Individual differences in emotion-induced blindness: Are they reliable and what do they measure?

The emotion-induced-blindness (EIB) paradigm has been extensively used to investigate attentional biases to emotionally salient stimuli. However, the low reliability of EIB scores (the difference in performance between the neutral and emotionally salient condition) limits the effectiveness of the paradigm for investigating individual differences. Here, across two studies, we investigated whether we could improve the reliability of EIB scores. In Experiment 1, we introduced a mid-intensity emotionally salient stimuli condition, with the goal of obtaining a wider range of EIB magnitudes to promote reliability. In Experiment 2, we sought to reduce the attentional oddball effect, so we created a modified EIB paradigm by removing the filler images. Neither of these approaches improved the reliability of the EIB scores. Reliability for the high- and mid-intensity EIB difference scores were low, while reliability of the scores for absolute performance (neutral, high-, and mid-intensity) were high and the scores were also highly correlated, even though overall performance in the emotionally salient conditions were significantly worse than in the neutral conditions. Given these results, we can conclude that while emotionally salient stimuli impair performance in the EIB task compared with the neutral condition, the strong correlation between the emotionally salient and neutral conditions means that while EIB can be used to investigate individual differences in attentional control, it is not selective to individual differences in attentional biases to emotionally salient stimuli.