Lobe size curves for several targets and their effect on visual search time

A series of experiments was conducted to obtain horizontal axis lobe size data for each of the six target characters in a complex background of 10 other characters, then two of these targets on the vertical axis, and finally the same two at five different heights on the horizontal axis. Secondly, a series of search time (ST) experiments was run for the same combinations but with larger background areas. The lobe size experiments gave linear regression lines for the probability of detecting a target versus its distance from the fixation point. But there was not much difference due to shape or size of the targets. From the ST data, a linear relationship was found between log ST and log (target height). The ST data were also used to calculate effective lobe sizes which were used with the hard shells of the initial lobe size data to find the effective diameter and effective probability of detecting a target. From these, the log mean ST was found to have a linear relationship with the log of the effective probability. Effective diameter of the lobe was a better estimate to use for estimating ST than data obtained from the hard shell approach, and the results suggest that factors other than lobe size, such as peripheral vision, information processing, and elements of pattern recognition had a major effect on the ST results. Relevance to industry: This research can help employers in industry to find which inspectors will be better suited to particular jobs, and also enable them to estimate better the time needed to perform such jobs at a good level of performance. (C) 2008 Elsevier B.V. All rights reserved.