Learning to read new words in children with Down syndrome: Testing the role of phonological knowledge Silvana Mengoni 1 , Hannah Nash 2 and Charles Hulme 2 1 The Open University 2 UCL silvana.mengoni@open.ac.uk  Individuals with Down syndrome (DS) often have better word than nonword reading (Roch & Jarrold, 2008). Phonological awareness tends to be poorer relative to reading ability and vocabulary has been found to be a stronger predictor of reading development than in typical development (Hulme et al., 2012).  Vocabulary tasks typically require phonological and semantic knowledge. Item-level phonological, but not semantic, knowledge has been found to predict irregular word reading in typically developing (TD) children, and the resolution of partial decoding attempts is argued to underlie this finding (Nation & Cocksey, 2009).  Pre-training on the phonological aspects of new words helps TD children when they learn to read these words (Duff & Hulme, 2012; McKague et al., 2001).  As individuals with DS have poor decoding skills, they may make a number of partial decoding attempts. If they have pre-existing phonological knowledge of a word, this may help them resolve their partial decoding attempts and read the unfamiliar words successfully. Introduction Individuals with DS were matched to a group of TD children on word reading. They were also matched to a subgroup of these children on decoding. Children were taught to read nonwords, with and without phonological pre-training. It was predicted that individuals with DS would show written nonword learning in line with their decoding skill but below their word reading level. It was predicted that individuals with DS would benefit more from phonological pre-training than TD children matched on word reading but to the same extent as TD children matched on decoding. Aims & Hypotheses Participants 16 individuals with DS aged 8-17 years were matched to 30 TD children aged 5-7 years on word reading and to a subgroup of these children on target decoding (1 st trial of written nonword learning in the control condition) Training programme 12 nonwords were taught (e.g. nirp, roak) in a written nonword learning procedure to each child. Each child was taught to read 6 nonwords that had not been pre-trained (control condition) and 6 nonwords that had been pre-trained. Method  Individuals with DS showed written nonword learning in line with their decoding skill but below that of their word reading skill.  Words that received phonological pre-training were read more successfully. There was a marginally significant effect for individuals with DS to benefit more from phonological pre- training than TD children matched for word reading but there was no such difference compared to the decoding matched subgroup.  The magnitude of the pre-training effect depends on decoding level. Pre-training is argued to support reading by providing individuals with a whole-word phonological representation, which can then provide a pronunciation match for partial decoding attempts.  The phonological aspects of vocabulary knowledge may help reading development. Therefore being presented with the spoken form of a new word before seeing it in print could be a valuable part of reading instruction for children with Down syndrome and other groups with poor decoding skills such as children with dyslexia. Conclusions & Implications Reading matched comparison A mixed design 2x4x2 ANOVA was conducted for the written nonword learning with condition (control vs. phonological pre-training) and trial (1-4) as within-participants factors and group (DS vs. TD) as a between-participants factor. There was a main effect of condition, F(1,44)=59.44, p<.001, a main effect of trial, F(3,132)=25.71, p<.001, and a main effect of group, F(1,44)=4.50, p=.040. The interaction between condition and trial was significant, F(3,132)=6.89, p<.001 as there was a greater improvement in the control condition between trials 1 and 2. The interaction between condition and group was marginally significant, F(1,44)=4.03, p=.051, due to a trend for a greater effect of pre-training for the individuals with DS (d=0.84) compared to the TD children (d=0.54) Decoding matched comparison There was a main effect of condition, F(1,30)=56.84, p.001, and a main effect of trial, F(3,90)=22.06, p<.001. There was no main effect of group. The only significant interaction was between condition and trial, F(3,90)=4.38, p=.006, and this was due to higher scores in the pre-training condition compared to the control condition on the first trial only. Results DS group TD group matched on reading TD group matched on decoding Chronological age 13;08 (2;11) 6;01 (0;06)* 6;02 (0:07)* Nonverbal mental age 5;00 (1;03) 6;00 (1;02)* 6;00 (1;02)* Word reading 26.44(13.88) 21.40 (9.08) 16.31 (7.74)* Nonword reading 6.69 (5.95) 10.20 (5.49) p=.051 6.50 (3.03) Phonological awareness 5.00 (3.16) 8.67 (2.31)* 7.56 (2.31)* * significantly different from DS group Phonological pre-training • 4 trials, each presenting all 6 nonwords • Repetition of the nonword with corrective feedback • Phonological consolidation exercise with corrective feedback e.g. sounding out, phoneme isolation Written nonword learning • 4 trials, each presenting all 6 nonwords • Reading practice with corrective feedback Pre-training condition Control condition Written nonword learning • 4 trials, each presenting all 6 nonwords • Reading practice with corrective feedback