The principal objective of this study was to develop a classification system for the elements of student and teacher evaluative dialogue which would account for (1) the content area--task centered versus peripheral, and open versus closed; (2) psychological tone--accepting versus rejecting, and neutral versus judging; (3) possibility for reflective thinking--long versus short, and repetitious versus singular; and (4) logical structure--"entry" versus "sustaining." The secondary objective was to determine what relationship may exist between patterns revealed by the classification system and (1) personality measures, (2) creative strategies, (3) predisposition for learning in art, and (4) learning in art as measured by strategy convergence. Twenty art education students participated in eight studio periods and 6 evaluative sessions, during which sessions, tapes were made which provided the primary source of data. The results of the testing of five hypotheses during the study show: (1) closed teachers use more "entries" than do open teachers; (2) "entries" or new teacher statements relate positively to aesthetic gains in art; (3) teacher open statements relate positively to gains in strategy; (4) there is no significant relationship between strategy of student or the teacher and any measure of verbal behavior used in the study; and (5) gains in aesthetic quality are positively related to the length of student statements and the percent of student talk. Data are presented in 26 tables. (DB)