Blind individuals rely on their sense of touch for pattern perception, much as the rest of us depend on vision. If a blind person has extra training in the use of touch for tasks such as Braille or spatial orientation, then we might expect increased skill as a consequence. This is the sensory compensation hypothesis, and there is evidence that practice can aid touch (Sathian, 2000).
Contents |
Most of us have excellent vision. However, if we are fortunate to live long enough, we are likely to suffer some degree of visual impairment, including, perhaps, blindness. A number of eye diseases increase in frequency with aging, notably cataracts, glaucoma and Age related Macular Degeneration (AMD). It is fortunate, however, that new very successful treatments now exist for cataracts. Diseases that inevitably led to blindness in the past, for example, wet AMD, are now treatable. Unfortunately, this is not the case for all eye disorders, and some individuals have to contend with blindness. The age of onset of blindness has implications for how well people use their sense of touch. For example, if people are Congenitally Blind (CB), that is, born without sight or lose it soon after birth, they will not benefit from visual experience or visual imagery. However, they may have an advantage conveyed by their education in mobility skills. People who lose sight later on in life, the Late Blind (LB) or adventitiously blind, retain the influence of visual experience and are likely to retain memory of visual imagery. Low vision may increase reliance on the sense of touch, but sight of the hand and crude large object perception may aid mobility and touch perception.
Several studies supported the sensory compensation hypothesis, suggesting that the superior tactile spatial acuity in the blind may be a form of adaptation (e.g., Sathian, 2000; Van Boven, Hamilton, Kauffman, Keenan, & Pascual-Leone, 2000). Short-term visual deprivation does not improve passive spatial acuity but long-term visual deprivation enhances tactile acuity (Wong, Hackeman, Hurd, & Goldreich, 2011). Blind participants are more accurate with the stationary fingertips than sighted participants, but the two groups show equivalent accuracy with the lips. Braille reading is related to enhanced fingertip acuity, suggesting that tactile experience drives tactile acuity enhancement in blindness (Wong, Gnanakumaran, & Goldreich, 2011).
This research area has been controversial, with some researchers emphasizing difficulties in the production and interpretation of tangible pictures. Many researchers have pointed out the difficulties involved in translating 3D information to a 2D display (Jansson and Holmes, 2003), while others have noted that it may be easy to name visual pictures, but not so easy to name haptic counterparts. Haptics involves the use of active touch to perceive objects and forms. The empirical data are variable, with some studies showing lower performance by the CB (Heller, 1989a; Lederman et al., 1990), while others have found better performance with raised-line pictures if they are larger (Wijntjes et al., 2008). Kennedy (2006) has argued that touch is suited for the understanding of pictures, even when they involve linear perspective. Perspective is a sort of illusory distortion that is found in vision. When we view railroad tracks receding in the distance, it looks as if they converge. Of course they do not. Kennedy has proposed that since perspective involves direction, it should be accessible to CB individuals and the sense of touch (see Heller et al., 2009; Heller & Kennedy, 1990).
Failures to name pictures could derive, at least in part, from lack of familiarity with the rules of depiction in the sense of touch. Thus, most blind people have had little experience with raised-line pictures, and if CB, then they won’t have seen pictures when younger. Experience with drawing, combined with increased tactile skill, probably aided the LB in a number of picture perception studies (e.g., Heller, 1989a). There are reports of excellent performance in a variety of picture perception experiments that required the understanding of depth relations and perspective (Heller et al., 2009). Some experiments indicate that CB individuals do not spontaneously follow the rules of perspective in their drawings, but may come to quickly understand aspects of perspective. Note that failures to name a picture could involve lack of access to categorical information, rather than a perceptual problem. If a young child calls a bus a “train,” that does not mean that the child cannot see the bus. Note that visual and haptic experience can alter whether responses to tangible patterns are global or local in blind individuals.
