Cognitive restructuring in the multilingual
mind: language-specific effects on processing
efficiency of caused motion events in
Cantonese–English–Japanese speakers
By: Wang & Wei
Talmy’s typological distinctions
S-languages:
Satellite-framed languages
V-languages
Verb-framed languages
Background
- Path of motion: where is it going
(into, out of)
- Cause: means of direction
(take, carry)
- Manner of cause: how did one move it (push, pull)
- Manner of the object: how did the object move (roll, slide)
E-languages:
Equipollent-framed languages
Linguistic Elements
- S-languages: Manner of cause is encoded with verb & the path is encoded outside the verb. Manner can be conflated w/ motion
The man pushed a box into a cave
- V-languages: Path is encoded with the verb. Motion is conflated w/ path, leaving manner unexpressed leaving interpretation to default
He moved up the goods
- E-language: manner & path are conflated and encoded w/ the verb; incorporated both V & S-languages
He pulled a car up a hill
Linguistic Elements
TASK 1 LINGUISTIC ENCODING
Task 1
- 48 animated cartoons w/ 36 test items & 12 control items.
- Each animation was shown for 6 seconds using four different types of manners of cause (pull, push, drag, and kick) & four different types of path (into, out of, across, and along)
- They were asked to watch the clip then describe what happened
- In this study participants were either monolingual, bilingual, or multilingual
- Monolinguals narrated in L1 (their only language), Bilinguals in English (L2), & multilinguals in Japanese (L3)
TASK 2
NON-LINGUISTIC CATEGORIZATION
Task 2
- 18 animated videos, 12 test rials, & 6 sets of filler items.
- Filler items were used as a distractor and to mask the contrast of interest.
- This task contained a target video then two other videos with manner and path as the contrast on interest.
- Target video was played by itself followed by the two alternatives playing simultaneously, side by side.
- Participants were asked to decide which video was more similar to the target video. Indications were produced by pushing either A or L key on a keyboard. They were asked to make a decision as quick as possible.
Example
A boy pulled a chair out of the room
Example
A boy pulled a chair into the room
A boy pushed a chair out of the room
Task 1 - Graph
- English monolinguals & bilinguals were equally likely to encode manner in the verb form & more frequently than Cantonese monolinguals
- English and bilinguals didn't encode path in the main verb & multilinguals used more path verbs
Manner verbs:
Positively correlated with English use
Negatively correlated with use of Japanese
Task 2 - Graph
Task 1 Discussion
- English (S-languages) expressed manner of cause (push, pull) more frequently than Cantonese (E-languages)
- Japanese (V-languages) displayed the lowest frequency of manner encoding - This is because the cause of motion is conflated with path & can easily be added or dropped
- There's no syntactic "slot" for encoding the manner of cause
- Japanese have a limited set of lexical devices & tend to encode a more general expression (e.g. took)
Discussion
Monolinguals
All three monolingual groups experienced a ceiling effect - indicating that path is the central element in motion events
Monolinguals
English monolinguals:
- Encoded manner w/ the main verb
- Reacted quicker in making manner-match vs. path-match choices
- Had the fastest reaction time (compared to Japanese monos, all multilinguals & bilinguals) for those manner-match choices
Japanese monolinguals:
- Reacted quicker in path-match compared to manner-match
- Had the quickest RT in path-match compared to all other groups
Cantonese monolinguals:
- Cantonese speakers had equal efficiency in path & manner-match choices
- Cantonese speakers encoded both manner and path in a verb-compound
English Bilinguals
Bilinguals
- Displayed the same patterns as English monolinguals in both the manner section & semantic distribution
- Bilinguals reacted to manner-match quicker than path-match
Japanese Multilinguals
Multilingual
- showed a tendency to reacted quicker to path-match rather than manner-match irrespective of the categorization preference
- The more of a language participants used, the more patterns for that language were present
- The amount of L1 (Cantonese) use wasn't a core predictor
- L2 & L3, however, did serve as a main predictor for the amount of cognitive reconstruction that was displayed