The effects of frequency, duration,

and intensity on L2 learning through

Duolingo
A natural experiment

Ekaterina Sudina and Luke Plonsky

East Carolina University, United States | Northern Arizona University,
United States

Instructed second language (L2) research has frequently addressed the

effects of spacing, or, alternatively, the distribution of practice effects. The
present study addresses Rogers and Cheung’s (2021) concerns about the
ecological validity of such work via a natural experiment (Craig et al., 2017).
Learners’ self-determined exposure and in-app behavior were examined in
relation to language gains over time. Duolingo learners of Spanish or French
(N = 287) completed a background questionnaire, scales measuring L2
motivation and grit, and two tests of L2 proficiency before and after a six-
month period of user-controlled app usage. Total minutes of app exposure
exhibited a correlation with written but not oral proficiency gains. More
dependable correlates of gains were frequency- and curriculum-oriented
measures. Additionally, L2 grit and motivation were weakly to moderately
correlated with several in-app behaviors. We conclude with implications for
how apps can best be leveraged to produce L2 gains.

Keywords: distribution of practice, instructed SLA, mobile-assisted

language learning, grit, Duolingo

1. Introduction

Second-language (L2) development is greatly influenced by the type and amount

of instruction that learners receive (e.g., Norris & Ortega, 2000; Saito & Plonsky,
2019). Less clear, however, are the effects of frequency, duration, and intensity of
target language exposure and practice. That is, are intensive periods of instruction
most effective, or is instruction more effective through more frequent exposure
(see Carpenter, 2020; Rogers & Cheung, 2021)?
https://doi.org/10.1075/jsls.00021.plo | Published online: 25 August 2023
Journal of Second Language Studies 7:1 (2024), pp. 1–43. ISSN 2542-3835 | E‑ISSN 2542-3843
Available under the CC BY 4.0 license. © John Benjamins Publishing Company
2 Ekaterina Sudina and Luke Plonsky

Informed by a quickly growing body of research in second-language acquisi-

tion (SLA) that has sought to address this question (e.g., Kasprowicz et al., 2019;
Li & DeKeyser, 2019; Yamagata et al., 2023), the present study builds on these and
other recent studies of distributed learning in SLA (e.g., Serrano & Huang, 2018;
Suzuki & DeKeyser, 2017) as well as several decades of theoretical and empirical
attention in educational psychology (see Carpenter, 2020; Rohrer, 2015) to shed
light on the effects of frequency and duration of Duolingo app usage on L2 learn-
ing in the context of the distribution of practice. In doing so, the study seeks to
make several unique and worthwhile contributions.
First, being the first to examine distributional effects for app-based language
learning, this study allows us to gain a better understanding of the effects of fre-
quency, duration, and intensity specifically for the Duolingo app and its users.
Second, previous studies on distributed learning have targeted predominantly
vocabulary and grammar knowledge in either traditional classroom or laboratory
settings (e.g., Kasprowicz et al., 2019; Nakata & Suzuki, 2019; Suzuki, 2019). The
present study, by contrast, considers overall proficiency in both oral and written
modes in a new and flourishing context of instructional technology. Third, Rogers
and Cheung (2021) raised concerns over the ecological validity of previous work
on distributed learning. In particular, the authors questioned whether existing
results, which are almost exclusively obtained in labs, would hold in less con-
trolled settings. The present study addresses this concern by conducting a natural
experiment (Craig et al., 2017), whereby learners’ self-determined exposure and
behavior (i.e., frequency, duration, and intensity of app usage) are examined in
relation to language proficiency gains made over a six-month period. Fourth, in
addition to exposure effects, two individual differences, L2 grit (Teimouri et al.,
2022) and motivation (Papi et al., 2019), are modeled in relation to learner behav-
ior and gains in proficiency to isolate and better understand the effects of fre-
quency, duration, and intensity of instruction. The literature review that follows
offers an overview of theories and studies on distributed learning in SLA and
neighboring disciplines, introduces the concept of natural experiment, surveys
recent studies on the role of app-based technology in instructed SLA, and dis-
cusses the concepts of L2 grit and motivation that are relevant to the present study.
The last section of the literature review provides an overview of the Duolingo
course structure at the time of data collection to better situate the study.
 The effects of frequency, duration, and intensity on L2 learning through Duolingo 3

2. Literature review

2.1 Distributed learning

Rogers and Cheung (2021) provided definitions of several concepts pertinent to

the present investigation. First, “distribution of practice, also referred to as input
spacing, refers to whether and how learning is spaced over multiple learning
episodes” (pp. 1138–1139; see also Rogers, 2017 for an overview of theoretical and
methodological differences in research on the distribution of practice in SLA and
cognitive psychology). Second, “massed practice refers to experimental conditions
in which learning is concentrated into a single, uninterrupted training session,
whereas distributed or spaced practice refers to learning that is spread over two
or more training episodes” (p. 1139). It is also important to differentiate between
spacing and lag effects as the latter were the focus of the current study. Spacing
effects (i.e., comparison of a massed condition versus a spaced condition) are typ-
ically robust; lag effects (i.e., advantage of a longer gap over a shorter gap) are
harder to obtain (see Rogers’ 2023 conceptual review for more); data from Kim
and Webb’s (2022) meta-analysis support this finding as an advantage of longer
spacing was found in delayed but not in immediate posttests. As noted by Rogers
(2023), “spaced-condition groups outperform the massed-condition group with
relatively large effect sizes (d ≥ 1.0), but there is little to no difference in perfor-
mance between those in the different distributed conditions” (p. 452).
According to Carpenter (2020), there is a consensus in educational research
that distributed practice, or spacing effect, is more conducive to learning when
repeated instruction is dispersed over time rather than happens in rapid succes-
sion (for a review of spacing and massing, see Rohrer, 2015). This stems in part
from deficient processing theory, which postulates that spaced instruction pro-
vides learners with more opportunities for noticing and ultimately leads to bet-
ter retention of the information presented. Another theory that might explain
the spacing effect is that of encoding variability; it emphasizes the role of con-
textual cues, which become more pronounced in spaced learning experiences.
Next, a study-phase retrieval theory suggests that retrieval of learners’ previous
experience, which typically happens during distributed instruction, aids in future
retention of the material. Finally, consolidation in the form of neural activation
processes is more likely to occur during spaced rather than massed practice. Crit-
ically, although the aforementioned theories have made great strides in explaining
spacing effects, none of them is fully satisfactory as a stand-alone theory; thus,
there is currently no consensus on the underlying theoretical mechanism respon-
sible for spacing and lag effects in learning (see Delaney et al., 2010, for more).
4 Ekaterina Sudina and Luke Plonsky

It should be noted that spaced instructional activities may be challenging to

implement in real-life classrooms as this approach requires considerable planning
on the part of instructors who need to create multiple lesson plans in advance and
adhere to spacing schedules consistently for longer periods of time. One could
even make the case that spacing effects are not as relevant in an ecological sense
to L2 classroom practice, or to mobile-assisted language learning (MALL). SLA
researchers are typically interested in lag effects, and comparisons of shorter gaps
(e.g., a 1-day gap between the time that the learner logs in to practice) versus
a longer gap (e.g., a 7-day gap). Notably, despite the consensus towards spac-
ing effects across the broader psychological literature (as suggested by Carpenter,
2020), there are fewer studies (and a lack of consensus) on lag effects (e.g., Rohrer,
2015; Rogers & Cheung, 2020).
Moreover, according to Cepeda et al. (2008), a true spacing study from a
cognitive psychology viewpoint includes “multiple periods of study devoted to
the same material, separated by some variable time gap, with a final memory
test administered after an additional retention interval” (p. 1095). Here, the “same
material” is an important feature of a spacing experiment. Thus, those SLA studies
that do not include the exact same content are arguably not truly spacing studies.
This questions the generalizability of cognitive psychology findings to instructed
SLA (and by extension MALL) where the language content is rarely copied verba-
tim from one session to the next.
Furthermore, Rogers and Cheung (2021) contended that experimental
research in the domain of distributed learning tends to overemphasize internal
validity of a study to the detriment of its external and ecological validity, which
limits the generalizability of tightly controlled experiments to authentic instruc-
tional settings. In fact, the enhanced ecological validity of Rogers and Cheung’s
(2021) conceptual replication was arguably one of the reasons why their study did
not lend support to the advantages of a more distributed L2 vocabulary learn-
ing practice (i.e., 8 days between training sessions) compared to a less distributed
one (i.e., 1 day between training sessions) among English-as-a-foreign language
(EFL) child participants in Hong Kong. This finding largely contradicted pre-
vious lab-based research with adult participants studying L2 vocabulary (e.g.,
Nakata & Suzuki, 2019). However, Rogers and Cheung’s (2021) results resem-
bled those of Kasprowicz et al. (2019) – another study with enhanced ecological
validity that examined the effects of distributed practice among young learners of
French studying L2 grammar (verb morphology). The interval being examined
by Kasprowicz et al. (2019), although quite constrained (3.5 vs. 7 days), emulated
“the most common lesson frequency in UK primary schools (one or two lessons
per week)” (p. 585); yet it did not yield differences in learning between groups. By
contrast, Li and DeKeyser (2019) found enhanced procedural knowledge reten-
 The effects of frequency, duration, and intensity on L2 learning through Duolingo 5

tion among adult learners of Mandarin Chinese exposed to instruction with

shorter rather than longer intervals between sessions (1 day vs. 1 week), but these
results did not hold for declarative knowledge. Similarly complex results were
reported by Serrano and Huang (2018). In their study, intense repeated L2 reading
sessions (1-day interval) were more beneficial for short-term gains in L2 vocab-
ulary for teenage EFL learners in Taiwan (following the immediate posttest),
whereas spaced sessions (1-week interval) resulted in higher retention long-term
(from the immediate to the delayed posttest). Nonetheless, the difference between
the two groups was negligible when vocabulary gains were compared from the
pretest to the delayed posttest.
To make the picture even more complex, Suzuki and DeKeyser’s (2017) study
of L2 Japanese morphology in adult learners found no advantages of distributed
practice (7-day interval) over massed practice (1-day interval) for utterance accu-
racy; moreover, it was less conducive to utterance fluency than massed practice.
However, the authors emphasized that the results of their lab-based study may not
be generalizable to real-life classrooms. Notably, Kim and Webb’s (2022) meta-
analysis of distributed practice revealed the advantage of spaced learning in SLA,
as indicated by small-to-medium and medium-to-large effect sizes (i.e., g = 0.58
and 0.80 for L2 learning and retention, respectively) across 37 eligible studies
(48 independent experiments) in their sample. Critically, the researchers argued
that the spacing effect varied depending on the methodological features of the
study design, which explains in part the complex and conflicting results of the
primary studies reviewed above. Moreover, the increasing maturity of spacing
research within SLA is evidenced in replication studies. To illustrate, Suzuki and
DeKeyser’s (2017) complex findings have been replicated by Suzuki (2017). Other
examples include Serrano and Huang (2018, 2023) as well as Rogers and Cheung
(2020, 2021). For an overview of a broader spacing literature, see Serrano (2022).1

2.2 Natural experiment

Despite the comprehensiveness of Kim and Webb’s (2022) meta-analysis, one

methodological variable that was not examined as a moderator in their sample
was that of ecological validity of primary studies on distributed spacing. Apart
from experimental designs, high ecological validity bears relevance for observa-
tional designs as well. One example of an observational study with increased eco-
logical validity is a natural experiment, which refers to “any event not under the
control of a researcher that divides a population into exposed and unexposed

1. We thank an anonymous reviewer for their considerable input into this section of the Liter-
ature review.
6 Ekaterina Sudina and Luke Plonsky

groups”; this allows natural experiments to “use this naturally occurring variation
in exposure to identify the impact of the event on some outcome of interest”
(Craig et al., 2017, p. 2). Critically, one of the major differences between random-
ized controlled trials, natural experiments, and nonexperimental observational
studies is how well the intervention has been defined (Craig et al., 2017, Table 1).
In randomized controlled trials, the intervention is thoroughly documented and
implemented; in natural experiments, the intervention is also happening, but
researchers have less control over it; finally, in nonexperimental observational
studies, there is no clear intervention at all. Although natural experiments have
been particularly embraced in public health research, they have immediate applic-
ability in instructed SLA and applied linguistics research more generally, espe-
cially in situations where experimental manipulations are not deemed reasonable
or ethical. As such, natural experiments have arguably higher ecological validity
compared to more traditional, rigidly controlled experiments. One example of a
natural experiment for SLA would be a study that examines the effect of partic-
ipating in a study abroad program on L2 learners’ oral proficiency using a pre-
and posttest design (unlike a quasi-experimental study, a natural experiment does
not strictly control the independent variables). However, to our knowledge, no
study to date has conducted a natural experiment to investigate the effectiveness
of mobile-assisted language learning.

