Data from: Decoding of baby calls: can adult humans identify the eliciting situation from emotional vocalizations of preverbal infants?

Preverbal infants often vocalize in emotionally loaded situations, yet the communicative potential of these vocalizations is not well understood. The aim of our study was to assess how accurately adult listeners extract information about the eliciting situation from infant preverbal vocalizations. Vocalizations of 19 infants aged 5-10 months were recorded in 3 negative (Pain, Isolation, Demand for Food) and 3 positive (Play, Reunion, After Feeding) situations. The recordings were later rated by 333 adult listeners on the scales of emotional valence and intensity. Subsequently, the listeners assigned the eliciting situations in a forced choice task. Listeners were almost perfectly able to discriminate whether a recording came from a negative or a positive situation. Their discrimination may have been based on perceived valence as they consistently assigned higher valence when listening to positive, and lower valence when listening to negative, recordings. Ability to identify the particular situation within the negative or positive realm was substantially weaker, with only three of the six situations being discriminated above chance. The best discriminated situation, Play, was associated with high perceived intensity. The weak qualitative discrimination of negative situations seemed to be based on graded perception of negative recordings, from the most intense and unpleasant (assigned to Pain) to the least intense and least unpleasant (assigned to Demand for Food). Parenthood and younger age, but not gender of listeners, had weak positive effects on the accuracy of judgments. Our results indicate that adults almost flawlessly distinguish positive and negative infant sounds, but are rather inaccurate regarding identification of the specific needs of the infant and may normally employ other sensory channels to gain this information.

前言语期婴儿（preverbal infants）常在情绪负荷场景中发出发声行为，但此类发声的交际潜力尚未得到充分阐释。本研究旨在探究成年听者能否准确地从婴儿前言语期发声中提取其触发场景相关信息。研究录制了19名5至10月龄婴儿的发声行为，覆盖3类负面场景（疼痛、独处、求食）与3类正面场景（游戏、团聚、喂食后）。后续由333名成年听者基于情绪效价（emotional valence）与强度量表对录制音频进行评分。随后，听者需在强制选择任务（forced choice task）中为每条音频匹配对应的触发场景。听者几乎能够完美区分音频来自负面还是正面场景。此类区分或基于感知到的情绪效价——听者在聆听正面场景的音频时始终给出更高效价评分，聆听负面场景的音频时则始终给出更低效价评分。而在同一效价类别内识别具体场景的能力则显著较弱，六种场景中仅三种的区分准确率高于随机水平。区分表现最佳的场景为游戏场景，其对应音频的感知强度较高。对负面场景的弱质性区分似乎基于对负面音频的梯度感知：从强度最高、最令人不适（被匹配为疼痛场景）到强度最低、最不令人不适（被匹配为求食场景）。听者是否为人父母以及年龄较轻，会对判断准确性产生微弱的正向影响，而听者性别则无显著关联。本研究结果显示，成年听者几乎可以无懈可击地区分婴儿发出的正负向发声，但在识别婴儿的具体需求方面准确率偏低，通常可能会借助其他感官渠道来获取此类信息。