In neurocognitive and physiological studies, context-independent (CI) and contextdependent (CD) properties have long been dichotomized in terms of their detecting, searching, retrieval and attention management. CI properties are always activated by the respective words on all occasions and is unaffected by contextual relevance. CD properties are the source of semantic encoding variability. Assisted by a small-sized corpus and with neurolinguistic theory of simultaneous translation, this paper investigated how CI information is processed in English-Chinese simultaneous interpretation. With naturalistic data pattern generated from the corpus, it argues that CI information in SI serves as one of the cognitive problem triggers as higher omission rate in seasoned interpreter’s recording, indicating possible cognitive overload. It is thus assumed that because CI information does not interfere with the meaning-based interpretation of the source language utterance, if it fails to be detected, no direct or obvious contextual connection can further facilitate its decoding and encoding. Thus, CI information in SL utterance may share higher rate of omissions or other kinds of delivering failures where cognitive overload appears.