Donders Institute for Brain, Cognition and Behaviour, Radboud University, Montessorilaan 3, 6500HE Nijmegen, The Netherlands; Department of Otolaryngology, Ohio State University Wexner Medical Center, United States.
Donders Institute for Brain, Cognition and Behaviour, Radboud University, Montessorilaan 3, 6500HE Nijmegen, The Netherlands; School of Psychology, Cardiff University, 70 Park Place, Cardiff CF10 3AT, United Kingdom.
Neuropsychologia. 2019 Mar 18;126:92-101. doi: 10.1016/j.neuropsychologia.2017.05.007. Epub 2017 May 6.
Infants are sensitive to structure and patterns within continuous streams of sensory input. This sensitivity relies on statistical learning, the ability to detect predictable regularities in spatial and temporal sequences. Recent evidence has shown that infants can detect statistical regularities in action sequences they observe, but little is known about the neural process that give rise to this ability. In the current experiment, we combined electroencephalography (EEG) with eye-tracking to identify electrophysiological markers that indicate whether 8-11-month-old infants detect violations to learned regularities in action sequences, and to relate these markers to behavioral measures of anticipation during learning. In a learning phase, infants observed an actor performing a sequence featuring two deterministic pairs embedded within an otherwise random sequence. Thus, the first action of each pair was predictive of what would occur next. One of the pairs caused an action-effect, whereas the second did not. In a subsequent test phase, infants observed another sequence that included deviant pairs, violating the previously observed action pairs. Event-related potential (ERP) responses were analyzed and compared between the deviant and the original action pairs. Findings reveal that infants demonstrated a greater Negative central (Nc) ERP response to the deviant actions for the pair that caused the action-effect, which was consistent with their visual anticipations during the learning phase. Findings are discussed in terms of the neural and behavioral processes underlying perception and learning of structured action sequences.
婴儿对连续的感官输入流中的结构和模式很敏感。这种敏感性依赖于统计学习，即检测空间和时间序列中可预测规则的能力。最近的证据表明，婴儿可以检测到他们观察到的动作序列中的统计规律，但对于产生这种能力的神经过程知之甚少。在当前的实验中，我们结合脑电图（EEG）和眼动追踪来识别电生理标记，这些标记表明 8-11 个月大的婴儿是否能检测到动作序列中学习到的规则的违反，并将这些标记与学习过程中的预期行为测量相关联。在学习阶段，婴儿观察到一个演员执行一个序列，其中嵌入了两个确定性对，而其他的则是随机的。因此，每对的第一个动作可以预测接下来会发生什么。一对会产生动作效应，而另一对则不会。在随后的测试阶段，婴儿观察到另一个包含偏差对的序列，违反了之前观察到的动作对。对事件相关电位（ERP）的反应进行了分析和比较，结果发现，对于引起动作效应的那对偏差动作，婴儿的负中央（Nc）ERP 反应更大，这与他们在学习阶段的视觉预期一致。研究结果从感知和结构化动作序列学习的神经和行为过程的角度进行了讨论。
