Earticle

성인학습에 대한 담론은 지난 반세기를 지나는 동안 학습의 이성적 인지 과정을 중시하 던 관점에서 학습이 이루어지는 상황에 집중하는 관점으로 점차 이동되어 왔다. 이 같은 추세 속에서 성인 학습의 연구자 중 인지적 관점을 취하는 이들은 습득을, 상황적 관점을 취하는 이들은 실천을 그려내고자 노력할 수 밖에 없다. 하지만 두 가지 측면을 분리하여 보고자 하는 노력은 어느 순간 한계를 드러내기 시작하였고, 최근 이런 극단의 벽을 허무 는 개인의 학습과 상호작용적 상황 학습을 모두 고려하고자 하는 제3의 장면이 등장하기 시작했다. 이런 시점에 지각, 의식, 정서, 정신으로 이어지는 Bateson의 인식론과 학습에 대한 논리를 탐색해 보는 것은 의미가 있을 것으로 간주된다. Bateson은 전통적인 이원론 적 구분을 지양하고 생태적 일원론을 추구하는 구성주의적 인식론을 주장한다. 또한 학습 에 있어서는 ‘논리유형이론’에 근거한 학습위계와 역설의 발생으로 형성되는 이중구속 적 모순을 중시한다. 때문에 이 같은 학습 관점은 성인학습 연구에 있어서 인지와 실천을 함께 논할 수 있는 근거를 제공해줄 수 있을 것으로 기대된다. 이에 본고에서는 Bateson의 사유체계와 학습이론을 검토하고 이를 바탕으로 성인학습에 적용, 학습의 지평을 확장할 수 있는 다양한 요소와 방법에 대한 논의를 이루고자 한다.

Over the last half century, discourse on adult learning has gradually shifted from the viewpoint of emphasizing the rational cognitive process of learning to the point of focusing on the situation in which learning takes place. In the same trend, those who take a cognitive viewpoint among researchers of adult learning are forced to try to draw learning, and those who take a situational point of view are trying to draw action. However, efforts to separate the two aspects have begun to reveal limitations at some point, recently, a third scene has begun to take into consideration both individual learning and interactive situation learning that breaks down this extreme wall. At this point it is deemed meaningful to explore the logic of Bateson's epistemology and learning that leads to perception, consciousness, emotion, and spirit. Bateson argues for a constructivist epistemology that avoids traditional dualism and pursues ecological monism. In addition, the emphasis is on the dual hierarchical contradiction formed by the learning hierarchy and paradox based on the 'logic type theory' in learning. This kind of learning viewpoint is expected to provide a basis for discussing cognition and practice in adult learning research. In this paper, we review Bateson 's reasoning system and learning theory, and discuss various factors and methods that can be applied to adult learning and extend the learning horizon.

이 연구의 목적은 지방자치단체의 비상대비업무 관련 학습조직화 수준과 팀원의 자기주도학습 능력 및 조직몰입 사이의 영향력 관계를 분석하였다. 이를 토대로 지방자치단체의 비상대비업무 관련 학습조직화 수준과 자기주도학습을 능동적으로 지향시키고 이를 통해 조직몰입도를 높힐 수 있도록 제언하고자 한다. 연구대상은 12개 시⋅도 민방위 비상대비팀 구성원 175명을 대상으로 설문조사를 실시하였다. 결과는 다음과 같다. 첫째, 비상대비 업무조직의 학습조직화와 자기주도학습, 조직몰입 변인 간에는 모두 유의한 상관관계가 나타났다. 둘째, 학습조직화가 구축되면 구성원의 자기주도학 습이 정적인 영향을 미치며, 직무몰입에도 긍정적 효과를 보인다. 셋째, 자기주도학습은 학습조직화 와 결과변수인 조직몰입과의 매개역할을 하는 것으로 나타났다. 이에따른 본 연구의 시사점 및 한계 점이 논의되었다.

The purpose of the study is to analyze the relationship between the level of organizational learning related to emergency preparedness of local governments and their members’ self-directed learning ability and organizational immersion. The study was conducted by carrying out a survey of 175 members of the Civil Defense Emergency Preparedness Team in 12 cities and provinces in South Korea. The results are as follows. First, there was a significant correlation among organizational learning, self-directed learning, and organizational immersion of the emergency preparedness organizations. Second, organizational learning had a positive effect on self-directed learning and job immersion of their members. Third, self-directed learning was shown to serve as a mediator between organizational learning and organizational immersion. Findings and limitations are discussed.

