Children's Comprehension of Sentences with Focus Particles and the Role of Cognitive Control: An Eye Tracking Study with German-Learning 4-Year-Olds

Abstract

Children's interpretations of sentences containing focus particles do not seem adult-like until school age. This study investigates how German 4-year-old children comprehend sentences with the focus particle 'nur' (only) by using different tasks and controlling for the impact of general cognitive abilities on performance measures. Two sentence types with 'only' in either pre-subject or pre-object position were presented. Eye gaze data and verbal responses were collected via the visual world paradigm combined with a sentence-picture verification task. While the eye tracking data revealed an adult-like pattern of focus particle processing, the sentence-picture verification replicated previous findings of poor comprehension, especially for 'only' in pre-subject position. A second study focused on the impact of general cognitive abilities on the outcomes of the verification task. Working memory was related to children's performance in both sentence types whereas inhibitory control was selectively related to the number of errors for sentences with 'only' in pre-subject position. These results suggest that children at the age of 4 years have the linguistic competence to correctly interpret sentences with focus particles, which--depending on specific task demands--may be masked by immature general cognitive abilities.

Fig 1. Two examples of the visual displays.

(A) Scenario for the test sentence Only the elephant has a kite (expected answer: yes). (B) Scenario for The duck has only a boat (expected answer: no). The images are for illustrative purposes only; They are very similar but not identical to the ones used in the experiment.

Fig 2. Mean accuracy scores for each sentence type and expected response in children and adults.

Error bars denote two standard errors.

Fig 3. Looks to the subject alternative set by children and adults.

Data shown are averaged over both testing blocks.

Fig 4. Looks to the subject alternative separated by testing block and the type of expected response.

Fig 5. Examples of the three conditions in the flanker task.

The target fish is always the one central on the screen. (A) Neutral condition without flankers. (B) Congruent condition: flanking fish face the same way as the target fish. (C) Incongruent condition: the direction of the flanking fish conflicts with the direction of the target fish. These examples show left-facing target fish, but there were right-facing ones as well. The images are for illustrative purposes only; They are very similar but not identical to the ones used in the experiment.

Fig 6. Mean accuracy scores for each sentence type and expected response.

Error bars denote two standard errors.

Fig 7. Looks to the subject alternative set plotted separately for block and expected response.

Fig 8. Relation between cognitive control and the comprehension of ‘only’.

The interference effect as a measure of cognitive control is plotted against the accuracy in the sentence-picture verification for the non-matching sentences containing only separated by position (pre-subject sentences on the left, pre-object sentences on the right).