Cite as:
Jordan K, 1998, "Defining the context in line-length illusion" Perception 27 ECVP Abstract Supplement

Defining the context in line-length illusion

K Jordan

In the parallel-lines illusion, a long contextual line (eg 105 mm) induces a robust overestimation of a proximal shorter line (eg 60 mm). We attempted to define the effective contextual stimulus for the induction of this line-length illusion. The variables of interest were the organisation of the contextual stimulus and contextual-stimulus motion. We employed solid and segmented contextual lines as well as contextual lines containing twelve non-collinear segments (experiment 1). The segmented non-collinear contextual lines differed in the organisation of the twelve segments. In experiment 2, we introduced motion of contextual stimuli. For solid and segmented collinear contextual lines, the entire contextual stimulus moved while the judged line was stationary. For the non-collinear segmented contextual stimuli, half of the contextual-stimulus elements moved while half remained stationary. Finally, we included three levels of motion of contextual-line segments (experiment 3).

The data indicate that line-length illusions can be induced by both solid and segmented collinear contextual lines. Additionally, a contextual stimulus composed of two non-collinear segments induced an illusion even when half of the stimulus moved and half was stationary. We conclude that (i) neither a solid contextual line nor collinearity of contextual elements are required to induce length illusion, (ii) the organisation of segments within the contextual stimulus is more important than the location of the endpoints of the stimulus in illusion induction, (iii) neither complete nor partial contextual-line movement diminishes the illusion, and (iv) the inferred organisation of the contextual segments is consistent with an object-based description of attention.

These web-based abstracts are provided for ease of seaching and access, but certain aspects (such as as mathematics) may not appear in their optimum form. For the final published version of this abstract, please see
ECVP 1998 Abstract Supplement (complete) size: 2107 Kb