Biomedical Engineering Reference
In-Depth Information
10.5.2 Identification of Spatial Properties
Shape
- can be studied as a whole complex body or by decomposing its constitutive
elements (edges, angles, curves, etc.). Although the haptic discrimination of shape is gen-
erally very much less efficient than visual discrimination (considering speed and number
of errors), the processing modes of the haptic and visual modalities appear similar in
numerous studies. Thus, both modalities are sensitive to the same dimensions of differ-
entiation [74] and to the same effect of complexity, evaluated by the number of sides of
each shape [75, 76]. This similarity has been confirmed by Garbin and Bernstein [77]
and Garbin [78] who, through studies of multi-dimensional scaling, found that the same
shape attributes (size, symmetry, and complexity) determine the subjective similarity of
shapes in vision and haptics. But only 67% of the stimuli occupy the same position on
the visual and haptic scales, in adults and at the age of six to seven years.
However, in other works, differences do appear between the visual and the haptic
processing of shape. Thus, due to the sequential nature of manual exploration and the
possibility of modifying the size of the tactual perceptual field at will, touch is less
sensitive than vision to the Gestalt laws of organization and spatial configuration. For
more information, refer to Chapter 5 which deals with a study of the sensitivity of touch
to the law of proximity [79]. On the other hand, Lakatos and Marks [80] found that
when adults were asked to make either visual or haptic similarity judgments on pairs of
geometrical forms differing in local features or global shape, the pairs with comparable
global shape, albeit with different local features, were judged to be less similar by touch
than by vision. This differential effect tended to decrease over time and was not related
to the haptic exploratory procedures used by the subjects.
In the same vein, the facilitating effect of symmetry, particularly vertical symmetry
which is always present in vision (e.g., [81 - 83]), is not observed in haptic shape per-
ception. Walk [83], and more recently Ballasteros, Manga, and Reales [84], found that
2D non-symmetrical shapes were detected more quickly using haptic methods and with
fewer errors. Conversely, with 3D objects, an advantage appeared for symmetrical shapes
in touch as well as in vision.
To interpret these results, Ballasteros, Millar, and Reales [85] assumed that the ref-
erence frames available in the tasks presented induced a specific spatial organization of
the stimulus and were responsible for the facilitating effect of symmetry. Thus, several
reference frames are available when manipulating 3D objects with both hands, such as
the body's median (Z) axis, the gravitational vertical, and the reference to the position of
the hand itself in relation to the head or trunk. But there are far fewer exterior perceptive
cues available for small, drawn figures, which are less well discriminated than 3D objects
without training. Sensory data in this case can only be related to a body-centered reference
frame [86]. As a result, reference to an egocentric Z axis should be easier when the two
index fingers are placed on either side of this axis in a two-handed exploration than in
a one-handed exploration in which the active index finger is not previously positioned
in the body's median axis. If such is the case, the facilitating effect of symmetry in the
two-handed condition and the absence of this effect in the one-handed condition should
be observed. Ballasteros, Millar, and Reales [85] obtained this result in a task in which
the subject was asked to judge if small, drawn figures (2
×
2 cm), symmetrical or not,
according to the vertical, horizontal, or oblique axes, are open or closed. Fewer errors
and shorter response times were observed when the two index fingers explored rather