2.3 Mobile-assisted language learning

As noted by Loewen (2020), the use of technology is now considered to be

one of four major contexts in instructed SLA along with traditional classroom,
study abroad, and immersion instruction (see Loewen, 2020). One of the advan-
tages of instructional technology is that it has “the potential to speed up or
enhance the process” of L2 learning and, most importantly, “deliver individualized
instructional materials that meet learners at their specific levels of proficiency”
(Loewen, 2020, p. 193). A second-order synthesis by Plonsky and Ziegler (2016)
found a small advantage of computer-assisted language learning over traditional
classroom instruction (based on the results of 14 meta-analyses in this domain,
which included a total of 408 primary studies and over 14,000 language learners).
Nonetheless, this synthesis was unable to examine the effectiveness of mobile-
assisted language learning due to the lack of primary studies in this area.
The situation, however, is rapidly changing, and research focusing on the
role of mobile- and app-based technology in instructed SLA is currently on the
rise (e.g., García Botero et al., 2019; Jiang, Rollinson, et al., 2021; Loewen et al.,
2019, 2020). In fact, several recent studies have focused specifically on the relative
effectiveness of Duolingo vs. university-based language instruction. The beginner
 The effects of frequency, duration, and intensity on L2 learning through Duolingo 7

and intermediate Duolingo learners attained similar – and in some cases, supe-
rior – L2 proficiency levels as university students who studied foreign languages
for four and five semesters, respectively (see Jiang et al., 2020; Jiang, Chen, et al.,
2021). Of particular relevance to the present investigation is a study by Jiang,
Rollinson, Plonsky, et al. (2021) that explored a possible relationship between app
usage and gains in proficiency and observed modest correlations between the two
(ρ = .02−.14 for L2 French listening and reading; ρ = .01−.06 for L2 Spanish lis-
tening and reading, respectively). However, only one of many possible temporal
or exposure-related indicators was used (total hours). Furthermore, the sample
included only novices, and no pretest data were collected – features the present
study seeks to improve on.
Additionally, recently published studies and existing reports on Duolingo
effectiveness have predominantly assessed specific language skills (i.e., listening,
reading, and speaking; see Jiang et al., 2020; Jiang, Chen, et al., 2021; Jiang,
Rollinson, et al., 2021; Jiang, Rollinson, Plonsky, et al., 2021).2 One exception is
Loewen et al.’s (2019) investigation of Duolingo users’ overall L2 Turkish profi-
ciency along with five subareas (listening, speaking, writing, reading, and lexi-
cogrammar); although comprehensive and informative, this study did not involve
a control or comparison group and included only nine participants. Clearly, the
field stands to benefit from more research on L2 learners’ overall proficiency in
the domain of mobile-assisted language learning.

2.4 L2 grit and motivation

Although research on individual differences has long established its niche in SLA
overall (Gass et al., 2020) and computer-assisted language learning in particular
(see Pawlak & Kruk, 2022), much remains uncertain about the role of individ-
ual differences in mobile-assisted language learning. For example, Loewen et al.
(2019) raised concerns over participants’ attitudes to some app-related features
(e.g., lack of interaction and limited variation in Duolingo tasks), which might
have affected learners’ motivation and persistence in language app use. In the
same vein, García Botero et al. (2019) noted inconsistencies between Duolingo
learners’ questionnaire responses, which pointed to students’ motivation in and
positive attitude towards using the app out of class, and their interview data,
which demonstrated students’ mixed views on engagement and lack of long-term
interest while using the app.

2. The first three publications are white papers published by Duolingo; Jiang, Rollinson,
Plonsky, et al., 2021 is a peer-reviewed article.
8 Ekaterina Sudina and Luke Plonsky

To investigate these issues further, research into mobile-assisted language

learning would benefit from examining language app users’ academic persever-
ance, or grit, as well as their motivated learning behavior. Grit has been defined as
“perseverance and passion for long-term goals” (Duckworth et al., 2007, p. 1087);
some researchers have also conceptualized it as a facet of the personality trait of
conscientiousness (Park et al., 2018; Schmidt et al., 2018). Nonetheless, it becomes
increasingly common in SLA research to conceptualize and measure grit as a
domain-specific construct by tailoring scale items and instructions to a specific
language learning context (see Teimouri et al., 2021 for more). Indeed, a growing
body of research into L2 grit has found evidence of a positive relationship between
language-domain-specific grit and achievement (e.g., Sudina & Plonsky, 2021a;
Teimouri et al., 2022).
One of the constructs that is conceptually related to L2 grit is intended effort,
defined as “the amount of time, effort, and energy L2 learners expend in the
process of L2 learning” (Teimouri, 2017, p. 686) and recently reconceptualized
into current L2 motivated learning behavior by Papi et al. (2019). This was done
in order to avoid bias in favor of promotion-focus rather than prevention-focus
learners and tap into learners’ actual motivational behavior rather than their
hypothetical disposition (see Papi et al., 2019).

2.5 The Duolingo course structure

All Duolingo courses are aligned with the Common European Framework of Ref-
erence (CEFR). Both French and Spanish courses start with a brief Intro section
(also known as A1.0, where A corresponds to the beginner or Basic User level;
see Council of Europe, 2001) followed by the A1 content, which has two sections
(A1.1 and A1.2) and covers both communicatively functional as well as grammati-
cal topics, and the A2 content, which also consists of two sections (A2.1 and A2.2)
and covers more advanced vocabulary and grammar. The last section of each
Duolingo course includes B1 content, where B corresponds to the intermediate or
Independent User level; see Council of Europe, 2001). The B1 content has four
sections (B1.1 through B1.4), at the end of which language learners are expected to
have mastered even more advanced communicatively functional and grammatical
topics (e.g., “World news,” “Learning,” “Subjunctive with common conjunctions,”
and “Past conditional” for French; “World news,” “Gossip,” “Imperfect subjunc-
tive,” and “Passive” for Spanish). In addition to allowing for multiple “oppor-
tunities for practice and repeated exposure to target language structures,” the
Duolingo courses combine “more implicit, comprehension‐based learning with
explicit feedback and explanations” (Jiang, Rollinson, Plonsky, et al., 2021, p. 981).
Notably, Duolingo encourages a high “degree of user autonomy in navigating the
 The effects of frequency, duration, and intensity on L2 learning through Duolingo 9

platform,” which translates into “substantial variation among individual learners

on both the percentage of content they complete before reaching the end” of the
B1.4 level and “on the total amount of time spent learning” (Jiang, Chen, et al.,
2021, p. 2).

2.6 The present study

Expanding on previous research on distributed learning in SLA, the present study

in the form of a natural experiment addressed the following research questions
(RQ) concerning Duolingo effectiveness and L2 development more generally:
RQ1. To what extent do learner gains differ when tested in the written vs. oral
mode?
RQ2. To what extent are frequency, duration, and intensity of Duolingo app
usage associated with gains in L2 Spanish and French?
RQ3. To what extent are L2 grit and motivation associated with the frequency,
duration, and intensity with which learners use Duolingo?
RQ4. To what extent are L2 grit and motivation associated with gains in L2
Spanish and French?
The present study meets the criteria for a natural experiment due to (a) “a clearly
identified intervention” which was not rigidly controlled as the goal was to exam-
ine the effects of learners’ self-determined exposure and in-app behavior, (b) a
lack of random assignment to intervention, and (c) a pre-posttest study design
(Craig et al., 2017, p. 19).

3. Method

3.1 Participants

In the Fall of 2021, a group of 787 participants studying Spanish (k = 406) or

French (k = 381) on Duolingo were invited to participate and completed a pretest
(completion rate = 34%). They were recruited at the beginning of the A1.2 section
among beginner-level learners (see The Duolingo Course Structure in the Liter-
ature Review section). More specifically, the participants were at Row 18 of the
French course tree structure (out of a total of 106 rows) and Row 21 of the Span-
ish course tree structure (out of a total of 121 rows), respectively. This suggests
that the Duolingo course participants were already familiar with the basics (e.g.,
family and travel-related vocabulary and expressions, the present tense) as well
10 Ekaterina Sudina and Luke Plonsky

as slightly more advanced topics (e.g., shopping and routines-related vocabulary,

grammatical agreement).
Six months later, in the Spring of 2022, a group of 288 participants completed
the posttest (out of a total of 427 of those who met the selection criteria and eli-
gibility requirements, see Procedure; response rate = 67%). One participant was
excluded due to completing a pretest in Spanish and a posttest in French. There-
fore, the final sample comprised 287 participants (Spanish: k = 148; French:
k = 139; age: M = 44.01, SD = 14.44, range: 19−77; gender: 61% female; 38% male;
1% other). Although all participants in the final sample were L1 English speakers,
10% of the respondents reported having been exposed to one or more other
languages at home in early childhood. Participants’ demographic and language-
related characteristics by group are summarized in Table 1.

Table 1. Participant characteristics

French Spanish
(k = 139) (k = 148)
Characteristic k % k %
Age
Mean 44.24 43.78
SD 14.53 14.40
Range 20–77 19–74
Gender
Male 49 35 59 40
Female 90 65 85 57
Other 4 3
Native language(s)
English 121 87 137 93
English + Other(s) 18 13 11 7
Other languages spoken (excluding French/Spanish)
No 90 65 110 74
Yes 49 35 39 26
Reasons for learning French/Spanish a
For travel 84 24 70 18
For school 4 1 7 2
For job-related purposes 18 5 40 10
For fun/leisure 121 34 112 28
For memory/brain acuteness 79 22 77 19
For social purposes 31 9 73 18
Other 16 5 21 5
 The effects of frequency, duration, and intensity on L2 learning through Duolingo 11

Table 1. (continued)
French Spanish
(k = 139) (k = 148)
Characteristic k % k %
Other languages studied (excluding French/Spanish)
No 31 22 34 23
Yes 108 78 114 77
Education b
Some high school 1 1
High school 12 8 10 7
Associate’s degree 4 3 10 7
Bachelor’s degree 55 38 58 39
Master’s degree 40 28 55 37
Ph.D. 18 13 6 4
Trade School 4 3 2 1
Other 11 8 7 5
Ethnicity
Asian 7 5 4 3
African American 4 3 7 5
Caucasian 117 84 126 85
Latino or Hispanic 7 5 7 5
Other 4 3 4 3
Self-rated level of French/Spanish when started using the
app c
Mean 2.38 2.52
SD 1.46 1.53
Self-rated level of French/Spanish at pretest c
Mean 4.18 4.29
SD 1.31 1.56
Self-rated level of French/Spanish at posttest d
Mean 4.43 4.44
SD 1.37 1.43
Skills learned through Duolingo the most e
Vocabulary 112 19 120 20
Grammar 87 15 102 17
Pronunciation 71 12 56 9
Listening 96 16 87 14
12 Ekaterina Sudina and Luke Plonsky

Table 1. (continued)
French Spanish
(k = 139) (k = 148)
Characteristic k % k %
Speaking 55 9 56 9
Reading 104 18 114 19
Writing 69 12 80 13
Using Duolingo resources other than regular lessons f
Stories 110 49 111 46
Podcasts 36 16 34 14
Tips 64 29 73 30
Nothing 14 6 25 10
Experience learning French/Spanish before using
Duolingo
No 42 30 32 22
Yes 97 70 116 78
Ways of learning French/Spanish before using Duolingo g
Being around native speakers 17 9 40 18
High school classes 65 36 79 36
Language apps 41 23 46 21
Internet-based materials (e.g., podcasts, YouTube) 9 5 11 5
Textbooks and other materials in print 3 2 4 2
Conversational language classes 25 14 22 10
Other 21 12 20 9
Taking French/Spanish classes in addition to Duolingo
No 134 96 140 95
Yes 5 4 8 5
Using other programs/apps in addition to Duolingo
No 116 83 130 88
Yes 23 17 18 12
Notes.
a. French: k = 353, Spanish: k = 400.
b. French: k = 144, Spanish: k = 149.
c.d 0 = Absolute beginner; 10 = Native speaker.
e. French: k = 594, Spanish: k = 615.
f. French: k = 224, Spanish: k = 243.
g. French: k = 181, Spanish: k = 222.
 The effects of frequency, duration, and intensity on L2 learning through Duolingo 13

3.2 Instruments and materials

Three different types of data and corresponding data sources were used in the
study.

