Data from: 'Hearing objects move: apparent motion in complex sounds'

How can we perceive objects as moving based on auditory information? This information is often not continuous, such as when objects are temporarily masked by other, louder objects. To perceive these auditory objects as moving along a continuous path, rather than as one object disappearing and another one reappearing, we need to establish correspondence between the correct object instances, resulting in an auditory apparent motion percept of only one object moving across space. However, correctly linking the object instances that belong to the same object is challenging, as auditory input is often ambiguous, such as when several objects follow overlapping trajectories. Previous studies with visual objects have shown that correspondence is influenced by spatiotemporal factors and object features, such as the color, size or shape. Recent studies have shown that correspondence in the auditory modality also depends on spatiotemporal and even feature information, i.e. the frequency of simple sinewave tones. This study aimed to determine whether this finding generalizes to more complex sounds that are meaningful to us and which may involve higher levels of processing. To test this idea, we created an ambiguous apparent motion display biased with sounds from two different musical instruments, a piano and a guitar. Additionally, we manipulated spatiotemporal information. Our results demonstrate that the auditory feature bias, which was previously observed with sinewaves, extends to complex sounds. Furthermore, spatiotemporal factors had a weaker influence on correspondence for complex than for simple sounds. This suggests that auditory correspondence might involve different types of correspondence mechanisms.我们如何基于听觉信息感知物体的运动状态？此类听觉信息往往并非连续的，例如当某一物体暂时被其他音量更大的声源遮蔽时。若要将这些听觉客体感知为沿连续路径运动，而非某一客体消失后另一客体重现，则需在正确的客体实例间建立对应关系，最终形成单一物体在空间中移动的听觉表观运动知觉（auditory apparent motion percept）。然而，正确关联同属一个客体的实例颇具挑战，因为听觉输入往往存在歧义，例如当多个声源沿重叠轨迹运动时。过往针对视觉客体的研究表明，客体对应关系会受到时空因素（spatiotemporal factors）以及客体特征（如颜色、尺寸或形状）的影响。近期研究则证实，听觉模态下的客体对应关系同样依赖于时空乃至特征信息，即简单正弦音调（sinewave tones）的频率。本研究旨在验证上述发现是否可推广至对人类具有认知意义、且可能涉及更高层级加工的复杂声音。为验证这一假设，我们构建了存在歧义的表观运动刺激范式，并辅以两种不同乐器——钢琴（piano）与吉他（guitar）——发出的声音作为偏向线索，同时对时空信息进行了操纵。研究结果显示，此前在正弦音调中观察到的听觉特征偏向效应，可推广至复杂声音场景。此外，相较于简单声音，时空因素对复杂声音场景下的客体对应关系影响更弱。这提示听觉客体的对应关系可能涉及不同类型的对应机制。