Item

Abstract

When sliding a nger across a bumpy surface, the nger follows the geometry of the bumps/holes providing positional cues for the shape. At the same time the nger is opposed by forces related
to the steepness of the bumps/holes. With a specic device Robles-de-la-Torre and Hayward [1]
dissociated positional and force cues in the haptic perception of small-scale bumps and holes:
Participants in this experiment reported to predominantly feel the class of shapes (bumps or
holes) indicated by the force cues. Drewing and Ernst [2] extended this research by disentangling
force and position cues to the perception of curves more systematically and by also
quantifying the perceived curvature. The result was that the perceived curvature could be predicted
from weighted averaging of the two cues. This is consistent with current models of cue
integration [e.g., 3].
These integration models further predict that the cue weight is proportional to the cue's
reliability. Here, we aimed at testing this prediction for the integration of force and position
cues to haptic shape by manipulating the shapes' material properties: high softness can be
assumed to decrease the reliability of the position cue as compared to low softness, and high
friction to decrease the reliability of the force cue. Using the PHANToM force-feedback device
we constructed haptic curve stimuli. We systematically intermixed force and position cues
indicating curvatures of 14 and 24 /m. Using the method of double-staircases, we measured
the point of subjective equality (PSE) of the curvature of these as compared to `natural' stimuli
(i.e., with consistent position and force cues). From the PSE data we determined the cue
weights. This was done under each combination of material properties (low vs high softness X
low vs high friction). We found that material properties affected the cue weights in a manner
consistent with our predictions. These results further conrm the validity of existing models of
cue integration in haptic shape perception.