Observers watched for 1 or 2 colored words as targets presented in lists of distractor strings (10 items/s). Identification of 1 target (T1) temporarily reduced the accuracy of reporting a 2nd target (T2). This attentional blink (AB) effect was most pronounced when T1 and T2 occurred close together in time. Use of recognition tests (instead of recall) improved performance but did not eliminate the AB effect. The AB effect was found with both word and nonword distractors, a smaller AB effect was found with consonant string distractors, and the AB effect was substantially attenuated with strings of unfamiliar characters (a false font). Analyses of errors indicated that the 2nd target is frequently replaced or corrupted by the following distractor during the blink. The AB effect appears to result from both attentional and mnemonic processes.