Processing reduced speech in the L1 and L2: a combined eye-tracking and ERP study

Figures & data

Figure 1. Example of a visual display containing a target, phonological competitor and two neutral distractors. In this display, the target word is fenêtre “window”, the phonological competitor is fourchette “fork” and the neutral distractors are coccinelle “ladybird” and montre “watch”.

Figure 2. Natives' (upper panel) and learners' (lower panel) proportions of fixations to the different lexical candidates for the sentences with full forms (left panel) and reduced forms (right panel). The dashed black lines show the total proportions fixations in each bin.

Figure 3. Differences between fixation proportions on the target and fixation proportions on the phonological competitor.

Table 1. Fixed effects from the growth curve analysis.

	Estimate	t value	Pr($>|t|$)
(Intercept)	1.809e−02	1.452	0.151664
poly1	1.091e−01	3.072	0.003024	**
poly2	4.321e−02	1.859	0.065904	.
native	8.682e−02	4.970	5.55e−06	***
reduced	4.028e−02	6.656	3.85e−11	***
poly1:native	1.764e−01	3.543	0.000707	***
poly2:native	6.284e−02	1.929	0.056516	.
poly1:reduced	5.446e−02	2.250	0.024609	*
poly2:reduced	2.997e−02	1.238	0.215828
native:reduced	−6.176e−02	−7.282	5.15e−13	***
poly1:native:reduced	−5.145e−02	−1.516	0.129615
poly2:native:reduced	−2.375e−02	−0.700	0.484017

Notes: The Dutch learners are on the intercept, so “native” refers to the French natives. Full stimuli are on the intercept, so “reduced” refers to the reduced stimuli.

Table 2. Average values of the individual eye features during the preview and target windows.

	Preview		Target
	Learners	Natives	Learners	Natives
Distance	1431	1080	1390	1199
# Saccades	3.2	3.2	2.8	3.1
# Pictures fixated	0.9	1.2	1.0	1.1
Fixation Time	34.5	44.4	38.4	38.5
Fixation Area	68	159	127	227

Notes: Count features are in real numbers; Fixation Time is in centiseconds; the remaining features are on arbitrary scales. Only the red number pairs are not significantly different between natives and learners.

Figure 4. The 35 EEG channels used in the analyses, and the three sets of five sensors analysed.

Figure 5. ERPs of three groups of five EEG channels covering the centre censors, posterior sensors and midline sensors (see Figure 4).

Figure 6. Significant t-values for the factors reduction and language and the interaction between these factors in each of the three sets of signals analysed.

Table 3. Example of a complete lme model (posterior sensors, target onset, window # 13).

	Estimate	Std. Error	t value	$\mathrm{Pr}(>|t|)$
(Intercept)	−3.588e+00	9.119e−01	−3.935	0.000118	***
Alday	−2.808e−01	9.589e−03	−29.281	< 2e−16	***
trial	4.598e−02	9.946e−03	4.623	3.80e−06	***
block	−1.083e+00	3.278e−01	−3.302	0.000960	***
reduced	−1.280e+00	4.440e−01	−2.883	0.003938	**
native	1.285e+00	8.635e−01	1.488	0.141144
Freq	2.087e+00	7.560e−01	2.760	0.006699	**
I(time−startTime)	−2.766e−03	2.947e+04	−0.517	0.605085
reduced:native	−1.654e+00	6.222e−01	−2.658	0.007861	**
reduced:Freq	−2.185e+00	4.226e−01	−5.171	2.34e−07	***
native:Freq	−2.117e+00	7.992e−01	−2.649	0.009888	**
reduced:native:Freq	2.977e+00	5.922e−01	5.027	5.02e−07	***

Table 4. Number of trials for each of the four stimulus types with time-lock on the onset of the target word, and time-locks on the moment the eyes enter a region of interest.

	L2-full	L2-reduced	L1-full	L1-reduced
onset	726	726	755	756
target	190	235	327	317
competitor	169	158	179	203
any	478	504	610	609
last	367	377	555	508

Figure 7. Comparison between significant t-values for the factors reduction and language and the interaction between these factors obtained with time-locks on the moment that the eyes entered the Region of Interest of one of the four pictures. Data shown are for the centre sensors. Note the annotation of the horizontal axis: time from time-lock. The corresponding physical time point is different for all trials.