When further learning fails: Stability and change following repeated presentation of text

Kay (1955) presented a text passage to participants on a weekly basis and found that most errors and omissions in recall persisted despite repeated re-presentation of the text. Experiment I replicated and extended Kay's original research, demonstrating that after a first recall attempt there was very little evidence of further learning, whether measured in terms of further acquisition or error correction, over three more presentations of the text passages. Varying the schedule of presentations and tests had little effect, although performance was better when intermediate trials included both presentation and test than when only presentations or tests occurred. Experiment 2 explored whether this 'failure of further learning' effect could be overcome by (a) warning participants against basing their recall on their previous recall efforts and specifically directing them to base their recall upon the passages, (b) making each presentation more distinctive, or (c) drawing participants' attention to areas that would benefit from further learning by requiring them to tally their omissions and errors. The effect persisted in all cases. The findings have serious implications for the learning of text material.