Position biases in sequential location selection: Effects of region, choice history, and visibility of previous selections

Abstract

In some situations, e.g., when filling out lottery tickets, it can be advantageous to select random locations. However, people usually have difficulties with this, because they are biased by preferences for certain regions, such as the center of an area. According to ideas from art theory, the preferred regions reflect the hidden structure of perceptual forces within an area. In the present study, these structures were investigated and modeled under different conditions for areas with square and rectangular shape. The general task was to sequentially place a number of dots at random locations in an area by clicking with the computer mouse at corresponding positions on the screen. Whereas in a single-dot condition each dot had to be placed in an empty area, the previously placed dots remained visible in a multiple-dots condition. In three experiments it was found that dots were preferentially placed at the center, the diagonals, and the principal axes. This preference was more pronounced in the single than in the multiple-dot condition. Moreover, sequential analyses revealed that dot placing was not only planned in advance, but that the participants also agreed to some extent in their sequential selections, which produced surprisingly similar sequential spatial patterns across participants, at least for the first dots. Altogether, the results indicate that people have great difficulties with the random selection of locations. Their selections are strongly affected by the attraction of specific regions, by previous selections, and by sequential habits.