Memory for Gist: Some Relevant Variables

Using sentences of the form: A is X, B is not X, C is X and D is not X (where X and X are antonyms) a study of gist memory was carried out concerned with evaluating two alternative proposals as to the way in which simple sentences might be remembered, viz. a " kernel " versus a " reading " hypothesis, and seeking to identify two possible variables that might modulate the gist effect, viz. the conceptual nature of the antonymy relation and the phonetic realization of this relation. With some variations in procedure the study was carried out in three separate replications which yielded mutually consistent results supporting the " reading " hypothesis, and indicating that, as expected, the gist effect was greater for contradictories than for contraries and for prefixed items (using the same root with and without negating prefix) than for unprefixed, phonetically distinct items ; the effect of these variables was interactive. The results were considerably clarified when, in addition to an a priori or dictionary specification of item type, procedures were used to provide a judgmental specification. This finding was taken to indicate the value of, and need for, such additional behavioural characterization of linguistic materials.