The samples in some studies.

This study investigates the impact of translation difficulty and translation experience on the processing of translation units (TUs) based on keylogging data. TUs are categorized into two types: production TUs (PTUs), which have linguistic equivalents in the source text (ST), and non-production TUs (N-PTUs), which lack such equivalents. N-PTUs typically include combinations of letters, morphemes, or non-text production keystrokes, such as backspaces, deletions, spaces, cursor movements, carriage returns, and mouse navigations. Data were collected from 55 fourth-year English major students and 35 semi-professional translators as they translated two STs of differing difficulty. The findings reveal no significant differences in the number of PTUs or the average size of PTUs (APTU size) between students and semi-professional translators. However, ST translation difficulty significantly influenced TU processing. For the more difficult text, the APTU size decreased notably, the number of 1–6-word PTUs increased substantially, and the frequency of ≥ 7-word PTUs remained relatively stable. Moreover, when translating the difficult text, semi-professional translators produced smaller APTU sizes, a greater number of N-PTUs, and a higher overall average number of PTUs compared to their performance on the relatively easy text. These results highlight the interplay between cognitive effort, ST translation difficulty, and translation experience in TU processing and have some implications for designing translation courses in academic contexts and developing professional translator training programs.