The spatial layout of visual displays was evaluated by means of three magnitude estimation tasks and one reproduction task. Each stimulus display consisted of 16 randomly-drawn, two-digit numbers. Six different spatial layouts, or display types, were used: systematic vs. random spatial order of numbers; regular vs. irregular boundaries of the stimulus pattern; and spatially grouped vs. ungrouped numbers, representing spatial order to different extents. A videotape recorder and television monitor were used for stimulus display presentations, with subjects seated at desks in a shielded room in front of the monitors. Presentation time for each stimulus display was 1.25 seconds. Decreased redundancy of the spatial configuration of elements was reflected in lower estimations and reproduction scores. The tasks referring to visual aspects were quite sensitive to variations in spatial configuration, but even the tasks referring to content attributes were affected in a similar way. The ease of application, the reliability, and the sensitivity of the paper and pencil methods make them suitable for use to complement operational tests of visual displays. Ten references are provided. (Author/LMM)