TACTILE FEEDBACK APPLIED TO COMPUTER MICE

Most computer systems use a mouse as the input device for menu selecti on, pointing, and manipulation of screen objects. The user can get vis ual information from the screen only about the position of the mouse a nd the action that was initiated. In contrast, during the manipulation of real objects, the visual channel is only responsible for giving br oader information about the action, whereas the motoric action itself is predominantly controlled by tactile information fed by interoceptiv e and exteroceptive sensory signals. Consequently, working with a stan dard computer mouse requires concentration primarily on the visual sys tem, leading to strain reactions. Furthermore, human motor action is s lower when only visual sense is used and tactile sense is ignored. In respect to this situation, it is proposed that a computer mouse be enh anced with additional tactile feedback to approximate more closely rea l object handling, thereby reducing strain reactions. To test this hyp othesis, a standard computer mouse was enhanced with a total of four a ctuators, two lying under the fingers that controlled the mouse sidewa ys and the other two under the mouse buttons. Thus, the approximation of the mouse cursor to an object or the crossing of an object can both be seen and felt, and should approximate real object handling. Three kinds of experiments with tracking and ballistic tasks were carried ou t by 22 subjects. The additional tactile feedback led to significant c hanges of the movement strategies. Whereas the overall performance dec reased if tactile feedback was present during tracking tasks, ballisti c tasks were carried out 9-19% faster. Although the subjects only had a few experiences with tactile feedback, all sequences of ballistic mo vements could be accelerated, in particular the movement preparation p hases. In consequence, a reduction of users' workload is indicated.