Maps are useful for blind people, but can be very difficult to use successfully. If a tangible map is too small, there will be problems with resolving fine detail. If a map is too large, there can be difficulties getting an overview of the map. Scale is often a problem for touch, just as in vision. Moreover, if straight ahead on the map does not conform to straight ahead in the world, people can be confused. For example, upward (straight ahead) on a path in the map needs to conform to the direction of locomotion, or people will have difficulty making directional judgments. There is little doubt about the possibility of CB individuals adopting a number of different vantage points, but maps can be difficult to interpret. Of course, individual differences play an important role for blind individuals, just as in the sighted.
The Braille system of embossed dots was developed because of difficulties with embossed print. Print must be much larger than Braille, before it can be comprehensible for touch. The Braille code is a two by three set of coordinate locations where dot patterns correspond to letters in the alphabet. Braille characters are just over 6 mm in length (vertical dimension), with the dots themselves about 1.44 mm in diameter. The spacing between adjacent centers of the dots is 2.34 mm. While there is little doubt about the utility of this reading scheme for blind people, most blind people do not read Braille, and reading speed is slower than for reading print in most sighted individuals. Skilled readers use both hands for reading Braille, with a variety of methods in evidence (see Davidson, Appelle & Haber, 1992). Some blind people use their right index finger to smoothly and rapidly scan lines of text and the left index finger for finding the beginning of the next line. Poor readers tend to make frequent pauses to attempt to identify individual patterns, and may read with a single finger. High-proficiency Braille readers read with both hands and manage to read at a faster rate with their left index finger, scanning almost twice as many Braille cells as low-proficiency readers (Davidson, Appelle & Haber, 1992).
The Braille reading rate using touch is variable and dependent upon a number of factors, including the difficulty of the text (Davidson, Appelle & Haber 1992). According to Foulke (1982) the average Braille reading rate is about 104 words per minute (wpm) for experienced adults, but there are reports of rates as high as 250-300 wpm for excellent Braille readers. The rate of 100 wpm for touch is much slower than visual rates for typical high school students reading print.
A major source of the difficulty in acquiring Braille reading skills derives from the late age of onset of blindness in most instances. Visual impairment is much more common in the aged, and older individuals are much less likely than young children to learn to read with their fingers. Tactile acuity declines with age, along with a number of other functions (Stevens et al., 1996).
It is easier to identify Braille patterns than corresponding letters by touch, if stimulus size is the same. Loomis (1981) demonstrated the superior performance for Braille. He argued that low-pass spatial filtering by the skin is a problem for letters, but Braille characters are far more distinctive.
Illusions occur in touch and in blind people, and are not solely visual. For example, the Mueller-Lyer illusion has been found in CB individuals (Heller et al., 2005). This indicates that explanations of the illusion in terms of size/constancy scaling may not suffice. Thus, one explanation of the illusion has been that the wings-out and wings-in versions of the illusion prompt perceptual differences in depth. If one sees two edges that are judged to be at different distances, but the same size, the “distant” one will be perceived as greater in extent. While this may contribute to the illusion in vision, it is an unlikely explanation for the CB participants. It was proposed that in touch, blind people may have difficulty noting where the straight line ends and the arrows/wings begin. This would lead to overestimation (d) or underestimation of the line (c), depending upon the endings.
The horizontal-vertical illusion (Figure 2) also occurs in CB touch, just as in sight. In the horizontal-vertical illusion, vertical lines are judged as longer than horizontals that are of equal length, when in the form of an inverted T shape. The illusion also occurs with curved shapes, as in the Saint Louis Arch (Heller et al., 2008, 2010, 2013). Heller et al. had blind and sighted participants make judgments about the height and width of curves, and found overestimation of vertical extents. This overestimation also occurs in touch when the height and width are equal, as with the St. Louis Arch. However, it is important to note that while some causal factors are similar in vision and touch, others may be different. For example, the horizontal-vertical illusion is affected by bisection, but it is also influenced by radial/tangential scanning in touch. If one makes radial movements towards the body, these movements are overestimated compared with tangential movements that do not converge on the body (Heller et al., 2010, 2013).
Heller et al. (2013) examined the horizontal-vertical curvature illusion with raised-lines, as well as solid objects. The illusion was found with both types of stimuli, indicating that the haptic illusion is not the result of the use of line drawings. In addition, the illusion was stronger when the curved stimuli were frontally placed, as on a computer screen or wall. These results indicate that the horizontal-curvature illusion is not entirely dependent upon radial-tangential scanning, and can occur in their absence. Radial scanning is not possible with frontal placement.
