Many parents have experienced incidents in which their preschool child spontaneously (i.e., without prompting of any kind) recall a previously experienced event. Until recently, such spontaneous memories had only been examined in non-controlled settings (e.g., diary studies). Using a novel experimental paradigm, a previous study has shown that when young children are brought back to a highly distinct setting (same room, same experimenter, same furnishing), in which they previously experienced an interesting event (a Teddy or a Game event), spontaneous memories can be triggered. However, exactly which cues (or combination of cues) are effective for the children's memory, remains unknown. Here, we used this novel paradigm to examine the possible impact of contextual cues at the time of retrieval. We manipulated whether the 35-month-old children returned to the same room (n = 40) or to a different, but similarly furnished, room (n = 40) after one week. The results revealed that although the children returning to a new room produced fewer spontaneous memories than the children returning to the same room, the difference was not significant. Interestingly, despite changing rooms, the children still produced spontaneous memories. Taken together the results may shed new light on the mechanisms underlying childhood amnesia.

The present study investigated the importance of Event Boundaries for 16- and 20-month-olds' (n=80) memory for cartoons. The infants watched one out of two cartoons with ellipses inserted covering the screen for 3s either at Event Boundaries or at Non-Boundaries. After a two-week delay both cartoons (one familiar and one novel) were presented simultaneously without ellipses while eye-tracking the infants. According to recent evidence a familiarity preference was expected. However, following Event Segmentation Theory ellipses at Event Boundaries were expected to cause greater disturbance of the encoding and hence a weaker memory trace evidenced by reduced familiarity preference, relative to ellipses at Non-Boundaries. The results suggest that overall this was the case, documenting the importance of Boundaries for infant memory. Furthermore, planned analyses revealed that whereas the same pattern was found when looking at the 20-month-old infants, no significant difference was found between the two conditions in the youngest age-group.

Following Event Segmentation Theory (EST) adult memory is enhanced at event boundaries (EB). The present study set out to explore this in infancy. Sixty‐eight 21‐month‐olds watched a cartoon with one of two objects (counterbalanced) inserted for 3s either at EB or between EB. Ten minutes later they watched both objects (familiar and novel) in a 10s Visual Paired Comparison (VPC) test while being eye‐tracked. Furthermore, they were asked to point to the previous object. Based on EST, we hypothesized that objects inserted at EB would be processed more fully, resulting in improved memory compared to objects inserted between EB. Only infants with objects at EB exhibited memory evidenced by a transient familiarity preference for the first 3s of the test. Only 18 infants completed the pointing test, but all infants presented with objects at EB (10/10) pointed to the correct (familiar) object, which was not the case for the infants presented with objects between EB (5/8).

An experimental paradigm has shown that it is possible to activate spontaneous memories in children by having them re-visit the setting in which they were introduced to a memorable event.Nevertheless, the most important cues for spontaneous recall remain undetermined. In response, we investigated the importance of the experimenter by introducing 35-month-olds (n = 62) and 46month-olds (n = 62) to the same or a new person after one week. We expected that altering the experimenter would result in fewer recollections through reducing the overlap of cues between encoding and testing. In contrast, the manipulation affected the two age groups differently: No effect of Condition was seen in the 35-month-olds, whereas the 46-month-olds performed better, when the experimenter had changed, suggesting a sensitivity to change and an ability to update their knowledge of the event. We replicated previous findings demonstrating that both age groups exhibited spontaneous recollections.

The visual recognition memory (VRM) paradigm has been widely used to measure memory during infancy and early childhood; it has also been used to study memory in human and nonhuman adults. Typically, participants are familiarized with stimuli that have no special significance to them. Under these conditions, greater attention to the novel stimulus during the test (i.e., novelty preference) is used as the primary index of memory. Here, we took a novel approach to the VRM paradigm and tested 1-, 2-, and 3-year olds using photos of meaningful stimuli that were drawn from the participants' own environment (e.g., photos of their mother, father, siblings, house). We also compared their performance to that of participants of the same age who were tested in an explicit pointing version of the VRM task. Two- and 3-year olds exhibited a strong familiarity preference for some, but not all, of the meaningful stimuli; 1-year olds did not. At no age did participants exhibit the kind of novelty preference that is commonly used to define memory in the VRM task. Furthermore, when compared to pointing, looking measures provided a rough approximation of recognition memory, but in some instances, the looking measure underestimated retention. The use of meaningful stimuli raise important questions about the way in which visual attention is interpreted in the VRM paradigm, and may provide new opportunities to measure memory during infancy and early childhood.