1. Exposure/behavioral data
The first data source shed light on learners’ exposure to the target language and
related behavior (in-app engagement). Of particular interest was (a) the duration
of app usage measured as total minutes per participant across the 6-month period
of study (i.e., “Minutes”), (b) the number of times the learner opened the app
in a given week (i.e., “Logins”) and the number of days the learner completed at
least one lesson (i.e., “Sessions”) – two frequency measures, and (c) the following
content-related/curriculum-oriented intensity variables: “Lessons” (i.e., the num-
ber of lessons completed), “Level reviews” (i.e., the final lesson for a given Level/
Skill combination), “Skill practice” (i.e., when a learner goes back to review skills
that they have already “gilded”), “Stories” (i.e., the number of stories completed),
and “Tests” (i.e., the number of tests completed). All of these indicators were used
in their raw forms as predictors of learner gains.

2. Self-report data
To understand learner demographics as well as participants’ language learning
history, an instrument that largely mirrored Jiang, Rollinson, Plonsky, et al.’s
(2021) background questionnaire was used. Additionally, we collected data using
scales for measuring two individual difference variables: L2 grit (adapted from
Teimouri et al., 2022) and L2 motivated learning behavior (adapted from Papi
et al., 2019). These variables, individually and in tandem, allowed the study to
examine these two individual differences as additional predictors of both in-app
engagement and gains in learning.
Teimouri et al. (2022) validated their instrument with a sample of 191 learners
of English in a foreign language context (Iran) and reported Cronbach’s α relia-
bility of .80 for the full L2 grit scale, .86 for the perseverance of effort subscale
(PE, five positively keyed items), and .66 for the consistency of interest subscale
(CI, four negatively keyed items). Papi et al.’s (2019) L2 motivated learning behav-
ior questionnaire (five positively keyed items) was first used with a sample of
257 learners of English in a second language context (the US) and had internal
consistency-reliability of .86 as measured by Cronbach’s α.
Both scales were employed after implementing minor adjustments. Specifi-
cally, the word English in the original scales was replaced with French or Spanish
in the present study in order to tailor the item wording to participants’ target lan-
guages. Additionally, for the sake of consistency, Papi et al.’s (2019) original 5-point
14 Ekaterina Sudina and Luke Plonsky

Likert-type scale (endpoints: 1 = never true of me; 5 = always true of me) was
replaced with Teimouri et al.’s (2022) 5-point fully verbal and numerical Likert-
type scale (endpoints: 1 = not like me at all; 5 = very much like me). An example
item for L2 grit: “I am a diligent French/Spanish learner”; an example item for L2
motivated learning behavior: “I work hard at studying French/Spanish.”

3. L2 proficiency
Two different types of language tests were used to measure participants’ L2 profi-
ciency: A C‑test (Spanish: Riggs & Maimone, 2018; French: Counsell, 2018) and
an elicited imitation test (EIT; Spanish: Solon et al., 2019; French: Gaillard &
Tremblay, 2016). These instruments were chosen based on a number of considera-
tions. First, Spanish and French C‑tests and EITs have undergone rigorous devel-
opment and possess strong validity arguments. To illustrate, Riggs and Maimone
(2018) reported a high correlation between Spanish C‑test scores and (a) self-
assessed proficiency (r = .81, p < .001) as well as (b) class level (r = .73, p < .001).
Counsell (2018) also reported sizeable and positive correlations between French
C‑test scores and (a) self-assessed proficiency (rs = .58–.67 for reading, writing,
listening, and speaking, respectively, with an overall r = .63, p < .01) and (b) pro-
gram level of study at the university (r = .85, p < .01). In the same vein, Gaillard and
Tremblay (2016) found that the strongest predictors of French EIT ratings in their
study were (a) C‑test scores (R2 = .79) and (b) class level (R2 = .69). Solon et al.
(2019), in turn, suggested that “the modified, 36-item EIT is, in fact, better able
to discriminate among learners at higher levels of proficiency than is the 30-item
EIT” (p. 14) and reported Cronbach’s α reliability ranging from .78 to .97 for the
30-item EIT and from .84 to .97 for the 36-item EIT for L2 learners at different
proficiency levels. Another benefit of using C‑tests and EITs is that they are con-
sidered to be good measures of explicit and implicit L2 knowledge, respectively
(e.g., Ellis, 2005; Heo, 2016).3
Moreover, the validity of these two groups of tests is supported not only by
primary studies but also by two recent meta-analyses. Synthesizing results across
239 studies, McKay (2019) found an almost perfect correlation between C‑tests
and tests of general language proficiency (r = .94). The evidence for EITs is like-
wise very strong. Kostromitina and Plonsky’s (2022) meta-analysis observed an
attenuation corrected correlation of r = .81 between EITs and other largely stan-
dardized tests of L2 proficiency.
Second are a set of practical considerations. These proficiency measures are
highly efficient and can be completed independently and online in approximately
20−30 minutes. Upon completion, these tests can then be scored quickly and

3. We thank another anonymous reviewer for pointing this out.

 The effects of frequency, duration, and intensity on L2 learning through Duolingo 15

accurately. The instruments are also freely available and do not carry any propri-
etary restrictions.
Finally, both C‑tests and EITs have been developed and are available in a
range of languages (Arabic, Chinese, German, Japanese, Russian). Therefore, the
present study could be replicated in other L2s without changing this critical
design feature (i.e., the dependent measure). The instruments used to collect self-
report and L2 proficiency data are available in Appendix A.

3.3 Procedure

Following IRB approval, the self-report and L2 proficiency measures were pilot-
tested, and the data were collected using an online survey platform Gorilla
(https://gorilla.sc). Eligible app users (see above) were invited to be part of the
study starting on August 3, 2021; the first round of data collection lasted until
November 5, 2021. Those who expressed interest were asked to begin the study by
completing an online survey that included a consent form followed by (a) instru-
ments for L2 grit and motivation (items for each scale were randomized to control
for order effect), (b) a language background questionnaire, (c) an EIT, and (d) a
C‑test for their chosen language (Spanish or French). This battery of instruments
was completed remotely, without a proctor, in about an hour. Upon completion,
participants were reminded of the minimum app engagement required for partic-
ipation (i.e., at least 2 logins to the app per week for the following 26 weeks).
Six months later (i.e., in February through May 2022), each participant who
had met the eligibility requirement was contacted again and invited to retake
the two individual difference scales as well as the two proficiency tests. Partici-
pants who met the selection criteria and the eligibility requirements received a
$ 100 Amazon gift card. The selection criteria included: (a) being a Duolingo user
studying either Spanish or French and (b) being a native speaker of English resid-
ing in the US. The requirements included: (a) completion of a survey and two
language tests (at the time of the pretest and posttest 6 months later) and (b) a
minimum of 52 logins on the Duolingo app (2 per week × 26 weeks) to ensure
minimally sufficient engagement with the target language and Duolingo content.
After all data were collected and de-identified, the C‑tests were scored auto-
matically, whereas the EITs were scored by four trained raters, all highly proficient
in the target language (two raters per language: the lead rater scored all of the
EIT items, whereas the second rater scored approximately 10% of the sample’s
EITs). Following rater training and norming sessions, which lasted approximately
two hours, the raters for each language (French vs. Spanish) got calibrated them-
selves and proceeded to independently score the EITs. To avoid potential rater
16 Ekaterina Sudina and Luke Plonsky

bias, raters were kept unaware of which audio files had come from the pretest and
which were from the posttest.

3.4 Data analysis

To calculate interrater reliability for the EIT scores, intraclass correlation coeffi-
cients (ICC, two-way mixed, consistency) were computed: (a) French: average α
for 28 items = .96; average by test type (14 items each): pretest = .95, posttest = .96;
(b) Spanish: average α for 30 items = .98; average by test type (15 items each):
pretest = .99, posttest = .97.
During the data clean-up, 30 cases (10% out of a total of 287; 17 Spanish,
13 French) were excluded listwise due to issues with recordings or participants’
misinterpretation of the task (several produced English translations rather than
French/Spanish imitations). There were no missing data on other variables.
To compare proficiency scores across languages and proficiency test modes
(written vs. oral), they were first converted to decimals separately by language.
Spanish EIT: 36 items, max possible score = 144 (4 per item). French EIT: 50
items, max possible score = 200 (4 per item). C‑test (both languages): 125 items,
max possible score = 125 (1 per item). Next, the assumptions for each statistical
analysis were checked and met (see Appendix B).

4. Results

4.1 Preliminary analyses: Scale data

The inspection of the scale data revealed two items with low corrected item-total
correlations (ITCs < .40) on the L2 Grit Consistency of Interest subscale: CI7R
and CI8R, which considerably affected reliability of the scale and were, therefore,
removed from further analyses. The rest of the corrected ITCs for all constructs
and subconstructs were > .40 on both the pretest and the posttest. The stabil-
ity of constructs over time was assessed by test-retest reliability: r(L2 grit) = .68;
r(L2 perseverance of effort) = .69; r(L2 consistency of interest) = .58; r(L2 motiva-
tion) = .61, p < .001. As demonstrated in Table 2, internal-consistency reliability of
the scales was also acceptable. Additionally, descriptive statistics indicated that the
participants had the highest mean score on L2 consistency of interest and the low-
est mean score on L2 motivation on both the pretest and the posttest.
 The effects of frequency, duration, and intensity on L2 learning through Duolingo 17

Table 2. Reliability analyses for scales (N = 287)

M SD α 95% CI M SD α 95% CI
Variable k Pretest Posttest
L2 grit 7 3.96 0.62 .84 [.81, .87] 3.79 0.66 .87 [.84, .89]
L2 perseverance of effort 5 3.71 0.73 .86 [.83, .88] 3.57 0.72 .86 [.83, .88]
L2 consistency of interest 2 4.58 0.63 .78 a [.72, .82] 4.35 0.77 .81 b [.76, .85]
L2 motivation 5 3.55 0.77 .88 [.86, .90] 3.30 0.81 .89 [.87, .91]

Notes.
a. k = number of items; M = mean; SD = standard deviation; 95% CI = 95% confidence intervals of
coefficient alphas; L2 = second language. Spearman’s rho = .59
b. k = number of items; M = mean; SD = standard deviation; 95% CI = 95% confidence intervals of
coefficient alphas; L2 = second language. Spearman’s rho = .65

4.2 Preliminary analyses: Proficiency data

As shown in Table 3, the participants’ mean scores on the EIT (i.e., oral profi-
ciency test) were higher in Spanish than in French, which was observed during
both the pretest and the posttest. However, Spanish EIT scores were more spread
out, as demonstrated by higher standard deviations. The participants’ average
C‑test scores (i.e., written proficiency test) were overall slightly higher than the
EIT scores, except for Spanish pre-test mean scores, which were virtually the
same in both modes. Nonetheless, the two groups’ proficiency in the written
mode appeared to be at about the same level on both the pretest and the posttest
(see Figure 1). The results of dependent-samples t-tests showed that learner profi-
ciency gains from the pretest to the posttest were significant, with small effect sizes
adjusted for the within-sample correlation (Plonsky & Oswald, 2014): (a) EIT
gains: t(256) = 13.38, p < .001, Cohen’s d = .29, 95% CI [.24, .33]; (b) C‑test gains:
t(286) = 11.00, p < .001, Cohen’s d = .36, 95% CI [.29, .43].
18 Ekaterina Sudina and Luke Plonsky

Table 3. Descriptive statistics for proficiency test scores (N = 287)

Proficiency Tests M SD Min Max
EIT pretest French .12 .08 .005 .47
EIT posttest French .17 .11 .03 .57
EIT pretest Spanish .27 .15 .02 .77
EIT posttest Spanish .30 .16 .01 .84
C‑test pretest French .28 .13 .02 .68
C‑test posttest French .32 .14 .00 .78
C‑test pretest Spanish .27 .12 .00 .68
C‑test posttest Spanish .32 .12 .07 .75

Notes. M = mean; SD = standard deviation. EIT (elicited imitation test) = oral proficiency;
C‑test = written proficiency.

Figure 1. Proficiency test gains in the written vs. oral mode and by language

4.3 RQ1: To what extent do learner gains differ when tested in the written
vs. oral mode?

The results of dependent-samples t-tests for RQ1 demonstrated that there were no
statistically significant differences in learner proficiency gains in the written vs.
oral mode (see also Figure 1), and the effect sizes adjusted for the within-sample
correlation were small (see Plonsky & Oswald, 2014): (a) French: t(125) = −.71,
p = .48, Cohen’s d = −.07, 95% CI [−.27, .13]; (b) Spanish: t(130) = 1.23, p = .22,
Cohen’s d = .13, 95% CI [−.08, .34].
As shown in Figure 1, the results of the independent-samples t-tests for RQ1
showed that the EIT gains (oral mode) did not statistically differ by language:
t(204.64) = −.85, p = .397, Cohen’s d = −.11, 95% CI [−.35, .14]. The same was true for
the C‑test gains (written mode) by language: t(285) = 1.52, p = .13, Cohen’s d = .18,
 The effects of frequency, duration, and intensity on L2 learning through Duolingo 19

95% CI [−.05, .41]. Of note, similar results were observed when the two tests were
re-run without outliers on the dependent variable.

