Hindsight bias is the phenomenon where people overestimate their original knowledge after outcome information is available. It was found in various domains and is known to be very hard to eliminate. However, despite extensive research on hindsight bias, this phenomenon in emotion recognition remains underexplored. Giroux et al. (2022) conducted a pioneering study, where participants first identified emotions as blurry faces clarified (foresight phase) and later recreated their recognition point after seeing the clear image (hindsight phase). They showed that participants exhibited hindsight bias for various emotions, except for happiness. The aim of our study was to replicate the findings of Giroux et al. (2022) for happy, sad, fearful and surprised faces using the same blurry-face procedure. We also tested how the perceived intensity of emotions and individual differences in depression and anxiety symptoms might influence hindsight bias. We could not replicate the pattern observed by Giroux et al. (2022); participants in our study recalled their original judgements quite accurately across all emotions. However, when considering the perceived intensity of emotions, participants showed hindsight bias for fearful and surprised faces rated as having low to medium intensity. Further, we found a significant link between depression scores and stimulus identification points, but it did not result in hindsight bias. In sum, our study sheds light on the complexities of hindsight bias in emotional recognition, highlighting the importance of emotional intensity in influencing this bias.