r. ebrahimpour

Learning has consistently been one of the aspects of human development since the beginning of human existence on the Earth, encompassing all aspects of human life and holding a special place in human life plans. On the other hand, technological advancements in recent decades have rapidly brought about significant changes in the realm of education and learning. One of the most prominent impacts of technological progress in the field of learning is the emergence of e-learning; a tool that enables learners to access educational resources at any time and place. Regarding the fact that every person has individual traits, preferences, and tendencies, traditional education's "one-size-fits-all" approach can be seen as one of its fundamental flaws. Learning styles, as one of the most important factors influencing learning, represent an individual's preferences and ways of processing and understanding information. Therefore, providing adaptive education based on learners' learning styles, with the aim of enhancing educational efficiency and reducing cognitive load during teaching, is both essential and inevitable. This research aimed to investigate the impact of adaptive education based on learners' learning styles. In order to achieve this, learners' learning styles were identified using an online platform and the Felder-Silverman Learning Style Indicator questionnaire, and educational content was automatically generated and presented to learners accordingly. Finally, the performance of the learners and cognitive load during instruction were examined.

A total number of 37 male and female undergraduate computer science students with an average age of 20.3, participated in this study. Initially, the participants were divided into two groups, and their learning styles were determined using the Felder-Silverman Learning Style Indicator questionnaire. Subsequently, one group received educational content tailored to their learning styles, while the other group received content not aligned with their learning styles. After studying the provided material, the cognitive load and learning outcomes of the participants were assessed using the NASA Task Load Index questionnaire and a designed performance test, respectively. Finally, the significance level of the results obtained from the two groups was evaluated using an independent t-test.

Based on the obtained results, no significant difference was observed in the test scores of the two groups' performance. However, when comparing the cognitive load between the two groups, the average cognitive load of the group that received content aligned with their learning style was significantly higher than the group that received incongruent content, with a value of 0.02 (p < 0.05).

According to the research findings, providing educational content based on learners' learning styles significantly reduce cognitive load during learning. Therefore, offering personalized education based on learning styles plays a crucial role as one of the adaptive teaching methods in e-learning, enhancing performance, and reducing cognitive load for learners.

Today, due to the increasing development of technology all over the world, e-learning systems are expanding rapidly. With the progress of electronic education, the movement from traditional education (the approach of providing one education for all) to personalized education began. Personalized education is an educational approach that aims to customize learning based on a learner's strengths, skills, interests, and needs. This method of education, like any other new method, has its strengths and weaknesses. In fact, increasing motivation and acquiring self-defense skills can be considered as one of the important benefits of this type of training. On the other hand, as the weaknesses of this method, we can mention the time-consuming training, the challenge in implementation, and the lack of clarity in the method of application. Due to the availability of many data from learners, the use of artificial intelligence to personalize education will both increase the quality and make education more attractive. Nowadays, one of the ways to personalize education is to provide it based on the preferences of learners. Learner preferences can be self-identified and explicitly identified and extracted by directly asking the learner or implicitly and collecting and monitoring data. Today, modeling user preferences is one of the most challenging tasks in e-learning systems that deal with a large amount of information. The aim of this research was to extract the implicit preferences of the learner by using an online interactive intelligent educational system that models the learner's preferences using conceptualization for learning objects through profile expansion and the use of artificial intelligence algorithms. The model was trained with the collected interactive data and provides new learning objects based on the learner's preferences. This research was practical in terms of purpose.

In this research, according to the society available to us, 29 male and female undergraduate students of computer sciences, with an average age of 21.5 years, who had not taken the machine learning course, were included as the participants. After registration, the participants were randomly divided into two control and experimental groups. The experimental group was presented with personalized content that matched their preferences, and the control group was presented with content that did not match their preferences. After the training, the learning rate and cognitive load of the participants were measured by the designed performance test and the NASA workload index questionnaire. At the end, the significance level of the obtained results of the two groups was evaluated using the independent t-test.

Based on the obtained results, the average performance test scores of the experimental group who received content matching their preferences had no significant difference compared to the average of the control group with a value of p=0.7 (while learning), but the cognition of the control group was significantly lower with p=0.00 compared to that of the experimental group.

Based on the findings of the research, providing personalized educational content based on learners' preferences using the profile expansion technique significantly reduced the cognitive load during learning. So, Providing educational content based on learners' preferences, as one of the personalized educational methods in e-learning, plays an important role in reducing the cognitive load of learners.

