Toddler Vocabulary and Object Processing: Data From a new Test of Visual Paired Associate Memory.

• In Event: Poster Session 10
  In Poster Session: PS 10 Topic - Attention, Learning, Memory

Abstract

Learning names for concrete objects clearly relies on visual processes such as object recognition and memory for previously-named objects. This is reinforced by recent research demonstrating that children’s first word productions are best predicted by the number of times an object is seen rather than the number of times a word is produced for the child (Clerkin et al. 2017), and by demonstrations performance on cross-situational word learning tasks is correlated with multiple object memory measures (Vlach and DeBrock, 2017). Furthermore, Collisson, Grela, Spaulding, Rueckl & Magnuson (2015) demonstrated that preschoolers with SLI, but not typically developing children, failed to learn simple visual paired associates and that individual differences in this ability predict the children’s attention to shape when generalizing novel nouns (i.e., the “shape bias”). The goal of the current work is to develop a simple visual paired associate task that can be used with much younger children – toddlers between 15- and 24-months-of-age, to examine the relation between visual paired associate learning and very early vocabulary development. Because children in our desired age-range are too young for the detailed instructions used by Collison et al. (2015) we used a preferential-looking procedure and an eye tracker. A pair of objects is presented side by side on a screen (e.g. a sun and a spoon, Figure 1). A probe object is then shown (e.g. the sun). The test screen shows the object previously paired with the probe (target) and a new item (distractor, e.g. a glass). Looks to the target on this screen were our primary dependent measure. There were four blocks of twelve trials. The first and second block presented common objects that 50% of 18-month-old children can name (Wordbank, Frank et al., 2016). The third and fourth block presented novel objects. The first and third blocks pitted targets against distractors the children had not yet seen in the study (“new” items), while the second and fourth blocks presented previously seen distractors (“old” items). We tested 31 20- to 26-month old children (12 boys), with a mean age of 23 months and a mean productive vocabulary of 270 (range 54 to 414) on the Oxford Communicative Developmental Inventory.
Preliminary findings suggest an unsurprising bias towards novelty such that children tended to look at the “new” distractors (left panel of Figure 2). Critically, on harder trials with old distractors previously seen in the study, children are more likely to look to the target, especially when presented with familiar objects (right panel of Figure 2). Furthermore, children with more words in their productive vocabulary looked more to the target during novel trials with old distractors (p =.00729), suggesting they remember more novel object pairs than children with fewer words in their productive vocabulary.
We are currently developing a scrambled version of the task that eliminates the blocked design and collecting data from younger children. In the future we hope to use this task to examine the relation between visual paired associate memory, vocabulary development and the development of word learning biases.

Authors

• Maria Eleni Bakopoulou, University of east anglia
  Presenting Author

• Eleanor Johns, University of East Anglia
  Non-Presenting Author

• Hollie Harrell
  Non-Presenting Author

• Larissa K Samuelson, University of East Anglia
  Non-Presenting Author