Subjects were asked to read sentences containing local structural ambiguity. The material was presented word-by-word, under subject control, in three different display modes. In the Cumulative condition words previously read remained visible. In the Sequential condition only one word was visible at a time but was presented in its appropriate spatial location. In the Central condition all words appeared in the same physical location. The results suggest that subjects may be less sensitive to ambiguity if prior text is not available for inspection, and that removing information on differential spatial location may influence comprehension.