Geometric transformations have played a crucial role throughout history in various aspects of human life. Symmetry is one of the important concepts in school mathematics. Students' academic performance is intricately connected to the knowledge and skills of their educators. Recognizing the importance of prospective teachers )PTs( as future educators, in the initial stage, the aim of this research is to assess and analyze the levels of geometric thinking among prospective elementary teachers )PETs( utilizing Van Hiele's theory. Subsequently, the research seeks to delve into the thinking process and gaze patterns of prospective mathematics education teachers (PMETs) using the cognitive science method of eye tracking.

This study focuses on investigating and evaluating the thinking of geometric transformations and problem-solving skills among prospective teachers (PTs(. The research method employed a combined survey method, encompassing two distinct tests conducted on two groups of PTs. The accessible statistical sample includes 50 participating PETs and 21 participating PEMTs from Iran. The PETs of Farhangian University of Isfahan were divided into two groups: 42 students who had not learned the concept of geometric transformations in their undergraduate program (NPGT), and 8 students who had learned this concept in their undergraduate program )PGT). To assess the level of geometric thinking among participants, a self-made geometric test based on Van Hiele’s theory was utilized. The test reliability was assessed using Cronbach's alpha coefficient, which yielded a value of 0.68. Additionally, the validity of the test has been confirmed by some professors. In evaluating geometric thinking, a cognitive science method was performed. This method involved designing a psychophysical experiment and recording eye movements of the PMETs. The psychophysical experiment part was conducted in the computer laboratory of Shahid Rajaee Teacher Training University, Tehran, and was performed by Eyelink device and MATLAB software on student teachers of mathematics education of this university.

The results of the research show that students recognize the shape with symmetry as a symmetrical shape, but they perform poorly in determining the type of symmetry of symmetrical shapes, especially when a shape has rotational symmetry or oblique axial symmetry or a combination of several types of symmetry. In the first stage, the evaluation of PETs responses showed that 34% of them were in the first level and 18% in the second level of Van Hiele. The cognitive findings revealed that PMETs demonstrated superior performance in recognizing symmetries characterized by a single type of symmetry, in contrast to shapes involving combinations of various symmetries. Examining the recorded eye-tracking images of the students revealed a difference in gaze patterns between the groups that gave correct and incorrect answers. In addition, this difference is also evident among images with different symmetries (reflection, central, rotational).

The current research confirms the weakness of students in identifying the type of symmetry in symmetrical shapes. It also emphasizes the need to pay more attention to the training of PTs during their academic years. To address this, it is suggested to revise the curriculum concerning geometric transformations in the university courses for PTs training, additionally, the utilization of software such as Augmented Reality (AR) and GeoGebra can .contribute to enhancing cognitive and visual abilities of PTs in comprehending the concept of symmetry

Methods of pedagogy, for a long time, have been a social practice based on direct experiences from the past, and many of teaching methods have been traditionally formed. These methods were not supported by a scientific theory-based system and failed to keep up with the rapidly changing social needs. Neuroeducation is an experimental science and an interdisciplinary field that applies the latest theoretical advances in the human brain and psychology to education. By studying the theories in neuroeducation, it enables us to improve optimal presentation of contents for a course, teaching strategy and teaching methods of new subjects, and simultaneously improve students' computational thinking ability. One way that these theories can be tested is to study how decision-making is formed in the human brain. Goal-based decisions and behaviors depend on both sensory evidence mechanisms that collect perceptual information from the outside and mechanisms that select appropriate behaviors based on that sensory information which is decision-making mechanisms. Behaviorism is one of the basic foundations of theories of learning and behavior. One way to study behavior in detail is to use computational models based on brain biology that have been developed by neuroscientists in recent years. In this paper we try to explore the relationship between neuroeducation and pedagogy by studying theoretical achievements in computational neuroscience, cognitive neuroscience and psychology.

To investigate this issue, a neural-computational model of brain-based for decision making was used. This model consists of two recurrent dynamic neurons that can explore how perceptual decisions are formed in complex behavioral spaces and show the key parameters of decision-making process. In this study, we designed three different experiments in the model that included the accuracy-speed trade-off when responding, the effect of attention on decision making, and the impact of the confidence of decision, and then analyzed how the parameters and model's behavior change then we map the parameters to the classroom and changes in student’s brain. Finally, we used linear regression model to study the relationships and correlations between the parameters of the model’s behavior.

