Interactive L2 vocabulary acquisition in a lab-based immersion setting

Figures & data

Table 1. Mean scores and standard deviations (between parentheses) on participant descriptives in the two conditions.

	Experimental n = 30	Control n = 15	Test statistics
Age	22.53 (2.47)	22.53 (2.50)	W = 228.5, p = .94
Years of learning Dutch	2.69 (1.78)	2.74 (1.96)	W = 230.5, p = .90
Self-rated proficiency*	3.07 (0.74)	3.27 (0.59)	W = 193, p = .41
Amount of daily exposure to Dutch*	3.07 (0.79)	3.29 (0.84)	W = 183.5, p = .32
Number of other languages known	2.33 (0.76)	2.47 (0.74)	W = 202.5, p = .56
Dutch vocabulary (LexTALE)	69.67 (7.75)	68.42 (8.27)	t(26.53) = 0.49, p = .63
Phonological working memory	80.17 (7.56)	81.71 (6.70)	t(28.53) = −0.68, p = .50

Note: For a description of the measurements, see Measures of individual differences. Variables marked with an asterisk were self-rated on a 1–5 Likert scale.

Figure 1. Mean scores across the four testing moments (EF = Exposure frequency). Error bars represent 95% confidence intervals based on a bootstrap.

Table 2. Percentage of correctly produced phonemes per target word during the price comparison task (i.e. the learning phase).

		Experimental group			Control group
		Mean	SD	95% CI	Mean	SD	95% CI
Cognate status	Cognate	75.26	12.19	70.71–79.82	8.47	8.20	3.93–13.02
	Non-cognate	54.93	18.43	48.05–61.82	1.50	4.34	−0.91–3.91
Exposure frequency	2 times	57.43	14.87	51.88–62.98	5.08	5.64	1.96–8.20
	4 times	72.77	14.07	67.51–78.02	4.89	4.05	2.65–7.13
Lag	3 trials	67.48	16.89	61.17–73.79	6.82	7.93	2.43–11.21
	7 trials	62.72	14.57	57.28–68.16	3.15	4.36	0.74–5.57
Total		65.10	13.60	60.02–70.18	4.99	4.73	2.37–7.60

Table 3. Percentage of correctly produced phonemes per target word in the two post-tests (i.e. the retention phase).

		Experimental group			Control group
		Mean	SD	95% CI	Mean	SD	95% CI
Cognate status	Cognate	59.06	14.97	53.47–64.65	8.04	6.91	4.21–11.87
	Non-cognate	31.27	17.24	24.83–37.71	1.50	4.34	−0.91–3.91
Retention interval	20 min	55.36	14.83	49.82–60.90	4.77	4.05	2.53–7.01
	6 months	23.43	12.68	17.13–29.73	N/A	N/A	N/A
Lag	3 trials	44.00	14.04	38.76–49.24	5.95	6.65	2.26–9.63
	7 trials	46.32	17.58	39.76–52.89	3.59	4.86	0.90–6.28
Total		55.36	14.83	49.82–60.90	4.77	4.05	2.53–7.01

Table 4. Outcomes of the learning model.

Fixed effects	Logit	Odds ratio	SE	z	p
(Intercept)	2.80	16.42	0.76	3.71	<.001
G = Control	−11.96	<0.001	2.24	−5.33	<.001
CS = Non-cognate	−3.25	0.04	0.75	−4.35	<.001
EF = 4 times	1.72	5.60	0.28	6.13	<.001
L = 7 trials	−0.74	0.48	0.60	−1.23	.22
G = Control: CS = Non-cognate	−2.86	0.06	2.55	−1.12	.26
G = Control: EF = 4 times	−2.26	0.10	0.71	−3.17	. 002
G = Control: L = 7 trials	−3.36	0.03	3.37	−1.00	.32
Random effects	Variance	SD
Participant (Intercept)	5.55	2.36
CS = Non-cognate	4.61	2.15
EF = 4 times	1.79	1.34
L = 7 trials	5.46	2.34
CS = Non-cognate: EF = 4 times	2.53	1.59
CS = Non-cognate: L = 7 trials	16.11	4.01
EF = 4 times: L = 7 trials	5.94	2.44
CS = Non-cognate: EF = 4 times: L = 7 trials	5.85	2.42
Word (Intercept)	11.47	3.39
G = Control	48.06	6.93
EF = 4 times	3.80	1.95
L = 7 trials	8.01	2.83
G = Control: L = 7 trials	133.27	11.54
EF = 4 times: L = 7 trials	6.37	2.52

Note: The intercept represents the following combination of variable levels: G [Group] = Experimental, CS [Cognate status] = Cognate, EF [Exposure frequency] = 2 times, L [Lag] = 3 trials. Colons (:) represent interactions but not lower-order effects, equal signs (=) signal the level of a categorical variable. Significant p-values are printed in bold.

Table 5. Outcomes of the retention model.

Fixed effects	Logit	Odds ratio	SE	z	p
(Intercept)	3.79	44.13	0.56	6.73	<.001
CS = Non-cognate	−3.07	0.05	0.58	−5.25	<.001
RI = 20 min	−1.62	0.20	0.26	−6.15	<.001
RI = 6 months	−6.15	0.002	0.72	−8.49	<.001
L = 7 trials	−0.18	0.84	0.33	−0.54	.59
Random effects	Variance	SD
Participant (Intercept)	2.31	1.52
CS = Non-cognate	1.75	1.32
RI = 20 min	0.64	0.80
RI = 6 months	5.19	2.28
L = 7 trials	0.91	0.95
CS = Non-cognate: RI = 20 min	0.69	0.83
CS = Non-cognate: RI = 6 months	21.67	4.66
Word (Intercept)	13.79	3.71
RI = 20 min	2.75	1.66
RI = 6 months	23.14	4.81
L = 7 trials	5.74	2.40
RI = 20 min: L = 7 trials	3.91	1.98
RI = 6 months: L = 7 trials	24.57	4.96

Note: The intercept represents the following combination of variable levels: CS [Cognate status] = Cognate, RI [Retention interval] = 4 exposures (i.e. participants’ scores after 20 minutes and six months are compared to their last score from the price comparison task), L [Lag] = 3 trials. Colons (:) represent interactions but not lower-order effects, equal signs (=) signal the level of a categorical variable. Significant p-values are printed in bold.

Data availability statement

The data and analysis script that support the findings of this study are available at https://github.com/johannadevos/NaturalisticL2WordLearning and at http://hdl.handle.net/11633/di.dcc.DSC_2017.00027_498.