As expected, the children in the RD group had poor
reading fluency and accuracy scores. Although their PPVT
scores were within the normal range, children in the RD
group had significantly poorer scores on this test than did the
children in the CG and CR groups. The group differences in
PPVT vocabulary scores paralleled the group differences in
reading scores. A possible explanation for this could be
what Stanovich (1986) referred to as the ‘Matthew Effect’
whereby ‘the rich get richer and the poor get poorer’. It is
likely that good readers read more and hence have greater
print exposure leading to increased vocabulary.
According to Castle and Coltheart’s model of reading,
learning to read requires development of precise representations
of speech sounds (phonemes) that are then mapped
onto the corresponding letter combinations (graphemes)
(Baldeweg et al., 1999). Because nonwords are unfamiliar,
memory is not helpful in decoding these words. In order to
read nonwords, children need to know the correspondence
between graphemes and phonemes. Fourteen of the RD
children scored poorly on the nonword subtest of the
word/nonword test (Edwards and Hogben, 1999), which is
consistent with these children having a phonological
awareness problem (DeMartino et al., 2001; Heiervang
et al., 2002; Marshall et al., 2001; Snowling, 1981; Tallal,
1980).
On average, children in the CR group could read regular
and irregular words as well as the CG but had significantly
poorer nonword reading scores. Thus, for unfamiliar
nonwords the CR children still had problems with the
rules governing grapheme–phoneme conversions,
suggesting that they have persistent phonological awareness
difficulties. This is consistent with Ingvar et al.’s (2002)
finding that compensated adults had slower reading speed
and different patterns of cerebral activity associated with
reading pseudowords (nonwords), but not with real words,
compared to good readers. Children in the CR group were
assessed only on a word fluency and accuracy task. It is
possible that a more comprehensive reading or language test
battery could have determined any residual language or
reading deficits in the CR group. Further studies investigating
visual perception, memory, sustained attention, and
receptive language could help in determining what enables
children with early reading difficulties to compensate and
become functional readers.