Skip to content

ch02-parallel-brain.md

Latest commit

 

History

History
41 lines (33 loc) · 2.34 KB

ch02-parallel-brain.md

File metadata and controls

41 lines (33 loc) · 2.34 KB
Raw

Chapter 2: 大脑：并行处理的超级有机体

Core Idea

The multilingual brain does not simply switch one language off and another on. Multiple languages remain active together, creating parallel activation and requiring powerful control systems.

Frameworks Introduced

  • Parallel activation
    • When to use: Explaining cross-language interference, eye movements, accent effects, or multilingual attention.
    • How: List all languages that could be activated by sound, spelling, sign, object, or translation.
  • Ripple effect
    • When to use: Explaining why a small word cue can trigger many related concepts.
    • How: Trace activation from word form to meanings, translations, neighbors, and associated concepts.
  • Dynamic activation threshold
    • When to use: Explaining why some languages intrude more after recent use or when languages are similar.
    • How: Consider proficiency, recency, age of acquisition, similarity, and context.

Key Concepts

  • Eye tracking: Measuring gaze to infer unconscious activation.
  • Cross-language competition: Activation of words or structures from a non-target language.
  • Domain-general transfer: Language-control experience influencing non-language cognition.

Mental Models

  • Think of the multilingual brain as a live network, not a set of separate drawers.
  • Treat interference as evidence of system design: the brain keeps options ready instead of waiting to identify language first.

Anti-patterns

  • On/off bilingualism: Assuming languages are fully disabled when not in use.
  • Noise framing: Treating cross-language activation as error rather than information.

Worked Example

When a Russian-English bilingual hears an English word such as speaker, gaze can drift toward an object whose Russian name overlaps in sound. The person is not choosing to translate; another language is active underneath the task.

Key Takeaways

  1. Parallel activation is efficient because it prepares the brain for unpredictable input.
  2. Cross-language activation spans sounds, spellings, signs, syntax, meanings, and translations.
  3. Managing competition trains attention and control.
  4. Multilingual cognition is a changing state, not a fixed trait.

Connects To

  • ch03: Parallel activation supports remote associations and creativity.
  • ch05: Competition management supports executive function.