4.4 RQ2: To what extent are frequency, duration, and intensity of Duolingo
app usage associated with gains in L2 Spanish and French?

A summary of descriptive statistics for the Duolingo app usage data is demon-
strated in Table 4. It shows meaningful differences among the frequency, duration,
and intensity variables. (A slight overlap between the two frequency variables was
addressed when conducting a follow-up multiple regression analysis, see Table 6.)

Table 4. Descriptive statistics for in-app usage data

Type App usage M SD Min Max
Frequency Logins 164.1 13.57 105 180
Sessions 154.7 18.82 50 180
Duration Minutes 3070 2812 252.7 22,753.8
Intensity: Content and Lessons 527.0 1395 9.0 19,701
Curriculum Level 45.06 46.81 0 309
reviews
Skill practice 93.22 157.4 0 1434
Stories 170.9 809.3 0 13,346
Tests 19.99 85.74 0 854

As demonstrated in Table 5, learners’ self-determined exposure and behavior

(i.e., frequency, duration, and intensity of Duolingo app usage) were weakly to
moderately correlated with their language proficiency gains made over a six-
month period. More specifically, the exposure-related variables most strongly
associated with gains measured in the oral model (i.e., via EIT) were the number
of lessons, level reviews, and logins. Raw duration measured in total minutes of
exposure during the 6-month period of study exhibited almost no association
with gains in the oral mode. As with EIT gains, in-app exposure/behaviors associ-
ated with gains in proficiency in the written mode included the number of lessons
completed and total number of level reviews (although the latter association was
neither particularly strong nor statistically significant, as shown by a confidence
interval that crossed zero). Unlike EIT gains, however, minutes of exposure was
found to be associated with C‑test gains.
While addressing RQ2, it occurred to us that some of the observed correla-
tions may have been artificially attenuated due to a restricted range of observed
20 Ekaterina Sudina and Luke Plonsky

values. In particular, requiring at least two logins per week, though necessary to
ensure regular exposure to the target language, seems to have yielded an unusual
level of homogeneity in login data across the sample: The mean number of logins
was 165 (6.35 logins per week) with a standard deviation of only 11.30. The rela-
tionships between logins and gains on the EIT and C‑test were then re-examined
using Thorndike’s formula for correction for range restriction, resulting in sub-
stantially larger correlations in both cases (i.e., .46 and .11, respectively).

Table 5. Pearson correlations between Duolingo app usage and proficiency gains
(N = 233)
Type App usage EIT gains C‑test gains
Frequency Logins a r = .14, BCa 95% CI [.03, .26] * r = .03, BCa 95% CI [−.09, .17]
Sessions r = .12, BCa 95% CI [.00, .24] r = .06, BCa 95% CI [−.07, .19]
Duration Minutes r = .01, BCa 95% CI [−.16, .18] r = .20, BCa 95% CI [.05, .35]
Intensity: Content Lessons r = .21, BCa 95% CI [.07, .35] r = .26, BCa 95% CI [.13, .40]
and Curriculum Level reviews r = .21, BCa 95% CI [.07, .34] r = .11, BCa 95% CI [−.04, .27]
Skill practice r = −.06, BCa 95% CI [−.19, .07] r = .08, BCa 95% CI [−.04, .22]
Stories r = −.02, BCa 95% CI [−.16, .14] r = .07, BCa 95% CI [−.04, .18]
Tests r = .09, BCa 95% CI [−.02, .25] r = .09, BCa 95% CI [−.06, .22]

Notes.
* Bias-corrected and accelerated 95% confidence intervals for correlation coefficients.
a. When adjusted for range restriction, the correlations for logins x EIT (r = .14) and logins x C‑test
gains (r = .03) increase to .46 and .11, respectively.

The results of the first multiple regression analysis suggested that when the
three variables most strongly associated with EIT gains were entered into the
model as predictors along with the target language, which was added as a covari-
ate, the model explained 4−6% of the variance in EIT gains and was statistically
significant: F(4, 238) = 3.56, p = .008, R2 = .06, adjusted R2 = .04. ‘Level Reviews’
emerged as the only meaningful positive predictor (see Table 6).4
The results of the second multiple regression analysis demonstrated that
when the two variables most strongly correlated with C‑test gains were entered
in the model as predictors along with the target language, which was added as a
covariate, the model explained 5−6% of the variance in C‑test gains and was sta-
tistically significant: F(3, 257) = 5.21, p = .002, R2 = .06, adjusted R2 = .05. ‘Lessons’
emerged as the only meaningful positive predictor (see Table 7).

4. Of note, the Sessions variable was excluded from the model due to a large correlation with
the Logins variable.
 The effects of frequency, duration, and intensity on L2 learning through Duolingo 21

Table 6. Regression analysis summary for variables predicting EIT gains

Variable B SE β t p
Target language −.008 .006 −.095 −1.49 .14
Logins .000 .000 .059 .887 .38
Lessons 8.820e-6 .000 .051 6.58 .51
Level reviews .000 .000 .158 2.052 .04

Notes. N = 243. Overall model: R2 = .06, adjusted R2 = .04.

Table 7. Regression analysis summary for variables predicting C‑test gains

Variable B SE β t p
Target language .012 .007 .101 1.651 .10
Minutes 7.784e-7 .000 .024 .255 .80
Lessons 5.152e-5 .000 .204 2.127 .03

Notes. N = 261. Overall model: R2 = .06, adjusted R2 = .05.

4.5 RQ3: To what extent are L2 grit and motivation associated with the
frequency, duration, and intensity with which learners use Duolingo?

As shown in Table 8, the individual difference variables of L2 grit and motivation

measured at the pretest were weakly to moderately correlated with learners’ fre-
quency, duration, and intensity of the Duolingo app usage. Although the mag-
nitude of effect sizes was small, meaningful positive relationships were observed
between the two frequency variables (i.e., “Logins” and “Sessions”) and L2 grit
perseverance of effort as well as L2 motivation. Additionally, duration (i.e., “Min-
utes”) and content-related intensity (i.e., “Lessons” and “Stories”) were positively
correlated with the two subcomponents of L2 grit as well as L2 motivation.

Table 8. Pearson correlations between Duolingo app usage and individual differences
(N = 260)
L2 Perseverance L2 Consistency
Type App usage of effort of interest L2 Motivation
Frequency Logins r = .16 [.06, .27] * r = .08 [−.05, .20] r = .20 [.08, .31]
Sessions r = .15 [.04, .26] r = .12 [−.02, .23] r = .20 [.07, .31]
Duration Minutes r = .18 [.05, .30] r = .13 [.02, .22] r = .24 [.12, .35]
22 Ekaterina Sudina and Luke Plonsky

Table 8. (continued)
L2 Perseverance L2 Consistency
Type App usage of effort of interest L2 Motivation
Intensity: Content Lessons r = .17 [.02, .31] r = .13 [.03, .22] r = .22 [.09, .33]
and curriculum Level reviews r = −.002 [−.11, .11] r = .01 [−.11, .11] r = .07 [−.05, .18]
Skill practice r = .003 [−.12, .13] r = −.01 [−.13, .11] r = .03 [−.10, .15]
Stories r = .16 [.03, .28] r = .12 [.02, .22] r = .17 [.05, .27]
Tests r = .09 [−.04, .20] r = .03 [−.09, .13] r = .06 [−.05, .15]

Note.
* Bias-corrected and accelerated 95% confidence intervals for correlation coefficients.

4.6 RQ4: To what extent are L2 grit and motivation associated with gains in
L2 Spanish and French?

Pearson correlations revealed the extent to which L2 grit and motivation mea-
sured at the pretest were associated with learner gains in written and oral profi-
ciency. The observed relationships were positive and constituted generally small
effect sizes (Plonsky & Oswald, 2014). As in response to some of the previously
conducted analyses, we applied a correction to the correlations that we had reason
to believe may have been attenuated due to range restriction. Specifically, the
standard deviations observed for both of the L2 grit subconstructs were substan-
tially smaller than in previous studies of foreign-language learners (e.g., Sudina
& Plonsky, 2021b) and were adjusted accordingly yet conservatively, shown in the
following results in parentheses following the corresponding uncorrected corre-
lations: r = .02, BCa 95% CI [−.10, .16] between EIT gains and L2 perseverance
of effort; r = .12 (rcorrected = .22), BCa 95% CI [−.01, .25] between EIT gains and L2
consistency of interest; r = .03, BCa 95% CI [−.10, .16] between EIT gains and L2
motivation; r = .16 (rcorrected = .19), BCa 95% CI [.04, .28] between C‑test gains and
L2 perseverance of effort (note that the confidence interval does not cross zero);
r = .09 (rcorrected = .17), BCa 95% CI [−.04, .23] between C‑test gains and L2 consis-
tency of interest; r = .17, BCa 95% CI [.05, .30] between C‑test gains and L2 moti-
vation (note that the confidence interval does not cross zero).
To examine the extent to which individual differences of L2 grit and moti-
vation predicted EIT/C‑test gains, two standard multiple regressions with four
predictors (i.e., L2 perseverance of effort, L2 consistency of interest, L2 motiva-
tion, and target language, which was added as a covariate) and EIT/C‑test gains
as outcome variables were performed. The two models explained 1−3% of the
variance in gains scores and were not statistically significant: (a) F(4, 244) = 1.76,
 The effects of frequency, duration, and intensity on L2 learning through Duolingo 23

p = .14, R2 = .03, adjusted R2 = .01 for EIT gains, with L2 consistency of interest
as the only contributing predictor: β = .14, B = .01, 95% CI [.00, .03], p = .05; (b)
F(4, 268) = 1.83, p = .12, R2 = .03, adjusted R2 = .01 for C‑test gains, with no meaning-
ful predictors.

5. Discussion

The current study sought to examine the predictive power of two sets of variables
on L2 gains made in app-based language learning via Duolingo. Specifically, we
were interested in better understanding L2 development as a function of both
(a) learners’ app-based exposure/behavior (e.g., instructional frequency, dura-
tion) as well as (b) learners’ L2 grit and motivation. On a broad, theoretical level,
these sets of variables represent the two main types of factors (learner-external
and learner-internal) known to influence L2 learning (Gass et al., 2020). On a
practical level, the results have the potential to inform the instructional design
of Duolingo’s curriculum and to provide implications for in-app experience that
increase learner efficiency.
The study is unique in at least two respects. First, to our knowledge, this is
the only study to consider distribution of practice effects in the context of mobile-
based language learning. Moreover, we have done so by means of a natural exper-
iment thereby greatly increasing the study’s ecological validity. Second, although a
growing body of evidence has begun to accumulate on the role of grit in L2 devel-
opment (see Teimouri et al., 2021), no study to date has done so with mobile lan-
guage learners. It is also the first study to employ a longitudinal design to examine
the power of grit in predicting gains over time.
One challenge to these goals, which we want to be upfront about, were the
relatively modest gains observed on both the written and oral proficiency tests
(i.e., C‑test and EIT) in both languages. The lack of target language gains that
were observed over time (i.e., our main dependent variable) imposed a limitation
on the study’s findings because less gains necessarily means less for the predictor
or independent variables to explain. These gains appear in conflict with previous
findings on the effectiveness of Duolingo (e.g., Jiang et al., 2021). However, there
are several alternate explanations. For example, unlike other standardized profi-
ciency tests (e.g., ACTFL’s Oral Proficiency Interview), the dependent measures
in the present study were not developed with lower proficiency levels in mind
and may have been too difficult, as noted to us by several participants. Another
explanation for the modest gains observed may be a lack of effort on pre- and
post-assessments on the part of learners. Finally, we need to account for the user
autonomy and the amount of the course content covered by the participants after
24 Ekaterina Sudina and Luke Plonsky

