Individual Differences in Learning Writing Systems

Abstract

Reading is a crucial skill and research on its acquisition has primarily focused on how children “crack the orthographic code” and whether the properties of one’s L1 influence how an L2 writing system is learned. The present study used artificial orthographies to investigate individual differences in learning new writing systems. Seventy-three adults were taught 2 novel orthographies pairing unfamiliar symbols to sounds in two learning conditions: explicit (ჺ-/b/) and implicit (ჺთჲ-/buv/). Results indicated that ability in paired associate learning significantly predicted learning in both conditions and that age inversely predicted ability in the implicit condition. Sex was not a significant predictor in either condition. Interestingly, almost half of participants performed better with explicit training than implicit training and vice versa. These findings indicate that variability in age and learning ability influences adult acquisition and demonstrates the utility of artificial orthographies in understanding reading acquisition.

Full text

Individual Differences in Learning Writing Systems
Kimberly Young1, Ariel M. Cohen-Goldberg1
1Department of Psychology, Tufts University, Medford, MA,
Introduction
•Most studies use composite reading measures to gauge learning and factors involved in
learning new writing systems.
•Composite reading measures don’t tell us about the learning process as it’s happening
and what factors effect learning.
•Across and within studies looking at learning there is a preponderance of heterogeneity.
•Readers can come from various backgrounds of spoken and reading experience,
different learning environments, and age of learning.
•A relatively new task, Artificial Orthography Learning (AOL), can be a helpful tool in looking
at aspects of reading acquisition.
•It can be conducted in a lab and can help control properties of learning such as
experience as well as properties of orthographies being learned.
•Less studies have focused on individual differences in learning new writing systems.
•Some studies have suggested differences in two learning abilities affect reading
acquisition more generally; Paired Associate Learning (PAL) and Statistical Learning (SL).
•Both of these abilities have been shown to be related to reading ability (Arciuli &
Simpson, 2012; Hulme et al., 2007).
•Less research has investigated both the roles of PAL and SL in learning new writing
systems.
•Few have looked at how individual differences in age and gender affect in vivo learning
of new writing systems in adults.
•This study addresses some of these issues by looking at individual differences in SL and PAL
as well as factors such as age and gender as predictors in learning new writing systems
•Learning of writing systems was investigated using an AOL task (Taylor et al., 2017).
Questions? E-mail: kimberly.louis_jean@tufts.edu
Objectives
•Investigate how individual differences predict ability to learn new writing systems
(paired associate learning, statistical learning, age, gender).
•Investigate how individual differences may interact with explicit and implicit learning.
•Investigate the use of artificial orthographies in reading research.
Methods & Materials
Participants
•N= 73; 35 women and 38 men
•Age: 19 - 36 years (M= 28, SD = 4.7)
•Monolingual American English speakers
•Participants recruited online through Prolific
Results & Summary
References
•Arciuli, J., & Simpson, I. C. (2012). Statistical learning is related to reading ability in children and adults. Cognitive science, 36(2), 286-304.
•Hulme, C., Goetz, K., Gooch, D., Adams, J., & Snowling, M. J. (2007). Paired-associate learning, phoneme awareness, and learning to read.
Journal of experimental child psychology, 96(2), 150-166.
•Schmalz, X., Schulte-Körne, G., de Simone, E., & Moll, K. (2021). What Do Artificial Orthography Learning Tasks Actually Measure? Correlations
Within and Across Tasks. Journal of cognition, 4(1).
•Taylor, J. S. H., Davis, M. H., & Rastle, K. (2017). Comparing and validating methods of reading instruction using behavioural and neural
findings in an artificial orthography. Journal of Experimental Psychology: General, 146(6), 826.
•Turk-Browne, N. B., Jungé, J. A., & Scholl, B. J. (2005). The automaticity of visual statistical learning. Journal of Experimental Psychology:
General, 134(4), 552.
Reading and Cognitive Battery
•Digit Span Task (forward and backward)
•Rapid Online Assessment of Reading (ROAR)
•Raven’s Matrixes
Paired Associate Learning (PAL)
•10 stimulus-response shape pairings
•Exposure: See all 10 pairs
•Training: 4 blocks of 2 alternative forced choice
task with feedback
•Testing: 1 block of 2 alternative forced choice task
with no feedback
Statistical Learning (SL)
•24 complex black shapes (8 triplets)
•Exposure: 10 minute stream (triplets appear 24 times each)
•Testing: 32 test items; 2 alternative forced choice task
between triplet and foil
Artificial Orthography Learning (AOL)
•2 novel writing systems
•Letters: Hungarian Runes, Georgian Mkhedruli
•Sounds (English):12 Consonants, 8 Vowels
•1:1 correspondence
•2 learning Conditions
•Explicit: Letter-Sound correspondences explicitly taught
•Implicit: Letter-Sound correspondences taught through whole
word pronunciations
•Procedure
•Familiarization: See symbol/word (24 words), hear
pronunciation
•Training: Pronounce symbol/word (24 words), hear feedback
•Testing: Read aloud 48 nonwords
= /b/
EXPLICIT LEARNING
= /buv/
IMPLICIT LEARNING
Turk-Browne et al. (2005)
Taylor et al. (2017)
Schmalz et al. (2021)
PAL strongly correlated with Explicit and Implicit
learning.
SL showed correlation with Explicit and Implicit learning.
This relationship disappeared when accounting for other
cognitive measures.
Gender showed no relationship in either condition. Age inversely predicted learning in the implicit
condition only.
Paired Associate Learning Statistical Learning
Gender
R= 0.48, p< 0.001 R= 0.40, p< 0.001 R= 0.3, p< 0.05 R= 0.3, p< 0.05
R= - 0.1, p= 0.38 R= 0.006, p= 0.96 R= -0.11, p= 0.34 R= -0.29, p= 0.01
Age
Conclusions
•Paired Associate learning is important for learning, regardless of the type of learning conditions.
•Implicit learning ability may decline as age increases, suggesting possible decline in working memory.
•Artificial orthography learning showed variation in learning in both the explicit and implicit learning
task.
•Indicating AOL can be a helpful tool in understanding different aspects of orthographic learning.
Women Men Women Men