Examining the effects of two cognitive styles (field dependence vs. field independence) on learners’ mobile-assisted vocabulary acquisition

2.1 Cognitive style

Cognition refers to the process of amassing and interpreting information gained from learning or experience, and then changing that information to knowledge to interpret the surrounding world (Te Molder & Potter, 2005). Individuals with different personal, behavioral, and psychological characteristics adopt different approaches, representing different styles, to engage their cognition. Style describes one’s proclivity towards the adoption of an approach in a variety of situations, and when this approach features intellectual and perceptual activities, it is referred to as cognitive style (Witkin et al., 1977). As Keefe (1979) defined, cognitive style refers to the link between the characteristic means for learning concepts and the usual approaches adopted to solve problems. As such, cognitive style can be construed as the connection between personality and cognition. Riding and Rayner (2013) characterized cognitive style as the frequent approach preferred for processing, restructuring, and representing information.

The disparities among cognitive styles concern different ways that learners engage in cognitive activities (e.g., intellectual activities). For example, when problem-solving, some learners deconstruct the task (i.e., the field) and analyze its components to find the causes and possible solutions. Conversely, other learners view the whole task to realize the causes and related answers. Another feature of cognitive styles is that they are not dichotomously valued as intelligence; one who has a higher intelligence quotient score can cognitively outperform the one with a lower score (Witkin et al., 1977). In fact, every cognitive style may be equally valuable and effective under a specific condition and in performing particular tasks.

Although there is consistency in individual learners adopting a cognitive style, there is inconsistency in this regard across learners, meaning that different learners adopt different cognitive styles when communicating with the learning environment (Witkin et al., 1977); hence, cognitive style can be used as a measure of learner differences (Messick, 1984).

2.2 Cognitive style: field dependence/independence

Learners can appreciate the field (i.e., the surrounding context) in different ways, which can be attributed to different cognitive styles. In this sense, learners can be categorized as field dependent (FD) or field independent (FI). FD learners view the whole field, rather than its components, with much less probability of restructuring it. Witkin et al. (1977) explained that FD learners are socially-orientated individuals whose perceptions and intellects rely on the social aspects of the field. In their research, Guo and Yang (2018) demonstrated that in communicative contexts, particularly when uncertainty arises, FD learners who are message-oriented show a higher proclivity to consider the social cues, as opposed to constituent cues, of communication. To illustrate, in FD-based perspectives, interaction is seen mainly as the medium of communication, and what linguistic cues^[1] (e.g., syntax cues, semantics cues) contribute to interaction is often overlooked. Moreover, FD learners show an inherent tendency towards being physically close to other learners, which can be indicative of their sense of dependency on external social referents (Zhang & Sternberg, 2001).

On the other hand, FI learners are analytical and self-perception oriented; they often deconstruct the field (e.g., learning task) and give their focal attention to particular parts of it rather than see the whole field (Rassaei, 2015). FI learners can analyze an item in a discrete fashion from its surrounding context. As such, Witkin et al. (1977) delineated that FI learners can detect a simple shape in a complex pattern in a short time. Another feature of FI learners is that they are impersonally-oriented individuals who often self-define learning goals; also, they prefer to learn and solve issues using internal cues and de-contextualized information (Garber et al., 2018; Pearson, 1992). For example, if ambiguity arises in communicative contexts, FI learners often consider the constituent cues (i.e., linguistic cues), as opposed to social cues, of communication to resolve that.

Zhang (2004) put forward that learners’field-dependence/independence can be determined by their susceptibility to external and internal cues while performing tasks. FI learners are not heavily influenced by the organization or the social aspects of the field when engaging in activities like problem solving, but rather they restructure the field, regardless of how dominating it might be. Conversely, FD learners are more conformist to the organization of the field and are not likely to restructure it; that is, they often accept and digest the material with its given structure (Chen, 2010; Jonassen & Grabowski, 1993).

