Schlegelmilch Karola, Wertz Annie E
Max Planck Institute for Human Development, Max Planck Research Group Naturalistic Social Cognition.
Infancy. 2019 Jul;24(4):636-662. doi: 10.1111/infa.12294. Epub 2019 May 5.
During infant eye-tracking, fussiness caused by the repetition of calibration stimuli and body movements during testing are frequent constraints on measurement quality. Here, we systematically investigated these constraints with infants and adults using EyeLink 1000 Plus. We compared looking time and dispersion of gaze points elicited by stimuli resembling commonly used calibration animations. The adult group additionally performed body movements during gaze recording that were equivalent to movements infants spontaneously produce during testing. In our results, infants' preference for a particular calibration target did not predict data quality elicited by that stimulus, but targets exhibiting the strongest contrasts in their center or targets with globally distributed complexity resulted in the highest accuracy. Our gaze measures from the adult movement tasks were differentially affected by the type of movement as well as the location where the target appeared on the screen. These heterogeneous effects of movement on measures should be taken into account when planning infant eye-tracking experiments. Additionally, to improve data quality, infants' tolerance for repeated calibrations can be facilitated by alternating between precise calibration targets.
在婴儿眼动追踪过程中，测试期间校准刺激的重复以及身体运动所引起的烦躁是测量质量的常见限制因素。在此，我们使用EyeLink 1000 Plus系统地研究了婴儿和成人的这些限制因素。我们比较了由类似于常用校准动画的刺激所引发的注视时间和注视点分散情况。成年组在注视记录期间还进行了身体运动，这些运动等同于婴儿在测试期间自发产生的运动。在我们的研究结果中，婴儿对特定校准目标的偏好并不能预测该刺激所引发的数据质量，但在中心呈现最强对比度的目标或具有全局分布复杂性的目标会带来最高的准确性。我们从成人运动任务中获得的注视测量结果受到运动类型以及目标在屏幕上出现位置的不同影响。在规划婴儿眼动追踪实验时，应考虑运动对测量结果的这些异质性影响。此外，为了提高数据质量，可以通过在精确校准目标之间交替来提高婴儿对重复校准的耐受性。