Background: Case-based learning (CBL) is becoming an important approach for improving interprofessional collaboration education. Previous studies have examined learners’ satisfaction with interprofessional education (IPE) in medical institutions. However, there are few studies on the implementation of university-led CBL interventions and their direct effects on learning outcomes. The aim of this study was to evaluate the effectiveness of CBL interventions on changes in the participants’ perception and knowledge acquisition ability. Methods: The CBL approach consisted of team-based case-based learning, self-directed learning, and post-feedback. It was conducted as a single course for pharmacy students in their 5th year in a university setting. Changes in the participants’ perceptions and self-assessments of competence levels were evaluated using survey responses. The effect of the CBL intervention on knowledge acquisition ability was directly evaluated using the exam score. Results: The majority agreed or strongly agreed that team-based case-based learning, and self-directed learning helped them to improve their knowledge and skills to a higher level and to increase the self-assessment of competency level. The average score of knowledge acquisition ability (average score of 75.0, p=0.0098) was significantly higher in the CBL intervention group than the lecture-based learning intervention group (average score of 52.0). Conclusion: The participants positively perceived that CBL intervention helped them to effectively improve their knowledge and the self-assessment of competency level. It also enhanced knowledge acquisition ability. These data, based on the survey responses, suggest that it is necessary to implement CBL interventions in a university-led single professional education.

본 연구에서는 제주도 동부 지역의 지하수위 예측을 위한 머신러닝 모델로서 ANN, ELM, LSSVR, LightGBM을 적용하였으며, 모델 평가지표와 도식적 비교를 통해 적용 모델의 지하수위 예측 성능을 평가하였다. 자기상관성이 강한 지하수위 예측에 있어서 비교 모델들은 1일 예측에서 모두 매우 우수한 예측 성능을 나타낸 반면, 7일 예측에서는 예측 성능이 저하되었으며, 비교 모델들간 예측 성능은 거의 비슷하게 나타났다. 또한, 해안지역의 경우 조위변동의 영향으로 인하여 내륙지역보다 지하수위 예측 성능이 다소 저하됨을 확인할 수 있었다. 자기상관성이 강한 지하수위 예측에 있어서 예측 결과가 비슷하 게 나타날 경우 ANN, ELM 등과 같이 상대적으로 복잡한 상세 튜닝을 필요로 하지 않는 머신러닝 모델의 적용성이 상대적으로 우수할 것으로 판단된다. 따라서 머신러닝 모델을 이용한 지하수위 예측은 제주도의 지하수자원을 위한 효과적인 관리도구가 될 수 있을 것이다.

This study applies ANN, ELM, LSSVR, and LightGBM as machine learning models for groundwater level forecasting in the eastern region of Jeju Island, South Korea. The groundwater level forecasting performance of the applied models is evaluated through model evaluation indices. In forecasting the groundwater levels with strong autocorrelation, the comparative models all show excellent performance in the 1-day forecasting. In contrast, the forecasting performance deteriorates in the 7-day forecasting. The results also show that the forecasting performances among the comparative models are almost similar. In addition, it is confirmed that in coastal areas, groundwater level forecasting performance deteriorates compared to inland areas due to the influence of tidal variation. When forecasting results are similar in forecasting the groundwater levels with strong autocorrelation, models that do not require relatively complex detailed tuning, including ANN and ELM, can have excellent applicability. Therefore, groundwater level forecasting using machine learning models can be an effective tool for groundwater resources in Jeju Island.

The fields of technical and vocational are fast becoming prime choices of the school leavers. Not only the fields are important industry-wise, they seem to be practical too as choice of related occupations. The challenge is to instil interest in students for these fields through relating technical and vocational with science, technology, engineering and mathematics (STEM) because they are intertwined with one another. The purpose of this study therefore was to develop a set of six STEM lesson plans with higher level questions inclusive. Two similar sets of pre and post-tests made up of higher level questions were developed as instruments to test the effectiveness of the lesson plans in increasing the achievement of elementary students in two schools and taught by two teachers. Quasi experimental design was used for this study involving 120 students from 4 classes. Half of them formed treatment groups and another half the control groups. The pre-test and post-test contained two sections, a and b which were objective and subjective sections consecutively. The data obtained were analysed using the Statistical Package for the Social Science (SPSS) and the findings were presented through frequency, mean, standard deviation and t-test. The results from the data collected showed the achievement of the treatment groups were significantly different from the control groups. And the means of the treatment groups had increased significantly compared to the control groups. Infusing science, technology, engineering and mathematics in lessons attracted the students to learn. Asking open and higher level questions to the students in schools are the foundation of training for the students to embark on technical work in the future or at least to come to understand how industries work.

