抄録
We had already reported that eye tracking movement characteristics are influenced by the attention with which a subject tries to look at a moving target. The target brightness feedback (TBF) method, which feeds back to a subject by using the brightness of a moving target whether he is tracking it precisely or not, was designed in order to raise his attention level. In the TBF condition, it was suggested to the subject that the target would be brighter if he fixated the target, and he was instructed to track the target so that it became as bright as possible. The target was moved sinusoidally (1Hz, ±10deg). Four saccadic and three smooth pursuit parameters were defined. Among them, the phase of smooth pursuit interested us. Although it was behind the target movement originally, it was delayed more by the TBF and the SD decreased simultaneously. This means that the phase delays and becomes stable in accordance with the rise of the attention level. The change ratio of each parameter was calculated by using the differences of both average and SD of the parameter between the non-feedback and the TBF condition by a specific algorithm. The result shows that the characteristics regularly change with the attention level independent of the subjects. We suggest that the attention level change, especially between one data-acquisition and the other acquisitions for each subject, can be estimated by the algorithm. However, the algorithm does not make it possible to estimate the attention fluctuation with time during tracking. So the average of the amplitude of smooth pursuit obtained from the 10 subjects (20 trials) was shown as a function of target period. It decreased gradually with the target period. This is an anticipated result. We suggest that the amplitude of smooth pursuit can express the attention level variation in time.