six months of learning. A follow-up analysis of participants’ maximum course

tree depth (i.e., the furthest row of each section of the courses the participants
were at) within seven days of completing the posttest suggested that our partici-
pants did not cover enough new content over six months of learning on Duolingo.
The results revealed that the majority of our learners remained at the beginning
level at the posttest based on the amount of material they covered. This finding,
along with considerable attrition rates in the present study, echoes Loewen et al.’s
(2019) observation that “learners may not persist long enough to make consid-
erable gains in their L2 knowledge, especially without any obligation or encour-
agement from peers and teachers commonly found in classroom environments”
(p. 308). In light of this finding, one suggestion for efficient use of Duolingo is
moving forward and learning new content despite the challenges along the way.
To increase motivation and accountability, Duolingo users might consider adding
and following friends and peers who are also using the app.
As stated above, one of our main interests in this study was to examine lan-
guage gains made in relation to learners’ in-app exposure (frequency, duration,
intensity) and associated behaviors (e.g., content-related choices). Somewhat sur-
prisingly, total minutes of exposure during the period of study only exhibited a
correlation with gains when measured with the C‑test. Rather, the more consistent
and dependable (i.e., across modes and dependent measures) correlates of gains
were the more frequency- and curriculum-oriented measures such as the number
of lessons, reviews, and logins.
This finding carries several important ramifications and interpretations. First,
these results generally confirm the lack of a relationship between hours of expo-
sure on Duolingo and language gains observed in Jiang et al. (2021). Such a find-
ing might be perceived as counter-intuitive or even disappointing in that greater
time spent by learners does not necessarily yield greater gains. However, we view
these findings more optimistically in that they indicate that gains can be made
even with shorter, more frequent and purposeful in-app engagement. Moreover,
the lack of a relationship between gains and time spent using the app is even more
noteworthy given the relative homogeneity of logins across the participants. In
other words, although the participants were free to log in as frequently as they
liked as long as they did so at least twice per week during the 26-week period of
study, most logged in on a daily or almost daily basis (>6 logins/week), thus pro-
viding a kind of built-in control for the role of frequency of exposure. This find-
ing also aligns well with Kim and Webb’s recent (2022) meta-analysis showing a
marked advantage for more frequent exposure to the target language as opposed
to longer (in minutes, hours) periods of exposure; in other words, the shorter
spacing gaps showed greater frequency effects. From a skill-acquisition perspec-
tive, this finding could be explained in terms of the intensity of exposure favoring
 The effects of frequency, duration, and intensity on L2 learning through Duolingo 25

proceduralization processes (see Serrano, 2011). Another study that demonstrated

a smaller frequency effect in spaced learning conditions compared to massed con-
ditions was Uchihara et al.’s (2019) meta-analysis of repetition effects in incidental
vocabulary learning. The major implication here for learners is fairly straightfor-
ward: Log in to the app frequently with the goal of completing entire lessons and
reviews, even if those sessions do not last for long periods of time.
Moreover, the fact that the total number of level reviews was positively corre-
lated with and emerged as a meaningful predictor of oral proficiency gains (RQ2)
deserves special attention. Based on the cognitive psychology definition of spac-
ing (Cepeda et al., 2008, see the literature review for more), a true spacing study
must involve the repetition of the same stimuli from the first study session to
the next. Level review appears to be where the participants return to review the
same content. Our findings indicate that engagement in level reviews is positively
related to and serves as a predictor of learning gains in the oral mode, which is in
line with the literature on distribution of practice effects (Nakata, 2015).
Research questions 3 and 4 both involved the two learner individual differ-
ences variables of L2 grit and motivation. RQ3 was concerned with the associa-
tion between these two variables and the learners’ in-app behaviors and exposure.
It is natural to expect that learners who possess higher levels of (L2) grit and/or
motivation would engage more often and/or more thoroughly with the Duolingo
curriculum. This supposition was indeed the case at least for some of the individ-
ual differences and for some of the learner behaviors. Not surprisingly, motiva-
tion exhibited some of the strongest associations with in-app exposure as well as
written proficiency gains; the latter is congruous with other app-based language
learning studies (e.g., Loewen et al., 2020 found that learner motivation was a sig-
nificant predictor of oral proficiency scores for learners who had studied Spanish
using Babbel for 12 weeks). However, we also observed meaningful and statis-
tically significant correlations between L2 grit and several in-app behaviors and
proficiency gains. In fact, L2 grit consistency of interest (and not motivation as in
Loewen et al., 2020) was the only contributing predictor of oral proficiency gains.
The findings for RQ3 are noteworthy on multiple levels. First of all, we have
to understand that, on a theoretical level, grit and motivation do not in and of
themselves induce greater learning. Rather, as this study shows, these qualities
are associated with the types of activities that lead to learning such as seeking
out target language interlocutors, spending time studying or reading in the tar-
get language, or – most pertinent to the present study – engaging with the tar-
get language instruction via a language learning app such as Duolingo. Thus, the
findings of the present study provide one of the first pieces of evidence of the pre-
dictive validity of the L2 grit scale of language learning behaviors. From an educa-
tional standpoint, these results demonstrate that greater motivation and grit may
26 Ekaterina Sudina and Luke Plonsky

lead to higher frequency of the types of activities shown in response to RQ2 to be

associated with language gains (see Figure 2). Duolingo may be interested, there-
fore, in seeking to foster learner motivation and grit as a means to enhance lin-
guistic development if only indirectly.

Figure 2. Hypothesized model of the effect of individual differences on app-based

learning, mediated by in-app exposure and engagement

RQ4 addressed the same relationship modeled in Figure 2 but without the
mediating effect of in-app exposure. The findings for this relationship were mod-
est but provide additional evidence of the predictive validity of L2 grit in the con-
text of app-based language learning.

6. Conclusion

The findings of this study carry relevance and potential benefits on multiple levels.
First, this study allowed us to gain a better understanding of the role of tech-
nology in instructed second-language acquisition. This is critical as technologi-
cal advances have the potential to make language learning not only “a lifelong
(spanning one’s lifetime) but also a lifewide (not confined to a particular loca-
tion, such as a school) activity” (Reinders & Stockwell, 2017, p. 372). Second, the
present investigation contributed to the growing line of evidence of Duolingo’s
effectiveness by assessing L2 learners’ proficiency in both written and oral modes
using high validity and high practicality measures. The study also shed further
light on our understanding of the individual and combined effects of frequency,
duration, and intensity of instruction on L2 development and, critically, on the
learner-internal factors that lead to choices to engage with the app. Finally, on a
practical level, the results of the present study may also inform Duolingo lesson
design and recommendations provided to learners with respect to the frequency,
duration, and intensity of app usage.

Funding

This article was made Open Access under a cc by 4.0 license through payment of an APC by or
on behalf of the authors.
 The effects of frequency, duration, and intensity on L2 learning through Duolingo 27

Acknowledgements

This project was funded by a Duolingo Efficacy Study grant. We are very grateful to the Learn-
ing Science team at Duolingo for seeing the value in this study and for all their support and
assistance. In particular we would like to thank Xiangying Jiang, Erin Gustafson, and Joseph
Rollinson for their patience, generosity, and help digging up the learner-usage data we needed
time and time again. We are also very grateful to the language learners who contributed their
time and energy (and data) to this study. In addition, our sincere thanks go to Kevin Hirschi,
Masha Kostromitina, Ben Brown, and Andrew Dennis, for their tireless assistance with scoring,
coding, piloting, and Gorilla-wrangling (the data collection platform, not the primate). Thanks
to Kate Yaw for help recording our test instructions. Last, our gratitude goes out to the C-test
and elicited imitation test authors who very kindly provided us with the materials needed to
employ their instruments in this study: Stéphanie Gaillard, Annie Tremblay, Corinne Counsell,
Daniel Riggs, Luciane L. Maimone, Megan Solon, Hae In Park, Carly Henderson, and Marzieh
Dehghan-Chaleshtori.

References

Carpenter, S. K. (2020). Distributed practice/spacing effect. In Li-fang Zhang (Ed.), Oxford

Research Encyclopedia of Education (pp. 1–20). Oxford University Press.
Cepeda, N. J., Vul, E., Rohrer, D., Wixted, J. T., & Pashler, H. (2008). Spacing effects in
learning: A temporal ridgeline of optimal retention. Psychological Science, 19(11),
1095–1102.
Council of Europe. (2001). Common European Framework of Reference for languages:
Learning, teaching, assessment. Retrieved from https://rm.coe.int/1680459f97
Counsell, C. L. (2018). The C‑test in French: Development and validation of a language
proficiency test for research purposes. In J. M. Norris (Ed.), Developing C‑tests for
estimating proficiency in foreign language research (pp. 203–230). Peter Lang.
Craig, P., Katikireddi, S. V., Leyland, A., & Popham, F. (2017). Natural experiments: An
overview of methods, approaches, and contributions to public health intervention
research. Annual Review of Public Health, 38, 39–56.
Delaney, P. F., Verkoeijen, P. P. J. L., & Spirgel, A. (2010). Spacing and testing effects: A deeply
critical, lengthy, and at times discursive review of the literature. In B. H. Ross (Ed.), The
psychology of learning and motivation: Advances in research and theory (Vol. 53,
pp. 63–147). Academic Press.
Duckworth, A. L., Peterson, C., Matthews, M. D., & Kelly, D. R. (2007). Grit: Perseverance and
passion for long term goals. Journal of Personality and Social Psychology, 92, 1087–1101.
Ellis, R. (2005). Measuring implicit and explicit knowledge of a second language: A
Psychometric Study. Studies in Second Language Acquisition, 27(2), 141–172.
Gaillard, S., & Tremblay, A. (2016). Linguistic proficiency assessment in second Language
acquisition research: The elicited imitation task. Language Learning, 66, 419–447.
García Botero, G., Questier, F., & Zhu, C. (2019). Self-directed language learning in a mobile-
assisted, out-of-class context: do students walk the talk? Computer Assisted Language
Learning, 32, 71–97.
28 Ekaterina Sudina and Luke Plonsky

Gass, S. M., Behney, J., & Plonsky, L. (2020). Second language acquisition: An introductory
course (5th ed.). Routledge.
Heo, Y. (2016). Heritage and L2 learners’ acquisition of Korean in terms of implicit and explicit
knowledge. PhD dissertation, Michigan State University, East Lansing.
Jiang, X., Chen, H., Portnoff, L., Gustafson, E., Rollinson, J., Plonsky, L., & Pajak, B. (2021).
Seven units of Duolingo courses comparable to 5 university semesters in reading and
listening. Duolingo Research Report DRR-21-03.
Jiang, X., Rollinson, J., Chen, H., Reuveni, B., Gustafson, E., Plonsky, L., & Pajak, B. (2021).
How well does Duolingo teach speaking skills? Duolingo Research Report DRR-21-02.
Jiang, X., Rollinson, J., Plonsky, L., & Pajak, B. (2020). Duolingo efficacy study: Beginning-
level courses equivalent to four university semesters. Duolingo Research Report
DRR-20-04.
Jiang, X., Rollinson, J., Plonsky, L., Gustafson, E., & Pajak, B. (2021). Evaluating the reading
and listening outcomes of beginning-level Duolingo courses. Foreign Language Annals,
54, 974–1002.
Kasprowicz, R. E., Marsden, E., & Sephton, N. (2019). Investigating distribution of practice
effects for the learning of foreign language verb morphology in the young learner.
Modern Language Journal, 103, 580–606.
Kim, S. K., & Webb, S. (2022). The effects of spaced practice on second language learning: A
meta-analysis. Language Learning, 72, 269–319.
Kostromitina, M., & Plonsky, L. (2022). Elicited imitation tasks as a measure of L2 proficiency:
A meta-analysis. Studies in Second Language Acquisition, 44, 886–911.
Li, M., & DeKeyser, R. (2019). Distribution of practice effects in the acquisition and retention
of L2 Mandarin tonal word production. Modern Language Journal, 103, 607–628.
Loewen, S. (2020). Introduction to instructed second language acquisition (2nd ed.). Routledge.
Loewen, S., Crowther, D., Isbell, D., Kim, K., Maloney, J., Miller, Z., & Rawal, H. (2019).
Mobile-assisted language learning: A Duolingo case study. ReCALL, 31(3), 293–311.
Loewen, S., Isbell, D., & Sporn, Z. (2020). The effectiveness of app-based language instruction
for developing receptive linguistic knowledge and oral communicative ability. Foreign
Language Annals, 53(2), 209–233.
McKay, T. (2019). More on the validity and reliability of C‑test scores: A meta-analysis of
C‑test studies (Unpublished doctoral dissertation). Georgetown University, Washington,
DC.
Nakata, T. (2015). Effects of expanding and equal spacing on second language vocabulary
learning: Does gradually increasing spacing increase vocabulary learning?. Studies in
Second Language Acquisition, 37, 677–711.
Nakata, T., & Suzuki, Y. (2019). Effects of massing and spacing on the learning of semantically
related and unrelated words. Studies in Second Language Acquisition, 41, 287–311.
Norris, J. M., & Ortega, L. (2000). Effectiveness of L2 instruction: A research synthesis and
quantitative meta-analysis. Language Learning, 50, 417–528.
Papi, M., Bondarenko, A., Mansouri, S., Feng, L., & Jiang, C. (2019). Rethinking L2 motivation
research: The 2 × 2 model of L2 self-guides. Studies in Second Language Acquisition, 41,
337–361.
 The effects of frequency, duration, and intensity on L2 learning through Duolingo 29