The results showed that using this decision-making computational model opened a way to study the speed-accuracy trade-off of students while answering exam questions and then, by using the model, an optimal trade-off could be found to answer the questions. Also, the analysis of model parameters showed that the level of students' attention in the classroom could be measured by the model and it had an important effect on decision making and answering the questions. Finally, the model could show the effect of students’ confidence on their performance and based on the fitted data of the model to students' behavioral data, we could make optimal suggestions from the perspective of educational psychology.

In this study, we show that by using decision-making neural-computational models, students' behavior in the classroom can be modeled. Educational science experts and psychologists in the field of pedagogy can use the findings to provide the best and most optimal teaching methods for teaching easily and the flourishing of students' creativity.

Today, e-learning is considered as a transformational technology and an important tool in the process of education and educational activities. On the other hand, the need to learn English as the first language in the world in order to exchange information and communicate with other nations in order to use up-to-date knowledge is undeniable, so the use of information technology to produce and provide educational services to improve English language teaching and learning is effective. Identifying the effective factors in achieving learning is one of the important and researched cases. Since the factors affecting learning are very wide and extensive, it is important to identify these factors in solving the problems and shortcomings of the educational system. One of these factors is cognitive style. People use different learning styles according to their individual differences. Cognitive style can be defined as the way people process new information and experiences in their minds; therefore, it is necessary to create a personalized environment based on the cognitive style of individuals in order to better adapt the educational strategy to the needs and abilities of the user and increase the efficiency of the learning process. In this research, Riding’s cognitive style, which divides people into two dimensions, verbal-imagery and wholistic-analytic, is used as an effective factor in learning. This study aims to predict the cognitive style of riding, based on the mouse movement of users in a language teaching software. In this regard, the language training software was designed and implemented, in which all the user's mouse movements are recorded on a millisecond scale when reading the English text and using the media designed in the software. Next, by using machine learning methods and interactive data stored from users while working with the software, an intelligent model was presented that categorizes people in two dimensions based on Riding’s cognitive style. This research is practical in terms of purpose.

In this study, Peterson’s cognitive style test is used to extract learners' cognitive style with the aim of constructing labeled data. Also, individuals’ mouse data is recorded when interacting with software, and artificial intelligence-based machine learning algorithms and models are used to build intelligent models for classifying and predicting individuals' cognitive styles. The process of training and building smart models is done through labeled data. Finally, the models used are evaluated by comparing the results of the cognitive style test and the outputs of the intelligent models. In the exams, male and female students aged between 22 and 35, with bachelor's and master's degrees familiar with English participated.

Users stored interactive data was used as the input to the five classifiers of the decision tree, neural network, nearest neighbor, support vector machine, and random forest. Patterson test results were also used as labels for these models; thus, individuals were categorized into two dimensions based on Riding’s cognitive style. The best classification was related to the decision tree with 90% accuracy in the verbal-imagery dimension and 87% accuracy in the wholist-analytic dimension of the results of this research.

According to the findings of this study, the designed language teaching system can intelligently extract the cognitive style of people when reading the English passage with appropriate accuracy. Therefore, in the future, the ability to provide personalized content in accordance with the cognitive style of people can be added to the designed software.

Visual attention is a high order cognitive process of human brain which defines where a human observer attends. Dynamic computational visual attention models are modeled on the behavior of the human brain and can predict what areas a human will pay attention to when viewing a scene such as a video. However, several types of computational models have been proposed to provide a better understanding of saliency maps in static and dynamic environments, most of these models are used for specific scenes. In this paper, we propose a model that can generate saliency maps in a variety of dynamic environments with complex scenes. We used a deep learner as a mediating network to combine basic saliency maps with appropriate weighting. Each of these basic saliency maps covers an important feature of human visual attention, and ultimately the final saliency map is very similar to human visual behavior. The proposed model is run on two datasets and the generated saliency maps are evaluated by different criteria such as ROC, CC, NSS, SIM and KLdiv. The results show that the proposed model has a good performance compared to other similar models. The proposed model consists of three main parts, including basic saliency maps, gating network, and combinator. This model was implemented on the ETMD dataset and the resulting saliency maps (visual attention areas) were compared with some other models in this field by evaluation criteria and their results were evaluated. The results obtained from the proposed model are acceptable and based on the accepted evaluation criteria in this area, it performs better than similar models.

Programmable logic devices, such as Field Programmable Gate Arrays, are well-suited for implementing biologically-inspired visual processing algorithms and among those algorithms is HMAX model. This model mimics the feedforward path of object recognition in the visual cortex.

