Perceprulzl and Motor Skills, 1986, 63,81-82. @ Perceptual and Motor Skills 1986 EXPLICIT SPEECH-SEGMENTATION ABILITY AND SUSCEPTLBILITY TO PHONOLOGICAL SIMILARITY IN SHORT-TERM RETENTION: NO CORRELATION ALAIN CONTENT, JOSB MORAIS, REGINE KOLINSKY, PAUL BERTELSON, AND JESUS ALEGRIA1 Universitk libre de Bruxelles Sammary.-Results of three earlier srudies were reexamined for the presence of correlations between the "rhyme effect" and speech segmentation performance. N o significant correlation was found. Relations between reading achievement and sensitivity to phonological properties of language have been shown in two different lines of research. On one hand, the capacity to analyse speech utterances into submorphemic units such as syllables and phonetic segments is correlated with reading acquisition (Content, 1984, 1985). On the other hand, susceptibility to phonological similarity in short-term retention, as measured by the well-known "rhyme effect" (poorer memory for sets of items with rhyming names) has been re- ported by some (Mann, 1984) to be related to reading ability, but see Hall, Wilson, Humphreys, Tinzmann, and Bowyer (1983) and Morais, Cluytens, Alegria, and Content (1986) for negative results. It has been suggested that effects of phonological similarity and segmentation performance reflect the properties of the same internal representations of speech. Campbell and Butter- worth (1985, p. 469) suggested that the same "phonological input buffer" may support item recall as well as phonemic segmentation. The notion of a common origin predicts consistent correlations between the rhyme effect and speech-segmentation performance. Surprisingly, no data are available in the literature. It happens that the rhyme effect has been measured together with some form of speech-segmentation performance in three studies carried out previously in our laboratory. W e here report the cor- relations which were not examined systematically before. Alegria, Pignot, and Morais ( 1982 ) applied two speech-manipulation tests and a short-term retention test to 64 first-graders, half of them taught by a - - - phonic method, the other half by a whole-word method. The speech-manipula- tion tests consisted in inverting the initial and the final consonants of an utter- ance, the other of inverting initial and final syllables. In the retention test, subjects had to recall the position of target items in a row of eight pictures 'This research has been supported by the Belgian "Ministere de la Politique et de la Programmation scientifique" ("Action de Recherche concertee: Processus cognitifs dans la lecture"). The first and third authors are Research assistants of the National Fund for Scientific Research. Our address: Laboratoire de Psychologie Exgrimentale, Av. A. Buyl, 117, B-1050 Bruxelles. Belgium.