banner image
No data available.
Please log in to see this content.
You have no subscription access to this content.
No metrics data to plot.
The attempt to load metrics for this article has failed.
The attempt to plot a graph for these metrics has failed.
Identification of walked-upon materials in auditory, kinesthetic, haptic, and audio-haptic conditionsa)
a)Portions of this research were presented at Acoustics’08 Paris ( Giordano et al., 2008 ).
Rent this article for


Image of FIG. 1.
FIG. 1.

Apparatus used for the presentation of the ground materials.

Image of FIG. 2.
FIG. 2.

Vertical and lateral acceleration spectra at the center of the toe for the vibrotactile noise (thick black line), and for the wood and small gravel grounds (thin gray and black lines, respectively; sampling frequency = 2 kHz; fast Fourier transform window size = 256 samples). Measurement equipment: PCB (Depew, NY) 352C42 accelerometer connected to a Model 480E09 PCB signal conditioner itself connected to a National Instruments (Austin, TX) USB-6218 board for analog-to-digital conversion.

Image of FIG. 3.
FIG. 3.

Average identification performance for each material category and experimental condition. Error bar: ±1 SEM.

Image of FIG. 4.
FIG. 4.

GRT models of the identification confusions for one of the participants in each of the experimental conditions. A bivariate normal distribution models the across-trials fluctuations in the sensory effects of a ground material (symbols = means; ovals = 0.05 equal-likelihood contours). Lines model the decision boundaries set by participants to carry out the identification task (solid lines = fixed boundaries; dashed lines = boundaries fitted to experimental data). Sensory effects that fall within the same region of the two-dimensional decision space receive the same identification response. Materials are labeled as in Table I.

Image of FIG. 5.
FIG. 5.

Average condition-specific GRT similarity between materials in the same category. Similarity = 0 and 1 for perfectly accurate and inaccurate discrimination, respectively. Error bar: ± 1 SEM.

Image of FIG. 6.
FIG. 6.

Metric MDS models of the ρc measures of between-condition similarity for each data type and material category (lines connect significantly similar conditions, p < 0.05). Conditions that lie closer within the MDS configuration yield highly similar data (see scale in figure). A Procrustes rotation was used to align the MDS models for identifications and discriminations for the same category (scale factor for rotation = 0). The MDS distances between different data types and categories are meaningless.

Image of FIG. 7.
FIG. 7.

Similarity (average ρc ) of identification confusions and GRT discriminabilities for solid and aggregate materials. Error bar: ±1 SEM.


Generic image for table

Matrices of identification confusions in the population of experimental participants (rows: stimuli; columns: responses). MA = marble; CE = ceramic; WO = wood; VI = vinyl; VG = very small gravel; SG = small gravel; MG = medium gravel; LG = large gravel. Number of repeated presentations of each material in each experimental condition = 120.


Article metrics loading...


Full text loading...

This is a required field
Please enter a valid email address
752b84549af89a08dbdd7fdb8b9568b5 journal.articlezxybnytfddd
Scitation: Identification of walked-upon materials in auditory, kinesthetic, haptic, and audio-haptic conditionsa)