Temporal Memory Development in Early Childhood: Insights From Novel Episodic and Semantic Memory Tasks

• In Event: 3-099 - Children’s Representation of Time in Memory and Future-Oriented Thought

Integrative Statement

Episodic memory (EM) allows us to remember past events from a particular time and place. Memory for the temporal features of past events is critical for EM, however relatively little is known about its development in early childhood. Even less is known about how the development of temporal memory relates to the development of memory for other contextual features (e.g., space), and how temporal memory for past events may be supported by developments in semantic memory (SM) for temporal information (e.g., conventional time knowledge). Thus, we examined 4- and 6-year-olds’ (N=64) EM and SM for temporal features using various novel tasks.

In the EM task, we tested children’s memory for events and their temporal, spatial, and inter-item associations. During encoding, children listened to several stories told by the experimenter while viewing images on the computer screen. Each story contained information about a child participating in different arbitrary events (Figure 1), each of which contained information that allowed for later testing of temporal order and spatial location (both fine contextual details), and spatial association and temporal association (both coarse contextual details). In addition, each story allowed for later testing of item associations within the event. After a break, in a recognition task, they were asked about old (“Did Mary eat an apple?”) and new events (“Did Mary feed the fish?”). Memory for contextual information was then tested for each event the participant responded “old.” See Figure 2 for sample images and questions.

Children also completed the Children’s Conventional Time (CCT) task, which we developed for young children who have not mastered conventional time scales used by adults. The CCT task consists of two subtasks assessing children’s knowledge of conventional time (‘time knowledge’) and the flexible retrieval of that knowledge (‘flexible retrieval’). The scales used were time of day, meals of the day, and seasons. For the time knowledge subtask, children were presented with shuffled cards that represented time-windows (e.g., morning, afternoon, night) and asked to order them. For the flexible retrieval subtask (adapted from Friedman, 1989), we asked questions that required flexibly moving from one time-window to another. For example, “If you’ve had lunch, what is the next meal going forward in time you’ll eat?” Children completed the Peabody Picture Vocabulary Test as a standardized measure of semantic knowledge.

For the EM task, repeated measures ANOVAs with pairwise comparisons revealed that a) four-year-olds were less accurate than 6-year-olds on all types of contexts, F(1,62)=11.55,p<0.01, b) children were more accurate on item-item associations than item-time or item-space associations, which did not differ from each other, F(2, 62)=4.22,p<0.05, and c) children were more accurate on coarse than fine-grained measures of memory for time and space, F(1,62)=32.06,p<0.001.

There were age-related improvements in CCT time knowledge, t(62)=4.8,p<.001, and CCT flexible retrieval, t(61)=5.9,p<.001. Regression analyses (F(4,58)=16.2,p<.0001,adjusted R2=.50) revealed that CCT flexible retrieval (β=.41,p<.009) and PPVT (β=.40,p<.001), but not age or CCT time knowledge (ps>.3), predicted EM temporal measures accuracy. The implications of this work to our understanding of temporal memory development in early childhood are discussed.

Authors

• Thanujeni Pathman, York University
  Presenting Author

• Margaret Scales, The University of North Carolina at Greensboro
  Non-Presenting Author