From the perspective of cognitive linguistics, metaphor and metonymy are not just figurative speech but an important reflection of people’s experiences, and they are prevalent in daily language use (Lakoff & Johnson, 1980; Zheng & Xiang, 2014). Difficulties and variations in translation tasks involving metaphor and metonymy are associated with factors such as cultural differences and the indirectness of language (Manipuspika, 2018). Metaphoric and metonymic expressions are thus characterized by the general problem of untranslatability (Schäffner, 2004). Extensive research has compared cognitive effort in processing metaphoric and literal expressions (Mashal et al., 2009; Rapp et al., 2012), metonymic and literal expressions (Rapp et al., 2011), and metonymic and metaphoric expressions (Jager & Cleland, 2015; Weiland et al., 2014). Controversies exist due to the variances in controlling variables such as syntactic structure, familiarity of the chosen materials, and contextual details of the stimuli in the experiments (Mashal et al., 2009; Schmidt & Seger, 2009). Despite the effort in capturing the cognitive mechanism behind processing metaphor and metonymy, specific focus on translating metaphor and metonymy-laden expressions is needed. Related research in translation (Sjørup, 2013; Wang, 2021) mainly discovered that translating metaphor is more cognitively demanding than literal expressions. However, translating metonymy itself and the comparison between translating metaphor and metonymy has remained underexplored. This research made use of the polysemous senses of English zero-derived denominal verbs (entailing different involvement of metonymy and metaphor) as stimuli in an English-Chinese sight translation experiment to investigate the level of cognitive load caused. Eye-tracking data, audio data, and NASA-TLX data were collected with complementary qualitative data from a semi-structured interview. Results suggested that the highest level of cognitive load was identified when interpreters translated stimuli with both metonymy and metaphor, followed by pure-metonymic stimuli, while the literal counterparts triggered the most negligible cognitive load.