The ability to distinguish between discrete emotions by monitoring autonomic or facial features has been an elusive "holy grail" for fields such as psychophysiology, affective computing, and human–computer interface design. However, cross-validated models are lacking, and contemporary theory suggests that emotions may lack distinct physiological or facial "signatures." Therefore, in this study, we propose a reorientation toward distinguishing between pleasant and unpleasant affective valence. We focus on the acoustic eyeblink response, which exhibits affective modulation but remains underutilized. The movement of the eyelid was monitored in a contactless manner via infrared reflectance oculography at 1 kHz while 36 participants viewed normatively pleasant, neutral, and unpleasant images, and 50-ms bursts of white noise were presented binaurally via headphones. Startle responses while viewing pleasant images exhibited significantly smaller amplitudes than those while viewing unpleasant images, with a large effect size (d = 1.56). The affective modulation of the eyeblink startle response is a robust phenomenon that can be assessed in a contactless manner. As research continues on whether systems based on psychophysiological or facial features can distinguish between discrete emotions, the eyeblink startle response offers a relatively simple way to distinguish between pleasant and unpleasant affective valence.

通过监测自主神经或面部特征来区分离散情绪的能力一直是心理生理学、情感计算和人机界面设计等领域难以捉摸的“圣杯”。然而，缺乏交叉验证的模型，当代理论表明，情绪可能缺乏明显的生理或面部“特征”。因此，在这项研究中，我们建议重新定位以区分愉快和不愉快的情感效价。我们专注于声学眨眼反应，它表现出情感调制但仍未得到充分利用。通过 1 kHz 的红外反射眼影以非接触方式监测眼睑的运动，同时 36 名参与者观看了通常令人愉快、中性和不愉快的图像，和 50 毫秒的白噪声突发通过耳机双耳呈现。观看令人愉快的图像时的惊吓反应比观看令人不愉快的图像时表现出的幅度明显更小，具有较大的效应量（d = 1.56）。眨眼惊吓反应的情感调制是一种可以以非接触方式进行评估的稳健现象。随着基于心理生理学或面部特征的系统是否可以区分离散情绪的研究继续进行，眨眼惊吓反应提供了一种相对简单的方法来区分愉快和不愉快的情感效价。