It is interesting that the Ponzo illusion does not occur in CB individuals or in blindfolded sighted participants. The Ponzo illusion involves making size estimates about two lines that are equal in extent, but appear between converging lines. The converging lines mimic railroad tracks that seem to converge in the distance. The failure to see the illusion in blind people using touch was explained in terms of a failure of the CB to make use of perspective cues. In vision, the higher line is perceived as further away than the second line, so it is judged as larger.
Do blind children perform better than sighted children of the same age in haptic spatial and memory tasks? In a study with 119 participants (59 blind) from 3 to 16 years of age, blind children performed significantly better than age-matched sighted children in a number of haptic tasks. These tasks were involved in different aspects of shape and spatial perception and cognition, including dimensional structure, spatial orientation, symmetry detection in raised-line and in raised surfaces, and dot spans (Ballesteros, Bardisa, Millar, & Reales, 2005). Testing dimensional structure involved a matching-to-sample task to assess whether the child can use different haptic dimensions (shape, size and texture) concomitantly. Spatial orientation measured the ability to recognize the spatial orientation of a shape in tabletop space. Symmetry detection in raised-line and in raised surfaces measured the accuracy of detecting bilateral symmetry (Ballesteros, Manga, & Reales (1997). The 3-D stimuli were constructed by extending the third dimension of figures (Ballesteros & Reales, 2004). Finally, dot span is a short term memory task consisting of a series of items that the child had to repeat correctly in the same order.
Considering older adults, declines in many cognitive domains during the aging process are well documented. Aging negatively affects cognitive processing and brain activity (Park et al., 2001; Park & Reuter-Lorenz, 2009). Declines in tactile perception with age could be related to the loss of tactile receptors, in addition to other factors (Bolton, Winkelmann, & Dyck, 1966; Bruce, 1980; Cauna, 1965). However, declines are not uniform across cognitive functions but follow different patterns of decline and stability across the lifespan (see Ballesteros, Nilsson, & Lemaire, 2009). An ability that does not decline with age is implicit memory, an unconscious memory assessed by showing faster and/or more accurate responses for repeated stimuli compared to new ones (repetition priming effects). Implicit memory for familiar objects that were presented to touch without vision was similar in young adults, older adults and in Alzheimer´s patients (Ballesteros & Reales, 2004; Ballesteros, Reales, Mayas, & Heller, 2008). Not only within-modal haptic priming, but also cross-modal priming (vision to touch and touch to vision)are spared with aging. Young and older adults show similar perceptual facilitation when the stimuli are presented twice in the same tactual modality (e.g., touch) than when the stimuli are presented first to touch and then to vision (Ballesteros et al., 2009).
When blindfolded participants explored raise-line convex curves with one finger and two fingers and judged the sizes of the curves (horizontal/vertical), it was found that the haptic curvature illusion is not only experienced in young adults but across adulthood and even into old age (Ballesteros, Mayas, Reales, & Heller, 2012). Young and older haptic explorers overestimated the vertical but adolescents did not show the haptic illusion. However, when adolescents performed the task visually, they showed a strong horizontal-vertical illusion.
In the elderly blind, extra training in the use of touch acts as a protective factor against the decline in tactile acuity. Until quite recently it was believed that tactile detection and discrimination are similar in blind and sighted older adults (Hollins, 1989). However, the use of modern and more precise psychophysical methods and passive touch have shown that tactile acuity is better in blind participants than in age-matched older adults (Stevens et al., 1996). Stevens et al. (1996) as well as Goldreich and Kanics (2003, 2006) found that, in passive spatial tasks, the acuity of blind individuals declined with age. Tactile acuity was better in the blind at any age but acuity of blind participants declined with age, in parallel with that of sighted participants (but see Heller & Gentaz, 2014.
