Direct comparison of the haptic and visual horizontal-vertical illusions using traditional figures and single lines
Date: 2012-06-21 01:30 PM – 03:00 PM
Last modified: 2012-04-27
Abstract
The horizontal-vertical illusion (HVI) has been widely and extensively reported as a visual phenomenon in which a vertical line is perceived as shorter than a horizontal line of the same length. Like a number of geometric illusions, the HVI has also been found to occur the haptically, though there is less agreement in the literature as to the extent and direction of the illusion. The relatively small number of haptic HVI papers coupled with a variety of stimuli and procedures used make it difficult to make direct comparison between the visual and haptic versions of the illusion.
After a brief critical literature review, the current paper reports a study in which the visual and haptic HVIs are directly compared. In a bid to reconcile previous shortcomings, three sets of stimuli were used: L-figures, inverted T-figures, and separated horizontal and vertical lines. The stimuli were presented in two lengths: 3 and 9 cm. The dependent variable was percentage error between the horizontal and vertical – no error represents an absence of illusion.
As expected, inverted T-figures produced an illusion significantly stronger than both the L-figures and single lines, which in turn did not differ from each other. Further, the illusion was present to the same extent in both modalities. Stimuli of 9cm produced relatively stronger illusions than those that measured 3cm, and stimulus size interacted with modality. The consequences of these findings for earlier research and proffered suggestions as to what causes this and other illusions are discussed.
After a brief critical literature review, the current paper reports a study in which the visual and haptic HVIs are directly compared. In a bid to reconcile previous shortcomings, three sets of stimuli were used: L-figures, inverted T-figures, and separated horizontal and vertical lines. The stimuli were presented in two lengths: 3 and 9 cm. The dependent variable was percentage error between the horizontal and vertical – no error represents an absence of illusion.
As expected, inverted T-figures produced an illusion significantly stronger than both the L-figures and single lines, which in turn did not differ from each other. Further, the illusion was present to the same extent in both modalities. Stimuli of 9cm produced relatively stronger illusions than those that measured 3cm, and stimulus size interacted with modality. The consequences of these findings for earlier research and proffered suggestions as to what causes this and other illusions are discussed.