Biomedical Engineering Reference
In-Depth Information
(e.g., rotating to one's left or right) flow patterns are laminar, moving across the retina
in the direction opposite to the motion. Combinations of radial and laminar optic flow
can signify relations among one's: (a) direction of travel (i.e., one's course), (b) head
direction, and (c) gaze direction. In so doing, the online interpretation of optic flow
provides information about the spatial relations in the environment that both are and
are not dependent on one's viewpoint [ 31 ].
A substantial body of research (as well as informal observation) indicates that
despite its availability and precision, visual information is often not necessary for
performing many rudimentary spatial tasks or even for developing detailed knowl-
edge of one's environment. The navigational abilities of congenitally blind people
[ 34 ] attest to this idea. The experimental literature too is rife with demonstrations
that visual information can be surprisingly unnecessary for accurate perception and
memory of space, and researchers frequently conclude that internal senses (discussed
below) are relativelymore influential than vision for the acquisition of online environ-
mental knowledge (e.g., [ 15 ]). On the other hand, it has been known for decades that
in theory, visual information can be sufficient for performing a variety of spatial tasks
[ 30 , 31 ]. For online perceptual tasks such as speed [ 59 ], distance [ 28 , 64 ], and head-
ing [ 58 ] estimation, optic flow alone appears to be sufficient for enabling accurate
spatial knowledge. However, when tasks require online tracking of one's rotational
changes, visual information by itself tends to be much less sufficient. Klatzky et
al. [ 55 ], for example, demonstrated that optic flow alone does not generally enable
people to keep track of their orientation in space and that additional body-based
information is required (see also [ 16 , 76 ]). There is some evidence, however, that
visual information can enable people to account for rotational changes, especially
if it provides information about landmarks, and is available in a sufficiently large
field of view [ 83 , 85 ]. Finally, for memory-based tasks, even very briefly presented
visual information can be sufficient for the acquisition of knowledge of spatial layout,
including relative directions and distances between objects in a single scene [ 118 ].
Indeed, static visual information from photographs [ 4 ], as well as dynamic visual
information from video [ 33 ] or desktop virtual environments [ 86 , 89 ] can be suffi-
cient for acquiring survey knowledge of relatively complex environments, although
such knowledge may be slow to develop [ 111 ] and show especially large differences
among users [ 87 ].
Other external senses such as audition, olfaction, and the somatosenses (i.e., pain,
pressure, vibration, and heat) can provide spatial information for both online and
offline purposes; however, the information from these sources is typically not as rich
as that provided by vision. Among these non-visual external senses, two are espe-
cially useful in generating environmental knowledge. First, audition can be used to
localize noise-producing objects [ 117 ] and to sense the scale of a local environment
[ 90 ]. Indeed, people can learn to echolocate well enough to aid nonvisual navigation
[ 91 , 98 ]. People can also gain accurate knowledge of an environment based solely
on auditory information such as verbal directions [ 21 ], spatial language [ 63 ]orenvi-
ronmental descriptions [ 36 ]. Second, a very common—though rarely investigated—
source of information about the external environment comes from somatosensory
pressure receptors that provide information about: (a) acceleration, through one's
 
Search WWH ::




Custom Search