In this paper, “the valuable diagnostic tool” Strategy Inventory for Language Learning (SILL), proposed by Oxford (1990), has been employed to conduct an empirical study on learning strategies of tertiary-level EFL learners in China on the basis of a brief survey of the research into L2 learning strategies. The paper examines and analyzes the overall frequency with which they employ learning strategies. The relationship between learning strategies and three other variables (gender, time spent in English learning, and English language proficiency) is also examined through the well-known SPSS 10.0. The findings of the research suggest that the learning strategies employed most frequently by the tertiary-level EFL learners are memory, cognitive and metacognitive strategies, followed by compensation strategies, with social and affective strategies the least frequently used. The study also shows that significant difference exists between the employment of learning strategies and gender in English learning, and that strategy use demonstrates a significant relationship, to varying degrees, with the time spent on English learning and English proficiency as well. Finally, the paper discusses the causes of such a relationship between the above-mentioned variables, and points out the major factors that affect the strategy use of the tertiary-level EFL learners in China.

This study aims to investigate the impact of topic familiarity and vocabulary learning tasks on vocabulary acquisition in an ESP (English for Specific Purposes) setting, focusing on 16 adult learners with varying proficiency levels (A1-B1), all training to be cooks. Participants worked with six cooking recipes, three on familiar topics and three on unfamiliar topics, with 30 target words assigned to one of three tasks: word list, flashcards, or fill-in-the-gaps. Immediate and delayed post-tests (three weeks later) were administered using the Vocabulary Knowledge Scale (VKS). Despite the small sample size, significant differences in retention are noted across proficiency levels and task types over time. Results from repeated-measures ANOVAs showed that topic familiarity alone did not significantly impact vocabulary acquisition. However, the interaction between topic familiarity and task type was significant, as were the effects of task types and proficiency levels individually. The interaction between task type and proficiency level also showed a significant effect on vocabulary learning. Participants generally performed better with the fill-in-the-gaps method, but these effects decreased over time. The study highlights the importance of considering both proficiency levels and task types for effective vocabulary instruction.

This study seeks to investigate the relationship between Prator's Hierarchy of L2 learning difficulty and segment-level pronunciation learning difficulties that were observed among Korean adult EFL learners in 2006. A total of 148 student participants at a college-level institute took part in an empirical study on the comparison between the effects of two different EFL pronunciation syllabi. Out of the 148, only 93 students yielded a complete set of pronunciation data, namely the pre- and post- test results of L2 sound identification and L2 word read-aloud tests. Two native English-speaking instructors of English and two non-native instructors of English each taught four 70-minute lectures to the participants. Based on the descriptive analysis, the participants showed identification accuracy results that generally affirmed the predictions suggested by Prator, while they showed somewhat puzzling results concerning their read-aloud accuracy scores. Two additional factors of L2 sound learning difficulty are suggested based on the results of the read-aloud tests: 1) the L1 orthographic system, which is relatively transparent compared to the L2 orthographic system, and 2) the phonological context in which a certain L2 sound occurs.

The study aimed to examine whether EFL extensive reading could lead to increased word knowledge of lower-level EFL Taiwanese learners. The participants were 45 lower-level proficiency English as a foreign language Taiwanese technological college freshmen. They read 30 English texts within a 15-week EFL extensive reading program. The data were collected through the vocabulary pre- and post-tests to measure learners’ incidental vocabulary learning gains in the 50 randomly selected target words achieved through the program. Results revealed that the significant vocabulary gains were achieved by the participants after the EFL extensive reading program, suggesting that the EFL extensive reading treatment had produced a beneficial effect on the incidental word learning gains of the participants with lower EFL competence. However, the improvement in word pick-up rate reached to a modest extent with at least 6% to at most 15% of the 50 measured words being moved from unknown to known and achieved mainly at the recognition level only. The possible factors which led the participants to achieve a significant advantage but at only a modest rate in the number of the target words recognized correctly were discussed. Also, some pedagogical implications for EFL vocabulary instruction in the Taiwanese educational context were provided.

