Browse By Day
Browse By Time
Browse By Panel
Browse By Session Type
Browse By Topic Area
Introduction: Reading development depends on multiple cognitive and linguistic (e.g. phonology, vocabulary) abilities. According to the Declarative/Procedural Hypothesis (Ullman, 2013), these abilities may rely on different memory systems. Procedural memory supports the discovery of linguistic structure during learning (e.g. phonology, aspects of syntax) while declarative memory supports aspects of language that involve the arbitrary mapping of form and meaning (e.g. lexicon). Both systems follow distinct maturational trajectories: procedural memory plateaus ~10 years-old (Finn et al., 2016), while declarative memory matures throughout adolescence (DiGiulio et al., 1994). The maturation of procedural and declarative memory systems yield different predictions concerning their roles in reading development. Here, we ask whether declarative and procedural systems differentially contribute to reading as a function of age of first reading exposure? Specifically, we examined reading development in rural Ivory Coast, where age of school enrollment is highly variable due to poverty and the need for child labor. Consequently, children begin reading anywhere between 4-12 years. We hypothesize that procedural memory (matures earlier) contributes more to emergent literacy in younger children, whereas declarative memory (matures later) contributes more to emergent literacy in older children. These hypotheses have largely remained untested -- most reading research involves children who start learning to read at the same time upon mandatory school enrollment.
Methods: 76 children ages 8-14 years (M=10.16, SD=1.51) participated in French language and literacy assessments and declarative and procedural memory tasks. Declarative (Hedenius et al., 2013): Participants encoded real and imaginary images which were tested for recognition (measured d’). Procedural (Nissen & Bullemer, 1987): Serial reaction time (RT) task presenting an encoding phase of random or sequential patterns which were tested for recognition (measured RT for random versus sequential).
Results: Overall word reading scores were low (M=48% correct, SD=37%), but younger children were better readers compared to older children who enrolled in school late (b=-5.21, t(67)=-2.665, p<.01). Procedural, but not declarative, memory predicted reading outcomes (b=-0.12, t(67)=-2.321, p<.05). Importantly, the effect of procedural memory on reading was dependent on child age (b=0.01, t(67)=2.472, p<.05): simple slope analyses revealed procedural memory was a significant predictor of reading for older children only (1SD above mean age: 11.7 years-old; b=0.027, t(67)=2.431, p<.05).
Discussion: Procedural memory continues to be an important predictor of reading for older children. We did not find evidence to indicate declarative systems support reading more among older first-time readers. This suggests that cognitive and linguistic abilities supported by procedural memory remain vital at the earliest stages of literacy development for older learners despite the availability of a more mature declarative memory system. Understanding age-dependent contribution of procedural and declarative systems to literacy provides us with valuable insight into how children of different ages learn to read and informs literacy curriculum and intervention for a significant proportion of the world’s children who enroll in school late.
Joelle Hannon, University of Delaware
Sayako Earle, University of Delaware
Axel Seri, Centre de Recherche et d’action pour la Paix (CERAP)
Fabrice Tanoh, Université Félix Houphouët-Boigny
Hermann Akpe, Réseau Ouest et Centre Africain de Recherche en Education (ROCARE)
Kaja K Jasinska, University of Delaware