Purchase this article with an account.
Julia Trommershäuser, Laurence T. Maloney, Michael S. Landy; The consistency of bisection judgments in visual grasp space. Journal of Vision 2003;3(11):13. doi: https://doi.org/10.1167/3.11.13.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
We study whether bisection in visual grasp space (the region over which eye and hand can work together to grasp or touch objects) depends on fixation or on the method of judgment employed (the task). We determined observer bias and sensitivity for bisection judgments (in a fronto-parallel plane as well as along contours slanted in depth). Significant biases were found that varied across observers both qualitatively and quantitatively. These biases were stable for a given individual (across a year between data collection intervals) and across tasks (method of adjustment vs. forced-choice). When observers maintained fixation (on an endpoint or in the neighborhood of the bisection point), fixation location had a small but significant effect on bias, although those effects were small compared with bisection uncertainty. We conclude that bisection judgments differ significantly between fixations, but that the effect of fixation location on bisection is not large enough to be detected reliably by the observer moving his or her eyes during a judgment.
Deviations of mean settings from the Euclidean bisection points for the three vertex pairs and each observer. Deviations are shown for the x direction (negative values indicate leftward biases), y direction (negative values indicate backward biases), and z direction (negative values indicate downward biases). Data are reported as mean ±1 SEM (40 data points per vertex pair). The Weber fraction (as a percentage) for the three-dimensional adjustment was computed by dividing the averaged SD (average of the x, y, and z SDs) by the length of the configuration (140 mm) and multiplying by 100.
A sequence of equally spaced points lying on a line connecting the Euclidean bisection point and the mean MOA setting from Experiment 1 was used as stimuli in Experiment 2; spacing between points is given in units of the MOA SD of Experiment 1 (Figure 4).
Results of the statistical analysis of comparing the MOA setting (Experiment 1) with the most preferred point in the 2AFC task (Experiment 2); see text and Figure 5 for details.
Bisection point relative to the Euclidean bisection point (± SD) and slope estimates resulting from the fit of the psychophysical function (see Figure 6 and text for details), displayed for each observer and configuration separately. Just noticeable difference (JND) estimated as the interval between the 25% and 75% level of the fitted psychophysical functions. SD of the MOA setting estimated based on 20 settings per observer and configuration.
Maximum distance Δ between bisection points under different fixations (see also Figure 6). A bootstrap analysis was performed to test whether Δ indicated a significant difference between different fixation points. The p values from this analysis are shown. Significant differences, employing a Bonferroni correction for the nine tests, are indicated by an asterisk.
This PDF is available to Subscribers Only