Measurement Scale and Zero Point Indicators Affect Object Display Separate and Integral Task Performance

Abstract

Object displays, compared to bar displays, usually show better performance on tasks that require the operator to integrate the information presented. The cost of this benefit is that performance on separate tasks, where the operator needs to extract individual pieces of information, is typically not as good with an object display as with a bar display. One potential explanation for the decrement in performance seen when using the object display for separate tasks is the lack of a frame of reference for the separate tasks. In this study the frame of reference was manipulated by providing varying degrees of units of measure (full, partial and no tick marks) and presence or absence of a zero point. The results indicate that, if a frame of reference is present reaction time (RT) on the separate task is improved, but at the expense of the integration task. Root-mean-square (RMS) error is reduced if a frame of reference is present. A speed-accuracy trade-off was observed for the integration task with unit of measurement present. Implications for the design of object displays are discussed in light of this trade-off.