Park, D., Yu, A., Baelen, R. N., Tsukayama, E., & Duckworth, A. L. (2018). Fostering grit:
Perceived school goal-structure predicts growth in grit and grades. Contemporary
Educational Psychology, 55, 120–128.
Pawlak, M., & Kruk, M. (2022). Individual differences in computer assisted language learning
research. Routledge.
Plonsky, L., & Oswald, F. L. (2014). How big is “big”? Interpreting effect sizes in L2 research.
Language Learning, 64, 878–912.
Plonsky, L., & Ziegler, N. (2016). The CALL-SLA interface: Insights from a second-order
synthesis. Language Learning & Technology, 20, 17–37.
Reinders, H., & Stockwell, G. (2017). Computer-assisted SLA. In S. Loewen & M. Sato (Eds.),
The Routledge handbook of instructed second language acquisition (pp. 108–125).
Routledge.
Riggs, D., & Maimone, L. L. (2018). A computerized-administered C‑test in Spanish. In
J. M. Norris (Ed.), Developing C‑tests for estimating proficiency in foreign language
research (pp. 265–294). Peter Lang.
Rohrer, D. (2015). Student instruction should be distributed over long time periods.
Educational Psychology Review, 27(4), 635–643.
Rogers, J. (2017). The spacing effect and its relevance to second language acquisition. Applied
Linguistics, 38(6), 906–911.
Rogers, J. (2023). Spacing effects in task repetition research. Language Learning, 73(2),
445–474.
Rogers, J., & Cheung, A. (2020). Input spacing and the learning of L2 vocabulary in a
classroom context. Language Teaching Research, 24(5), 616–641.
Rogers, J., & Cheung, A. (2021). Does it matter when you review? Input spacing, ecological
validity, and the learning of L2 vocabulary. Studies in Second Language Acquisition, 43,
1138–1156.
Saito, K., & Plonsky, L. (2019). Measuring the effects of second language pronunciation
teaching: A proposed framework and meta-analysis. Language Learning, 69, 652–708.
Schmidt, F. T., Nagy, G., Fleckenstein, J., Möller, J., & Retelsdorf, J. (2018). Same same, but
different? Relations between facets of conscientiousness and grit. European Journal of
Personality, 32, 705–720.
Serrano, R. (2011). The time factor in EFL classroom practice. Language Learning, 61(1),
117–145.
Serrano, R. (2022). A state-of-the-art review of distribution-of-practice effects on L2 learning.
Studies in Second Language Learning and Teaching, 12(3), 355–379.
Serrano, R., & Huang, H. Y. (2018). Learning vocabulary through assisted repeated reading:
How much time should there be between repetitions of the same text? TESOL Quarterly,
52(4), 971–994.
Serrano, R., & Huang, H. Y. (2023). Time distribution and intentional vocabulary learning
through repeated reading: A partial replication and extension. Language Awareness, 32(1),
1–18.
Solon, M., Park, H. I., Henderson, C., & Dehghan-Chaleshtori, M. (2019). Revisiting the
Spanish elicited imitation task: A tool for assessing advanced language learners? Studies
in Second Language Acquisition, 41, 1027–1053.
30 Ekaterina Sudina and Luke Plonsky

Sudina, E., & Plonsky, L. (2021a). Academic perseverance in foreign language learning: An
investigation of language-specific grit and its conceptual correlates. Modern Language
Journal, 105, 829–857.
Sudina, E., & Plonsky, L. (2021b). Language learning grit, achievement, and anxiety among L2
and L3 learners in Russia. ITL – International Journal of Applied Linguistics, 172, 161–198.
Suzuki, Y. (2017). The optimal distribution of practice for the acquisition of L2 morphology: A
conceptual replication and extension. Language Learning, 67(3), 512–545.
Suzuki, Y. (2019). Individualization of practice distribution in second language grammar
learning: The role of metalinguistic rule rehearsal ability and working memory capacity.
Journal of Second Language Studies, 2(2), 169-196.
Suzuki, Y., & DeKeyser, R. (2017). Effects of distributed practice on the proceduralization of
morphology. Language Teaching Research, 21(2), 166–188.
Teimouri, Y. (2017). L2 selves, emotions, and motivated behaviors. Studies in Second Language
Acquisition, 39, 681–709.
Teimouri, Y., Sudina, E., & Plonsky, L. (2021). On domain-specific conceptualization and
measurement of grit in L2 learning. Journal for the Psychology of Language Learning, 3,
156–164.
Teimouri, Y., Plonsky, L., & Tabandeh, F. (2022). L2 Grit: Passion and perseverance for
second-language learning. Language Teaching Research, 26, 893–918.
Uchihara, T., Webb, S., & Yanagisawa, A. (2019). The effects of repetition on incidental
vocabulary learning: A meta-analysis of correlational studies. Language Learning, 69(3),
559–599.
Yamagata, S., Nakata, T., & Rogers, J. (2023). Effects of distributed practice on the acquisition
of verb-noun collocations. Studies in Second Language Acquisition, 45, 291–317.

Appendix A. The full instrument used in the study

(After signing the consent form). Thank you! Please set aside 60 minutes of undisturbed time
to participate in the study. You should have access to a computer with a working microphone
and speakers and reliable internet service. Please, do not use a mobile phone.

Thank you! Please provide the information below:

Language learned on Duolingo: French/Spanish

Now think about your foreign language learning experience on Duolingo (in your case, French/
Spanish) and respond to the following items by selecting the statements that best describe you.
This is not a test and there are no right or wrong answers, so please be honest.

L2-GRIT SCALE
(adapted from Teimouri et al., 2022)

Perseverance of effort
1. I will not allow anything to stop me from making progress in learning French/Spanish*.
2. I am a diligent French/Spanish learner.
 The effects of frequency, duration, and intensity on L2 learning through Duolingo 31

3. Now that I have decided to learn French/Spanish, nothing can prevent me from reaching
this goal.
4. When it comes to French/Spanish, I am a hard-working learner.
5. I put much time and effort into improving my weaknesses in learning French/Spanish.

Consistency of interest
6R. I think I have lost my interest in learning French/Spanish.
7R. I have been obsessed with learning French/Spanish in the past but later lost interest.
8R. My interests in learning French/Spanish change from year to year.
9R. I am not as interested in learning French/Spanish as I used to be.

Note. ‘R’ indicates negatively keyed items that have been reversed.
*The original scale was developed for L2 English, which in the present study was replaced
with French/Spanish.

L2 motivated learning behavior

(adapted from Papi et al., 2019)

1. I work hard at studying French/Spanish *.

2. I spend a lot of time studying French/Spanish.
3. I put a lot of effort in studying French/Spanish.
4. I constantly think about my French/Spanish learning activities.
5. Studying French/Spanish is very important to me these days.

Note. * The word English in the original scale was replaced with French/Spanish in the present
study.

Response options and scoring

Not like me at all Not much like me Somewhat like me Mostly like me Very much like me
1 2 3 4 5

Note. The mean scores on each scale indicate the levels of L2 grit and L2 motivated learning behavior,
respectively. The items were randomized.

Background questionnaire
(based on Jiang et al., 2021)
This form asks for background information about you. Although we ask for your name and
email, we do so only because we want to associate your answers to this questionnaire with your
other data. Your answers will be treated confidentially. Only the researchers will have access to
the information you provide.

1. Name:
2. Email (the one that is associated with your Duolingo account):
3. Age (please put a number):
4. What language(s) was/were spoken in your home before you were 6 years old?
5. What other languages do you speak, if any?*
6. Why are you learning French/Spanish? (Check all that apply)
For travel For school For job-related purposes For fun/leisure
For memory/brain acuteness For social purposes Other (please specify)
32 Ekaterina Sudina and Luke Plonsky

7. What other languages have you studied?

8. What is your highest level of education?
Some high school High school Associate’s degree Bachelor’s degree
Master’s degree Ph.D. Trade school Other (please specify)
9. What gender do you identify as?
Male Female Other (please specify)
10. Please specify your ethnicity.
Caucasian African American Latino or Hispanic
Asian Other (please specify)
11. How much French/Spanish do you think you knew on Duolingo when you started using
the app?
0 1 2 3 4 5 6 7 8 9 10
Absolute beginner Native speaker
12. How much French/Spanish do you think you know now?
0 1 2 3 4 5 6 7 8 9 10
Absolute beginner Native speaker
13. In which area(s) do you think Duolingo helped you the most? (Check all that apply)
Vocabulary Grammar Pronunciation Listening
Speaking Reading Writing
14. How much time (in hours) per week did you use Duolingo to learn French/Spanish?
15. In addition to the Duolingo French/Spanish lessons, what other Duolingo resources did
you use to learn the language? (Check all that apply)
Duolingo French/Spanish Stories
Duolingo French/Spanish Podcasts
Duolingo Tips in French/Spanish
Nothing else

16. What do you like about learning French/Spanish on Duolingo?

17. What do you want to see changed on Duolingo?
18. Did you have experience learning French/Spanish before using Duolingo? Yes/No
19. (If yes) How did you learn French/Spanish before using Duolingo? (Check all that apply)
Being around French/Spanish speakers
High school French/Spanish classes
College French/Spanish classes
Language apps
Internet-based materials such as podcasts and YouTube
Textbooks and other materials in print
Other (please specify)

20. Did you take French/Spanish classes during the time you used Duolingo? Yes/No
21. Did you use other programs or apps to learn French/Spanish during the time you used
Duolingo? Yes/No

Note. *Any other Ls, whether L1 or L2.

 The effects of frequency, duration, and intensity on L2 learning through Duolingo 33

L2 proficiency

Spanish elicited imitation test

(Solon et al., 2019)

Introduction
This language test will ask you to listen to several short audio files in Spanish and make a
recording in response. (Please be patient as recordings may take time to load.)
Note that some of the items might be quite challenging. Please try to complete each of them
to the best of your ability.
<Click here to start>
Instructions-1
You are going to hear several sentences in English (6 in total). After each sentence, there will be
a short pause, followed by a tone sound {TONE}. Your task is to try to repeat exactly what you
hear. You will have only one attempt to do so. You will be given sufficient time after the tone to
repeat the sentence. Repeat as much as you can. Remember, don’t start repeating the sentence
until after you hear the tone sound {TONE}. Now let’s begin.
<I’m ready>

Practice stimuli
1. We drove to the park.
2. I’ll call her tomorrow night.
3. You can buy meat at the butcher shop.
4. My brother just bought a brand new computer.
5. Sometimes they take their dog for a walk in the park.
6. We’re going to play volleyball at the gym that I told you about.

Instructions-2
Now, you are going to hear a number of sentences in Spanish (36 in total). Once again, after
each sentence, there will be a short pause, followed by a tone sound {TONE}. Your task is to try
to repeat exactly what you hear in Spanish. You will have only one attempt to do so. You will be
given sufficient time after the tone to repeat the sentence. Repeat as much as you can. Remem-
ber, don’t start repeating the sentence until after you hear the tone sound {TONE}. Now let’s
begin.

