In this paper, we describe our approach to subtask 2 of the Emotional Impact of Movies task from the Mediaeval 2018 Challenge. We compared the performances of LSTM ensembles to single LSTM models for predicting the fear-inducing seconds in movies. We also compared the performance of an LSTM model trained on audio feature data to the performance of an LSTM model trained on the outputs of a pretrained VGG16 model. Ultimately, we found that a single LSTM trained on VGG16 outputs achieved the highest F1 score on the test set.