The visual system separates every scene into a figure in front of a background

Figure/ground (or figure/background) is the Gestalt principle that the visual system automatically assigns every visible region either to a ‘figure’ (the object, foreground, thing attended to) or to ‘ground’ (the background surface behind it). The figure appears in front, is more ‘thing-like’, and attracts attention first; the ground extends behind and feels continuous. We prefer stability — in unambiguous images we see the figure first and immediately. When intentionally destabilised (Rubin’s Vase) figure and ground flip (triggering multistability). In interface design: high-contrast text is figure on a ground background; modal dialogs dim the ground to force figure status on the dialog. Misapplication: when foreground and background have insufficient contrast, users cannot identify which is figure, increasing cognitive load.

Examples

Light text on a dark background (or dark text on light): text is figure. Switching to dark mode reverses the colours but not the figure/ground assignment — the text is still figure.

Assessment

Take a monochrome photograph and identify five distinct figure/ground boundaries. Then design a UI card in which figure/ground is deliberately ambiguous and explain the usability consequence.