2.3 Second language vocabulary acquisition

Vocabulary is an important element of language (Nation & Newton, 2008). There are mixed beliefs about vocabulary acquisition. While some people might think that for English as a second language (ESL) learners, vocabulary acquisition can occur via guessing the meanings of vocabulary items from context, research suggests that ESL learners need direct exposure to and explicit instruction on vocabulary to learn the target word(s) (Folse, 2004). As Chapelle and Jamieson (2008) explained, ESL learners need to devote their time to studying vocabulary, because (a) the guessing-from-context strategy, commonly used in reading comprehension activities, often leads to inaccurate guesses, and (b) unknown content words in a reading passage are very likely to remain unlearned if no attempt is made to consult their meanings from a dictionary, teacher, or computer/mobile application.

Even a small number of unknown content words in a reading text can make reading comprehension problematic for ESL learners. As Hu and Nation (2000) posited, the ESL learner’s reading comprehension can be affected by vocabulary to the extent that two unknown content words in each line of a text – assuming that each line holds approximately ten words – would make reading comprehension problematic for the learner. Studies indicate that an ESL learner needs to understand between 95 % (Laufer, 1992) and 98 % (Hu & Nation, 2000; Schmitt et al., 2011) of the content words in an authentic text (e.g., a novel) to comfortably read and successfully comprehend the whole text. These percentages require a vocabulary size between about 5,000 words for complex texts like novels (Hirsh & Nation, 1992) and 8,000–9,000 words for academic texts (Nation, 2006). Such vocabulary size estimates signify that importance needs to be placed on vocabulary retention and instruction to enhance the depth and size of ESL learners’ vocabulary.

The term retention is defined as “the ability to provide the meaning of a new word after a given period of time” (Ramezanali, 2017, p. 6). To have ESL learners retain a large vocabulary (e.g., 8,000 words) can create some barriers for teachers as well as students. For example, classroom instruction is time bound; therefore, teachers cannot cover all the vocabulary items. Also, for the same reason, students cannot learn all the vocabulary items in the classroom context because of a limited lexical exposure taking place in the classroom (Chapelle & Jamieson, 2008). According to Nation (2013), any vocabulary item requires a minimum of 10–15 exposures to be retained. If such a frequency of lexical exposure does not happen, ESL learners are less likely to leverage repetition, which is an important technique necessary for vocabulary recall and retention (Webb, 2007).

However, advancements in MAVA have easified lexical exposure, and therefore, repetition in and out of the classroom context. MAVA provides learners with different means through which they can interact with word definitions, synonyms, sample sentences, etc., each of which can lead to vocabulary development (Tozcu & Coady, 2004). For example, in a MAVA setting, ESL learners can: Access word meanings while reading a passage simply by voicing “what does [X] mean?” to their smart device, favorite the word’s meanings on a dictionary app, create a flashcard with the word immediately, add mnemonics to flashcards, and the like. Chapelle and Jamieson (2008) postulated, these interactions go beyond word guessing; they increase the kinds of support (e.g., verbal support, imagery support) that learners need when engaged in different language activities. Research suggests that more kinds of support would result in a better vocabulary recall and retention (Yoshii & Flaitz, 2002). Therefore, in line with Thornbury’s (2002) criteria for claiming to have learned a word, it can be deduced that MAVA has the capacity to help ESL learners earn a good command of vocabulary because it promotes (a) vocabulary retention, and (b) the use of the vocabulary items in different communicative contexts.

2.4 Frayer Model of vocabulary development

To examine whether cognitive style acts as a function of vocabulary recall among ESL learners who engage in MAVA, this study used a popular vocabulary achievement model named the Frayer Model (FM) vocabulary development. Developed by Frayer et al. (1969), FM is one of the commonly employed concept mapping models in vocabulary achievement (Greenwood, 2002; Palmer et al., 2014) and concept comprehension (Bowe, 2019). FM includes five sections: keyword holder, definition, fact and/or characteristics, examples, and non-examples. Keyword holder is the center of the model, which is reserved for the target vocabulary item (e.g., “glance” – the word is adopted from the Vocabulary Level Test (VLT); Webb et al., 2017). Definition is a place where a friendly, easy-to-understand definition of the target word is given. In the examples section, synonyms, collocations, and in the non-examples, antonyms can be written, with or without example sentences. Plus, using schematic knowledge, students should write in the facts/characteristics section one or more cues that associate the word meaning to a mental entity established in their mind. In this study, as the flashcards were built on a mobile app – Memrise, the image upload feature of the app allowed us to add pictures to the facts/characteristics section of some, but not all, flashcards to help participants recall (and learn) the meaning of the target word. Figure 1 shows the components of the Frayer Model.