In a recent study, using tactile-acuity charts that required active exploration Legge et al. (2008) found that sighted subjects showed an age-related decrease in tactile acuity of nearly 1% per year. In contrast, blind individuals did not show an age-related decline using active touch. What can account for the superiority of blind individuals in tasks involving shape and spatial perception and cognition and the enhanced tactile acuity in old age? Enriched tactile experience may explain these differences between blind and sighted people across the lifespan. However, there are reports that with passive spatial tasks, the acuity of blind individuals also declines with age, but acuity remains higher than that for sighted individuals (Goldreich & Kanics, 2003, 2006).
The findings discussed above agree with the sensory compensation hypothesis (Sathian, 2000). Practice can aid touch. However, recruitment of the visual cortex in blind individuals might explain their better tactile and haptic perception (Cohen et al., 1997). However, when sighted individuals wore blindfolds for two hours, this resulted in significant deactivation in intermediate regions in visual shape processing, e.i., V3A and ventral intraparietal sulcus – vIPS. (Weiser et al., 2005). This was not found in controls. These results suggest that short-term blindfolding induces changes in the neural processing of tactile form perception and may reflect short-term neural plasticity. Merabet et al. (2007) investigated the involvement of early visual cortical areas in normally sighted, briefly blindfolded subjects as they tactually explored and rated raised-dot patterns using fMRI. They found that tactile form exploration produced activation in the primary visual cortex (V1) and deactivation of extrastriate cortical regions V2, V3, V3A, and hV4, with greater deactivation in dorsal subregions and higher visual areas. These researchers interpreted the findings as suggesting that tactile processing affects the occipital cortex via a suppressive top-down pathway descending through the visual cortical hierarchy, and an excitatory pathway arising from outside the visual cortical hierarchy that drives area V1 directly.
These findings from transcranial magnetic stimulation (TMS), structural and functional imaging studies suggest that the human brain reacts dynamically in response to visual deprivation. Brain regions usually involved in visual processing are involved in processing inputs from other sensorial modalities (for a review see Noppeney, 2007). The brain clearly can benefit from reorganization as a function of experience, and this cortical plasticity could mediate sensory compensation (Amedi et al., 2005). In addition, there is evidence that there is plasticity within the somatosensory cortex. Tactile experience could alter the organization and sensitivity of these regions of the cortex (Pascual-Leone & Torres, 1993; Sterr et al., 1998).
In passive touch, the individual is immobile and stimulation is imposed on the skin (Gibson, 1966). Gibson argued that passive touch is atypical and prompts subjective sensations; however, many useful sensory aids for blind people make good use of passive touch. Also, people who are both deaf and blind may communicate via Print-On-Palm (POP), where letters and words are printed on the skin of the palm. Performance can be high if the rate of presentation is slow enough to avoid the interfering effects of after-sensations, where people continue to feel a stimulus after it has disappeared (see Heller, 1980, 1989b). Performance in congenitally blind individuals is lowered when patterns are tilted, but this could be a consequence of lack of familiarity and experience with letter and number shapes (Heller, 1989b). Congenitally blind individuals may not be familiar with number patterns, since they are taught Braille from an early age. Circumstances are very different, of course, for people who are both deaf and blind or LB. The deaf-blind are generally taught to read POP and achieve high rates of speed and accuracy.
It is possible to present Braille patterns in a passive mode via a tachistometer, allowing precise control over stimulus timing. Legibility thresholds are less than 100 ms using passive touch with skilled readers of Braille (Foulke, 1982).
Many studies demonstrate advantages in pattern perception in the late blind and the beneficial role of visual experience and imagery, but this is context dependent. If a task involves reading Braille, one might expect that age of onset matters, with much later onset of blindness a negative predictor of skill. If tasks involve situations that control for differential familiarity, one may see comparable or lower performance by CB and sighted individuals compared to the LB, when using their sense of touch. Late blind persons may have the combined benefits of haptic and visual experience. Again, there are large individual differences and the effects of visual experience can be negative, when one is blinded very late in life. Under typical circumstances in the sighted, vision is used to guide touch, but this advantage is lacking in the totally blind, or blindfolded individual. Blind participants tend to be much faster than blindfolded sighted individuals using their touch for pattern perception (Heller and Ballesteros, 2006; Heller & Gentaz, 2014).