Many children with developmental language disorder (DLD) have higher-order cognitive deficits (i.e. executivefunction problem solving and planning) in addition to language impairment. Establishing the relation betweenlanguage and cognition is important for our understanding of the nature of DLD and for developing approachesto improve cognition which is crucial because higher-order cognitive skills predict later academic and careersuccess. One possibility is that children with DLD have cognitive deficits because they are less successfulrelative to peers with typical language in utilizing self-directed speech (SDS; i.e. speech to oneself) to regulatetheir behaviors and thoughts. For example children with DLD may be less likely to articulate steps whileproblem solving or verbally rehearse task rules. As such language impairment may play a causal role inhigher-order cognitive deficits in DLD via impaired SDS. Sociocultural models suggest that SDS begins asexternal regulatory speech from caregivers that children internalize to develop SDS. In this model expressivelanguage impairment would negatively affect a childs ability to produce SDS and receptive languageimpairments would reduce a childs ability to understand caregivers external regulatory speech which iscentral to SDS development. Few studies have examined SDS in DLD but previous research indicates thatchildren with DLD have immature SDS relative to typically developing peers. However there is also promisingevidence that SDS can be trained in children with DLD. Although children with DLD show immature SDSrelative to typical peers longitudinal research in this area is lacking making it difficult to determine thedevelopmental trajectory and discern why children with DLD lag behind typical peers. The first goal of thisresearch is to document the longitudinal development of SDS in preschoolers with and without DLD in order tocompare the groups in their SDS and planning abilities and to assess the relation between these variables overtime. This goal will be achieved in Study 1 in which preschoolers with and without DLD will complete the Towerof Hanoi planning task four times over one year. The second goal is to elucidate the effects of SDS on planningperformance in preschoolers with and without DLD. This will be addressed in Study 2 in which preschoolerswith DLD and typical language will complete Tower of Hanoi four times in different experimental conditions thatmanipulate (i.e. prevent or encourage) SDS production. If task performance differs across the conditions itsuggests that planning is affected by increasing or decreasing SDS supporting a causal relationship betweenSDS and cognitive performance. The third goal is to test the predictions of the sociocultural model with respectto the effects of expressive versus receptive-expressive language deficits on SDS. This goal will be addressedwith data from children with DLD from Studies 1 and 2 by comparing SDS development between children withexpressive language impairment versus those with receptive-expressive impairment. Sociocultural models ofSDS would predict an asymmetry with larger SDS deficits associated with receptive-expressive impairment.