HMAX includes several layers and its most computation intensive stage could be the S1 layer which applies 64 2D Gabor filters with various scales and orientations on the input image. A Gabor filter is the product of a Gaussian window and a sinusoid function. Using the separability property in the Gabor filter in the 0° and 90° directions and assuming the isotropic filter in the 45° and 135° directions, a 2D Gabor filter converts to two more efficient 1D filters.

The current paper presents a novel hardware architecture for the S1 layer of the HMAX model, in which a 1D Gabor filter is utilized twice to create a 2D filter. Using the even or odd symmetry properties in the Gabor filter coefficients reduce the required number of multipliers by about 50%. The normalization value in every input image location is also calculated simultaneously. The implementation of this architecture on the Xilinx Virtex-6 family shows a 2.83ms delay for a 128×128 pixel input image that is a 1.86X-speedup relative to the last best implementation.

In this study, a hardware architecture is proposed to realize the S1 layer of the HMAX model. Using the property of separability and symmetry in filter coefficients saves significant resources, especially in DSP48 blocks.    The author(s). This is an open access article distributed under the terms of the Creative Commons Attribution (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, as long as the original authors and source are cited. No permission is required from the authors or the publishers.

Today, it is common to use multimedia in foreign language teaching. There are some principles for designing multimedia that would reduce task cognitive load. These principles are based on the cognitive load theory. The methods of cognitive load measurement are divided into two categories, namely the subjective and objective measurements. NASA-TLX is an example of the subjective measurements; methods such as electroencephalography and eye-tracking are among the objective measurements. Due to the advantages of objective measurements, using these methods is common in cognitive studies. Eye-tracking technology can record different eye-movements of humans such as pupil dilation, saccades, fixations, blinks and microsaccades with a high sampling rate. These measurements are being widely used in cognitive and mental workload studies. In this paper, the cognitive load in multimedia language learning has been evaluated, using eye-tracking data analysis.

Two multimedia versions for teaching English were produced with the same narration and the length of 342s. In one version, the principles in designing multimedia were applied whereas in the other version, they were violated so that more cognitive load in comparison to the former version could be imposed. Ten subjects whose English listening comprehension was assessed with a simulation of the International English Language Testing System (IELTS) participated in the experiment and were randomly divided into two equal groups of five. The two groups were homogeneous with respect to their listening proficiency. One group watched the multimedia without principles while the other group watched the multimedia with principles. Then, each individual answered 12 multiple choice questions about the concepts presented in the multimedia as a performance test. During watching the multimedia and taking the performance test, the participants’ eye movement data were recorded. Then, each person filled out the NASA-TLX Questionnaire. Based on the results of the performance test and the NASA-TLX, the difficulty level of the multimedia without principles as compared to its version with principles was evaluated. The collected data were divided into blocks of 30 seconds.

Based on the NASA-TLX, the group who watched multimedia without principles experienced more cognitive load in comparison to the group who watched multimedia with principles, which approved our assumption about the higher load of the multimedia without principles. However, no significant difference was found in the results of the performance test between the two groups. According to statistical analyses, the pupil diameter, saccade length, saccade velocity, blink latency, and microsaccade amplitude in the multimedia blocks of both groups were significantly different. Nevertheless, no significant difference was found between the two groups in terms of the fixation time, the fixation rate, and the microsaccade rate.

Based on the findings of this study, pupil dilation, saccade length, saccade velocity, blink latency, and microsaccade amplitude have a significant relationship with the amount of the load imposed by the instructional multimedia which corresponds to the literature review of the study. Based on the results of this study, along with the subjective methods, eye movement data can also be considered as an appropriate tool for assessing the cognitive load imposed by multimedia learning and qualifying the multimedia instructional content. A significant difference was also found between the two groups in the study in terms of their blinking rate.  More investigation and different experiments are needed for examining other eye movement criteria that have been investigated in this study, including fixation time, fixation rate, and microsaccade rate so that a more definitive conclusion would be reached regarding a significant relationship between these parameters and the mental load imposed by the multimedia English teaching.     

This paper proposed a new method for face recognition with principal component analysis in the feature extraction phase, and devised a modified version of Mixture of Experts in which each expert is an MLP, instead of linear networks in order to improve the performance of the expert networks, and consequently the whole network performance; Therewith, we use a Momentum term in training the MLP experts, which speeds up the adjustment of weight greatly. We explore three different Mixture of Experts constructing a neural network. Our proposed model, achieved a correct recognition rate on Yale and ORL datasets. Comparisons with other algorithms demonstrate that our method performs better in terms of higher recognition rate, with smaller number of epochs in human face recognition.