| Number | 704 |
|---|
| Year | 1990 |
|---|
| Drawer | 13 |
|---|
| Entry Date | 11/15/1999 |
|---|
| Authors | Lukatela, G., & Turvey, M. |
|---|
| Contact | |
|---|
| Publication | Memory & Cognition, 18 (2), 128-152. |
|---|
| url |
|
|---|
| Abstract | Ten experiments involving visually presented Serbo-Croatian words & pseudowords are reported. The stimuli were used to make up phonemically similar & dissimilar context-target sequences. Experiment 1 (N = 76 young adult native Serbo-Croatian speakers) sought to verify the hypothesis that successive visual presentation of two phonemically similar words affects lexical decision to the second word, independently of graphemic similarity. Experiment 2 (N = 44) compared effects of phonemic similarity on lexical decision where stimuli differed either in initial or middle consonant & differed graphemically in being written in either Roman or Cyrillic script. Experiment 3 (N = 22) examined the degree to which a phonemic similarity effect depends on frequency of the target word. Experiment 4 (N = 20) examined the possible interaction of context familiarity with target familiarity/frequency. Experiment 5 (N = 52) was essentially a replication of experiment 1, examining the phonemic similarity effect for context-target sequences differing in the middle letter. Experiment 6 (N = 44) repeated experiment 5 but additionally manipulated type of phonemic similarity. Experiment 7 (N = 22) examined the contributions of context familiarity & target familiarity to the phonemic similarity effect. Experiment 8 (N = 52) examined stress pattern effects whereas experiment 9 (N = 48) manipulated the stress relationship between context & target. Experiment 10 (N = 44) sought to show that a pseudoword context phonemically similar to its target facilitates lexical decision on the target. Overall findings regarding phonemic similarity effects are discussed in relation to a model in which word-processing units are activated by phoneme-processing units & in which compositionally similar word units inhibit one another based on their relative familiarity. The model fits into
an account in which phonology is computed prelexically & automatically. |
|---|
| Notes | |
|---|
