or work roughened fingers would have the same effect.
Shear force, speed of movement, already studied by Holland (1934), and the effect of temperature may also be involved. Lederman included these aspects when she made a series of systematic studies on roughness discrimination using "aluminium plates with linear gratings of rectangular cross section cut into the surface" (Lederman, 1982, p.136). Concerning the surface of the plates, results showed that the ratio of groove to ridge width does not affect perceived roughness and neither does the fundamental spatial frequency of the stimulus grating. Finger force proved to be the second most influential factor, perceived roughness increasing with increases in force applied perpendicular to the surface; and for hand speed, perceived roughness decreased slightly with increasing speed, but this was negligible relative to groove width and finger force effects. Lederman (ibid., 136-137) argued that if the effect of hand speed was negligible the actual movements of the skin are unimportant and therefore temporal pulse frequency to ratio of skin displacement plays no role.
Lederman herself questioned whether such results could be used in comparison when other types of surface are involved (ibid., p.142). This aspect must surely be taken into account before comparison can be made with the sensing of a braille surface, for a metal surface with slits and ridges will give a very different "feel" from the use of plastic or paper surface covered with domed dots. For example, in the one the finger will slightly penetrate the slits, whereas in the other it will tend to slightly fold over the protuberances.
It has been said here that roughness is a part of the structure of texture. For braille, texture has added meaning, and therefore this aspect will be addressed in the section on central mechanisms.
When the neural inputs from peripheral regions involving touch reach the brain they join the cerebellum which lies under, and towards the back of the brain. Overlying the cerebellum and anterior to it is the mid-brain and overlying that is the cerebral cortex. The latter, consisting of much convoluted soft tissue, and therefore giving a much enlarged total surface, is divided into 2 hemispheres connected by fibres which are together known as the corpus callosum. From the cerebellum the information is linked with a network of intercommunications between different, specialized regions of the brain.
Each of the hemispheres consists of the occipital lobe at the back, the parietal lobe further forward with the temporal lobe beneath, and anterior to these is the frontal lobe. When the neural impulses reach the corpus callosum most of them diverge to opposite sides of the brain, so that information from the left side of the body is controlled by the right hemisphere and the left hemisphere deals with information from the right side of the body. There is considerable specialization in different parts of the brain. Sherrick and Craig (1982, 60-63) showed that the touch sensitivity in monkeys is located in the parietal lobes of both hemispheres, and Millar has stated that in humans touch information is represented in the anterior part of the parietal cortex with spatial coding represented in the posterior part of the parietel lobe (ibid., p.53).
Some of the information concerning the relative functioning of the 2 hemispheres has been gained from observations of malfunctioning due to illness, operations, gunshot wounds and the like. For example, it was found that when the left hemisphere was damaged, speech was sometimes affected, although "some aspects of language are also represented in the right hemisphere". In a similar way it has been found that the right hemisphere is involved in space recognition, yet not exclusively so. Concerning activity in the hemispheres, Millar (1994, p.58) wrote, "The systems do not simply duplicate each other; they are sufficiently specialized to provide additional functions, as well as forming a basis for fail-safe multiple representation".
Before 1971, investigations concerning the best hand or hands for reading braille had proved inconclusive (American Association of Workers for the Blind, 1913, both hands; Burklen, 1932, left hand; Fertsch, 1947,l right hand). Teachers were, therefore, inclined to leave preference to the readers. Hermelin and O'Connor (1971) were the first to apply a psychological explanation for these different findings. They quoted Kimura (1971) who had shown that numbers of dots shown visually were identified better by the left than the right visual field, that is, by the right hemisphere. Conversely, letters were identified better by the right visual field, that is by the left hemisphere (Hermelin and O'Connor, 1971A). As braille reading involves both dots and representations for letters, that is, space as well as language, Hermelin and O'Connor questioned which hemisphere and therefore, which hand, was best for reading punctiform characters (ibid., 1971A).
Fourteen children, aged between 8 and 10 years read separately with the index and centre fingers on the left and right hands respectively. The centre finger reading had been included to obviate the practice factor. Results showed that left-handed reading (right hemisphere - space) was significantly faster than right-handed reading (left hemisphere - language). This fact suggested that "the brain treats input such as braille reading material, as spatially arranged items, to be more efficiently analysed by the right hemisphere before or while verbal coding of the material occurs in the left". A similar experiment was carried out with adults (ibid., 1971B). No difference in speed was found between the use of each hand but fewer mistakes were found in reading with the left middle finger. It was therefore assumed by teachers that pupils who read with the left hand had more advantage than those who used right handed reading. Further experimental work concerning right, left, or both handed reading of braille is explored in Chapter 8.
So far only a general account has been given of the parts of the brain involved in touch perception. Input from the peripheral regions is co-ordinated in the cerebellum and sent on to specific regions, mostly in the cerebral cortex, where it is encoded, and responses when required are sent to the motor output neural systems. The paradigm of separate specialised inputs being the sole means of coding inputs must be rejected as being too simple an explanation. For example, it has been seen how information concerning space and word recognition combine to help determine hand use when reading braille. Major specialised regions of the brain have been recognised (e.g. Longman, 1982, p.656, diagram) and these are linked up by a complicated network of neural pathways; indeed Millar (1994, p.54) has hypothesised that as knowledge of the cross-modal functioning of the brain becomes better known, it is likely that a finer more inter-relating network of pathways will be discovered, and that the information centres will be recognised to be sub-divided into smaller, more specialised units.
