Repeated Visual Sequence Learning and Transfer to new Surface Features in 7-year-olds and Adults

• In Event: Poster Session 02
  In Poster Session: PS 02 Topic - Attention, Learning, Memory

Abstract

Tracking and using sequential order information in the environment has been considered as a key mechanism that allows children and adults to acquire skills like the grammar of a language. Previous studies on sequence learning over multiple sessions across days or even months have mainly focused on phonological associative learning to model word-form learning. Yet the developmental trajectory of sequential learning with other stimulus materials such as visual stimuli and learning of more complex sequence rules is still largely unknown.

To address this question, we utilized a modified Artificial Grammar Learning (AGL) task with visual stimuli and a complex sequence rule to compare learning over several sessions as well as transfer to a new set of visual stimuli between school-aged children and young adults.

Twenty-eight healthy children (7.05 ± 0.07 years) and 29 healthy adults (23.01 ± 3.49 years), recruited from the local community, completed four sessions of visual sequence learning on a tablet computer, which took place on separate days within one week (see fig. 1). The first three sessions used a first stimulus set, while the fourth session displayed the second stimulus set but with the same AG rule. This experimental protocol allowed us to test learning speed and transfer of learning. To assess visual sequence learning, we implemented a modified AGL task that consisted of 5 blocks with alternating learning and test phases. During learning phases, participants watched grammatical sequences with animal pictures (“circus trains” adapted from Rosas et al., 2010) or color segments (“team flags” adapted from Witt & Vinter, 2012; counterbalanced), whose order was governed by a complex rule system (artificial grammar developed by Reber, 1967). The subsequent test phases, which included trial-by-trial feedback, asked them to select the grammatical sequence out of two displayed sequences in each trial (two-alternative forced choice). We additionally assessed explicit knowledge of the AGL sequence rules at the end of the last session as well as general grammar and memory skills.

The trajectories in sequence learning performance across sessions, measured as percent correct in test phases, were indistinguishable between children and adults, but adults outperformed children in all sessions (see fig. 2). Initial as well as overall learning performance in the last (transfer) session was higher than in the first session in both groups, indicating successful transfer to a new stimulus set. Adults acquired more explicit sequence knowledge than children and only for adults the amount of explicit knowledge correlated with learning success. The latter result suggests different learning mechanisms in children and adults, that is, a shift from more implicit to more explicit rule learning between age 7 years and adulthood. Greater working memory skills were associated with better learning in children.

The present results confirm developmental changes in the mechanisms of visual sequence learning from more implicit to more explicit learning, which might indicate age-related differences in overall learning capacity but not speed.

Authors

• Daniela Schoenberger, University of Hamburg
  Presenting Author

• Maike Fuchs, University of Hamburg
  Non-Presenting Author

• Patrick Bruns, University of Hamburg
  Non-Presenting Author

• Brigitte Roeder, University of Hamburg
  Non-Presenting Author