Novel Phonological Sequence Learning in Early Typical and Atypical Development - Project Summary/Abstract
Children with developmental language disorder (DLD; aka specific language impairment) are primarily
diagnosed on the basis of grammatical performance. However, preschoolers with DLD often present with a co-
occurring speech deficit, specifically related to difficulty sequencing syllables in novel words (e.g., Benham et
al., 2018). Productions are characterized by variable sounds and syllables associated with the same target,
despite no overt articulatory or speech motor deficits. Currently, we do not have a basis from which to judge
whether this observed speech production variability is a result of a delay in establishing a rule-governed
phonological system, an activity in which young toddlers are immersed, or whether it relates more broadly to a
deficit in sequential learning affecting the motor and phonological organization of sounds and syllables.
Positioned within a lexical, phonological, and articulatory interface, the broad aim of this training plan is to
determine features of typical and atypical speech production of preschoolers with DLD as compared to typically
developing 2- and 4-year-olds. Group differences in variability may reveal disparate contributions of
phonological and articulatory factors to sound learning. We test two hypotheses regarding the nature of speech
production variability in novel word learning. The first hypothesis is that word forms may be variable in children
with DLD due to deficits in the sequencing of sounds and/or movement. If this is supported, children with DLD
will demonstrate qualitative differences in sound and movement organization compared to typical 2- and 4-
year-olds. Alternatively, children with DLD may have a delay in establishing a rule-governed phonological and
a stable articulatory system, related to shifting lexical and phonological representations (Ingram, 1974; Vihman,
1996). In this case, speech production of children with DLD will largely pattern with that of typical 2-year-olds.
To test these competing hypotheses, we will determine the effect of phonological complexity on sound and
movement sequencing in words and nonwords. Both phonological complexity and lexical status are known to
influence production variability. In Aim 1 we assess phonological sequencing using conventional analyses as
well as novel network science methods. In Aim 2 we assess movement sequencing using acoustic and/or
kinematic analyses. In the proposed project and training plan, we predict that the methods used in Aims 1
(phonological sequence variability) and 2 (movement sequence variability) will reveal important and potentially
divergent effects on phonological and articulatory aspects of speech production in typical and atypical learners,
which will provide significant insights into sources of variability in DLD. Critically, the inclusion of phonological
and articulatory parameters in the diagnosis of DLD has the potential to provide a means for early identification
(since phonology emerges before morphosyntax), diagnosis, and treatment for this underserved population.