Measuring icon complexity: An automated analysis

Integrating information from two pictorial animations: Complexity and cognitive prerequisites influence performance

“…By using this strategy, all corresponding elements of the two visualizations are exactly at the same spatial location, thereby avoiding split attention effects. However, on the other hand, the visual complexity of the display, defined as the absolute number of subcomponents of the image, is substantially increased (Donderi, 2006;Forsythe, Sheehy, & Sawey, 2003) thus resulting in high-complex displays. Cognitive load theory identified inherent complexity or element interactivity as one source for intrinsic cognitive load (Ginns, 2006;Sweller & Chandler, 1994).…”

Confounds in pictorial sets: The role of complexity and familiarity in basic-level picture processing

“…The system should be robust and should be able to determine shapes in much the same way as a human observer, it should be stable, and there should be clarity about the ways in which measurement, identification, and description are attained. Although some of the measures reported by Forsythe et al (2003b) failed on these criteria, automated measures based on psychophysical evidence proved more stable.…”

“…Following this line of argument, Forsythe, Sheehy, and Sawey (2003b) have pursued one implication-namely, that a computer-based system capable of processing visual primitives might offer a valid measure of complexity for all 2-D stimuli.…”

“…This difference in processing speed would seem to be a function of image complexity: When an object is of a detailed nature, its global attributes are processed much more quickly than its local ones (Hoeger, 1997;Parker, Lishman, & Hughes, 1997). Forsythe et al (2003b) showed that these basic perceptual components (i.e., edges) are important in the measurement of complexity. Two edge detection techniques were tested: the Canny edge detection algorithm and perimeter detection.…”

“…Simple image properties are extracted from an image in parallel, and these properties are then combined to form objects of a particular shape, color, and size (see, e.g., Treisman, 1986;Treisman & Gelade, 1980;Treisman & Souther, 1985). Forsythe et al (2003b) tested the proposition that the number of discrete objects miithin a picture and the number of holes within those objects can be used as a measure of visual complexity. It was considered that a picture rated as complex would contain more elements than an abstract picture and that these elements would themselves be complex in nature, having more local detail (i.e., holes).…”