Figure 1:

Frayer Model (adapted from Frayer et al., 1969).

The FM enables learners to develop an in-depth grasp of new vocabulary items because it is not merely meant to focus students’ attention on dictionary definitions; rather students using this model are encouraged to think about, analyze, and know the characteristics (or concept) of new vocabulary items, for example, by being provided with some facts about the vocabulary item.

2.5 Memrise

To operationalize vocabulary acquisition via mobile learning, the researcher employed the term MAVA. Adhering to Traxler (2010) view of mobile learning, MAVA refers to gaining vocabulary knowledge which is assisted, encouraged, or enhanced by handheld mobile devices such as smartphones. In this respect, Memrise, a digital flashcard app, was chosen to provide participants (ESL learners) with MAVA.

Developed in 2010, Memrise is a popular mobile app (also a web app) – with more than 50 million users to date – designed to assist users/learners to improve their vocabulary recall and retention via digital flashcards. Not only does this app aim for vocabulary acquisition, but it also assists learners to memorize concepts and characters in other disciplines (e.g., science). Memrise offers pre-defined vocabulary courses in the form of digital flashcards across different language proficiency levels (e.g., beginner). Each level exposes learners to a certain amount of vocabulary items and their definitions. Moreover, Memrise has testing features including multiple choice tests, audio tests, and type-the-right-word tests. Additionally, using speech synthesis and speech recognition technology, the app offers a pronunciation feature that checks and improves the learner’s pronunciation. Taken together, the tests and flashcards of Memrise provide learners with affordances (e.g., increased lexical exposure) that resonate well with vocabulary acquisition (Walker, 2015).

A key aspect of Memrise is that it is inspired by the repetition theory emanated from behaviorism. This learning theory was introduced by Watson (1958) who believed that the more regular occurrence of an associated stimulus and response, the mightier habit one will develop. As such, humans’ memory is like a sieve that needs continual refills of an item to retain that item. Likewise, using the spaced repetition system, Memrise exposes learners to flashcards at regular intervals. Intervals are determined by the frequency with which the learner responds to a flashcard correctly; flashcards which are frequently marked as true will appear with longer intervals and vice versa.

Another feature of Memrise is its customizable flashcards, catering to learners’ vocabulary acquisition style (i.e., cognitive style). Creating Memrise flashcards, learners can add word translations and mnemonic aids like an image to enhance vocabulary recall. Furthermore, Memrise has a timing feature. That is, each individual Memrise flashcard is set a timeframe that motivates the learner to respond to vocabulary prompts within a shorter time. A quicker response will result in a higher score and also a higher rank in the leaderboard of that specific lesson. The timer can be disabled at the user’s discretion – to date, the timer is configurable on the web app only. Figure 2 shows an example of a multiple-choice question on Memrise. Figure 3 shows a VLT flashcard, created on Memrise, that has a meme-based prompt.

Figure 2:

Multiple-choice question (Memrise).

Figure 3:

A sample VLT flashcard with a meme on Memrise.

While Memrise is one of the top-ranked language learning applications, it is essential to acknowledge that no tool is without its occasional shortcomings. Noteworthy among critics is Lotherington’s (2018) identification of areas for improvement in content-based mobile apps like Memrise. These areas include concerns about instructional content, exercises that might be perceived as less engaging, and a course progression that may appear somewhat inflexible in its structure (Lotherington, 2018, p. 211). However, Memrise offers a multitude of benefits and features that have contributed to its popularity among language learners and researchers. Its incorporation of fun and interactive features, alongside affordances that enhance learner engagement in vocabulary acquisition, has earned it a well-deserved reputation (Smith et al., 2018; Walker, 2015). The research has shown that Memrise, when used in conjunction with traditional learning methods, can serve as a valuable tool for enhancing students’ vocabulary performance (Walker, 2015). Moreover, its adoption of a spaced repetition system reflects its dedication to effective language learning practices. Thus, our decision to focus on Memrise for this study is rooted in the recognition of its strengths and affordances (see Section 2.6 for further details on affordances) while also acknowledging areas that may benefit from further development.