The purpose of this learning situation needs analysis is to investigate the perceived and desired language skill needs and wants of Japanese university EFL learners in comparison with their actual proficiency data to inform researchers, practitioners, and teachers about the current state and needs of EFL learners. This study involved 350 first-year Japanese EFL learners who completed an online questionnaire about their current self-estimated proficiency and their language skill needs at the time of graduation and throughout their lifetime according to the Common European Framework of Reference self-assessment descriptors. The participants also answered an open-ended question regarding what they want to be able to do in English at the time of graduation. Their perceived receptive proficiency levels were compared with their scores on English proficiency tests. The results showed that while participants could selfestimate their proficiency to some extent, they tended to overestimate their current proficiency and aspired to achieve even higher levels throughout their lifetime. The results of the open-ended question implied that there were needs and wants for engaging in real-time communication, which possibly reflected the participants’ lowest evaluation in their current interacting skills among the five skills.

本研究は，日本人大学生英語学習者を対象に学習者が認識する現状の言語技能レベルと実際 の習熟度テストの成績を比較し，学習者が目標とする言語技能レベルと英語使用ニーズを調 査する学習状況ニーズ分析である。大学1年生350名を対象に，ヨーロッパ言語共通参照枠 （CEFR）の自己評価記述文に基づき，現在の言語技能レベルを自己推定し，卒業時および生 涯にわたる言語技能のレベル目標を回答する多肢選択式15項目，大学卒業時に英語でできる ようになりたいことを自由記述形式で回答する1項目からなるオンライン質問調査を実施し た。また，参加者の現在の受容技能の自己推定レベルを英語受容技能習熟度テストの得点と 比較した。その結果，参加者はある程度の自己推定を行うことができたものの，現在の言語技能を過大評価する傾向があり，生涯を通じてさらに高いレベルの達成を目指していること が示された。また，自己推定した言語技能では，やり取りして話す技能が最も低く評価され たものの，自由記述の回答では，リアルタイムでのコミュニケーションに関するニーズが存 在することが示された。

The multi-criteria decision-making model is an important model for infor-mation and training systems. This paper aims to raise concerns associated with problem-oriented learning management, which is based on the multi-criteria decision-making model that adjusts to the level of doubt the users are likely to encounter. In the proposed method of measuring the level of doubt, the user provides a clear, "transparent situational picture" for more objective decision-making. Additionally, the method makes it possible to reduce the probability of guessing the correct answer, which increases the objectivity of knowledge in diagnostic systems for process control training on remote technology.

다중 에이전트 강화학습의 발전과 함께 게임 분야에서 강화학습을 레벨 디자인에 적용하려는 연구가 계속되 고 있다. 플랫폼의 형태가 레벨 디자인의 중요한 요소임에도 불구하고 지금까지의 연구들은 플레이어의 스킬 수준이나, 스킬 구성 등 플레이어의 매트릭에 초첨을 맞춰 강화학습을 활용하였다. 따라서 본 논문에서는 레 벨 디자인에 플랫폼의 형태가 사용될 수 있도록 시각 센서의 가시성과 구조물의 복잡성을 고려하여 플랫폼 이 플레이 경험에 미치는 영향을 연구한다. 이를 위해Unity ML-Agents Toolkit과MA-POCA 알고리즘, Self-play 방식을 기반으로2vs2 대전 슈팅 게임 환경을 개발하였으며 다양한 플랫폼의 형태를 구성하였다. 분석을 통해 플랫폼의 형태에 따른 가시성과 복잡성의 차이가 승률 밸런스에는 크게 영향을 미치지 않으나 전체 에피소 드 수, 무승부 비율, Elo의 증가폭에 유의미한 영향을 미치는 것을 확인했다.

As multi-agent reinforcement learning advance, it has been utilized for level design in game with multiple players and complex interactive environments. Despite the significant impact that the platform's design has on game level design, player-related metrics including behavior, skill level, and skill composition have been the primary focus of most reinforcement learning studies. In this paper, we study the impact of platform design on playing experience by utilizing multi-agent reinforcement learning. we use the Unity Engine to create various platform designs based on a 2vs2 shooting game and visual sensor visibility and structural complexity are taken into consideration for analysis. With Unity ML-Agents Toolkit, we construct a reinforcement learning environment based on the MA-POCA algorithm and the self-play method. Though the agent, we analyze changes in play experience with win rate, the number of episode, draw rate, and the Elo rating value. With the study, we figure out that visibility and complexity of platform design exhibit minimal impact on win rate balance in self-play method, but does influence the total number of episodes, draw rate, and the increase in Elo rating value.

