Self-regulated monitoring in undergraduate translation: a think-aloud study of SRL processes.

Introduction: This exploratory qualitative study examines patterns of articulated monitoring depth in concurrent think-aloud protocols from undergraduate translation students, focusing on qualitative variation between surface monitoring (SM) and evaluative monitoring (EM). Methods: Twelve third-year undergraduate translation majors completed an English-Chinese literary translation task while verbalizing their thoughts concurrently. Participants were purposively selected from higher- and lower-scoring bands on a recent programme-wide translation assessment. Monitoring episodes were analyzed according to whether checking remained at local lexical or formal confirmation (SM) or involved testing candidate solutions against meaning, discourse relations, rhetorical purpose, or target-language fit (EM). Results: The central contrast concerned the regulatory endpoint of focal episodes rather than a uniform group difference. Several higher-scoring cases articulated sustained EM, whereas several lower-scoring cases included checks that ended once immediate lexical or structural uncertainty had been reduced. The distinction was neither categorical nor person-fixed: both forms occurred within the same participant, and P12 moved from dictionary-driven SM to discourse-based EM during the task. Within-case sequences further revealed narrowing, recovery, and maintenance of monitoring depth. Prolonged lexical search, repeated re-parsing, and unresolved uncertainty co-occurred with narrowing in some episodes. Discussion: Because cognitive load was not directly measured and concurrent verbalization may become less complete as task demands rise, these patterns are interpreted as hypothesis-generating rather than as evidence of a causal mechanism. Pedagogically, the findings support instruction that makes evaluation criteria, deliberate deferral, global rereading, and diagnosis-led revision explicit.