In recent decades, individualized instruction has become a very popular form of teaching and education. The use of individualized instruction, as well as the learning opportunities it offers, significantly improve almost all spheres of education, making it simpler and more efficient. The subject of this research is the efficiency of individualized instruction based on the learning styles of students. The efficiency of instruction is observed through different learning outcomes (the acquired knowledge, the level of motivation to learn and attitude towards the learning material). The aim of this research was to identify the differences in the efficiency of instruction when, on the one hand, the individualized model of instruction was applied and when, on the other hand, the non-individualized instruction model was applied. The individualized instruction model implied the adaptation of learning in the classroom to the specific learning styles of students. The Myers-Briggs model was used as the theoretical basis and the main conceptual model for this research (Briggs-Myers& McCaulley, 1992), by authors Isabela Myers & Katherine Briggs. This learning style model is based on Jung’s personality typology. The personality profile is determined based on the four dimensions: extraversion - introversion, sensing - intuition, thinking - feeling, judgment – perceiving. By combining these four dimensions, 16 possible characteristic learning styles can be obtained. In this paper, the focus is on two dimensions that were taken into account during the individualization of instruction: extraversion - introversion and sensing - intuition. By combining these two dimensions we can get 4 characteristic types of learning styles: ES (extravert sensing), EN (extravert intuitive), IS (introvert sensing), IN (introvert intuitive).

The results of numerous studies (Ranđelović, 2013) that researched the issue of efficiency of individualized approaches to instruction are quite consistent and indicate the advantage of respecting individual differences in the learning management process compared to traditional, frontal types of instruction, which do not focus on the individual differences between students who all have different abilities, personality traits and characteristic cognitive styles.

Although a number of research studies have confirmed the greater efficiency of individualized versus non-individualized instruction (Bachari, Abelwahed, Adnani, 2010; Randjelovic, 2012; Randjelovic, 2020; Ober, 2018; Cuevas, 2015; Ishak & Awang, 2017; Oberer , 2003), there are authors who have provided empirical data in which no significant difference in efficiency was found between these two models of instruction (Shepherd & Alpert, 2015; Benjamin, 2006).

The presented research starts from the initial hypothesis that there are differences in the achieved efficiency (acquired knowledge, level of motivation, attitude towards the learning material) in the individualized and non-individualized learning model. In order to get an answer to this hypothesis, the following research questions were analyzed:

- Determine whether there is a statistically significant difference in the quantity of the material learned by students, when, on the one hand, the individualized instruction model was applied and when, on the other hand, the non-individualized instruction model was applied.

- Determine whether there is a statistically significant difference in the quantity of the learning material learned by students, when, on the one hand, the individualized instruction model was applied and when, on the other hand, the non-individualized instruction model was applied from the aspect of the tasks with different levels of difficulty.

It should be determined whether there is a statistically significant difference in motivation among students, when the individualized model was applied and when the non-individualized model of instruction was applied

- It should be determined whether there is a statistically significant difference in the attitude towards learning among students, when the individualized model was applied and when the non-individualized model of instruction was applied

The experimental scientific - research method was applied, that is, the repeated pretest-posttest experiment. The following instruments were used: Serbian language and literature knowledge test; Student motivation scale in class (Ranđelović, 2012), Inventory of learning styles for primary school students (ILSPSS-O, Randjelovic, 2020), which is intended to examine learning styles among 4th to 8th graders, and relies on the MB model of learning styles (according to: Shindler, 2002); Learning Materials Attitude Scale. All these instruments have a satisfactory reliability that ranged from 0.721 to 0.889 expressed by the Cronbach's alpha coefficient. The sample included 60 fourth grade students from the “Dr Dragisa Misovic” Elementary School in Čačak. The experimental program lasted for four school lessons, more precisely four working days, and the teaching units taught (two of them) were selected from the Serbian language and literature course learning material. One teaching unit was taught by using the individualized instruction model (based on their learning styles), and the other was taught by using the non-individualized instruction model (not paying attention to students’ learning styles). Three parameters of teaching efficiency were taken into account as dependent variables in this research: - The quantity of the material learned by students: expressed as a score on a specially made knowledge test. - Level of student motivation: expressed as a score on the Student motivation scale in class; - Attitude towards the learning material: expressed as a score on the Learning Materials Attitude Scale Type of learning in the classroom was the independent variable in this research, and it is a categorical variable with two levels: individualized and non-individualized instruction model. Moreover, in order to answer the second research question, the level of difficulty of tasks was used as an independent variable, a categorical variable with 3 levels. This classification of the level of difficulty of learning material is in line with Bloom's taxonomy of cognitive domain: N1 - refers to knowledge and understanding; N2 - refers to the application and analysis; N3 - refers to the synthesis, excluding the evaluation category. The experimental program was conducted by the researcher - a master's student of psychology, who completed the training seminar on the individualized instruction model entitled “Individualization of teaching by showing respect for the differences in student learning styles”, as well as the received guidelines on how to apply the tests of four subgroups (ES, EN, IS, IN) (according to Shindler, 2004).