본 연구는 레벨이 다른 일본어학습자의 조건형식 ｢たら｣, ｢と｣, ｢ば｣의 습득과정에 나타나는 중간언어의 양상에 대해 해당 형식의 전후에 위치하는 항목과의 접속에 의해 형성되는 유니트를 중심으로 분석한 것이다. 분석 결과는 다음과 같다. (1) 본 연구의 학습자들의 조건형식의 습득은 ｢たら｣를 축으로 이루어진다. ｢たら｣는 3개의 조건형식 중에서 가장 빈번히 사용되며 전후에 다양한 형식의 문법항목과 접속한다. 이러한 의미에서 학습자들에게 있어 ｢たら｣는 가장 범용성이 높은 조건형식으로 인식될 가능성이 있다. (2) 조건형식에는 특정한 형식과 품사와 빈번히 접속하여 사용하는 경향이 보인다. 이번 분석에 의해 밝혀진 유니트로는 ｢ナ형용사＋だったら(でしたら)｣,｢명사＋だと(ですと)｣,｢ある＋ば｣,｢ば＋いい｣,｢イ형용사＋と(특히｢ない＋と)｣등이 있다. (3) 이와 같은 유니트 형성의 양상은 학습자의 레벨에 따라 다르게 나타난다. 조건표현 유니트는 학습초기부터 보이고 학습중기에 가장 많이 사용되며, 학습후기에는 다소 감소되는 공통점이 보인다. (4) 일단 형성된 유니트는 소멸하지 않고 학습레벨의 상승에도 그대로 유지되거나 유사한 유니트로 변화･분화하는 등 그 모습이 다양하다.

The study analyzed the common points and differences in the level of the unit seen in the process of learning the condition expressions. The following is the summary of the points clarified by this study. 　From the above analysis, the following points have been clarified : (1)The acquisition of the learner's condition expressions in the study is the axis of study. “tara” is the most commonly used in three conditional formats, and is connected to the various forms of grammar in the form of a variety of forms. In this sense, it is possible for learners to be recognized as the most versatile type of condition expressions. (2)The condition tends to be used frequently in conjunction with specific forms and parts of speech. Examples of these studies include “na-adj+dattara(deshitara)”, “noun+ dato(desuto)”, “aru+ba”, “ba+ii”, and “i-adj.+to(nai+to)”. (3)The state of the unit formation is different depending on the level of the learner. The unit of condition expressions in terms of learning were commonly used in the early stages of learning, and was commonly used in the middle of the study period in the latter half of the study period, and it was common in the latter half of the study. (4)The unit formed once does not disappear, and varies from one kind to the same shape, even if the level is changed and the same shape is changed.

本研究は、レベルの異なる日本語学習者の条件形式｢たら｣、｢と｣、｢ば｣の習得過程に現われる中間言語の様相について、前後に位置する項目との接続によって形成されるユニットを中心に分析したものである。分析の結果は次の通りである。 (1)学習者の場合の条件形式の習得は、｢たら｣がその軸となっている。｢たら｣は3つの条件形式の中で最も多く使用されており、前後に多様な形式の文法項目と接続されている。その意味で学習者にとって｢たら｣は最も汎用性の高い条件形式として認識されている可能性がある。 (2)条件形式には特定の形式や品詞と頻繁に接続して使用する傾向が見られる。本研究の分析によって明らかになったユニットとしては、｢ナ形容詞＋だったら(でしたら)｣、｢名詞＋だと(ですと)｣、｢ある＋ば｣、｢ば＋いい｣、｢イ形容詞＋と(とりわけ｢ない＋と)｣などが挙げられる。 (3)このようなユニットはの形成の様子は学習者のレベルによって異なる。条件表現のユニットは学習初期から見られ、学習中期(本研究では中上級)に最も多く使用され、学習後期には少し減少するという点で共通していた。

