Visual recognition memory and cross-modal transfer skills among monolingual and bilingual 18-month-olds

• In Event: Poster Session 05
  In Poster Session: PS 05 Topic - Attention, Learning, Memory

Abstract

Aim: Marian et al. (2018) proposed that bilingualism affects how one modulates the process of audio-visual integration suggesting that individuals’ linguistic experience plays a role in how they bind sensory information. Our aim is to investigate whether these differences of binding sensory information based on previous language experience extends to a tactile-visual pairing.

Methods: In the current study we tested 49 monolingual and 30 bilingual 18-month-olds in their homes. Infants were categorized as bilingual or monolingual according to parent estimates of language input to their child using a systematic interview (LEAT, DeAnda et al., 2016). Infants were classified as bilinguals if they heard a second language (L2) for at least 20% of the time or monolinguals if their L2 was less than 20%.
We used an Intra-modal and a Cross-Modal Transfer Task (Rose et al., 1978, 2005, 2009) to assess infants’ visual recognition skills with three visual paired-comparison trials. In each of the intra-modal trials, infants were visually familiarized with a geometric piece (familiar target) for 15 seconds and then tested for recognition by pairing the familiar target (FT) with a novel target (NT). To measure their cross-modal transfer skills, we followed the same structure with the only difference being that infants were familiarized with the FT by only touching the pieces (see Figure 2). Recognition memory for both intra-modal and cross-modal transfer is measured by the Novelty Score (the percentage of time spent looking at the NT). Infants were included in the final analysis if they completed 3 intra-modal and 3 cross-modal trials.

Hypotheses: Singh et al. (2015) have demonstrated how bilingual infants are more efficient in recognizing familiar stimuli compared to their monolinguals peers. Therefore, we predict that our bilingual infants will show higher novelty scores than monolinguals will on intra-modal trials. We also expect that infant’s experiences at processing multiple languages and multimodal cues would allow them to more flexibly encode and transfer information between modalities resulting in higher Novelty Scores compared to their monolingual peers. A reliable coding protocol has been developed, and data coding is a priority for both Fall 2020 and Spring 2021 with expected completion by February 2021.

Analysis Approach: To test both intra- and cross-modal recognition, the mean of the novelty score will be tested against a chance value of 50% using two-tailed t-tests.
To compare the differential effects of representational mode on intra-modal and cross-modal performance by Language Group, the novelty percentages will be analyzed using a language group (Monolingual vs. Bilingual) x trial type (Intra-modal vs. Cross-modal) repeated-measures analysis of variance.
To test the predictive effects of L2 exposure as a continuous factor, we will fit a linear model, as our DV variable is continuous. We will include an interaction with trial type in the fixed effects and estimate corresponding random slopes where appropriate. This will be necessary to estimate separate parameters for the Intra-modal and Cross-modal trials (i.e., to model whether the Novelty Scores vary as a function of trial type).
Novelty_Score ~ L2_perc * trial_type + (trial_type|id)

Authors

• Joscelin Rocha Hidalgo, Georgetown University
  Presenting Author

• Madeline Lui, Georgetown University
  Non-Presenting Author

• Mary Feller, Georgetown University
  Non-Presenting Author

• Olivia Blanchfield, Georgetown University
  Non-Presenting Author

• Kate Freda, Georgetown University
  Non-Presenting Author

• Caroline Telesz, Georgetown University
  Non-Presenting Author

• Rachel F Barr, Georgetown University
  Non-Presenting Author