Poster #7 - Retention of Self-Derived Information

• In Event: 3-159 - Poster Session 12
  In Poster Session: PS 12 Section - Attention, Learning, Memory

Integrative Statement

How children build a knowledge base is an important question. One way they do this is by combining previously acquire knowledge to self-derive new knowledge. Children as young as 4 combine novel “stem” facts to self-derive “integration” facts (Bauer & San Souci, 2010). Children integrate facts in classroom (Esposito & Bauer, 2017) as well as laboratory (Bauer & San Souci, 2010) environments, and bilingual children integrate information acquired in two different languages (Esposito & Bauer, 2018). Whether newly self-derived facts remain accessible in the long term has yet to be tested. Using data collected from a collaborating school offering Spanish-English dual-language as well as traditional English-only educational programs, we investigated how self-derivation performance was affected by a one-week delay, and whether this varied by educational program.

Students (n = 265) in 1st-2nd grade (ages 6-9 years) were read four pairs of stories, each containing one stem fact to allow for four possible integration facts. The first sessions were conducted in the students’ own classrooms with the class. For students in traditional classrooms, all facts were presented in English. For students in the dual-language classrooms, half of the story pairs were presented in English and half in a cross-language condition: one fact in English and the other in Spanish. After a buffer activity, the students were first asked integration questions, then stem-fact questions in an open-ended format. They then completed forced-choice questions in the same sequence. One week later, participants were asked open-ended questions to probe recall of the integration facts.I If they did not produce a correct response, they were provided with three responses to choose from. We examined children’s open-ended and total performance (open-ended plus unique correct responses in forced-choice).

We conducted mixed-factor ANOVAs with Week (1,2) as within-subjects and educational program (traditional, dual-language) as between-subjects. Open-ended responses were significantly higher in Week 2 (M =1.28, SD = 1.13) than Week 1 (M = 1.11, SD = 1.02), F(1,261) = 4.36, p=.04. Neither the main effect of education program (p=.69) nor the interaction of delay and educational program (p=.76) was significant, meaning dual-language students did not differ from English-only performance at either time point. Total scores showed the same pattern: performance after one week (M = 3.09, SD = 1.01) was greater than performance at Session 1 (M = 2.86, SD = 1.16); the effect was significant: F(1,261) = 15.98, p < .001. Again, there was no main effect of educational program (p=.87) and the interaction of delay and educational program (p=.13) was not significant. Thus, whether measured by open-ended or total performance, there as a significant increase in performance over one week.

Our findings support the conclusion that self-derived information is retained with high accuracy over at least one week, irrespective of the language of fact presentation. Higher performance after the delay could be due to retrieval practice at Session 1 (forced-choice testing provided children a second opportunity to self-derive) or to post-encoding processes that took place over the delay (e.g., sleep-dependent consolidation).

Authors

• James Daly, Emory University
  Presenting Author

• Alena G Esposito, Clark University
  Non-Presenting Author

• Patricia Bauer, Emory University
  Non-Presenting Author