2.6 Affordances

Affordances, as elaborated by Gibson (1977), refer to the inherent potential properties and functions that a technological tool offers to its users. In the context of technology-assisted learning, affordances pertain to the different ways that digital tools and mobile applications present for fostering interaction, engagement, and learning (Norman, 1999). Within this framework, Memrise provides numerous affordances to language learners. Notably, it facilitates multimodal learning through text, audio, video, and interactive exercises. Additionally, Memrise employs gamification principles, incentivizing language learning through point systems, competitive features, and engaging challenges. Another key affordance is Memrise’s adaptive learning, where exercises and challenges adapt based on individual learner performance, ensuing optimal level of engagement. By examining the affordances offered by Memrise, we can gain useful insights into how language learners perceive and use this application. This understanding sheds light on how learners harness the application utilized herein to enhance their vocabulary acquisition and retention, providing valuable information on the effectiveness of this learning tool in vocabulary development.

3.1 Research questions

The present study explores the relationship between the two cognitive styles studied herein, short-term vocabulary recall among English language learners, and a widely-used vocabulary learning application – Memrise. This study is grounded in two research questions (RQs):

4.1 Participants

To carry out the research, ESL learners (n = 201) aged between 15 and 22 years old, 53 % male and 47 % female, from three English language centers in Johor Bahru, the second biggest state of Malaysia, participated in this study. They were chosen using non-probability, convenience sampling. The selected participants (n = 138; for further details on sampling, see the Procedure section) consisted of school or college students who desired to extend their ESL studies by attending enrichment classes whose curricula were different than those of their school or college.

4.2 Measures

4.3 Procedure

Participants were informed of the research objectives, instruments, and phases designed for the purpose of answering the research question of this study. It was explained that the whole treatment would be completed within 5 weeks, two sessions per week and each session for 1 h in their own classroom. In total, the treatment comprised ten sessions. Figure 4 demonstrates the treatment design of the present research.

Figure 4:

Treatment design.

To begin with, two proficiency tests were conducted in the first session. After conducting the tests, 192 participants were identified as intermediate-level ESL learners. To identify participants’ cognitive style, the GEFT was administered (session 2). According to the format of GEFT, participants were required to find a simple shape in a complex figure in 20 min. Following the GEFT, participants were categorized as either FD (n = 69) or FI (n = 78) learners based on their final scores. However, 45 participants were excluded due to exhibiting a combination of both FD and FI cognitive styles. To ensure an equal distribution of participants for analysis, we randomly selected 69 participants from the FI group, matching the existing number in the FD group.

In the third session, the pretest instructions and timing were explained to participants; pretest included 150 matching questions adopted from the VLT. For the next six sessions, participants learned and practiced the target vocabulary items using digital flashcards of Memrise. In each of those sessions, participants were exposed to a level of the VLT clusters, including 30 vocabulary items. Participants could consult each other or their teacher about the content of flashcards or ask for further VLT-based clarification if they needed any. Figures 5 and 6 show a sample content using the FM and a sample VLT flashcard on Memrise, respectively.

Figure 5:

Sample content; target word: Glance.

Figure 6:

Sample VLT flashcard on Memrise; target word: Glance.

By session nine, participants finished practicing and reviewing 150 vocabulary items of the VLT. Finally, to measure participants’ performance in short-term vocabulary recall, a posttest with identical questions and timing to the pretest was conducted in the tenth session. The results of the posttest were compared with those of the pretest to make interpretations on participants performance and on the possible relationship between participants’ cognitive styles and their vocabulary recall.