Research suggests that particular classes of stimuli are associated with attentional biases, with cues linked to threat, self, and current goals all being found to be effective at attracting attention. However, it is not currently clear which of these stimulus types are prioritised in the attentional hierarchy. In the current project, we used dot probe tasks to determine how attention is allocated between pairs of neutral, threatening, self and goal-relevant face images. In Experiment 1, participants were presented with self, angry and neutral face images in the dot probe task. Response latencies revealed a clear attentional bias for self-images over both angry and neutral images. In Experiment 2, each participant was assigned a specific goal image to monitor while completing the dot probe task. Again, self-cues were found to produce a strong attentional bias, but images associated with the temporary goal were found to be the most effective source of attentional bias. The datasets for Experiments 1 and 2 are stored in the attached file.