PDF《Eastern Illinois University (USA)》

Psicológica (2005), 26, 161-171.

Pattern Perception and Pictures for the Blind Morton A. Heller* , Melissa McCarthy and Ashley Clark Eastern Illinois University (USA) This article reviews recent research on perception of tangible pictures in sighted and blind people. Haptic picture naming accuracy is dependent upon familiarity and access to semantic memory, just as in visual recognition. Performance is high when haptic picture recognition tasks do not depend upon semantic memory. Viewpoint matters for the ease or difficulty of interpreting haptic pictures of solid objects. Top views were easiest for sighted and blind persons when geometrical solids had constant cross- sections in the vertical axis. The presence or absence of viewpoint effects depends upon the nature of the solids that are represented. Congenitally blind people do not spontaneously produce perspective drawings, but recent data suggests that depictions including linear perspective can be understood after minimal experience. The results suggest that two-dimensional configurations are not necessarily problematic for touch. Touch is a remarkable sense, and it is just beginning to reveal its secrets to persistent researchers. Much more time and energy have been devoted to the study of picture perception in vision than in touch. Until recently, most researchers have questioned whether touch is capable of getting useful spatial information from pictures, and have argued that the sense of touch is far more suitable for the understanding of three-dimensional objects and the surface qualities of objects (Revesz, 1950;

but see Kennedy, 1993). Many blind people around the world have had little formal education in graphics or maps, and most have had no exposure to tangible pictures. In our current society, blind people are increasingly dependent upon the use of computers for communication, as are most of us. Current computer operating systems typically use graphical user interfaces, and so an understanding of spatial perception of blind people and two-dimensional displays take on added importance. Most researchers in this area have assumed that congenitally blind people only differ from the sighted and late blind in their lack of visual experience, and presumably lack of visual imagery. This notion has been used as a rationale for evaluating the impact of a lack of visual imagery on * Acknowledgment: Preparation of this manuscript was undertaken with support from NSF grant C RUI 0317293. Adress: Morton A. Heller. Eastern Illinois University.

1151 Physical Sciences Building.

600 Lincoln Ave. Charleston, Illinois

61920 (USA). E-mail: [email protected] M.A. Heller et al.

162 haptic perception. Lower performance by congenitally blind subjects has been interpreted as an indicator of the impact of a lack of visualization (see Heller, 2000, 2002). Unfortunately, this position is problematic for a number of reasons, and this manuscript will try to clarify some of the issues. Blind people vary greatly in their educational experience and in their perceptual skills. We have tests of spatial skills for sighted persons, since we are aware of their great variability. However, we don't have these norms for visually impaired persons, nor do we even know what normal touch might be. This suggests a great deal of caution in the interpretation of the results of any research with limited samples of blind participants. Picture Naming Studies The results of studies on naming tactile pictures are very mixed. Late blind participants performed at a much higher level than the congenitally blind or blindfolded sighted controls (Heller, 1989). The results of this research were interpreted as showing that late-blind participants had the combined advantage of increased haptic skill and the impact of prior experience with pictures. Congenitally blind people are generally unfamiliar with tangible pictures, and so lower performance could simply reflect a relative inexperience with the rules governing picture perception. There is a further difficulty interpreting the results of experiments involving naming pictures. Failure to name a picture could indicate an inability to perceive the pattern properly, or a failure to imagine the configuration, or a failure in word finding. Thus, we do not assume that a young child can not see a cat, just because the child calls the cat a "doggy." Naming failures do not necessarily tell us much about perception, but may represent failures in accessing semantic memory (Heller et al., 1996). In an attempt to distinguish perceptual failures from problems in word finding and semantics, experiments were conducted to eliminate labeling and the influence of semantic memory from picture perception tasks. Heller et al. (1996) asked subjects to feel a tangible picture, and name it. They were either given the relevant superordinate category or this information was denied. For example, when people felt a picture of a table, they were given the superordinate category name of "furniture." Providing the superordinate category significantly aided picture naming and doubled naming accuracy scores. This indicated the importance of categorical information. One may be able to recognize something without knowing the name of the object. Picture Recognition Without Naming In an attempt to rule out the impact of semantic memory, a task was devised that tested recognition, but did not require subjects to name a picture (Heller et al., 1996). Participants felt three picture choices and had to select a designated target;