The attentional boost effect (ABE) refers to the counter-intuitive finding that the detection of infrequent targets in a divided-attention condition enhances memory of images co-occurring with targets, as compared with images co-occurring with frequent distractors (Swallow & Jiang, 2010). Previous studies have shown that the ABE also applies to verbal materials but had small or no effect on low-frequency (LF) words or orthographically distinctive (OD) words. The present study is to test whether the ABE be moderated by the unusual properties of stimuli, by manipulating another distinctive property of the to-be-remembered stimuli: negative emotion. The present study includes two experiments. In Experiment 1, participants encoded a long sequence of neutral or negative words or images (500ms/ item), while simultaneously monitored the color of a small circle (100ms) located under each word or a small square (100ms) located at the center of each image. Participants were required to remember the words or images and pressed a spacebar whenever they detected occasional targets (red circles or black squares) among more frequent distractors (green circles or white squares). In a later recognition test, stimuli co-occurring with targets and distractors were mixed with new stimuli for an old/new judgment. Experiment 2 further manipulated the arousal of the negative stimuli, and was divided into two conditions: low-arousal and high-arousal negative stimuli paired with targets and distractors. Thirty-four undergraduates participated in Experiment 1 and thirty undergraduates participated in Experiment 2. All experimental procedures were programmed in Presentation 0.71, and run on a DELL Dimension 8200 computer with a 19? monitor (1024×768 pixels). The main results were as following: (1) In Experiment 1, the ABE was significant for neutral stimuli [F(1,33)=39.45, p<.001] and for negative stimuli [F(1,33)=9.99, p=.003], with the recognition being better for target-paired stimuli than for distractor-paired stimuli. More important, the ABE was smaller for negative-emotional stimuli than for neutral-emotional stimuli [F(1,33)=9.808, p=.004, ηp2=.229], indicating the moderating influence of negative emotion on the ABE. (2) In Experiment 2, the ABE was significant for low-arousal negative stimuli [F(1,29)=14.04, p=.001], whereas there was no hint of an ABE for high-arousal negative stimuli [F(1,29)=.10, p=.754], that is, the ABE had no effect on the high-arousal negative stimuli. These results confirm and extend previous results suggesting that the ABE could be moderated by the distinctiveness of memory materials. The unusual nature of negative stimuli attracts greater attentional resources during the encoding. As a consequence, the transient attentional enhancement occurring as a consequence of target detection in the ABE paradigm might have minimal or no effects on these distinctive items.