Abstract
Anaphora has been defined as a word or phrase that is substituted for and refers back to a specific antecedent, and which serves as either a cohesive element in language or an element that must be resolved in order to comprehend what has been read. The role of anaphora has been primarily studied in children rather than adults. The purpose of this study was to investigate college students' ability to resolve anaphoric relationships and to empirically test whether anaphora aids or hinders the comprehension of connected discourse. Two passages were taken from Timed Readings (1975) and manipulated to produce two versions of each. One version contained twenty anaphoric substitutes and the other contained only noun referents. Each passage remained identical except for anaphoric manipulation. One hundred and forty-one college students were asked to read both passages, one anaphoric and one referentially specific. Two measures of anaphoric resolution were used; free-recall and probe questions. A third passage, adapted from the Dutka (1978) study was administered. The cloze technique was used as the measure of anaphoric resolution. A 2 x 4 factorial repeated measure design was used with passage type (anaphoric vs referentially-specific) and order of presentation as the independent variables. The dependent variables were the number of facts that could be recalled from the passage, and the number of correct responses to a probe task. Anaphora was found to neither aid nor hinder reading comprehension on a free-recall task but hindered comprehension on a probe question task. Using the anaphoric characteristics of distance (the number of words between an anaphoric item and its referent) and length (the number of words in the referent replaced by the anaphoric item), it was found that both affected performance on the cloze task. Length affected performance on the free-recall but not the probe-question task. Distance did not affect performance on the free recall task but approached significance on the probe question task. The two passages were equivalent in length and number of words, but not equivalent based upon performance. Additional research is needed to assess the conditions under which anaphora aids or hinders reading comprehension.
Milam, Daneen Add (1982). Anaphoric resolution in text comprehension and memory. Texas A&M University. Texas A&M University. Libraries. Available electronically from
https : / /hdl .handle .net /1969 .1 /DISSERTATIONS -515401.