Main stimuli
1. Quiero cortarme el pelo. (7 syllables)
2. El libro está en la mesa. (7 syllables)
3. El carro lo tiene Pedro. (8 syllables)
4. Él se ducha cada mañana. (9 syllables)
5. ¿Qué dice usted que va a hacer hoy? (9 syllables)
6. Dudo que sepa manejar muy bien. (10 syllables)
7. Las calles de esta ciudad son muy anchas. (11 syllables)
8. Puede que llueva mañana todo el día. (12 syllables)
9. Las casas son muy bonitas pero caras. (12 syllables)
34 Ekaterina Sudina and Luke Plonsky

10. Me gustan las películas que acaban bien. (12 syllables)

11. El chico con el que yo salgo es español. (13 syllables)
12. Después de cenar me fui a dormir tranquilo. (13 syllables)
13. Quiero una casa en la que vivan mis animales. (14 syllables)
14. A ustedes les fascinan las fiestas grandiosas. (14 syllables)
15. Ella sólo bebe cerveza y no come nada. (15 syllables)
16. Me gustaría que el precio de las casas bajara. (15 syllables)
17. Cruza a la derecha y después sigue todo derecho. (15 syllables)
18. Ella ha terminado de pintar su apartamento. (14 syllables)
19. Me gustaría que empezara a hacer más calor pronto. (15 syllables)
20. El niño al que se le murió el gato está triste. (15 syllables)
21. Una amiga mía cuida a los niños de mi vecino. (16 syllables)
22. El gato que era negro fue perseguido por el perro. (16 syllables)
23. Antes de poder salir él tiene que limpiar su cuarto. (16 syllables)
24. La cantidad de personas que fuman ha disminuido. (16 syllables)
25. Después de llegar a casa del trabajo tomé la cena. (17 syllables)
26. El ladrón al que atrapó la policía era famoso. (17 syllables)
27. Le pedí a un amigo que me ayudara con la tarea. (17 syllables)
28. El examen no fue tan difícil como me habían dicho. (17 syllables)
29. ¿Serías tan amable de darme el libro que está en la mesa? (18 syllables)
30. Hay mucha gente que no toma nada para el desayuno. (18 syllables)
31. Son ellas las que acaban de decorar la sala de espera. (19 syllables)
32. ¿Sabe usted si el tren de las once y media ya ha salido de la estación? (20 syllables)
33. Nunca me divertí tanto como cuando fui a la pista de hielo. (20 syllables)
34. Cuanta más prisa tenía en su trabajo, menos calidad producía. (21 syllables)
35. Ellos lo organizaron el año pasado en la universidad cercana. (23 syllables)
36. Acabamos de volver del supermercado donde las ofertas eran muy interesantes. (27 sylla-
bles)

Thank you! Your responses have been recorded.

French elicited imitation test

(Gaillard & Tremblay, 2016)

Introduction
This language test will ask you to listen to several short audio files in French and make a record-
ing in response. (Please be patient as recordings may take time to load.)
Note that some of the items might be quite challenging. Please try to complete each of them
to the best of your ability.
<Click here to start>

Instructions-1
You are going to hear several sentences in English (6 in total). After each sentence, there will be
a short pause, followed by a tone sound {TONE}. Your task is to try to repeat exactly what you
hear. You will have only one attempt to do so. You will be given sufficient time after the tone to
repeat the sentence. Repeat as much as you can. Remember, don’t start repeating the sentence
until after you hear the tone sound {TONE}. Now let’s begin.
<I’m ready>
 The effects of frequency, duration, and intensity on L2 learning through Duolingo 35

Practice stimuli
1. We drove to the park.
2. I’ll call her tomorrow night.
3. You can buy meat at the butcher shop.
4. My brother just bought a brand new computer.
5. Sometimes they take their dog for a walk in the park.
6. We’re going to play volleyball at the gym that I told you about.

Instructions-2
Now, you are going to hear a number of sentences in French (50 in total). Once again, after each
sentence, there will be a short pause, followed by a tone sound {TONE}. Your task is to try to
repeat exactly what you hear in French. You will have only one attempt to do so. You will be
given sufficient time after the tone to repeat the sentence. Repeat as much as you can. Remem-
ber, don’t start repeating the sentence until after you hear the tone sound {TONE}. Now let’s
begin.

Main stimuli
1. Dans cette grande ville, les rues sont larges.
2. Je doute qu’il sache si bien conduire.
3. Qu’est-ce que tu as dit que tu faisais?
4. Il est possible qu’il pleuve des cordes.
5. Les maisons sont trè s belles mais trop chè res.
6. Le livre rouge n’était pas sur la table.
7. Ni lui ni moi ne les avions comprises!
8. Il prend une douche tous les matins à 7h00.
9. Je n’aime pas les films qui sont à l’eau de rose.
10. Aprè s le dé jeuner, as-tu fait une bonne sieste?
11. Tu aimes é couter la musique techno, n’est-ce pas ?
12. Est-ce que tu penses que je dois me faire couper les cheveux?
13. Traverse la rue au feu et puis continue tout droit!
14. Y-a-t-il beaucoup de gens qui ne mangent rien le matin?
15. On en avait une petite noire qui s’appelait minouche.
16. J’espè re que le temps se ré chauffera plus tôt cette anné e.
17. Le petit garç on dont le chaton est mort hier est triste.
18. Quand Sophie reç ut sa collè gue, elle lui proposa du thé .
19. Ce restaurant est sensé avoir de la trè s bonne nourriture.
20. Je veux une belle et grande maison dans laquelle mes enfants puissent vivre.
21. La chatte que tu as nourrie hier é tait celle de ma voisine.
22. Le nombre de fumeuses en France ne cesse d’augmenter chaque anné e.
23. Gabriel, en é pousant sa patronne, a fait d’une pierre deux coups.
24. N’ê tes-vous pas fatigué s aprè s ce voyage en voiture de trois jours?
25. Nous aurions dû faire des ré servations avant d’aller au thé âtre.
26. Prenons deux semaines pour visiter New York pendant les vacances d’été !
27. Qu’allez-vous faire demain soir aprè s lui avoir dit la vé rité ?
28. Est-ce qu’elle vient de finir de peindre l’inté rieur de son appartement?
36 Ekaterina Sudina and Luke Plonsky

29. La personne avec qui je sortais n’avait pas un grand sens de l’humour.
30. Elle commande uniquement des plats de viande et ne mange jamais de lé gumes.
31. Vous pensez que le prix des maisons en ville va redevenir abordable?
32. Une bonne amie à moi s’occupe toujours des trois enfants de mon voisin.
33. Avant de pouvoir aller dehors, il doit finir de ranger sa chambre.
34. La police a arrê té le terrible voleur qui é tait grand et mince.
35. Auriez-vous la gentillesse de me passer le livre qui est sur la table ?
36. Elle a dé cidé de suivre des é tudes d’arts plastiques à l’Ecole des Beaux-Arts.
37. Dè s que la pré sidente eut signé le document, son secré taire l’emporta.
38. Excusez-moi, savez-vous si le train de 11h30 a dé jà quitté la gare?
39. Je ne me suis jamais autant amusé e que lorsque je suis allé à la patinoire.
40. Ce sont eux qui l’ont organisé l’an dernier à l’Université de l’Illinois.
41. Plus elle se dé pê chait dans son travail, moins elle ré alisait un travail de qualité .
42. Dè s que l’on aura dî né , on regardera attentivement le documentaire sur France 3.
43. Ne penses-tu pas que les ré alisatrices du film souhaiteraient lire les scé narios le plus tôt
possible?
44. L’examen n’était pas aussi difficile que celui de Monsieur Durand en cours de litté rature.
45. Laura et Julie, ce sont elles qui viennent de finir de dé corer é lé gamment la chambre d’amis.
46. Il est possible que ses parents soient arrivé s en France avant le dé but de la guerre d’A lgé rie.
47. On vient juste de rentrer du supermarché où les promotions é taient particuliè rement
inté ressantes.
48. Les é tudiantes Laure et Sté phanie vont continuer à l’étudier à l’Université de Montré al.
49. Marie, prenez votre courage à deux mains et vous verrez que cet entretien passera comme
une lettre à la poste!
50. Les é tudiants sortant de l’université avec un Master en poche ont plus de chance de trouver
un travail que les autres.

Thank you! Your responses have been recorded.

Notes on EIT administration and scoring

For the sake of comparability of the findings and the testing procedures, several modifications
were made:

1. The original French stimuli were amplified in Audacity.

2. Sample items (practice stimuli) for both tests were taken from the Spanish version of the
EIT and provided in English.
3. Both Spanish and French stimuli were presented in increasing length (not randomized as
was the case with the original French stimuli).
4. An introduction for both French and Spanish EIT was added to explain the nature of the
test.
5. The original instructions had to be slightly modified (due to the self-paced nature of both
tests in the present study).
6. All instructions were recorded by a female speaker with a standard American dialect. After
recording in Audacity, the peak amplitude was normalized to −1.0 dB (to help with the vol-
ume); a noise reduction for extraneous background noises was performed; and the beep
sounds were added where indicated in the script.
7. For both French and Spanish EITs, a tone sound (.25s) from the original French EIT was
used. A 3-second pause was inserted between the auditory sentence and the tone sound
 The effects of frequency, duration, and intensity on L2 learning through Duolingo 37

(as in the original French test). The total wait time between the auditory stimulus and the
onset of sentence repetition was, therefore, 3.25s (as in the original French EIT study).
8. There were no breaks between the trials as in the Spanish EIT study.
9. As both EITs were administered as self-paced tests, the maximum recording time was esti-
mated based on the formula by Solon et al. (2019; see supplementary materials) and set to
19s/19,000 ms (rounded based on the calculations below).
Spanish sentence #36 (Acabamos de volver del supermercado donde las ofertas eran muy
interesantes) = 27 syllables, 6.248s (the longest sentence recorded by a native speaker)
27 syllables = 6.248s → native speaker time
7 syllables = 6.248 + 2 → nonnative speaker time
27 syllables = (6.248 + 2) + (20 syllables *.5) = 8.248 + 10 = 18.248s → max. recording time for
nonnative speakers
French sentence #49 (Marie, prenez votre courage à deux mains et vous verrez que cet
entretien passera comme une lettre à la poste!) = 28 syllables, 6.238s (the longest sentence
recorded by a native speaker)
28 syllables = 6.238s → native speaker time
7 syllables = 6.238 + 2 → nonnative speaker time
28 syllables = (6.238 + 2) + (21 syllables *.5) = 8.238 +10.5 = 18.738s → max. recording time for
nonnative speakers
10. Finally, for both French and Spanish tests, a rubric developed by Solon et al. (2019) was
used.

Spanish C‑test
(Riggs & Maimone, 2018)

Introduction
In this language test, you will be presented with short Spanish texts in which parts of words are
deleted. The deletions correspond to the final portions of the words. Please do your best to fill
in the missing part of the word.
Complete the words as accurately as possible, paying attention to the spelling and gram-
matical features like accents or agreement in gender and number.
You may put a zero in the blank if you do not know the answer and do not want to guess.
There will be a total of 5 texts, each taking about 3 to 5 minutes to complete. Please try to finish
each text in under 6 minutes.

Main part
Below you will be presented with short Spanish texts in which parts of words are deleted. The
deletions correspond to the final portions of the words. Please do your best to fill in the missing
part of the word. Complete the words as accurately as possible, paying attention to the spelling
and grammatical features like accents or agreement in gender and number. You may put a zero
in the blank if you do not know the answer and do not want to guess. There will be a total of 5
texts, each taking about 3 to 5 minutes to complete. Please try to finish each text in under 6 min-
utes.
Spanish accents (if you do not have a Spanish keyboard): á, é, í, ó, ú, ñ, ü.

Example
On Sunday, the weather was beautiful, and we went for a walk.
On Monday, it was raining, and we stay at home.
38 Ekaterina Sudina and Luke Plonsky

Text 1. 17 de agosto, Marbella

Laura todavía está en la playa y yo estoy ya en la habitación del hotel. No quiero
sa______, hace mu______ calor y no de______ tomar m______ el s______ . Todas
l______ mañanas va______ a l______ playa, a u______ playa peq______ pero
m______ bonita ce______ del hotel. Allí no h______ mucha ge______ . Después
com______ juntos en u______ bar. Hay muc______ en esa zo______ . ¡Cómo
m______ gusta l______ comida de aq______ , sobre to______ el pes______ ! Por las
tar______ vamos a______ centro de Marbella. A veces volvemos muy pronto al hotel.
Nos gusta mucho escuchar música o leer un buen libro.
Text 2. Descripción de mi mamá Lauri
Mi personaje favorito es mi mamá, su nombre es Lauriana de Jesús Pinta. Tiene
cuar______ y ocho añ______ de ed______ y tra______ de prof______ en el
cen______ educativo “Fiscal Francia” de ni______ . Ella vi______ en Calvas c______
mi pa______ y mi her______ , a qui______ quiero y apr______ mucho. A mi madre
la adm______ porque e______ una per______ luchadora, respe______, solidaria,
soci______ , etc. Le gu______ el dep______ , en espe______ el basqu______ . Está
siempre ale______ , es m______ amorosa y comuni______ , y trata de darme y
enseñarme lo mejor que tiene. Los momentos que he compartido con ella han sido
inolvidables.
Text 3. Así es el día a día de una cantante famosa
Me levanto a las 8 y desayuno un panecillo con salmón ahumado. Mi hor______ suele
s______ frenético, pe______ mi ma______ sabe l______ que hay que ha______
y pu______ repasarlo c______ ella. Me gu______ sorprenderme, a______ que
nu______ miro mi calen______ la no______ anterior. Me enc______ la sens______
de levan______ por l______ mañana y te______ que mi______ por la ven______
para desc______ en q______ ciudad es______ . Como norma______ paso la
may______ del tiempo en el hotel, mi madre no se preocupa demasiado. Pocas veces
discutimos, porque quiero muchísimo a mi madre.
Text 4. Perfil de alimentación de los argentinos
El último censo realizado en la República de Argentina contabiliza a su población
en algo más de 40 millones de habitantes. El pa______ produce una cant______
suficiente pa______ alimentar a 442 millones de pers______ , sin emb______ , por
u______ lado s______ observan indiv______ que pres______ déficit de nutri______
en s______ alimentación, y por ot______ lado, tam______ excesos. A los argentinos
les so______ comida pero les fa______ variedad. Hay homoge______ en l______
cocina y e______ la me______ de los argentinos. Se cons______ pocos alim______
de bu______ calidad nutri______ , mientras que el exceso de con______ de
ot______ agrega grasas de mala calidad, sodio y azúcares.
Text 5. Un cubano en Kiev, recién llegado y sintiéndose agobiado
Creo que este correo que les escribí a mis padres fue el más sentimental que he hecho
en mi vida. Yo ha______ salido ha______ dos dí______ de Cuba, y el______ no
hab______ tenido noti______ mías, no sab______ dónde est______ , ni có______
iba, ni q______ había si______ de m______. Yo me sen______ muy tri______ , pero
no po______ decirles e______ . Empecé dici______ que to______ me iba bi______
y que ha______ salido estu______ el vi______ , y s______ querer m______ lágrimas
empe______ a salir y no paraban de rodar por mis mejillas mientras escribía. Sabía
que los estaba engañando, pero consideraba injusto preocuparlos; total, no resolvería
nada.
 The effects of frequency, duration, and intensity on L2 learning through Duolingo 39

