Abstract: With the increase of information in human-computer interfaces,shape encoding method for highlight is proposed to guide the sequence of visual attention effectively,and to improve the efficiency of information search and target recognition.Firstly,basic characteristics of shape are described according to the shape similarity.The difference investigation of every two shapes of different subjects is carried out with the semantic differential method.High discriminating characteristics of shape are found with the correlation analysis and regression equation.Secondly,with the attributes of the three characteristics,the solid,strip and annular shapes are separated as three types.The visual search experiments under different shape encoding combinations are performed.Thirdly,combining with analysis of variance and the least significant difference test,the effects of shape characteristics,the attributes and complexity of characteristics on attentional capture and cognitive performance are analyzed.The results show that,the circular arc shapes capture attention faster than other contour shapes.The topological difference and its levels have important influence on cognitive performance.With the increase of shape complexity,the growing rate of the cognitive performance changes with the shape characteristic.Finally,Take the multimedia unified supervision control system for example,it was testified that the shape combinational designs of different complexity could make for reducing visual clutter and improving cognitive performance.