Holmes Emma, Kitterick Padraig T, Summerfield A Quentin
Department of Psychology, University of York, York, UK.
The Brain and Mind Institute, Natural Sciences Centre, Western University, Room 120, London, Ontario, N6A 5B7, Canada.
Atten Percept Psychophys. 2018 Aug;80(6):1520-1538. doi: 10.3758/s13414-018-1531-x.
Endogenous attention is typically studied by presenting instructive cues in advance of a target stimulus array. For endogenous visual attention, task performance improves as the duration of the cue-target interval increases up to 800 ms. Less is known about how endogenous auditory attention unfolds over time or the mechanisms by which an instructive cue presented in advance of an auditory array improves performance. The current experiment used five cue-target intervals (0, 250, 500, 1,000, and 2,000 ms) to compare four hypotheses for how preparatory attention develops over time in a multi-talker listening task. Young adults were cued to attend to a target talker who spoke in a mixture of three talkers. Visual cues indicated the target talker's spatial location or their gender. Participants directed attention to location and gender simultaneously ("objects") at all cue-target intervals. Participants were consistently faster and more accurate at reporting words spoken by the target talker when the cue-target interval was 2,000 ms than 0 ms. In addition, the latency of correct responses progressively shortened as the duration of the cue-target interval increased from 0 to 2,000 ms. These findings suggest that the mechanisms involved in preparatory auditory attention develop gradually over time, taking at least 2,000 ms to reach optimal configuration, yet providing cumulative improvements in speech intelligibility as the duration of the cue-target interval increases from 0 to 2,000 ms. These results demonstrate an improvement in performance for cue-target intervals longer than those that have been reported previously in the visual or auditory modalities.
内源性注意通常通过在目标刺激阵列之前呈现指导性线索来进行研究。对于内源性视觉注意，随着线索-目标间隔的持续时间增加至800毫秒，任务表现会得到改善。关于内源性听觉注意如何随时间展开，或者在听觉阵列之前呈现的指导性线索提高表现的机制，人们了解得较少。当前的实验使用了五个线索-目标间隔（0、250、500、1000和2000毫秒）来比较关于在多说话者听力任务中准备性注意如何随时间发展的四种假设。年轻成年人被给予线索，以关注在三个说话者的混合声音中讲话的目标说话者。视觉线索表明了目标说话者的空间位置或其性别。在所有线索-目标间隔下，参与者同时将注意力导向位置和性别（“对象”）。当线索-目标间隔为2000毫秒时，参与者在报告目标说话者所说的单词时，始终比间隔为0毫秒时更快且更准确。此外，随着线索-目标间隔的持续时间从0增加到2000毫秒，正确反应的潜伏期逐渐缩短。这些发现表明，准备性听觉注意所涉及的机制会随着时间逐渐发展，至少需要2000毫秒才能达到最佳配置，但随着线索-目标间隔的持续时间从0增加到2000毫秒，语音清晰度会不断提高。这些结果表明，对于线索-目标间隔而言，表现得到了改善，其时长超过了之前在视觉或听觉模态中所报告的时长。