French C‑test
(Counsell, 2018)

Introduction
In this language test, you will be presented with short French texts in which parts of words are
deleted. The deletions correspond to the final portions of the words. Please do your best to fill
in the missing part of the word.
Complete the words as accurately as possible, paying attention to the spelling and gram-
matical features like accents or agreement in gender and number. Words with hyphens or apos-
trophes like “celui-ci” or “l’ami” count as one word.
You may put a zero in the blank if you do not know the answer and do not want to guess.
There will be a total of 5 texts, each taking about 3 to 5 minutes to complete. Please try to finish
each text in under 6 minutes.

Main Part
Below you will be presented with short French texts in which parts of words are deleted. The
deletions correspond to the final portions of the words. Please do your best to fill in the missing
part of the word. Complete the words as accurately as possible, paying attention to the spelling
and grammatical features like accents or agreement in gender and number. Words with hyphens
or apostrophes like “celui-ci” or “l’ami” count as one word. You may put a zero in the blank if
you do not know the answer and do not want to guess. There will be a total of 5 texts, each tak-
ing about 3 to 5 minutes to complete. Please try to finish each text in under 6 minutes.
French accents (if you do not have a French keyboard): é, à, è, ù, â, ê, î, ô, û, ç, ë, ï, ü.

Example
On Sunday, the weather was beautiful, and we went for a walk.
On Monday, it was raining, and we stay at home.

Text 1. Edda, la gourmande

Je m’appelle Edda. J’ai 37 ans. Je su_______ née à Rome e_______ vis dep_______
neuf a_______ à Paris av_______ mon ma_______ italien. D’ori_______ italienne
(m_______ père) e_______ française (m_______ mère), j_______ baigne
dep_______ toute pet_______ dans c_______ deux cult_______. Gourmande
e_______ curieuse dep_______ toujours*, c’e_______ tout nature_______ que
j’_______ commencé à m_______ passionner po_______ la cui_______. Elle
fa_______ vraiment par_______ de ma vie**. La magie est toujours là.
Note. *The answer key was slightly different from the test version and had “comme
toujours” instead. **The answer key version (“la cuisine. Elle fait vraiment partie
de ma vie”) was preferred to the test version (“la d_______ cuisine. El_______ fait
vrai_______ partie de ma vie”).
Text 2. La Martinique
La Martinique est une île de 64 km de long qui s’étale sur à peine 20 km de large, et
est dominée par la montagne Pelée qui culmine à 1397 m. Elle présente une grande
diversité de paysages. Le S_______ est cons_______ de coll_______ à l_______
végétation p_______ abondante. L_______ Nord e_______ montagneux. L_______
plages surpr_______ par le_______ beauté e_______ leur incro_______ diversité,
av_______ des coul_______ qui vo_______ du sa_______ blanc lumi_______ au
40 Ekaterina Sudina and Luke Plonsky

no_______ volcanique. L’e_______ est transp_______ et da_______ les fo_______

marins o_______ trouve beau_______ de pois_______* colorés. Il y a les plages tran-
quilles du Sud-caraïbe, bordés de cocotiers, et celles plus tumultueuses, de la côte
atlantique.
(Note. *Based on the author’s suggestion in personal correspondence, “des bancs de
poissons” in the answer key was replaced with “beaucoup de poissons.”)
Text 3. Rester chez ses parents
En Île-de-France, les jeunes restent un peu plus longtemps chez leurs parents que
dans les autres régions françaises. Mais pourquoi cette tendance des jeunes à rester
chez leurs parents s’accentue-t-elle en région parisienne ? Cette diffé_______ avec
l_______ reste d_______ la France s’exp_______ essentiellement p_______ la
prox_______ des unive_______. En ef_______, il y a beau_______ de facu_______
à Paris e_______ dans s_______ région. P_______ conséquent, l_______ jeunes
d_______ l’Île-de-France q_______ font d_______ études univers_______ ne
so_______ pas obl_______ de qui_______ le domi_______ familial. Ma_______ en
prov_______, par con_______, les universités sont souvent éloignées du domicile des
parents et les jeunes doivent quitter leur famille pour poursuivre leurs études.
Text 4. La science-fiction
La science n’a-t-elle pas de quoi considérer avec mépris les œuvres de science-fiction
souvent basées sur des faits irréalistes et des connaissances approximatives ? La NASA
a décidé d’en finir avec cette situation. La cél_______ agence spat_______ américaine
vi_______ de lan_______ un nou_______ projet inti_______ « œuvres d_______
fictions insp_______ par l_______ NASA », e_______ collaboration av_______ un
édi_______ de science-_______. L’idée e_______ de met_______ en con_______
des écri_______ avec l_______ scientifiques d_______ l’institution, af_______ qu’ils
le_______ offrent u_______ expertise e_______ corrigent le_______ éventuelles
err_______. L’objectif est simplement de produire des œuvres scientifiquement
approuvées.
Text 5. L’humour belge
L’humour est belge, assurément ! Voilà une affirmation qui ne manquera pas de
faire s’esclaffer la France entière. Mieux enc_______, elle se_______ l’objet d_______
quolibets div_______. C’est qu’_______ adorent ç_______, nos ch_______ voisins,
ridic_______, critiquer, cho_______. À Paris rè_______ la gra_______ tyrannie
d_______ persiflage. S_______ montrer d_______ et méc_______ est dev_______
gage d_______ réussite po_______ un humo_______. Goût d_______ scandale.
Esca_______ de l_______ méchanceté. No_______ les Bel_______, avons
l’hu_______ plus tendre, plus bon enfant. Rire ne signifie nullement gouailler, railler.
Humour gentil, candide, humour belge.

The final message at the end of the pretest

This is the end of Part I of the study. As a reminder, please do your best to use Duolingo at least
twice a week for a total of 26 weeks. Six months later, you will be invited to participate in Part
II of the study. Thank you!
 The effects of frequency, duration, and intensity on L2 learning through Duolingo 41

The final message at the end of the posttest

This is the end of Part II of the study. If you are eligible for compensation, you will be con-
tacted by researchers within a month. Thank you!

Notes on C‑test administration and scoring

1. Each test included five short texts with 25 blanks in each (i.e., 125 blanks in total).
2. Both tests were timed (30 minutes).
3. There was only one correct answer for each blank in the French C‑test. In the Spanish
C‑test, 16 blanks had two correct answers.
4. For both tests, instructions and example sentences were provided in English (rather than
in the target language as was in the original French C‑test).
5. Small adjustments were made to the French C‑test when three minor inconsistencies were
revealed between the answer key and the test version.

Appendix B. Assumption checking

RQ1a. The assumptions for paired samples t-tests by language were satisfactorily met (i.e.,
the dependent variable of proficiency gain scores was continuously scaled; the distrib-
ution of the differences in proficiency gain scores followed the normal curve and con-
tained no extreme univariate outliers; the independent variable of test mode – oral vs.
written – consisted of categorical data from two related groups).
RQ1b. There were no major violations of assumptions for the two-sample t-tests. For the EIT
gains in French vs. Spanish, the dependent variable was approximately normally dis-
tributed for each language group, but Levene’s test was statistically significant, sug-
gesting the lack of homogeneity of variances; nonetheless, the sample sizes for the
two language groups were roughly equal, which does not require equal population
variances. For the C‑test gains in French vs. Spanish, the dependent variable was,
again, approximately normally distributed for each language group, and Levene’s test
was not statistically significant (i.e., equal variances assumed). However, univariate
outlier analysis revealed two extreme outliers (|z| > .3.29) on the EIT gains variable
and six additional extreme outliers on the C‑test gains variable. A close inspection of
these scores did not indicate any red flags in participants’ performance. Therefore, the
analyses were conducted twice, with and without outliers, to allow for comparisons.
RQ2a. To meet the assumptions for Pearson correlations, all extreme outliers (|z| > .3.29)
were removed from the variables of interest (i.e., 6 from the Login and C‑test gains
variables, 4 from the Session, Minutes, Level reviews, Skill practice, and Tests vari-
ables, 2 from the Lessons and EIT gains variables, and 1 from the Stories variable)
as they were found to affect the correlation estimates. The assumption of linearity
was satisfied as indicated by the matrix scatterplot. Q-Q plots and histograms sug-
gested minor deviations from normality. Therefore, bootstrapped Pearson correla-
tions (based on 1,000 samples) with bias-corrected and accelerated confidence
intervals were performed (final N = 233).
RQ2b. Prior to performing multiple regression analyses, all extreme univariate outliers were
removed from the variables of interest. The strongest predictors were chosen based
on correlational analyses (see RQ2a). However, in the model predicting EIT gains,
42 Ekaterina Sudina and Luke Plonsky

the Sessions and Logins variables were highly correlated. The ensure the absence of
multicollinearity, the Sessions variable was removed from the model because it had
a weaker correlation with EIT gains than the Logins variable. Following the removal
of 10 multivariate outliers on three continuous predictor variables in the model pre-
dicting EIT gains and the removal of 19 multivariate outliers on two continuous pre-
dictor variables in the model predicting C‑test gains, the assumptions of linearity;
absence of multicollinearity; absence of autocorrelation; and normality, linearity, and
homoscedasticity of residuals were met.
RQ3. The assumptions for Pearson correlations between Duolingo app usage variables (i.e.,
frequency, duration, and intensity) and individual differences (i.e., L2 grit and moti-
vation) were satisfied after removing extreme outliers (|z| > .3.29) from the variables
of interest. The inspection of the matrix scatterplot supported the assumption of lin-
earity. To account for occasional deviations from normality, which were indicated by
Q-Q plots and histograms, bootstrapped Pearson correlations (based on 1,000 sam-
ples) with bias-corrected and accelerated confidence intervals were performed (final
N = 260).
RQ4a. Concerning the assumptions for Pearson correlations, first, eight extreme outliers
(|z| > .3.29) were removed from the gains variables (i.e., two from the EIT and six from
the C‑test gains), and four outliers were eliminated from L2 grit consistency of inter-
est because they were found to affect the correlation values. A matrix scatterplot did
not indicate any violations of linearity. To account for minor deviations from normal-
ity (as suggested by Q-Q plots and histograms), bootstrapped Pearson correlations
(based on 1,000 samples) with bias-corrected and accelerated confidence intervals
were performed (final N = 248).
RQ4b. To check the assumptions for multiple regression analyses, first, all extreme univariate
outliers were removed from the variables of interest (see RQ4a). Five multivariate out-
liers on three continuous predictor variables were removed as well. All the assump-
tions of linearity, the absence of multicollinearity, the absence of autocorrelation, and
normality, linearity, and homoscedasticity of residuals were met.

Address for correspondence

Luke Plonsky
Northern Arizona University
United States
lukeplonsky@gmail.com

Co-author information

Ekaterina Sudina
East Carolina University
sudinae22@ecu.edu
 The effects of frequency, duration, and intensity on L2 learning through Duolingo 43

Publication history

Date received: 3 March 2023

Date accepted: 8 August 2023
Published online: 25 August 2023