How do human infants perceive events that occur in time? The current project examines this question by studying sequence learning, a critical skill that underlies event perception. The theoretical assumption guiding this project is that as development progresses, infants become capable of learning increasingly more complex sequential relations. The goal of the project is to determine at what age infants become capable of learning particular forms of sequential complexity. To answer this question, the investigator plans to use a habituation/test procedure together with measures of visual attention. Infants are first habituated to sequences consisting either of moving/sounding objects, moving/silent objects, or only sounds arranged in a particular order. Then, they are given test trials during which the order of sequence components is changed and their ability to detect that change is measured. The investigator hopes to determine when specific sequence learning abilities emerge and improve during infancy through these experiments that vary (1) sequence complexity, (2) the particular ordinal relations of sequence components, and (3) the specific nature of sequence components (i.e., whether they are audiovisual, visual, or auditory). The predictions are, as infants grow, that they become capable of learning more complex sequential relations, that they come to depend less on their redundant specification in multiple sensory modalities, and that they are able to generalize their learning across sensory modalities.
Sequence learning is a fundamental skill that is involved in many higher-level skills such as speech, language, reading, event representation, social interaction, and the acquisition of complex motor skills. As a result, this project will provide critical new insights into the underlying developmental foundation for the emergence of the many higher-level behavioral skills that children must master in order to function adaptively. In addition, by carefully characterizing and charting the developmental emergence of sequence learning skills in infancy, the current project offers the possibility of discovering whether the failure to develop age-appropriate behavioral skills, as is the case in many types of developmental learning and communication disorders, may be related to a failure to develop basic sequence learning skills.
This project investigated the development of sequence perception and learning in human infants. Sequences are everywhere and play a critical role in our daily life. Speech, language, everyday events, and our behavioral actions all derive much of their adaptive value and meaning from the way that their constituent elements are organized over time. For example, in language the same words arranged in different sequential orders produce different meanings (e.g., "the boy kissed the girl" vs. "the girl kissed the boy"). In music, the same notes can be arranged differently across time leading to different melodies. Finally, in behavioral action, we must learn to perform the proper sequence of motor behaviors to succeed in nearly everything we do (e.g., opening the door, playing tennis, playing the piano, typing on a keyboard, etc.). Humans must acquire the ability to perceive and understand sequences as early as possible in development in order to function adaptively. For example, if infants are to acquire speech-production capacity and language comprehension skills, they must acquire basic sequence processing skills first so that they can gain access to the perceptual structure underlying the sequences that they must decode and learn. In this project we investigated the development of sequence perception and learning guided by a theoretical model which holds that initial sequence perception skills emerge from infants’ general perceptual sensitivity to temporal pattern information. The model postulates that infants must master four specific types of sequential attributes and that their ability to perceive and learn each of them emerges in turn such that once they learn the lower-level attribute this then facilitates the learning of the next, more complex one. The four attributes are: (a) adjacent (i.e., statistical) relations of sequence elements, (b) ordinal relations, (c) rule-bound (i.e., syntactic) relations, and (d) hierarchical (i.e., recursive) relations. In the current project, we investigated the emergence of the perceptual skills that enable infants to detect each one of these attributes. We did so by allowing infants to first learn sequences of moving and impacting objects and then testing them to determine whether they could detect changes in these sequences based on manipulation of one of the four sequence attributes that infants are assumed to master during the first year of life. Successful detection of these changes was interpreted as indicating that the infants were able to perceive that particular sequential attribute. Overall, we made several important new discoveries in this project. These discoveries are consistent with the general principles of the proposed theoretical model and provide new insights into the development of a fundamental human cognitive skill. Briefly, we have discovered that: Adjacent-relation (i.e., statistical) learning is the earliest to emerge in development – by 4 months of age - and permits infants to begin learning about the most basic properties of audiovisual sequences Ordinal-relation learning emerges later than adjacent-relation learning – by 10 months of age – but even then it still depends on adjacent relations & even at this age infants are unable to generalize their learning of specific ordinal relations to novel sequences Rule learning is difficult even for older infants when they are required to learn relatively long sequences composed of arbitrary audiovisual sequences Learning of hierarchical sequential relations emerges by the end of the first year of life but, like learning of ordinal relations, is restricted to familiar sequences first and only later generalizes to novel sequences Overall, the findings from this project show that the acquisition of the ability to perceive and learn four basic attributes of audiovisual sequences begins early in infancy and that by the end of the first year of life infants already possess some impressive sequence learning skills. At the same time, however, our studies showed that event though the sequence perception abilities that infants acquire during the first year of life, they are still not fully mature, suggesting that they improve as children grow. For example, we found that even older infants have difficulty generalizing their knowledge of the specific sequences that they are able to learn to novel situations. The ability to generalize our knowledge to new situations is a hallmark of mature congitive functioning. Thus, even though by the end of the first year of life infants possess the essential cognitive tools to begin acquiring speech and language and to begin comprehending the meanings behind the events that they witness in their daily lives, these tools require further development during childhood. How they develop after infancy is a fascinating topic that needs further study. In the meantime, however, the results from this project, together with previously published findings on this topic, provide important new and critical insights into the mechanisms underlying the development of various fundamental domains of cognitive functioning.
