Alexithymia is a personality trait associated with impairments in emotional processing. This study investigated the influence of alexithymia and sex in the ability to recognize emotional expressions presented in faces, voices, and their combination. Alexithymia was assessed by the Toronto Alexithymia Scale (TAS-20) and participants (n = 122) judged 12 emotions displayed uni- or bimodally in two sensory modalities as measured by the Geneva Multimodal Emotion Portrayals Core Set (GEMEP-CS). According to their scores, participants were grouped into low, average, and high alexithymia. The results showed that sex did not moderate the relationship between alexithymia and emotional recognition. The low alexithymia group recognized emotions more accurately than the other two subgroups, at least in the visual modality. No group differences were found in the voice and the bimodal tasks. These findings illustrate the importance of accounting for how different modalities influence the presentation of emotional cues, as well as suggesting the use of dynamic instruments such as GEMEP-CS that increment ecological validity and are more sensitive in detecting individual differences, over posed techniques such as still pictures / Genetic and neural factors underlying individual differences in emotion recognition ability