It could be thought that for those lacking the sense of sight all the enrichment that comes via the visual pathways is lost. This impression may be enhanced by the fact that many psychologists have used comparisons of those without vision as controls, in order to find out more about visual conditions, thus strengthening the negative side of the deprivation. Millar (1994, p.84) suggests that, being without the information coming via the visual pathways not only causes greater dependence upon the remaining senses, as would be expected, but that extra neural links may, in consequence, be formed between the specialised areas.
Information will usually be selected from several inputs. For example, when moving across a letter U some of the inputs which combine for recognition of this braille configuration, may include the number and position of the dots, the outline shape of the letter and the likelihood of it being U because the previous letter was Q. Sometimes the information from more than one specific area will result in redundancy, making the impression stronger, and "one condition in which correlated information from another source facilitates recognition, is when perception from one of the sources is difficult, or clarity is reduced" (ibid. p.43). Millar (1994, p.15) suggests that this cross-modal functioning can lead to a partial overlap of information, that is, "the sense modalities are sources of specialised, but complementary and convergent information" (ibid., p.15). Since this partial overlapping information comes from differently specialised inputs, theoretically this partial redundancy should lead not only to more accurate input, but even to "improved tactual recognition of otherwise difficult patterns (ibid., p.44).
Warren's opinion (1982, p.123), though writing in general terms, may well apply to braille reading: "... there is much research to be done on the encoding and retention of haptically gained information as with most developmental research, careful attention will have to be paid to the comparability of experimental tasks across age groups. Converging operations, in which the same issue is approached from several experimental paradigms, will be necessary before firm conclusions can be reached".
Contrary to early beliefs, there is no one centre of the brain labelled "memory". J.J. Gibson (1968, p.264) wrote "... a kind of memory in a new sense of the term is definitely required if we are to explain not apprehension over time, but repeated apprehension over time. For the fact is that an observer learns with experience to isolate more subtle invariants during transformation and to establish more exactly the permanent features of an array". Memory is part of the whole process of input from peripheral regions, and encoding (that is organising the information so that it is available for synthesising with other input), makes retrieval possible when required.
Problems arise when attempts are made to discover more about haptic memory, together with speech, and, at a higher level of thought. Warren asked the questions, how is the information encoded, and what is stored, suggesting four possible approaches (Warren, 1982, 118-122):
Warren (ibid., p.122) considered the third option was promising, but had not yet been sufficiently tested to give a good indication of its potential. Even so, Millar (1974) assumed that this programme might give results for blind children's reading. First (1974) she tested subjects aged approximately 10 years using four duplicated sets of three-dimensional nonsense shapes; the interval activity involved unfilled delay, verbal distractor, movement distractor and movement rehearsal respectively. The last condition used finger tracing of the shapes from memory. Both blind and sighted children were included in the samples. It was argued from the results that "tactile short-term memory involves both decay of tactile impressions with time, and interference by additional activities with a longer-term process" (ibid., p.263). Further tests were carried out in 1985 involving braille letters, which are described in the next chapter. Both experimental interference and instructions given in an attempt to influence coding were involved.
So far the mechanisms of perception in both the peripheral and central regions have been described, but what makes the mechanisms work? A car cannot drive itself; intelligence and ability are required by the driver of a car; but these qualities are still not enough. A driver has to learn how to use the mechanism before competence can be established, and so it is with learning how to recognise braille characters and use them in words and sentences. For children, there is the added complication concerning the rate of development at different stages during the learning process. These aspects all interact and need to be understood for the most successful teaching and learning situations.
The use of standardized tests is one way to monitor progress. The Williams Intelligence Test for children with defective vision (1956) has been proved to have a satisfactory overall test/retest correlation over a two-year period (Tobin, 1994, p.40). Gomulicki (1961) investigated the basic learning capacities of blind children, and his tests included observations of manipulative ability and tactile discrimination. In each case (p.22 and p.24 respectively) intelligence played a significant role. In his concluding remarks concerning all the investigations he stated (ibid., p.52) "... the net correlation between intelligence and performance is nevertheless significant at a higher level for the blind than for the sighted". Nolan and Kederis (1969, p.44) considered that "mental ability is a limiting factor" and suggested (ibid., p.48) that "for students whose IQ is below 85 braille is an extremely inefficient medium of communication and that the necessity of mastering it may constitute an additional handicap". Tobin (1971, p.52), when considering some teaching and psychological variables, wrote "the correlational part of the study was aimed at uncovering something of the interlocking structure of organismic and personality variables associated with success in learning braille" and as an integral part of this interlocking, has stated (personal comment) that the correlation between intelligence and braille is higher than the correlation between intelligence and print reading.
Aspects of braille reading can be confidently assessed using standardized tests such as those devised or adapted by Tooze (1962), Lorimer, J. (1962), and Lorimer, J. (1977). These tests are all referred to in more detail in the next chapter. They are all used individually and scores gained may be compared with the norms provided for each age group. Such tests give general markers of progress achieved and some also have diagnostic value. However, for unstandardized but more detailed information on specific aspects, such as hand use and strategies used in character recognition, it is necessary to become familiar with investigations which have been carried out by educational psychologists over the years. A selection of the more important of these is presented and reviewed in Chapter 8.
The American Association of Workers for the Blind made the first studies of how braille reading is carried out (1907-1913), and from then until the present day most investigators have been aware of the necessity for comparing the achievements of different age groups and differing abilities and linking this information with stages of development. It would seem to the writer that more information is needed to show in detail the effects of learning without the major sense of sight; how the individual copes with this deficit; and, more specifically in relation to the use of braille, the effect of age of onset has on individual progress. "It should be obvious, then, that there are some important unanswered questions in the development of haptic perception. The questions are both theoretical and practical. Vigorous focal research is needed to answer them" (Warren, 1982, p.126).