본 연구에서는 일본어 학습자의 가능표현의 사용이 모어, 레벨, 환경이라는 다 양한 요인과 어떻게 관계되어 있으며, 그 결과를 일본어 모어화자 데이터와 비교 검토함으로써 학습자의 가능표현 습득의 보편성과 특수성에 대해 밝히고자 하였 다. 이상의 분석을 통해 다음과 같은 사실을 알 수 있었다. (1) 학습자들의 일본어 가능표현의 사용은 ‘デキル형’이 중심이 되며, 이는 모 어･레벨･환경 요인에 관계없이 공통된다. 그리고 이는 모어화자의 사용과는 다른 학습자만의 특수성을 띠고 있다. (2) 학습자의 모어 요인은 가능형식의 사용 빈도, 접속 경향 등 가능표현의 사용 등에 특별한 상관관계를 나타내지 않는다. (3) 학습자의 레벨 요인은 3종류의 가능 형식 중에서 ‘デキル형’의 사용과 상관 관계를 지닌다. (4) 학습자의 환경 요인은 ‘デキル형’의 접속, 「ない」와의 사용 등과 상관관계를 지닌다. (5) 학습자와 모어화자의 가능표현 사용 양상을 비교 검토한 결과, 학습자의 중 간언어는 사용 형식의 측면에서 언어 외적 요인에 관계없이 보편성이 나타나며, 학습 레벨과 환경에 따른 특수성이 관찰된다.

This study attempted to clarify the universality and specificity of learners' acquisition of possible expressions by comparing and reviewing the results of the use of possible expressions of Japanese learners with various factors such as language, level, and environment. Through the above analysis, the following facts were found. (1) Learners' use of Japanese-speaking expressions is centered on the ‘Dekiru type’, which is common regardless of language, level, and environmental factors. And it has a unique characteristic of learners that is different from the use of more speakers. (2) The learner's more factors do not show any special correlation with the frequency of use of the possible form and the use of possible expressions such as access tendency. (3) The learner's level factor has a correlation with the use of the ‘Dekiru type’ among the three possible forms. (4) The learner's environmental factors have a correlation with the connection of the ‘Dekiru type’and the use with ‘Nai’. (5) As a result of comparing and reviewing the use of possible expressions by learners and native speakers, learners' intermediate languages show universality regardless of language-external factors in terms of usage type, and specificity according to learning level and environment is observed.

본 연구는 일본어 학습자들이 일본어 가능표현을 어떻게 습득해 가는지에 대해 밝히기 위하여 의미가 비슷한 유형의 문법항목 내에서 학습자들이 각각의 항목을 어떻게 구분하며 사용하고 있는지에 대해 운용분석의 관점에서 분석하는 것을 목적으로 한 것이다. 이를 위해 대규모 코퍼스 자료를 분석하여 JFL 환경의 학습자들이 가능표현을 습득함에 있어 3개의 가능형식을 레벨에 따라 어떻게 나누어 사용하는지, 그리고 그러한 사용 차이가 무엇을 의미하는지에 대해 살펴보았다. 분석 결과를 요약하면 다음과 같다. (1) 일본어 가능표현의 습득은 ‘デキル형’을 중심으로 이루어지며 중급 이후 ‘가능동사형’과 ‘(ラ)レル형’으로 확대되어 간다. (2) ‘가능동사형’은 레벨에 따른 사용 비율의 차이가 크지 않지만 ‘(ラ)レル형’은 레벨이 상승함에 따라 사용 비율이 증가한다. (3) ‘デキル형’의 경우, 학습 초기에는 「동사+こと(ガ)できる」의 형태로 주로 사용되지만 학습 중기 이후에는 「명사+(조사)+できる」의 형태로 분화된다. (4) 가능형식이 포함된 유닛은 특징적으로 나타나지 않지만 중위군의 학습자들에게서 「가능동사형+ない」, 「(ラ)レル+ない」라고 하는 정형적 사용 경향이 일부 관찰된다.

The purpose of this study is to analyze from the perspective of operational analysis how learners classify and use each item within grammar items of similar meaning in order to reveal how Japanese learners acquire Japanese expressions. To this end, large󰠏scale corpus data were analyzed to examine how learners in the JFL environment use the three possible forms divided by level in acquiring possible expressions, and what such differences in use mean. The analysis results are summarized as follows. (1) The acquisition of possible Japanese expressions is centered on the ‘Dekiru type’ and after the intermediate level, it expands to ‘Potential verbs type’ and ‘(Ra)Reru type’. (2) The ‘Potential verbs type’ does not have a large difference in the ratio of use according to the level, but the ‘(Ra)Reru type’ increases as the level increases. (3) In the case of the ‘Dekiru type’, it is mainly used in the form of ‘verb+koto(ga)dekiru’ in the early stages of learning, but after the middle of learning, it is divided into the form of ‘noun+(postposition)+dekiru’. (4) Units containing possible forms do not appear characteristically, but some formal use trends such as ‘Potential verbs type+nai’ and ‘(Ra)Reru type+nai’ are observed in the middle group of learners.