In two experiments, audio re-recordings of texts transcribed from tele
vision newscasts were presented to independent groups, one group in ea
ch case hearing a 'bulletin' with original text structures (as broadca
st), the other with key stories revised in accord with story-grammar n
otions and restoring chronological sequence in the narrative. In Exper
iment 1 the middle item in a three-story bulletin was subjected to tex
t structure manipulation. After hearing the bulletin, subjects were qu
estioned on information in the texts. In Experiment 2, subjects heard
two stories with original or restructured text. In addition to being t
ested for retention they also rated for bias and quality of writing. T
ext restructuring improved learning in both studies, especially of inf
ormation central to the main points of stories. In Experiment 2, such
text revision also affected judgements of bias without altering assess
ment of writing quality. It is suggested that attention to assuring mo
re coherent story structures in newscast texts could counter the probl
ems of poor comprehension and perceived bias on the part of audiences
which result from current production priorities and practices.