Saeed Ketabi et al./ Elixir Social Studies 53 (2012) 12082-12085 12082 1. Introduction One of the most important factors determining future achievement in learning an L2 is mastery of new words (Cunningham, 2006). Students learn most of the new words through written texts (Graves, 2006). Hart and Risley (1995) contend that most learners do not know how to make use of texts in order to gain vocabulary knowledge. They argue that there is a need to distinguish strategies which can promote attaining new words. A common strategy in dealing with unfamiliar words is skipping them in context and trying to understand the meaning through local and global context and replacing synonyms for those words (Graves, 2006). But this strategy has been questioned by scholars arguing that new words are not learnt in this way, although it may facilitate text understanding (Stanovich, 1996). A key step in learning new words is grasping phonological information. Words in the form of print (printed graphemes) must be changed to phonemes in a process called phonological recoding and this process needs to be considered in reading and vocabulary achievement (Ziegler & Goswami, 2005). An important technique for successful reading is making visual connections to understand word meanings (Stuart & Coltheart, 1988). The two mechanisms in visual connections entail phonological decoding and visual-orthographic operations. Ehri (1998) contends that for successful mastery of vocabulary in texts, the pronunciation of that word must be figured out in a way that letter-sound relationships are recognized. Syntactic features of words need to be learned from the context and finally these characteristics are to be linked together and be kept in memory. The Self-Teaching Hypothesis proposed by Share (1995) links successful phonological recoding to learning orthographic patterns of words which can lead to proficiency in reading. In this model, when graphemes are changed to phonemes in learners’ mind, visual word identification dexterities are reinforced. Learners come to an oral imagination of the written form through phonological recoding (Share, 1995). It has been mentioned that direct teaching and use of context for guessing meaning are inefficient ways to learn vocabulary since in direct teaching teacher time is limited and unforeseeability of content words deters learning of a considerable number of words (Gough, 1983). The Self- Teaching Hypothesis can play a great role for readers who need an apparatus for acquiring the pronunciation of unknown words (Share, 1995). A study done by Rosenthal and Ehri (2008) corroborates the role of connections of word identities for keeping the word meaning in memory mentioned previously. In this study one group of learners (experimental) were exposed to spelling and pronunciation of a number of words and for the other group (control) only pronunciation was given. The results indicated that spelling facilitated memory for meaning and pronunciation. Cunningham et al. (2002) did a study in which English students read aloud texts containing pseudo-words and found evidence for orthographic learning. This study supported Self-Teaching Hypothesis since phonological recoding proved to be effective in promoting orthographic lexicon (Share, 1995). In another study, Cunningham (2006) provides strong evidence for the role of phonological recoding in the mastery of orthography. In this study, phonological recoding was significantly lowered by making students have concurrent articulation in reading (since it obstructs pronunciation). As expected, orthographic knowledge was diminished in the learners. Also de Jong et al. (2009) came to the conclusion that learners in both silent reading and oral reading improved knowledge of orthography. They argued that because the pseudo-words were monosyllabic phonological decoding had been acquired for both conditions. The current research tries to provide evidence for the Self- Teaching Hypothesis. As mentioned above, this hypothesis Vocabulary learning during reading: A comparison of a word-based strategy versus a text-based strategy Saeed Ketabi 1 , Ehsan Alijanian 2 and Mohammad Ahmadvand 2 1 English Department, Faculty of Foreign Languages, University of Isfahan, Isfahan 81746-7344, Iran. 2 University of Isfahan, Isfahan, Iran. ABSTRACT The present study is an attempt to provide evidence for the Self-Teaching Hypothesis according to which phonological recoding supports learners’ achieving word recognition skills and keeping in mind the connections amongst pronunciation, spelling, and meaning of new words in reading. For this purpose, 46 Iranian EFL learners (2 classes of 12 to 15 years old) were randomly selected. These learners were given 6 to 8 sentence long texts and in each text an unknown vocabulary item was underlined and repeated 2 times. Participants in the experimental condition were asked to read the text silently but the underlined word aloud and those in the control condition were asked to read the text (including the new words) silently. After this experiment learners were tested on text recall, spelling, and meaning of new words. The results show significant gains for the experimental group on spelling and meaning of new words. Pedagogical implications of this finding will be discussed at the end. © 2012 Elixir All rights reserved. ARTICLE INFO Article history: Received: 6 October 2012; Received in revised form: 7 December 2012; Accepted: 17 December 2012; Keywords Self-Teaching Hypothesis, Phonological recoding, Vocabulary knowledge, Text recall. Elixir Social Studies 53 (2012) 12082-12085 Social Studies Available online at www.elixirpublishers.com (Elixir International Journal) Tele: E-mail addresses: ehsanalijanian@gmail.com © 2012 Elixir All rights reserved