The effectiveness of context before, after, and around a missing word
The probability of correctly guessing a missing word was measured using lour different kinds of context: all words before the missing word (forward context), all words after the missing word (backward context), all words before and the one word after the missing word (surround context), and just the one word after the missing word. The probability of correctly guessing a missing word was greater with the forward than with the backward context. The probability of guessing correctly with the surround context was much greater than would be predicted from the independent combination of its forward and one word after components. The results provide evidence that expectations are formed continuously during comprehension, but not in a strict word-by-word order. Implications for information theory are noted. © 1976 Psychonomic Society, Inc.
Duke Scholars
DOI
Publication Date
Publisher
Related Subject Headings
- Experimental Psychology
- 5204 Cognitive and computational psychology
- 5202 Biological psychology
- 1702 Cognitive Sciences
- 1701 Psychology
Citation
DOI
Publication Date
Publisher
Related Subject Headings
- Experimental Psychology
- 5204 Cognitive and computational psychology
- 5202 Biological psychology
- 1702 Cognitive Sciences
- 1701 Psychology