The results have shown that there is a significant difference (t = 23.525, df = 59, p <0.05) in the progress of students regarding the acquired knowledge of the Serbian language (concerning teaching units taught in experimental lessons) when lessons were organized by using the individualized instruction method (based on students' preferred learning styles) and when they were organized by using the non-individualized instruction method. The effect size (0.9037), expressed by the Eta-squared value, according to Cohen's criteria (Cohen, 1988) implies a large impact. Students advanced more in knowledge when the individualized instruction method was used in the classroom. With regards to the parameters Motivation and Attitude towards the learning material, it has been shown that there is no significant difference in these aspects of instruction efficiency when working in an individualized and non-individualized way. The results have also shown that there is no significant interaction between the level of knowledge factors (N1, N2, N3) and the instruction model (individualized and non-individualized). Although there are partially significant effects of instruction model factors, as well as knowledge level factors, their interaction is not significant. Such results are partly in line with previous research findings and speak in favor of greater efficiency of individualized compared to non-individualized instruction, primarily in terms of knowledge acquisition. There are two significant contributions of the conducted experimental research: - A model of individualized instruction was successfully applied, which took into account the learning styles of students. - Individualized instruction based on student learning styles (according to the MB model of learning styles) was implemented during regular classes (not additional, as in most research so far) and with “average students” (not gifted), and this suggests the higher value of the obtained findings as well as the wider applicability of the tested model in primary school education.

Bachari, E.E., Abelwahed,E.H., Adnani, M. E. (2010). An Adaptive Learning Model Using Learner's Preference. Ubiquitous Computing and Communication Journal, 5(3): p1-8. Benjamin, A. (2006). Valuing differentiated instruction. The Education Digest, 72(1), 57-59. Briggs-Myers, I. & McCaulley, M. (1992). Manual a Guide to the Development and Use of the Myers-Briggs Type Indicator. Consulting Psychologists Press. Cuevas, J., (2015), Is learning styles – based instruction effective? A comprehensive analysis of recent research on learning styles, University of North Georgia, USA, 13(3) 308 – 333. Ishak, NB., Awang, MM (2017). The Relationship of Student Learning Styles and Achievement in History Subject. The International Journal of Social Sciences and Humanities Invention 4(3): 3372-3377. Myers, I.B. and McCaulley, M.H. (1985). Manual: A Guide to the Development and Use of the Myers-Briggs Type Indicator. Palo Alto, Ca.: Consulting Psychologists Press Oberer, J. J. (2003). Effects of learning-style teaching on elementary students’ behaviors, achievement, and attitudes. Academic Exchange Quarterly, 7, 193–200. Ober J., (2018). Differences in high-school student learning by instruction type and MBTI personality type. VA: Liberty University. Ranđelović , D. (2012), Stilovi učenja i modeliranje procesa učenja. Neobjavljena doktorska disertacija. Banja Luka: Filozofski fakultet Ranđelović , D.(2013). Mogućnosti modeliranja procesa učenja u razrednoj nastavi. U prof. dr B. Jovanović (glavni i odgovorni ur) Pojedinac, porodica, društvo u tranziciji. Tematski zbornik međunarodnog značaja. Kosovska Mitrovica: Filozofski fakultet, str 283-303. Ранђеловић, Д. (2020). Практикум: Стилови учења ученика – претпоставка успешне индивидуализације наставе. Косовска Митровица: Филозофски факултет Универзитета у Приштини са привременимн седиштем у Косовској Митровици и Регионални центар за професионални развој запослених у образовању Ниш. [Practicum: Student Learning Styles - a Prerequisite for Successful Individualized Instruction. Kosovska Mitrovica: Faculty of Philosophy, University of Pristina with a temporary seat in Kosovska Mitrovica and Regional Center for Professional Development of Education Employees Nis.] Shepherd, C. & Alpert, M. (2015) Using Technology to Provide Differentiated Instruction for Deaf Learners. Journal of Instructional Pedagogies, v16: 1-7. Shindler, J. (2002). Exploring various structural options for performance assessment scale design: Which rubric is best? National Forum of Teacher Education Journal, 12, 3–12. Shindler, J. (2004): Teaching for the Success of all Learning Styles: Five Principles for Promoting Greater Teacher Effectiveness and Higher Student Achievement for all Students, California State University, Los Angeles, downloaded from the website 20/01/2011, http://www.calstatela.edu/faculty/jshindl/plsi/teachingacrosstype.htm