Reason: Under embargo until May 2017. After this date a copy can be supplied under Section 51(2) of the Australian Copyright Act 1968 by submitting a document delivery request through your library
A multimodal multidimensional investigation of the horizontal-vertical illusion
thesis
posted on 2017-01-30, 23:12authored byHowell, Jacqui
Current theories that attempt to explain the horizontal-vertical illusion (HVI) are inconsistent with all empirical evidence, largely due to the remaining gaps in our knowledge. This thesis expands our understanding of the HVI by addressing two main research issues – whether the HVI works in a similar manner across the spatial senses, and the effect of embedding single lines into two-dimensional shapes.
The first research issue was addressed by presenting single lines sequentially to visual, haptic and auditory senses allowing a fair comparison by providing each modality the availability of equal information. In Experiment 1 participants used vision, haptics, audition, and a combination of the three senses to estimate lengths presented in the x-, y- and z-axes. No significant difference in length estimations was found between the modality conditions. A strong correlation was observed between the length estimations for each pair of senses. In Experiment 2 auditory lengths were presented using different sound to that used in Experiment 1 (i.e., white noise instead a buzzer sound) and a reverse HVI was found. Despite this, the findings show that length perception operates in a similar manner across the modalities.
The second main research issue was broken into two parts. The first part (Experiments 3 and 4) assessed whether the HVI existed in the creation of squares, and the second (Experiment 5) compared the degree of illusion in traditional HVI figures with that of single lines. In Experiment 3 participants altered the dimensions of overlapping L-shaped cut-outs so the inner segment formed a square. As the sides of the squares were frequently made shorter than the bases (and/or the base longer than the side) it was found that the HVI does influence the creation of squares. In Experiment 4 participants were asked to construct squares using a cardboard cut out. The creation of squares demonstrated the error of the standard with changes to the side resulting in a greater degree of illusion.
In the final experiment, Experiment 5, in which participants used vision and haptics to make length estimates of lines presented separately or as part of an L or inverted-T figure, it was found that the latter produced the greatest degree of illusion where the side was overestimated compared to the base, and the strength of illusion produced from L figures and single lines was relatively similar. All three stimulus types produced a significant illusion in that vertical lines were, on average, judged to be longer than horizontal lines of the same size.
The findings reported here point towards a more appropriate explanation for the illusory distortion. It is likely that the HVI is the result of flawed spatial cognition that impacts all modalities; however it remains unclear why this phenomenon occurs.